Halliburton - Surface Data Logging Manual
Halliburton - Surface Data Logging Manual
Course 237
January 2001
This manual contains CONFIDENTIAL AND PROPRIETARY INFORMATION and is the property of Sperry-Sun, a Halliburton Company. Neither this manual nor information contained herein shall be reproduced in any form, used, or disclosed to others for any purpose including manufacturing without the express written permission of Sperry-Sun. Manuals are company property and non-transferable to other employees, unless authorized by Management. You are responsible for this manual. DO NOT leave this manual where it may be photocopied by others. This manual is designed to provide information useful for the optimal utilization of Sperry-Sun equipment. Charts, descriptions, tables and other information contained herein may have been derived from actual tests, simulated tests, or mathematical models. Although information has been carefully prepared and is believed to be accurate, Sperry-Sun cannot guarantee the accuracy of all information contained herein. Sperry-Sun reserves the right to modify equipment, software and documentation, and field equipment and/or procedures may differ from those described herein. Trained Sperry-Sun personnel act as consultants to Sperry-Sun customers. Practical judgement and discretion must be used, based upon experience and knowledge, to review the circumstances for a particular job and then to perform the job in a professional manner. Accordingly, the information contained herein should be used as a guide by trained personnel, and no warranties, expressed or implied, including warranty of merchantability or fitness for use, are made in connection herewith. In no event will Sperry-Sun be liable for indirect or consequential damages arising from the use of the information contained in this manual, including without limitation, subsurface damage or trespass, or injury to well or reservoir. Users are responsible for ensuring that they have the latest version of this manual. To verify the latest version, contact Sperry-Sun Drilling Services at (281) 871-5166.
2001 Sperry-Sun, a Halliburton Company Unpublished work, all rights reserved. Printed in the U.S.A.
ii
Contents
Table of Contents
Chapter 1 - Fundamentals of Petroleum Geology 1-1
Physical Geology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1 The Earths Size and Shape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2 The Earths Relief Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2 Gravity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3 Isostasy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4 Minerals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-6 Cleavage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-6 Fracture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-6 Form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7 Color . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7 Streak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7 Luster . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7 Hardness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7 Specific Gravity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-8 Other Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-8 Rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-9 Igneous Rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-10 Metamorphic Rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-13 Sedimentary Rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-14 Deposition of Sedimentary Rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-16 Classification of Sedimentary Rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-16 Textures of Sedimentary Rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-18 Characteristics of the Common Sedimentary Rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-18 Structural Features of Sediments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-23 Stratigraphic Relations of Sediments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-24 Isostatic Control of Sediments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-28 Diastrophism and Structural Geology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-29 Attitude of Strata . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-29 Warps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-30 Folds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-31 Kinds of Faults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-31 Historical Geology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-33 Petroleum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-35 Chemistry of Petroleum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-35 Chemical Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-35 Aliphatic Hydrocarbons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-36 Carbocycles or Aromatic Hydrocarbons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-37
SDL Applied Fundamentals January 2001 2001 Sperry-Sun, a Halliburton Company iii
Contents
Origin of Petroleum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-38 Transformation of Organic Material into Petroleum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-41 Migration and Accumulation of Petroleum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-42
2-1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1 Collecting Cuttings Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2 Shaker Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3 Settling Box Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3 Collecting the Cutting Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4 Collecting Wet Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-5 Washing the Cuttings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-6 Washing Cuttings From Water-Based Muds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-6 Cleaning Samples From Oil-Based Muds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-6 Processing the Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-8 Sieve Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-8 Logging While Coring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10 Sidewall Coring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10 Sample Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-11 Sample Quality and Examination Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-11 Use of Transmitted Light . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-12 Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-12 Order of Written Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-13 Rock Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-13 Color . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-13 Texture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-14 Grain or Crystal Sizes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-14 Shape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-14 Sorting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-15 Cement and Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-15 Fossils and Accessories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-16 Sedimentary Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-17 Porosity and Permeability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-17 Hydrocarbon Shows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-17 Some Criteria and Procedures For Rock and Mineral Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-18 Testing Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-18 Test for Specific Rocks and Minerals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-21 Porosity and Permeability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-23 Detection and Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-23 Choquette and Prays Carbonate Porosity Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-25 Archie's Classification of Porosity in Carbonate Rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-25
iv 2001 Sperry-Sun, a Halliburton Company SDL Applied Fundamentals January 2001
Contents
Hydrocarbons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Routine Hydrocarbon Detection Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Other Hydrocarbon Detection Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Solid Hydrocarbons and Dead Oil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Generalizations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Problems in Interpreting Drill Cuttings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contamination from Previously Penetrated Beds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Other Contaminants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Miscellaneous Interpretation Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Geological Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Unconformities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Equipment, Special Techniques and Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Equipment and Supplies for Routine Sample Examination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Thin Sections from Drill Cuttings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Staining Techniques for Carbonate Rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Detailed Insoluble Residue Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Versenate Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Limestone-Dolomite Differentiation Using Fairbanks Solution . . . . . . . . . . . . . . . . . . . . . . . . . . . Sperry-Sun Calcimeter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shale Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shale Density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2-26 2-27 2-30 2-32 2-32 2-33 2-33 2-34 2-35 2-37 2-37 2-38 2-39 2-39 2-41 2-42 2-43 2-45 2-47 2-47 2-53 2-56
3-1
Gas Determination from the Drilling Mud . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1 Aeration Gas Trap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2 Transportation Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4 Gas Detection Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-5 Origins of Gas Shows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-6 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-6 Sources of Gas in Mud . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-8 Gas from Drilling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-8 Free Gas and Liquefied Gas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-9 Dissolved Gas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-9 Effective Porosity and Permeability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-10 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-14 Factors Influencing the Amounts of Gas in Mud . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-15 Flushing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-15 Fluid Invasion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-19 Mode of Penetration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-26 Circulating System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-29
SDL Applied Fundamentals January 2001 2001 Sperry-Sun, a Halliburton Company v
Contents
Recovery at the Surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Borehole Contamination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Surface Influences on Gas Shows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Flowline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gas Trap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gas Detection Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Show Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Background Suction Pit Mud Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Flowline Mud Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Steam Still . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Show Evaluation of Gas-In-Mud Chromatographs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Determination of Oil in the Cuttings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Evaluation of a Cuttings Oil Show . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Determination and Evaluation of an Oil Show from Mud . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3-29 3-31 3-32 3-32 3-33 3-34 3-35 3-35 3-36 3-36 3-39 3-48 3-49 3-52
4-1
Causes of Kicks and Warning Signs of Kicks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1 Causes of Kicks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2 Warning Signs of Kicks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-10 Basic Kill Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-15 Shut-In Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-15 Well Control Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-20 Kill Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-27 Slow Pump Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-27 Shut-In Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-29 Influx Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-30 Kill Weight Mud . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-32 Kick Killing Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-32 Wait and Weight Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-33 Drillers Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-37 Concurrent Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-38 Kill Procedures with Subsurface BOP Stacks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-41 Other Considerations In Deepwater Drilling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-48 Comparison of the Three Methods of Well Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-49 Gas Kicks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-49 Kick Tolerance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-56 Saltwater or Oil Kicks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-57 Special Problems in Well Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-60 Lost Returns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-60 Drill Pipe Plugged or Bit Plugged . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-66 Washed or Plugged Choke . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-67
vi 2001 Sperry-Sun, a Halliburton Company SDL Applied Fundamentals January 2001
Contents
Hole in the Drill Pipe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Drill Pipe Off Bottom or Out of the Hole . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shallow Gas Hazards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shut-In Versus Diversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Use of the Marine Riser . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Diverter Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gas Bubble Migration and Expansion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Appendices Appendix A Appendix B Appendix C Appendix D Appendix E Appendix F Appendix G SPWLAR-Recommended Abbreviations for Lithology Descriptions A-1 Classification Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-1 Buoyancy Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-1 Formation Evaluation by Analysis of Hydrocarbon Ratios (Pixler Paper) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-1 Wellsite Formation Evaluation by Analysis of Hydrocarbon Ratios (Ferrie Paper) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .E-1 Factors Affecting the Surface Expression of Hydrocarbon Shows (George Paper) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .F-1 Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G-1
vii
Contents
viii
Chapter 1
1-1
that is concerned with the order and sequence of the rocks that make up the Earths crust.
1-2
relation to the radius of the Earth, it dwindles into insignificance. In relation to its size, the Earth is about as smooth as a billiard ball.
T A L C O N T I N E N
F S H E L
Figure 1.1 Diagram Showing the Continental Shelf and the Gulf of Mexico
1.1.1 Gravity
True, the variations of the Earths surface may seem insignificant in relation to the Earths size; yet, understanding them, and finding the forces which caused them, necessarily involve the basic concept of geology. The forces responsible for shaping all of the Earths features are part of the universal Law of Gravitation. This law states that every particle in the universe attracts every other particle with a force that is directly proportional to the product of their masses, and inversely proportional to the square of the distance between them. Because of the Earths great size and density, gravity has a strong attraction for objects on or near its surface. It holds the atmosphere and hydrosphere (water) on the Earths surface. It holds the Earth together and attracts all objects to it. Gravity then is responsible for the falling of the rains, which cut away at the mountains; for the flow of streams and rivers which erode the land masses; and for the deposition of sediments in lakes and oceans.
1-3
Three factors - altitude, latitude, and variations in density of nearby rocks - affect the force of gravity at any point on the Earths surface. The third factor, variation in rock density, is responsible for the major differences of the Earths features. Because the force of gravity is greater on rocks of high density than on rocks of low density, it is concluded that heavier rocks would normally tend to occur at a lower elevation than adjacent lighter rocks. A principle generally accepted by geologists and geophysicists is that the continents are composed of a lighter rock than that which underlies the oceans. Careful examination of rocks taken from continents and ocean basins indicates that this principle is well founded.
1.2 Isostasy
The fact that the lighter continental blocks stand higher than the heavy oceanic segments suggests that the two units are in equilibrium. The term isostasy (from the Greek isos equal and stasis standing) is used to define this condition of balance. Such condition means that the pressure at some depth beneath large units of the Earths crust must be substantially the same, and that any specific differences which develop because of processes in operation at the surface, must be adjusted by slow rock movement in the Earths plastic interior to maintain balance. Hence, if a heavy load is placed on a certain area on the Earths surface a gradual sinking of the area will follow. Conversely, if a heavy load is removed, the area will rise. The isostatic movement of one area is necessarily offset by an opposite isostatic movement of another area (Figure 1.2). Isostatic movement has not been confined to forming the continents and ocean basins. It has been active throughout the geologic past, creating shallow seas and mountains. In North America, isostic downwarpings have caused many different geologic seas to deposit sediments across the entire continental area except the area of the Canadian Shield.
1-4
Elsewhere, areas which have undergone glaciation in the recent geologic past are still rising even though the glaciers and ice sheets melted back many thousands of years ago. It is estimated that the Scandinavian area has risen 900 feet since the ice sheet melted about 12,000 years ago. Since this area is still rising, it can be concluded that it has not reached the elevation at which it stood prior to glaciation. In areas receiving tremendous deposits of sediments, the land is gradually sinking under the increasing loads. The Mississippi River Delta is sinking at a rate of more than 8 feet per century. However, the accumulation of deposits from the river keep the depth of the Gulf of Mexico and the land area of Louisiana about constant. It can be concluded that gravity, through the principle of isostasy, has been responsible for forming the major features of the Earths surface such as continents, oceans, basins, mountains, plains, etc. Also, it can be concluded that erosion and deposition, being a continuing process, are means by which isostatic movement never reaches equilibrium. Thus gravity is the most dynamic force acting on the Earths surface.
1-5
1.3 Minerals
Everyone realizes the importance of minerals in the nations industry and economy. Iron ore minerals are required to keep the steel industry operating just as barite and bentonite are required by the drilling mud industry. The future of both of these industries will be determined by the amount of mineral reserves available to them for preparing products. Geologists, through the science of mineralogy, are constantly seeking deposits of vital minerals in order to keep industrial reserves high. Mineralogy is the study of minerals. It includes their chemical compositions, crystal structure, physical properties, and occurrence. A mineral is a naturally occurring substance which has a definite chemical composition and internal structure with characteristic physical properties. Over 2,000 minerals have been recognized and described. The vast majority are rare. Many of them have never been found in more than one location. Others are found only as precious metals, gemstones, or valuable ores. Only about 20 are found abundantly in the Earths crust. These are called the rock and soil forming minerals because they comprise all but a small fraction of the Earths rocks and soils. Many minerals can be identified on the basis of a single physical property. As an example, halite, or rock salt, is identified by taste. However, most minerals require a combination of two or more physical properties for positive identification. The more important physical properties for mineral identification are discussed below.
1.3.1 Cleavage
Many minerals cleave or part along smooth planes. Some minerals such as mica, have a perfect cleavage in one direction only, while other, such as galena, have perfect cleavage in three directions. Terms such as perfect, uneven, hard, and easy, are used to describe cleavage planes.
1.3.2 Fracture
Minerals which have no cleavage fracture or break irregularly. Fracture faces are described as being conchoidal (like glass), rough, smooth, even, splinter, or fibrous.
1-6
1.3.3 Form
Minerals tend to crystallize into definite, characteristically shaped crystals, bounded by smooth planes called crystal faces. If crystal faces are present, their shapes and interfacial angles are diagnostic.
1.3.4 Color
All specimens of some minerals, such as magnetite and galena, have a constant or uniform color, but others, such as quartz and calcite, may vary in color because of impurities.
1.3.5 Streak
The color of the powder of a mineral is determined by scratching the surface of the mineral with a knife or file, or it it is not too hard, by rubbing it on an unpolished porcelain surface. The streak of a mineral may be similar, or entirely different from the color of the mineral itself.
1.3.6 Luster
The luster of a mineral refers to the way ordinary light is reflected from its surface. Metallic luster is like that of polished metals; vitreous luster is like that of glass; adamantine like that of diamonds. Other self-explanatory terms used to describe luster are resinous, silky, pearly, dull, Earth, oily, and waxy.
1.3.7 Hardness
The relative hardness of two different minerals can be determined by pushing a pointed corner of one firmly across the flat surface of the other. If the mineral with the point is harder, it will scratch or cut the other. The hardness of minerals is usually recorded in terms of Mohs Scale of Hardness ranging from 1 to 10. The numbers refer to the hardness of 10 minerals, arranged in order of increasing hardness.
1-7
1 2 3 4 5
6 7 8 9 10
When the minerals to make up this series are not available, it is convenient to know that a pocket knife blade is about 5.5, a copper penny 3.5, and the thumbnail about 2.5. Since most minerals have a hardness of less than 6, these tools are usually adequate for determining the hardness of an unknown specimen.
Specific Gravity is stated as a number indicating the ratio of the weight of the substance to that of an equal volume of water. Specific gravity can be determined by several different instruments in the laboratory.
1-8
1.2 Rocks
One of the basic principles of geology is the Uniformetarian Principle. It may be stated as follows: The present is the key to the past, or applied more specifically to our present subject: Rocks from long ago at the Earths surface may be understood and explained in accordance with processes presently going on. It assumes that, in the geologic past, water collected in streams and carried loads of sediments to the sea; that marine animals lived and died in the ancient seas, and that their shells were buried in the deposits accumulated on the sea floor. It also assumes that ancient volcanoes erupted and extruded lava flows, just as they do today. These and other similar assumptions are accepted truths as there is no reason to believe that the physical laws and natural processes of the geologic past have changed. Therefore, if features in solid rocks can be recognized as identical to those now being formed by volcanoes, streams, and beaches, it is reasonable to conclude that they were formed by the same type processes which are presently occurring. The Uniformetarian Principle is the underlying theme for all geologic studies. To evaluate any rock, which is defined as an aggregate of minerals, it is essential to know its origin, occurrence, mineral and chemical composition, and the process or processes by which it was formed. All this information can usually be obtained through the interpretation of the significant features contained within the rocks themselves. There are three major classes of rocks: igneous, metamorphic, and sedimentary. This classification is based on origin. Igneous rocks are formed by the cooling and solidification of molten or liquid rock. Metamorphic rocks are formed by the alteration, through heat and pressure, of existing rocks. Sedimentary rocks are formed by the accumulation of sediments. Each of these three classes is important in a fundamental study of geology because each class has a different significance in the Earths history. Each class contains minerals and ores which may not be found in the other two. In a study of petroleum geology for instance, sedimentary rocks are given much more attention than the other two because petroleum is found almost exclusively within them.
1-9
1-10
Based on their shapes and sizes, plutonic rock bodies are classified as dikes, sills, laccoliths, volcanic necks, stocks, or batholiths (Figure 1.3).
1-11
Dikes
Dikes are tabular bodies of igneous rocks that fill former fractures in the Earths crust. They may cut across formations, or they may cut into masses of older igneous rocks. They vary in width from less than 1 inch to many feet, and in length from a few yards to many miles.
Sills
Sills, like dikes, are tabular intrusive masses. They differ from dikes in that they lie parallel to the formations of the enclosing rocks. Some sills are small, covering areas of only a few acres, but others are very large. Most are less than 100 feet thick.
Laccoliths
Laccoliths are large lenticular masses of igneous rock similar in origin to sills. They are formed when intruded masses lift up overlying beds into domelike structures. Laccoliths may range from 1/2 to 4 miles in diameter.
Volcanic Necks
The igneous rock solidified in the conduits that once fed volcanoes often remain as remnants after erosion has removed the rest of the volcano. These cylindrical masses are termed necks or plugs. They may be several thousand feet in diameter.
Batholiths
Batholiths are the largest and originally the deepest intrusive bodies of igneous rock known. They are believed to have been the feeder sources of liquid material for the igneous masses formed at a higher level. Batholiths are so large that they are never sufficiently exposed to permit measurement of all three dimensions. Many are 50 to 100 miles wide and more than 1,000 miles long, and they extend downward to great but unknown depths. Most batholiths occur as the cores of a folded mountain system. Igneous activity is generally considered to be the origin of most metallic mineral ores. Gold, silver, and copper are often found as native metals in veins of plutonic rocks. Ores of other metals, such as lead, zinc, and nickel, are believed to have been deposited by hot solutions from igneous rocks. Understandably, certain
1-12
igneous rocks, especially plutonic rocks, are constantly being sought for their mineral content.
1-13
water, the halogens, sulfur, carbon dioxide, iron, silica, etc. They carry many metallic elements into rocks being metamorphosed. Rocks metamorphosed by chemical activity are characterized by the presence of new minerals.
Clastic Material
The deposition of clastic materials produces clastic rocks such as shales, sandstones, and conglomerates. The essential difference between these rocks is in the size of the fragments of which they are composed. A size scale of clastic materials is shown in the following table. These sizes are the ones commonly accepted by geologists.
1-14
/16 to 2
/256 to 1/16
Below 1/256
Clastic fragments are sorted by the action of various transporting agents. Thus, if the materials have about the same specific gravity, fragments of about the same size will be deposited together. If the materials differ in specific gravity, large fragments of light material will be mixed with small fragments of heavy material in deposition. The shapes of fragments may be described as angular, sub-angular, and rounded. Rounded fragments usually result from wear during prolonged transportation. Sharp, broken fragments generally have been deposited near their source. Thus, the shape and size of clastic materials are important guides to geologists.
Chemical Materials
The most abundant soluble salts are calcium carbonate, silica, sodium chloride, and compounds of magnesium, potassium, iron, and aluminum. These salts are of varying solubilities in river water. Some of them are very soluble in sea water, while others are not. The manner in which these compounds form sedimentary rocks will be discussed later.
Organic Materials
The organic materials which form a small part of the sedimentary rocks are derived from land and marine plants and animals. They contribute organic material in the form of carbon. Under very special conditions, depositions of carbonaceous materials become coal or petroleum. Swamps and lagoons along shores are ideal sites for the deposition of carbonaceous material. The hard skeletons of shells of marine animals are relatively insoluble in sea water. They are composed largely of calcium carbonate and silicon dioxide. In some areas, these materials have formed thick deposits of sedimentary rocks.
1-15
1-16
Formed through weathering Mechanical weathering Produces fragments of minerals and rocks that are removed mechanically and deposited as the Clastic Sedimentary Rocks. Chemical weathering Produces insoluble products that are Produces soluble products that are Removed in solution by streams and deposited as the
The rocks formed from substances slightly soluble in sea water; hence deposited soon after reaching the sea.
The rocks formed from substances very soluble in sea water; hence deposited only as a result of evaporation. Calcareous Limestone
Limestone...CaCO3 Chalk Dolomite... MgCO3 Chert..........Soluble silica Flint Hematite ......... Iron Limonite ......oxides
in order of deposition Gypsum .CaSO42H2O Anhydrite........ CaSO4 Salt (halite) ........NaCl Potassium and Magnesium minerals Siliceous Diatomite
The carbon is derived dominantly from CO2, hydrogen from water, other constituents, indirectly, through weathering and alteration.
Figure 1.4 Sedimentary Rocks Chart
1-17
Conglomerates
A conglomerate may be made up of any kind of rock fragments held together by some cementing material such as shale or clay. Its distinguishing characteristic is rounded coarse fragments. Conglomerates are necessarily younger than the fragments of which they are composed.
Sandstones
Sandstone is composed of fragments and grains from a size smaller than those of a conglomerate down to grains of about the size of ordinary granulated table sugar. The principal distinction between coarse sandstone and fine conglomerate is that
1-18
the sandstone is more apt to contains grains or fragments of rather uniform size, without any unduly large pebbles. Sandstone can be described as fine, coarse, or medium. It should be described as soft if it can be crumbled in the fingers, hard if it breaks with difficulty with a hammer and medium-hard if it breaks under an ordinary hammer blow. Sandstones are usually deposited in water relatively close to a seashore. Such an area along the shore is today, and has been in the past, the home of numerous shellfish, such as clams, oysters, conch shells, scallops, sand dollars, and related forms of life. These shells are composed of calcium carbonate or calcite. As previously mentioned calcite is the principal natural cement and the presence of numerous shells in the sandstone is a source of calcite which may in time, dissolve and be redeposited again among the sand-grains cementing them together into a solid rock. Complete fossil seashells are also common in sandstones.
Wind-Deposited Sandstones
Sandstones are also formed by the solidification of wind-blown sands on land. A wind-blown sand deposit might be found underneath marine water deposited sediments if an area such as the Sahara Desert should be suddenly submerged beneath the sea. Wind-deposited sandstones are, on the whole, rare among oilfield sediments and do not warrant much discussion. However, their present is significant if they are found, as it indicates that the area was dry land and not ocean at the time of deposition. This might be of great importance, since the source of oil is almost universally marine organic material. A well known wind-deposited sandstone formation is the Navajo sandstone of the Navajo-Grand Canyon-Zion National Park area.
Siltstone
Siltstone is a rock composed of material whose small particles are larger than the fine material of true shale and too small to be called sandstone. When broken in the fingers, it is gritty rather than slippery, and often it is composed of a mixture of fine sand grains and mud. For general purposes, a coarse siltstone could be included under sandstone or a very fine siltstone might be included with shale.
Shale
Shale is the rock formed from compaction and solidification of the fine materials of sedimentation which originally settled out in the water as mud. Consequently, it is a deposit formed at greater distances from the shore than a conglomerate or a sandstone, the material of which drops to the bottom as soon as the speed of the
1-19
depositing current is reduced where a river runs into the ocean. The tiny particles which make up shale will remain suspended for a long time even in still water. Fine material washed into an ocean by a river may be caught in some of the slow-moving ocean currents and transported for many miles. Hence, deposits of shale of decidedly uniform character may be deposited over thousands of square miles of ocean bottom. Often the deposition of shale is periodic, not continuous. This may be due to several causes, and much remains yet to be learned on the subject. Seasonal floods, for example, may be a cause. Rivers often carry large loads of erosion products during floods, and are relatively clear and carry nothing for other periods of the year. In such cases, a film or layer of mud would be deposited on the bottom after each storm or flood, and in long geological time, many thousands of such films or layers would pile up. Shale formed by the solidification of such a deposit of many layers, would be banded or bedded rock, and a core or surface exposure of it would show bedding or stratification. There would actually be visible fine line or bands in the rock. According to local conditions which caused the bedding or stratification, the beds or strata might be a fraction of an inch thick or several feet. All material that is fine is not necessarily of the same composition. The fine materials which accumulates to form shale is naturally ground up (and perhaps altered) material of the rocks of the land from which it was eroded. Since numerous kinds of rocks exist at different localities, their erosion products are different, even though ground up to the same size particles. For this reason, all the shale in the world is not alike. The particles of which it was composed are alike, but only in size. Therefore, there are many different kinds of shale. As there is no particular advantage in using new names for the different varieties, they are all generally called shale, without another qualifying word added to indicate what kind. A shale containing an important amount of calcium carbonate (line) is referred to as a calcareous shale or simply limey shale. Likewise, shale containing a large amount of silica, such as much of the Monterey shale of California, is called siliceous shale. Many local characteristics are also used, such as as nodular shale, poke chip shale (when the cores split into smooth wafers or plates resembling poker chips), paper shale, and others.
Limestones
Limestone is different from any of the above described sedimentary rocks, because it is a chemically deposited sediment and not from mechanical settlings. The relation of erosion to the formation of limestone is not the carrying fine particles of calcite (CaCO3 - the mineral of which limestone is composed) but it is to carry instead, calcium carbonate to the ocean in solution.
1-20
This calcium carbonate is in solution in sea water as is salt. Certain animals living in the sea, principally corals, and small forms of life known as bacteria, are able to extract calcium carbonate from the sea water and, by rather vaguely known like processes, to resolidify or precipitate it as small, solid crystals of calcite. By this process, deposits of limestone rock accumulate on sea floors where there is an abundance of coral or peculiar bacterial life. Certain plants such as algae can also extract calcium carbonate from sea water and deposit it as limestone. The life forms which are able to deposit limestone live in warm tropical or sub-tropical waters, such as the Coral Sea north of the Great Barrier Reef, which extends for more than a thousand miles along the northeast coast of Australia. These animals require, in addition to warm temperatures, clear waters, fairly free from stifling turbidity. Such conditions existed in past geological time in the limestone area from the Gulf of Mexico far north to Canada. If the small, oil-containing forms of life are also present and accumulate with the limestone, a pertroliferous or oil-bearing limestone will be deposited. Such rocks have produced oil fields in many parts of the world.
Dolomite
If a large part of the calcium in a limestone is replaced by magnesium, the rock is dolomite. Dolomitization is a common process in limestones of all ages, and it is often accomplished during the process of sedimentation. In fact, many so called limestones are dolomites.
Chalk
Chalk is a special type of limestone composed of small shells, or fragments, cemented together. Forminifera shells constitute a large part of the material, but the presence of shells or other organisms is common. Chalk usually is soft, porous, and white or grey, and some of it is massive in appearance. The chalk cliffs of Dover, England, are an example. Some of the chalks of the Southwest, particularly those of Texas, grade into denser beds that are as well consolidated as ordinary limestone.
Marl
The porous masses of shells and shell fragments that accumulate on the bottoms of may freshwater lakes form marls. The term marl is used also to designate calcarious shales in which clay and finely divided particles of calcium carbonate are mixed.
1-21
Coquina
This term usually is applied to the more recent deposits of cemented shell accumulations.
Reefs
Fossilized corals and associated marine life form another type of limestone known as coral reefs. These limestones possess the skeletal features of the organism of which they are formed. Reefs are formed in tropical waters along the shore of land masses and around islands. They were probably formed in all of the ancient inland seas of North America. Reefs are more technically known as bioherns.
Chert
Chert is a hard, compact, dense, siliceous material that occurs as distinct layers, or as pebbles, in the beds of other rocks. Either colloidal silica was deposited with the other sediments, or after deposition, silica-bearing waters partially replaced the associated sediments.
Diatomaceous Earth
Diatoms are minute plants that live in great numbers in the sea and in freshwater lakes. When they die, their siliceous skeletons accumulate to form diatomaceous Earth. At many places diatomaceous Earth is interbedded with shales.
Coal
Coal is formed by the compacting and partial decomposition of vegetable accumulations. The alteration of vegetation into peat, lignite, and various other grades of coal is a long process. The grade of coal is dependent upon the kind of material deposits and the amount of alteration that has taken place.
1-22
Sodium Chloride, NaCl Magnesium Chloride, MgCl2 Magnesium Sulfate, MgSO4 Calcium Sulfate, CaSO4 Potassium Sulfate, K2SO4 Calcium Carbonate, CaCO3 Magnesium Bromide, MgBr2 Total
When sea water is evaporated to dryness, the salts come and are deposited. The least soluble salts are deposited first. Calcium carbonate and iron oxide, if present in the water, are the first to be deposited. Gypsum follows, and often with it some anyhdrite is deposited. After gypsum, sodium chloride, or common table salt, is deposited. The bitter salts consisting of sulfates and chlorides of potassium and magnesium are deposited last. These are so soluble that they are not always deposited with salt and gypsum. The rocks formed in this manner are called evaporites. Thick deposits of evaporites were probably formed in evaporating bodies of sea water which intermittently received influxes of fresh sea water. Associated with salt and gypsum in many places are red beds, composed mainly of red sandstones and shales. These are red because they contain small amounts of iron oxide. It is believed that they have been formed under arid conditions.
Cross-bedding
Sediments that show parallel bedding at an angle to the planes of general stratification are cross-bedded. Wherever steep slopes are produced by the rapid deposition of sediments (as at the front of a delta or on offshore bars, barriers, etc.)
1-23
cross-bedding occurs. Wind-lain deposits, such as sand dunes, are characteristically cross-bedded.
Graded Bedding
When a mixture of particle grains is brought to the site of deposition, the coarser and heavier grains settle more rapidly than others. It follows then that the bed of sediment finally accumulated shows a segregation of the grains as determined by their relative rates of setting. Thus the bottom portion of a bed may consist of coarse or heavy particles, whereas the upper portion is made up of relatively fine or light particles. Such an arrangement is called graded bedding. The presence of graded bedding in rocks indicate seasonal deposition within a relatively still body of water.
1-24
A complete reversal of this occurs when the sea is regressing. Sediments carried to the sea by rivers and streams are deposited progressively farther from the original shore line. As the sea regresses, young deposits of sediments are exposed, eroded, and carried once more into the sea and redeposited progressively further from the original shore line. The arrangement of layers of sedimentary rocks is called offlap (Figure 1.7).
1-25
1-26
shifted. Repeated lateral shifting of the depositional sequence causes conglomerates to be deposited over sands, and sands to be deposited over shales, when the shore line shifts seaward. The reverse occurs when the shore line shifts inland. This results in an interfingerng, or interbedding, of lateral facies. Detailed studies of such lateral changes are of utmost importance in the search for petroleum (Figure 1.8).
Figure 1.8 Diagram Showing Ideal Lateral and Vertical Variation Within Sedimentary Rocks
1-27
1-28
The character of sedimentation depends in part upon the balance between the rate of subsidence and the rate of filling of a basin. Contemporaneous changes in the source areas also affect the character. Notable uplift adjacent to a basin would supply a great bulk of coarse sediment, whereas long-continued erosion of stable land would change the sediment to fine mud and solutions. If little water reached a basin, evaporation would begin, causing the deposition of evaporites. Hence, sedimentary rocks, by their composition, texture, thickness, area distribution and other characteristics, reflect the complex interplay of a number of factors. Perhaps the most influential of these is isostatic movement.
1-29
Figure 1.9 Diagram Illustrating the Dip and the Strike of a Tilted Bed
1.6.2 Warps
Rocks which have been warped form gently sloping structures, such as irregular shaped basins and domes. The beds of rocks in such structures are gently tilted. Uniformly tilted beds are homoclines. This name indicates that the strata of this structure are inclined in the same direction. Broad downwarped structures are called geosynclines (Figure 1.10).
1-30
1.6.3 Folds
Where beds of rock have been subjected to extreme horizontal forces, they bend into folds with alternating crests and troughs. The principal types of folded structures are anticlines, synclines, and domes and basins. Where the beds of rock are arched up like the roof of a house, they form an anticline, i.e. the bed dip away from each other. Downfolds, or troughs, where the beds dip toward one another, are call synclines.
1-31
Some faults are neither normal or reverse, but involve longitudinal movement parallel to the fault plane, as in the San Andreas fault of California. Such a fault is a rift or tear fault. A block depressed between two faults is a graben, and a block raised between two faults is a horst (Figure 1.11).
1-32
1-33
The standard geologic time scale is divided into eras, periods, epochs, ages, stages, and substages. Because many local variations in nomenclature this presentation will be limited to a discussion of the eras, periods and epochs as we are concerned with them as surface data loggers. The following table represents the geologic column and time scale used by the U.S. Geological Survey.
Table 1.4 Geologic Column and Time Scale Era Period Epoch
Cenozoic
Quaternary Tertiary
Mesozoic
Cretaceous Jurassic Triassic Permian Pennsylvanian Mississippian Devonian Silurian Ordovician Cambrian Keweenawan Huronian Timiskamingian Keewatin
Paleozoic
Proterozoic
Archeozoic
Often a geologic cross-section accompanies a geologic column. A geologic cross-section is a graphic representation, over an extended area, of the attitudes of the subsurface formations.
1-34
1.8 Petroleum
It is evident that each carbon atom is united with one or more additional carbon atoms, and the remaining unsatisfied valences are united with hydrogen. The formula for the closed-chain series is arranged in a ring as shown by the structural formula for benzene (C6H6).
1-35
H C H H C C C H
Figure 1.13 Structural Formula for Benzene
C C
H H
Olefine Series
The Olefine series members contain two less hydrogen atoms than those in the paraffin series. The basic formula for this series is CnH2n. Some of the members of this series are ethylene, propylene, butylene, etc. The Olefines are similar to the paraffins in physical properties but they are different in chemical properties. The olefines have double bonds between some of the carbon atoms, as propylene (C3H6). H C H C H C H H H
1-36
Acetylene Series
The basic formula for the acetylene series is CnH2n-2. Members of this series have two carbons united by triple bonds. There are two less hydrogen atoms in compounds of this series compared with corresponding members of the olefine series.
Diolefine Series
The diolefines are unsaturated hydrocarbons having the same basic formula as the acetylenes, CnH2n-2. However, the structural formula differs in the the diolefines have two double-bonded carbon atoms instead of one triple-bonded carbon atom.
Cycloparaffin Series
This series is known as alacyclic because it possesses both the properties of aliphatic and cyclic hydrocarbons. It resembles the paraffin series in chemical and physical properties except for density, which is greater. The basic formula is CnH2n, and the structural formula is cyclic, but the members are saturated hydrocarbons since they have single bonds between the carbon atoms.
Napthalene Series
Compounds of this series have the basic formula CnH2n-12. In the structural formula napthalene, C10H8, the nucleus is composed of two rings, which is a typical structural formula for the series.
1-37
HC HC HC HC C C
CH CH CH CH
Crude petroleum is composed, then, almost entirely of a mixture of aliphatic or carbocyclic hydrocarbons. There is little information as to the actual ratio of the two types of hydrocarbons in oil, but it seems probable that the cycloparaffins predominate. The low boiling fraction of practically all petroleum is composed of paraffin series hydrocarbons. However, the differences in oil from various sources are exhibited in the higher boiling fractions. If residue after the removal of the volatile members consists of large amounts of paraffin or wax, the petroleum is classified as a paraffin base oil. Similarly, if napthalenic hydrocarbons predominate the petroleum is an asphalt base oil.
Inorganic Theories
Inorganic theories were the first advanced to account for the formation of petroleum. Betheolot, in 1866, suggested that mineral oils were formed by the action of water on metallic carbides. He based his idea on the assumption that the interior of the Earth contained free alkaline metals with which carbon dioxide could react at high temperatures to form carbides and acetylides. The carbides or acetylides would then react with water to form acetylene, which when heated to
1-38
approximately 900oC, polymerizes to form benzene, one of the hydrocarbon series. Medeleef, about this time, also showed that the action of carbon dioxide and water upon the alkali metals (sodium and potassium) gave off small quantities of hydrocarbons. Both of these theories would be acceptable, but for the fact that neither pure carbides nor pure alkali metals are known to exist in the Earths crust. If they do exist in the pure state they can do so only at the high temperatures associated with volcanic phenomena. As a greater part of the Earths oil fields are far removed from any center of igneous activity, these theories were never seriously accepted by geologists. Other unacceptable inorganic theories of the origin of petroleum concern the reaction of a volcanic gas (ammonium chloride) with native iron, or the reaction between limestone and gypsum at very high temperatures which forms disassociated water and carbon.
Organic Theories
The organic origin of petroleum is generally accepted by scientists. But there remain many problems as yet unsolved. It is generally believed that petroleum originated by a series of complex processes from plant and animal substances. The exact nature of the original organic material is not known, although much valuable data on this has been assembled. The biological, chemical and geological processes necessary for the conversion of the organic matter of plants and animals into hydrocarbons are not completely understood. It has been reasonably established that petroleum is of organic origin because: Some petroleums are optically active, i.e., most oils have the power of rotating the plane of polarization of polarized light. Only matter derived from organic origin could have this power. 2. Petroleum contains nitrogenous compounds. All such compounds found in nature are either of plant or animal origin. 3. Some petroleum contains chlorophyll porphyrins, which are derivatives obtained from the chlorophyll of plants or from the blood cells of animals. 4. Some petroleums contain hydrogen sulfide gas which results from bacterial decomposition of plants and animals.
1.
Despite considerable research there is still a wide divergence of opinion as to the type of organic material which can be changed into petroleum.
1-39
Organic matter that might be considered as potential source material for petroleum occurs in a wide variety of both animals and plants. This fact alone may explain the great variation in petroleums found in nature. On the other hand, the source material may have consisted predominantly of a single type of organic matter. The variations in composition may have developed later as a result of migration, bacterial action, metamorphism, etc. As yet there is no conclusive evidence indicating whether the primary source material of petroleum consists of many types of organic matter or one predominant type. The primary source of the organic matter in sediments may be either animal or vegetable remains, or both. Some of this matter is carried to areas of sedimentation by streams, waves, or currents, and some of it remains where it occurred. Since most petroleum deposits are closely associated with marine sediments, it follows, then, that petroleum very likely originated in marine sediments. Consequently, the organic matter of the oceans is of utmost importance in the study of the origin of petroleum. Most of the organic matter in sea water is either dissolved or is in a colloidal form. The rest is contained in the plant and animal like of the ocean, chiefly in plankton, the microscopic and semi-microscopic free-swimming organisms. It is difficult to estimate the rate at which organic matter is produced in the sea. Plankton, for instance, is produced at a rate as high as several hundred grams per cubic meter of sea water per year. Photosynthesis (the process whereby plants convert carbon dioxide and water into carbon compounds under the influence of light) has been estimated to produce 12 million tons (80 million barrels) of hydrocarbon material annually in the ocean. A minute fraction of this material, preserved in sedimentary rocks, could be transformed into all known petroleum deposits, plus those that we can expect to discover in the future. Organic matter is formed not only by plants and animals in the ocean but also by those on land. Much of it formed on land eventually reaches the ocean by streams and rivers in solution or in colloidal dispersion. In fact, of 50% of the sedimentary materials carried by streams and rives may be organic matter. In this regard, humic substances are probably the most important organic materials formed on land. They are formed by the slow decomposition of lignins in peat. They are found in soil highly charged with decaying vegetation. Vast quantities of humic acid are forming constantly in swampy regions, especially in the tropics. These substances include humic acid, geic acid, and ulmic acid. There is a close similarity between these substances and petroleum, as illustrated by the deposits of asphalt and other hydrocarbons formed of humic substances along the coast of Florida. The precipitation of such deposits might be caused by the mingling of fresh and salt water.
1-40
1-41
deposited sediments. It is known that microbial activity affects some of the properties of recent sediments, such as oxygen tension, oxidation-reduction potential, hydrogen-ion concentration, sulphide and sulphate content, carbonate content, and state of iron and manganese. Certain metallic elements (such as lead, nickel, vanadium, iron, and copper) are commonly found in petroleum, apparently in some form of organic combination. Some of these elements may act as catalysts in the generation of petroleum.
Conditions
All available evidence suggests that organic materials might have been transformed into petroleum under the following conditions:
1. 2. 3. 4.
5.
Deposition of organic material in fine sands and silts in fairly shallow marine water. Rapid burial preventing destruction by bottom dwellers. Normal decomposition with burial and the beginning of anaerobic bacterial activity. Conversion of material toward hydrocarbon material. The transformation continues until the mixture becomes so foul by the accumulation of hydrogen sulfide gas that it kills off all bacteria. Migration and accumulation of oil as sediments are compacted.
Migration
The movement or migration of petroleum from the source beds into reservoirs can be divided into two parts: (1) transverse migration from the source beds into a carrier bed; and (2) longitudinal migration through the carrier bed to a suitable trap. The movement of petroleum through rocks is apparently caused by several types of energy, including compaction, capillarity, differential specific gravity, hydrostatic pressure and gas pressure.
1-42
It is believed that compaction within the source beds is the principal force causing the movement of petroleum from source beds into carrier beds. Compaction is also regarded as an important force in the migration of petroleum through the carrier beds. Obviously the fluids occupying the pore space will be driven out by the compaction of the clay, mud or ooze of the source bed. These fluids move in the direction of least resistance into non-compacting porous formations, such as sandstone or porous limestone. Although direct proof that compaction has been an important factor is not to be expected, the direct association of most oil-producing regions with structural basins is an indication that it does play an important part. Capillarity is action, due to surface tension, by which the surface of a liquid where in contact with a solid, is elevated or depressed. Surface tension of a liquid causes it to act as an elastic enveloping membrane, so that it tends to compact to the minimum area. The surface tension of water is approximately three times that of oil. Capillary action, therefore, would tend to draw water into the finest openings, displacing the oil and gas. In a slow transfer of liquids between shales and sandstone, oil would be displaced from the shales into the sandstones because the water enters fine pores three times as easily as oil and has three times as much difficulty in leaving. Every oil field is evidence of migration caused by the action of gravity. If present, water occupies the lowest position in the reservoir. Oil floats on water and it occupies the next highest position above the water. Any gas present will occur above oil and will occupy the highest position in reservoirs. Other forces may cause petroleum to migrate great distances, but gravity is responsible for the final arrangement of water, oil, and gas in reservoirs. The theory behind the action of hydraulic pressure in the migration of petroleum suggests that moving water under hydraulic pressure has been an important agent in the migration and accumulation of petroleum. According to this theory, hydrocarbons are carried along by the flow of underground water. However, the movement of petroleum through rocks is probably faster than the movement of water through rocks. Yet, it is conceivable that oil migration could be aided or hindered, depending on the direction of flow, by the movement of underground water. Differential gas pressure has been suggested as a factor in the migration and accumulation of petroleum. However, it usually is considered only an aid to other factors, such as capillarity, differential specific gravity, or hydraulic movement.
Transverse Migration
Migration directions are considered in terms of the stratification planes of rocks. Oil either migrates in a longitudinal (parallel) or transverse (vertical) direction to the stratification planes. Generally, the primary migration from source rock to
1-43
reservoir rock is transverse, and the secondary migration through the reservoir to the trap is longitudinal. Transverse migration can be downward or upward. If movement is taking place because of differences in the specific gravity of oil and water, the migration direction of oil will be upward. But if the oil is being squeezed from a rock by compaction it will move in a path of least resistance, whether that be upward, downward, or sideways. The prerequisite for transverse migration is that a receptive layer must be present to receive the flow. Downward transverse migration is responsible for the occurrences of oil in basement igneous rocks of buried hills. Other examples of transverse migration are the accumulations of oil beneath unconformities, especially those occurring in the leached upper surfaces of thick limestones.
Longitudinal Migration
Longitudinal migration is possible where a porous and permeable rock layer occurs in the sedimentary section. Longitudinal migration is by no means confined to widespread sandstones or regional porous limestones. Sand-filled channels and bars in thick shale sections also may be used. The confinement of oil accumulations to the highest levels in the reservoir rock is presumptive evidence that oil moved through the rock until those levels were attained. Unless it is assumed that by some strange coincidence oil entered the reservoir where there were traps, it must be concluded that the oil migrated laterally until trapped. It can be concluded that petroleum has traveled by both longitudinal and transverse migration in moving from the source to the trap.
Accumulation
Many different classifications have been proposed to include the wide range of geologic conditions under which oil and gas pools occur. But because of the many different types of oil and gas pools, it has been difficult to establish a classification which covers all types. One generalization applies to all types; oil and gas accumulate in pools because their upward or lateral migration is stopped by a trap or closure. These traps are formed by stratigraphic conditions which were formed at the time of deposition of the sediments, by later changes in the sediments, by structural deformation, or by a combination of two or more of these factors. The following is an outline classification of traps or reservoirs.
1-44
I.
Closed Reservoirs A. Reservoirs closed by local deformation of strata. 1. Reservoirs closed by folding a. Reservoirs in closed anticlines and domes b. Reservoirs in closed synclines and basins 2. Reservoirs developed through the off-setting of strata by faulting of homoclinal structures 3. 4. Reservoirs defined by combinations of folding and faulting Reservoirs formed through the cutting of strata by intrusions a. Intrusions of salt b. Intrusions of igneous rock 5. Reservoirs developed in fault and joint fissures and in crush zones. B. Reservoirs closed because of varying porosity of rocks 1. Reservoirs in sandstone caused by lensing of sandstone or by varying porosity 2. Lensing porous zones in limestones and dolomites 3. Lensing porous zones in igneous and metamorphic rocks 4. Reservoirs in truncated and scaled strata a. Closed by overlap of relatively impervious rock b. Closed by seal of viscous hydrocarbons C. Reservoirs closed by a combination of folding and varying porosity.
II. Open reservoirs None of commercial importance Figures 4-16 through 4-24 show the types of traps and where in the trap that oil and gas is likely to accumulate. Note: In the following illustrations, the symbol at the right shows hydrocarbon accumulation.
1-45
1-46
1-47
1-48
1-49
1-50
1-51
1-52
1-53
Note the change in the dip of axis of fold (line GH). BC indicates the width of the productive area for the upper sand; EF that of the lower sand. Axes of folds (at A and D) lie near the left edge of the productive area. Well 1 is productive; well 2 only a short distance away, is barren. Well 4 produces from the upper sand only; and well 3 from both the upper and lower sands.
Figure 1.32 Asymmetric Anticlinal Fold Along the Flanks of a Major Uplift
Figure 1.33 illustrates how greater accumulations petroleum may be found on the basinward side of an anticline. Note the difference in the level of the edge-water lines on the opposite flanks of the anticline. All anticlines are long narrow domes in the sense that they are closed structures. However, domes are usually spoken of as closed structures in which the length does not exceed three times the width.
1-54
Figure 1.33 Dome Structure, illustrated in plane view by the structure contours and by the vertical sections through the major and minor axes
Oil accumulates in the porous formations above and on the flanks of the salt core. A monocline is formed when the crest of an anticlinal fold is eroded away and a partial cross section of the rock strata is exposed as an outcrop.
SDL Applied Fundamentals January 2001 2001 Sperry-Sun, a Halliburton Company 1-55
The shallow well, number 1, produces heavier, more viscous oil than number 2 owing to evaporation of the lighter constituents at the outcrop. Well 3 encounters edge water. Oil that has migrated to the surface is lost; however, as it accumulates on the Earths surface, the lighter fractions evaporate leaving a residue of asphaltic-like material. This residue will plug the pore spaces in the rock and prevent further loss. Such surface indications of a bituminous nature have resulted in the discovery of many important oil reservoirs. Faulting will many times place a permeable strata against a shale strata. If conditions are favorable for petroleum accumulation, the oil would be trapped at the fault line. Faults may also allow migration of petroleum from stratum to stratum across fault lines where the permeable beds are adjacent.
1-56
This figure illustrates oil accumulation on both the upthrown and downthrown sides of a fault and show how faulting may leave barren places in an anticlinal structure. Wells 1, 3, and 4 are productive; whereas well 2 encounters edge water; and well 3 intersects the fault plane. The sealing of tilted, eroded beds by deposition of new sediments form favorable traps for oil accumulation in the older rocks against an unconformity. In other cases, the oil might migrate across the unconformity and ultimately be trapped in beds not related to those in which the oil was originally stored.
1-57
The impervious stratum at the base of the upper series prevents the escape of the oil. Oil seeks out and accumulates in lenses of porous sands imbedded in dense less porous rock strata. Most sedimentary rocks are laid down at or near the shore line so the channels or lenses of sand would be roughly parallel to the shore lines of the period in which they were formed.
1-58
Lenses of coarse sand embedded oil bearing shales serve as local centers of concentration. Such conditions are common in California fields. Well A encounters four zones of production; whereas well B is barren.
Another important trap generally referred to as a stratigraphic trap occurs from irregularities in bedding and to some extent on structural conditions. A lateral variation of porosity or a pinching out of a porous strata between two impermeable beds will provide favorable conditions for segregation and accumulation of petroleum. In some regions porous coral limestone has been formed in relatively shallow water and subsequently covered by impermeable strata. Gravitational segregation of the oil, gas and water results in the oil and gas migrating to the upper portion of the reef. Local variations in porosity also determine the areas of accumulation of oil and gas. Sediments adjacent to and above coral reefs usually dip slightly away from the reef due to differential compaction of sediments accumulating on the sloping surfaces.
1-59
1-60
Chapter 2
Sample Examination
2.1 Introduction
An accurate sample description is the basic function of geologic work - the foundation on which the entire structure of subsurface analysis rests. This manual has been assembled in an effort to furnish a convenient reference on standard stratigraphic procedures. Techniques of collecting, preparing, examining, and describing well cuttings and core samples are set forth. At one time the primary responsibility of the surface data logger was to provide correlation for structural mapping, now it has become increasingly important for him to also provide stratigraphic data. The source, transporting medium, environment of deposition, and post-depositional history of the sediments all can be determined by sample examination. There are two elements are involved: (1) logging what is physically present in the samples, and (2) interpretation of the geologic history from the material in the samples. A description can become so detailed as to obscure important characteristics of the samples; the surface data logger should learn to be selective and report only the important details. Sample analysis should be made carefully and attentively. The accuracy of a study is dependent upon the quality of the samples and the proficiency of the surface data logger. Careful initial examination and description of the samples will save time and prevent the necessity for re-examination. There will be times when it is impossible due to well conditions for the surface data logger to accomplish this the first time. It is more important that the samples be caught first. There are two general methods of sample description and logging, the interpretive system and the percentage system. The interpretive log is preferable but its accuracy depends in some measure on the quality of the samples, and the surface data loggers familiarity with the local stratigraphic section of the area. Sloughed cuttings must be disregarded, and only the lithology felt to be represent the interval drilled is to be logged. If several different rock types are present in the sample, all are assumed to be derived from the drilled interval, they are logged as discrete beds, interbeds, intercalations, lenses, or nodules, rather than as percentages. The interpretation in this case is based on the surface data loggers knowledge of the area. On interpretive logs, lithologic contacts are drawn sharply, and the entire width of the log column is filled with the suitable lithology plot types. Two hazards in this form of logging are unexpected recurrence of lithologic types and wildcat wells where there are no lithologic histories available for comparison.
2-1
Experience and good training are essential for making a good interpretive log. Generally the surface data logger examining the samples is best qualified to recognize lithologic and formational contacts. Although formation contacts should be picked on the basis of sample data rather than on mechanical logs, the latter, as well as the drill rate, can also be useful in defining boundaries of specific lithologic units, and zones of interest. In percentage logging, the surface data logger, after eliminating the obvious foreign matter and unquestionable sloughed cuttings, logs each rock type with the percentage of that rock type that is present in the sample. This system of logging may be used to advantage in areas where:
1. 2. 3. 4. 5. 6.
The details of the stratigraphy are unknown. The samples are of very poor quality. No mechanical logs are available. The sampled interval contains many thin beds. The stratigraphic sequence is interrupted by structural complications. The person studying the samples is inexperienced or is not an experienced surface data logger.
The principal disadvantages of this system are that lithologic breaks do not show up sharply on the log, and the logging of each rock percentage often gives a confused picture of the in place rock formation.
2-2
2-3
If the amount of cuttings coming across the shaker decreases rapidly, or the cuttings disappear altogether, this could mean that a fine grained sand has been encountered. If this happens, you should check the sand content of the mud. 4. This same problem could also mean that a salt section has been encountered. Check the chlorides of the mud, and watch for deterioration of the mud properties. 5. Check the desander and desilter discharges for the possible presence of sample material which has gone through the shaker screens. A periodic check from these sources will provide a basis for comparison.
3.
2-4
30-foot samples and the drill rate is 30 ft/hr, then catch 5- or 10-foot samples). There is no such thing as too much information.
2-5
2-6
Method One
Set up two containers, such as two 5-gallon buckets. One should contain a non-fluorescent solvent (preferably Varsol or naphtha). This should be used for the first washing to remove the outer coating of drilling mud from the cuttings. In the other container, mix a solution consisting of a commercially available detergent into 5 gallons of water. Wash the cuttings in the detergent solution to remove the solvent. After this they can be washed in water as usual. To make a good inspection for lithology and staining, the cuttings must then be broken or crushed. Sometimes, cuttings that have been subjected to oil-based mud turn to paste if any attempt is made to wash them with water as described in method one. Methods two and three can overcome this problem.
Method Two
Wash the cuttings in only a non-fluorescent solvent, such as Varsol or naphtha. One means of accomplishing this with a minimum use and waster of solvent is to prepare 3 or 4 baths for processing the samples. A closed container (without an internal lip, as some cuttings may hang up at this point) is ideal since this allows shaking of the solvent/cuttings mixture. First, dump the fresh cuttings into the first container containing solvent. Secure the lid on the container and agitate by shaking for approximately 30 seconds. Remove the lid and pour the solvent into another container for use again on the next sample. Refill the container with fresh solvent and repeat the entire process. When the solvent remains clear after agitation, the sample is clean. Again, pour off the solvent for later use and empty the cuttings into the perforated basket in the Sperry-Sun spin-dryer. Remove the solvent from the cuttings by spinning the assembly under the wall mounted motor for 30 seconds. Remove the perforated basket from the spin-dryer and empty the cuttings into a dry sieve assembly. Vigorously shake the sieves side to side to force cuttings through the various screens.
Method Three
Wash the cuttings in the base fluid of the mud (i.e., with Petrofree use Ester). Then wash the cuttings as described above until clean.
2-7
2-8
The 80-mesh screen, with.177-mm openings, will retain most grains of the fine size (.125 mm to .250 mm) and smaller cuttings. An increase in volume here may indicate a softer formation (which should be reflected in the drill rate) since most of the cuttings are generally in the upper 20-mesh screen. The same process of inverting this sieve after some of its catch has been slid onto the sample tray is also recommended. The sieve receiver is an essential component of the sieve set because it catches everything that goes through all the screens. Generally, we would be looking for very fine grain sand in this portion of the sample. Experience should show that if you expect sand, due to a drilling break, but do not see much in the upper sieves, the sieve receiver will hold what you are looking for. This will also be where dissolved cutting would be collected. After the sieving, a small portion of the washed sample should be put onto one of the trays provided for microscopic inspection and then drained. A larger sample should be placed on another tray, then drained and dried before it is placed into a labeled envelope and boxed for the oil company laboratory analysis. The tray for immediate examination should contain only a single layer of cuttings. This is important when considering the relative percentages of the different materials contained in the sample.
2-9
2-10
However, because compaction occurs as the bullet enters, the results are inevitably less reliable than those from a conventional core.
2-11
samples bring out rock characteristics that are not apparent in dry samples. Leave the samples covered by a thin film of water. After the cuttings have been logged, they are set aside to dry and then placed in dry sample envelopes. The technique of scanning samples before logging them in detail has many advantages. In addition to helping the surface data logger pick tops and lithologic changes it may also aid him in determining the extent of porous and hydrocarbon bearing intervals. However, the principle advantage of this technique is that it provides the surface data logger the opportunity to observe and interpret depositional sequences. When sample intervals are laid out in sequence subtle changes in texture, mineralogy, color and facies often become apparent even before microscopic examination. Thus the surface data logger is alerted to look for these changes when making the detailed sample examination. This method of examining samples encourages surface data logger to observe and log lithologic rather than sample interval units. It is still important that the surface data logger do a complete and thorough description of each sample. Textures in carbonate rocks can be clearly observed with the aid of wetting agents such as mineral oil, glycerine, clove oil, etc. A further improvement of this technique is the use of transmitted light as described below.
2.6.3 Abbreviations
Abbreviations should be used for all descriptions recorded on surface data logs. Appendix A contains a list of the recommended abbreviations from the SPWLA manual.
2-12
Rock type - followed by classification Color Texture - including grain size, roundness, and sorting Cement and/or matrix materials Fossils and accessories Sedimentary structures Porosity and oil shows
2.6.6 Color
Color of rocks may be a mass effect of the colors of the component grains, or result from the color of the cement or matrix, or staining of these. Colors may occur in combinations and patterns, e.g., mottled, banded, spotted, variegated. It is recommended that colors be described on wet samples under ten-power magnification. If is important to use the same source of light all the time and to use constant magnification for all routine logging. General terms, such as dark grey and medium brown, generally suffice. Ferruginous, carbonaceous, siliceous, and calcareous matter are the most important staining or coloring agents. From limonite or hematite come yellow, red, or brown shades. Gray to black color can result from the presence of carbonaceous or phosphatic material, iron sulfide, or manganese. Glauconite, ferrous iron, serpentine, chlorite, and apatite impart green coloring. Red or orange mottlings are derived from surface weathering or subsurface oxidation by the action of circulating waters. Bit or pipe fragments in samples can rust and stain the samples. Drilling mud additives may also cause staining.
2-13
2.6.7 Texture
Texture is a function of the size, shape, and arrangement of the constituent elements of a rock.
2.6.9 Shape
Shape of grains has long been used to decode the history of a deposit of which the grains are a part. Shape involves both sphericity and roundness. Sphericity refers to a comparison of the surface area of a sphere of the same volume as the grain, with the surface area of the grain itself. For practical purposes, distinction is usually made in large particles on the basis of axial ratios, and in grains by visual comparison with charts. Roundness which refers to the sharpness of the edges and corners of a fragment, is an important characteristic that deserves careful attention in detailed logging. Five degrees of rounding are described as follows: Angular edges and corners sharp, little or no evidence of wear. Subangular faces untouched but edges and corners rounded. Subrounded edges and corners rounded to smooth curves; areas of original faces reduced. Rounded Original faces almost completely destroyed, but some comparatively flat faces may be present; all original edges and corners smoothed off to rather broad curves. Well rounded no original faces, edges, or corners remain; entire surface consists of broad curves, flat areas are absent.
2-14
2.6.10 Sorting
Sorting is a measure of the size frequency distribution of grains in a sediment or rock. It involves shape, roundness, specific gravity, and mineral composition as well as size. A classification given by Payne (1942) that can be applied to these factors is: Good: Fair: Poor: 90% in 1 or 2 size classes 90% in 3 or 4 size classes 90% in 5 or more size classes
More precise values may be determined by direct comparison with sorting comparators.
2-15
may cause loss of porosity either by compaction, or by swelling when water is introduced into the formation. Argillaceous material can be evenly distributed in siliciclastic or carbonate rocks, or have laminated, lenticular detrital or nodular form. Compaction and the presence of varying amounts of secondary quartz, secondary carbonate, and interstitial clay are the main factors affecting pore space in siliciclastic rocks. While there is a general reduction of porosity with depth due to secondary cementation and compaction, ranges of porosity vary considerably due primarily to extreme variations in amounts of secondary cement. For instance, coarse-grained sandstones have greater permeability than finer ones when the same amount of cementing material is available to both. However, the same thickness of cement will form around the grains regardless of their size, therefore the smaller pore spaces, which occur in finer grained sandstones, will be cemented earliest.
2-16
2-17
Sample Examination Some Criteria and Procedures For Rock and Mineral Identification
2.10 Some Criteria and Procedures For Rock and Mineral Identification
Etching is accomplished by sawing or grinding a flat surface on a specimen, which is then submerged for a short time (10 to 30 seconds) in dilute acid with the flat surface parallel to the surface of the acid. After etching the surface is carefully washed by gentle immersion in water, care being taken not to disturb the insoluble material adhering to the surface of the specimen. Limestone specimens etched in HCl usually develop and acid polish. Insoluble materials such as clay, silt, sand,
2-18
Sample Examination Some Criteria and Procedures For Rock and Mineral Identification
chert, or anhydrite will stand out in relief against the soluble matrix. Dolomite crystals usually stand out also, inasmuch as they are attacked by the acid more slowly than is calcite. The internal structures of fossils, oolites, and detrital fragments are commonly revealed on an etched surface. If the appearance of the etched surface is so diagnostic that permanent record is desired, an acetate peel can be made, or the surface can be photographed.
Hardness
Scratching the rock fragment surface is often an adequate way of distinguishing different lithic types. Silicates and silicified materials, for example, cannot be scratched, but instead will take a streak of metal from the point of a probe. Limestone and dolomite can be scratched readily, gypsum and anhydrite will be grooved, as will shale or bentonite. Weathered chert is often soft enough to be readily scratched, and its lack of reaction with acid will distinguish it from carbonates. Caution must be used with this test in determining whether the scratched material is actually the framework constituent or the cementing or matrix constituent. For example, silts will often scratch or groove, but examination under high magnification will usually show that the quartz grains have been pushed aside and are unscratched, and the groove was made in the softer matrix material.
Parting
Shaly parting, although not a test, is an important rock character. The surface data logger should always distinguish between shale, which exhibits parting or fissility and mudstone, which yields fragments which do not have parallel plane faces.
Thin Sections
Certain features of rocks may not be distinguishable even under the most favorable conditions without the aid of thin sections. Thin sections adequate for routine examination can be prepared without the use of the refined techniques necessary to produce slides suitable for petrographic study. Some of the questions of interpretation which might be clarified by the use of thin sections include the following: mineral identification, grain-distribution, grain
2-19
Sample Examination Some Criteria and Procedures For Rock and Mineral Identification
sizes, source rock quality. Although wetting the surface of a carbonate rock with water, or mineral oil, permits in depth observation of the rock, some particles, or particle-matrix relationships still remain obscure until the rock is examined by transmitted light, plane and/or polarized. Once these features have been recognized in thin sections, they are frequently detectable in whole fragments, and only a few thin sections may be needed in the course of logging a particular interval. It is important to have polarizing equipment available for use in thin section examination - many features of the rock texture, and some minerals, are most readily recognized by the use of polarized light.
Insoluble Residues
Carbonate rocks may contain significant percentages of chert, anhydrite, sand, silt, or argillaceous materials that are not readily detected in the untreated rock fragments. The study of cherts and associated residues has been a common practice for many years in certain areas. For routine logging of micro-insoluble residues, symbols for accessory minerals, should be used.
Versenate Analysis
Versenate analysis is a relatively fast and inexpensive method for determining quantitatively the calcite/dolomite ratios of given carbonate rocks. The method has shown merit in the mapping of intimately associated limestone and dolomite. It is based on the color reactions of a reagent on crushed and sieved carbonate samples.
2-20
Sample Examination Some Criteria and Procedures For Rock and Mineral Identification
Clay
Shales and clay occur in a broad spectrum of colors, mineral composition, and textures. Generally, their identification is done with ease; however, light colored clay is commonly mistaken for finely divided anhydrite. The two may be distinguished by a simple test. Anhydrite will dissolve in hot dilute hydrochloric acid and, when cooled, will recrystallize out of solution as acicular needles. Clay remains insoluble in the hot dilute acid.
Chert
Recognition of the more common varieties of chert and siliceous carbonates generally is not a problem. Weathered chert, however, is often found to be soft enough to be readily scratched and mistaken for clay or carbonate. Lack of reaction with acid generally distinguishes this type of chert from carbonates. Clay and tripolitic chert may require petrographic techniques for differentiation. In thin sections under polarized light, chert commonly has a characteristic honey-brown color.
Evaporites
1.
Anhydrite and gypsum are usually readily detected in cuttings. Anhydrite is more commonly associated with dolomites than with limestones, and is much more abundant in the subsurface than gypsum. At present, there appears to be little reason to distinguish anhydrite from gypsum in samples. Anhydrite is generally harder and has a pseudo-cubic cleavage; the cleavage flakes of
2-21
Sample Examination Some Criteria and Procedures For Rock and Mineral Identification
gypsum have swallow-tail twins. Anhydrite can be readily recognized in thin sections by its pseudo-cubic cleavage, and under polarized light, by its bright interference colors. The dilute hydrochloric acid test is a valid and simple test for anhydrite or gypsum in cuttings. Place the cuttings in a watch glass and cover with acid. Heat on a hot plate to 250oF (120oC) and wait for the sample to start dissolving. If anhydrite or gypsum is present, acicular gypsum crystals will form around the edge of the acid solution as it evaporates. If the sample contains much carbonate, a calcium chloride paste may form and obscure the acicular gypsum crystals. Dilute the residue with water, extract and discard the solution and report the test. A simple method of distinguishing finely divided anhydrite from silt is a scratch test. This can be done by two methods: a b Rub a glass rod on the residue in the bottom of a glass test plate and listen for a gritty sound. Place a drop of liquid containing the residue on a glass cover-slip, and cover it with another slip. Rub them together between the thumb and forefinger. Examine the slips under the microscope for scratch marks, or listen for a gritty sound.
2.
Salts are rarely found at the surface and generally do not occur in well samples. Unless a salt-saturated or oil-base mud is used, salt fragments or crystals dissolve before reaching the surface. The best criteria for detecting a salt section are: (a) the occurrence of salt hoppers (molds of dissolved salt crystals in other rock fragments), (b) marked increase in salinity of the drilling mud, (c) a sudden influx of abundant caved material in the samples, (d) a sharp increase in the drilling penetration rate, and (e) mechanical log character, particularly the sonic, density, and caliper log. Cores are the most direct method of determining whether salt is present, but they are not usually cut in salt sections. Salts are commonly associated with cyclical carbonate sections and massive red bed sequences. In the former, they are usually thin bedded and often occur above anhydrite beds. Potassium-rich salts, the last phase of an evaporation cycle, are characterized by their high response on gamma ray log curves.
Phosphate
Place on the suspected mineral (either on the hand specimen or on an uncovered thin section) a small crystal of pure white ammonium molybdate. Allow one or two drops of dilute nitric acid to fall on the crystal. If the rock contains phosphate, the crystal rapidly takes on a bright yellow color.
2-22
Siderite
Siderite is usually readily distinguished by its characteristic brown color and slow rate of effervescence with dilute HCl. The mineral often occurs as buckshot-sized pellets. The presence of siderite or iron dolomite in the same rock with calcite may be difficult to recognize and the following stain procedure is recommended for use when such cases are suspected. The polished face of the chip is immersed for 5 to 10 minutes in a hot, concentrated solution of caustic potash to which a little hydrogen peroxide is added at intervals during treatment. The surface is finally washed and dried in the air. Siderite is stained brown while ferrous dolomite (ankerite) takes a weaker stain and ordinary dolomite remains colorless; calcite is roughened but is not destroyed and chamosite retains it green color unless carbonate of iron is present. This method is equally applicable to powders.
Feldspar
The presence, quantity and type of feldspar constituents can be important in the study of reservoir parameters in some sandstones, particularly the coarse arkosic sands or granite washes. Staining techniques, operationally applicable to rather large etched core (or surface) sample surfaces, allows a better estimation of the amount and distribution of feldspar grains. The use of sections to make these estimates is expensive, and often difficult because of the small surfaces provided.
Bituminous Rocks
Dark shales and carbonates may contain organic matter in the form of kerogen or bitumen. Carbonates and shales in which the presence of bituminous matter is suspected should be examined by thin section and pyrolysis-fluorometer methods for possible source rock qualities. Dark, bituminous shales have a characteristic chocolate brown streak which is very distinctive.
2-23
the void space in the rock; permeability is a measure of the capacity of a rock for transmitting a fluid. Permeability is dependent on the effective porosity and the mean size of the individual pores; it has a direct bearing on the amount of fluid recoverable, whereas porosity determines the amount that is present. Generally the smaller the grain or crystal size, the lower the permeability. The ability to estimate porosity accurately comes through practice and experience in examining samples. Although magnification of about 10X is frequently adequate to detect porosity, higher magnification is often necessary. Pores are easier to recognize in dry samples than in wet ones. Qualitative estimates of the pore size range and mean should be made for all porous intervals. Quantitative permeability measurements are not possible by microscopic examination, but qualitative indications often may be seen. The speed with which water is absorbed by a rock fragment is an indication of its relative permeability. Conversely, water will stand up in a bead on a completely impermeable fragment. If porosity of any category is observed, it should be thoroughly described using the proper symbols to denote its relative quality. Additional comments about it should be made in the remarks column. Samples with porosity should always be checked for hydrocarbons regardless of whether or not staining is observed on the rock surface. High gravity oils may leave little or no visible staining on the rock. A chlorothene or other nontoxic solvent cut should dissolve any trapped hydrocarbons from the inner pores that have not been previously dissipated. In siliciclastic rocks three types of porosity are common: intergranular, moldic and fracture. Intergranular is by far the most common type and the most readily seen in cuttings. Normally it is difficult to detect moldic or fracture porosity in cuttings. Moldic porosity, the result of leaching of soluble grains is often difficult to differentiate from plucked grains. The presence of fragments of coarsely crystalline vein calcite in cuttings is often the only indication of the occurrence of fractures. Porosity in carbonate rocks is generally classified in one of the following categories: interparticle, intercrystal, vuggy, moldic and fracture. These broad categories can be further subdivided into such specific types as inter-oolitic, leached fossil, pel-moldic, etc. These have genetic significance and should be described in detail on the log. Interparticle porosity, the pore space between particles of the rock, and intercrystal, that between crystals, usually is not larger than the particles or crystals. Vuggy porosity, comprising pore space equal to or larger than the particles of the rock, commonly results from the leaching of particles, and may have the form of irregular shaped voids. It is important to record the sizes of vugs, as well as their presence. Where vugs are as large as, or larger than, the cuttings, the only evidence of their presence may be the occurrence of crystals, either free, or cemented to a surface which is actually a portion of the wall of a vug. In this situation it may be impossible to differentiate vuggy from fracture porosity in carbonate rocks.
2-24
Two carbonate porosity classifications are listed below. One developed by Choquette and Pray emphasizes geologic or genetic interpretation. The other by Archie deals primarily with physical properties used for evaluating or exploiting the fluid content of a rock.
Genetic or descriptive classification Sample 1 - limestone, oolitic grainstone, fine grained, well sorted with interparticle porosity Sample 2 - dolomite, finely crystalline, sucrosic, porous
2-25
2.
Archie - III FB10 a b c d e III = sucrosic or granular texture F = fine grain size B = visible pores less than 0.125 10 = 10% porosity in B pores B size pores with 10% porosity + A size matrix pores with 7% = total 0 17%.
The Archie terminology defines both lithic rock types as having identical potential as reservoir rocks although the stratigraphic implications as to the origin of the rocks are entirely different. Data emphasizing the petrophysical characteristics of carbonates derived from Archies classification can be included on sample logs in conjunction with lithologic descriptions.
2.12 Hydrocarbons
Although petrophysical analysis may give conclusive determination of the presence of commercial quantities of oil, it is the surface data loggers responsibility to report and log all shows, and to see that good shows are evaluated. Positive indications of hydrocarbons in cuttings can be a decisive factor in the petrophysicists evaluation of a well. Unfortunately, no specific criteria can be established as positive indications of whether or not a show represents a potentially productive interval. The color and intensity of stain, fluorescence, cut, cut fluorescence and residual cut fluorescence will vary with the specific chemical, physical, and biologic properties of each hydrocarbon accumulation. The aging of the shows (highly volatile fractions dissipate quickly), and flushing by drilling fluids or in the course of sample washing, also tend to mask or eliminate evidence of hydrocarbons. The presence or absence of obvious shows cannot always be taken as conclusive. In many cases, the only suggestion of the presence of hydrocarbons may be a positive cut fluorescence. In other cases, only one or two of the other tests may be positive. Hence, when the presence of hydrocarbons is suspected, it is very important that all aspects be considered: the porosity and thickness of the interval, the petrophysical evaluation, and the quality of the hydrocarbon tests. Listed below are some of the most common methods of testing for hydrocarbons in samples and cores that should be used by the surface data logger during routine sample examination.
2-26
2-27
on the bottom. In the case of oil-bearing noncalcareous sandstone, large lasting bubbles form on the surface but may not float the fragment. The large bubbles result from the surface tension caused by the oil in the sample, which tends to form a tougher and more elastic bubble wall. It should be pointed out that this test is very sensitive to the slightest amount of hydrocarbons, even such as found in carbonaceous shale; therefore, it is well to discount the importance of a positive test unless the bobbing effect is clearly evident or lasting iridescent bubbles are observed. The test is very useful, however, as a simple and rapid preliminary check for the presence of hydrocarbons. A positive oil-acid reaction alerts the observer to intervals worthy of more exhaustive testing.
Fluorescence
Examination of mud, drill cuttings and cores for hydrocarbon fluorescence under ultraviolet light often indicates oil in small amounts, or oil of light color which might not be detected by other means. All samples should be so examined. Color of fluorescence of crudes range from brown through green, gold, blue, yellow to white; in most instances, the heavier oils have darker fluorescence. Distribution may be even, spotted, or mottled, as for stain. The intensity range is bright dull, pale, and faint. Pinpoint fluorescence is associated with individual sand grains and may indicate condensate or gas. Mineral fluorescence, especially from shell fragments, may be mistaken for oil fluorescence, and is distinguished by adding a few drops of a solvent. Hydrocarbon fluorescence will appear to flow and diffuse in the solvent as the oil dissolves, whereas mineral fluorescence will remain undisturbed. When using the Sperry-Sun API gravity chart to determine the API gravity from the fluorescence, it must be taken from the unwashed cuttings mixed with water. By washing the drilled cuttings, some of the oil is washed from the cuttings, resulting in a brighter and lighter color fluorescence than the actual formation. With sidewall cores and conventional cores this problem is not as pronounced.
2-28
contaminate the solvent and render it useless. Chlorothene is recommended for general use although it too may become contaminated after a long period of time. Acetone is a good solvent for heavy hydrocarbons but is not recommended for routine oil detection.
Caution:
Carbon tetrachloride is a cumulative poison and should not be used for any type of hydrocarbon detection.
To test cuttings or cores, place a few chips in a white porcelain evaporating dish or spot plate and cover with reagent. The sample should be dried thoroughly at low temperature, otherwise water within the sample may prevent penetration by the reagents, thus obstructing decisive results. The hydrocarbon extracted by the reagent is called a cut. It is observed under normal light and should be described on the basis of the shade of the coloration, which will range from dark brown to no visible tint. A faint residual cut is sometimes discernable only as an amber-colored ring left on the dish after complete evaporation of the reagent. A very faint cut will leave a very faint ring, and a negative cut will leave no visible color. The shade of the cut depends upon the gravity of the crude, the lightest crudes giving the palest cut, therefore, the relative darkness should not be taken as an indication of the amount of hydrocarbon present. A complete range of cut colors varies from colorless, pale straw, straw, dark straw, light amber, amber, very dark brown to dark brown opaque. The most reliable test for hydrocarbons is the cut fluorescence or wet cut test. In this test the effect of the reagent on the sample is observed under ultraviolet light, along with a sample of the pure solvent as control. The sample should be thoroughly dried before applying the reagent. If hydrocarbons are present, fluorescent streamers will be emitted from the sample and the test is evaluated by the intensity and color of these streamers. Some shows will not give a noticeable streaming effect but will leave a fluorescent ring or residue in the dish after the reagent has evaporated. This is termed a residual cut. It is recommended that the cut fluorescence test be made on all intervals in which there is even the slightest suspicion of the presence of hydrocarbons. Samples that may not give a positive cut or will not fluoresce may give a positive cut fluorescence. This is commonly true of the high gravity hydrocarbons which give a bright yellow cut fluorescence. Distillates show little or no fluorescence or cut but commonly give positive cut fluorescence, although numerous extractions may be required before it is apparent. Generally low gravity oils will not fluoresce but will cut a very dark brown and their cut fluorescence may range from milky white to dark orange. An alternate method involves picking out a number of fragments and dropping them into a clear one or two ounce bottle. Petroleum ether, chlorothene, or acetone is poured in until the bottle is about half full. It is then stoppered and shaken. Any oil present in the sample is thus extracted and will color the solvent. When the color of the cut is very light, it may be necessary to hold the bottle against a white
2-29
background to detect it. If there is only a slight cut, it may come to rest as a colored cap or meniscus on the top surface of the solvent.
Caution:
Proper ventilation is important when using petroleum ether as it may have a toxic effect in a confined space. In addition, petroleum ether and acetone are very inflammable and must be kept away from open flames. Wettability
Failure of samples to wet, or their tendency to float on water when immersed, is often an indication of the presence of oil. Under the microscope, a light-colored stain which cannot be definitely identified as an oil stain may be tested by letting one or two drops of water fall on the surface of the stained rock fragment. In the presence of oil, the water will not soak into the cutting or flow off its surface, but will stand on it or roll off it as spherical beads. Dry spots may appear on the sample when the water is poured off. This, however, is not useful in powdered (air drilled) samples which, because of the particle size and surface tension effects, will not wet.
Hot-Water Test
Place 500 ccs of fresh, unwashed cuttings in a tin or beaker which has a capacity of 1,000 ccs. Pour in hot water with a temperature of at least 170oF (77oC) until it covers the sample to a depth of 1 cm. Observe the oil film thus formed under ultraviolet light and record the amount of oil released using the scale illustrated.
2-30
Iridescence
Iridescence may be associated with oil of any color or gravity, but it is more likely to be observable and significant for the lighter, more nearly colorless, oils where oil staining may be absent. Iridescence may be observed in the wet sample tray. Iridescence with oil coloration or staining may indicate the presence of light oil or condensate.
Pyrolysis Test
When well samples of kerogen-rich rock are pyrolyzed in a thick walled test tube placed over a propane torch, oily material may be generated and condensed as a brown residue around the walls of the tube. This simple technique may be used to identify source rocks capable of generating liquid oil. However, the pyrolysis technique cannot distinguish between oil source rocks and those sediments rich in humic matter (carbonaceous shales and coals) which are considered to be dominately sources for gas. This test is also not responsive to post mature source rocks. The artificial test-tube generating process is believed to be similar to that associated with natural time-temperature dependent processes accompanying rock burial in depositional basins. Hydrocarbons in organic rich sediments may be determined semi-quantitatively with a Turner fluorometer. One hundred milligrams of rock is pyrolyzed as above and the resulting condensation is diluted with 3 milliliters of chlorothene. The fluorescence of the solution thus produced is recorded in fluorometer units. For a more
2-31
comprehensive discussion of pyrolysis and pyrolysis techniques, refer to R. L. Heacock and A. Hood (1970).
2.12.4 Generalizations
No rules of thumb can be used to relate the evidences of the presence of hydrocarbons to potential production. However, there are some generalizations that are worth noting. Lack of visible stain is not conclusive proof of the absence of hydrocarbons. (Gas, distillates and high gravity oils ordinarily will have no visible stain.) 2. Lack of fluorescence is not conclusive proof of the absence of hydrocarbons.
1.
2-32
Bona fide hydrocarbon shows will usually give a positive cut fluorescence (wet cut). High gravity hydrocarbons will often give a positive cut fluorescence and/or residual cut, but will give negative results with all other hydrocarbon detection methods. (Minerals which fluorescence will not yield a cut.) 4. The oil acid reaction test will give positive results when oil is present, but it is very sensitive and may give positive results in the presence of insignificant amounts of hydrocarbons.
3.
Recirculation
Recirculation chiefly refers to sand grains and microfossils from previously drilled rocks which re-enter the hole with the mud stream and contaminate the rising sample.
2-33
Cement
Cement fragments in cuttings are easily mistaken for sandy, silty, or chalky carbonate. However, most cements are of an unusual texture or color, frequently have a glazed surface, tend to turn yellow or brown when immersed in dilute HCl, and are usually full of fine black specks. The latter are sometimes magnetic, in which case the fragments of cement can be removed from the cuttings with the aid of a small magnet. If the identification of cement is questionable, the well record should be examined to determine where the casing was set or cement plugs were set.
Drilling Mud
In examining unwashed or poorly washed cuttings, it is often important to be able to recognize the drilling muds which were used. An inexperienced surface data logger may confuse drilling mud with soft clay, bentonite, or sometimes gypsum or a carbonate. Thorough washing and rinsing in a pan of water will generally remove most mud contamination. If necessary, lithic fragments can be broken open to see if the interior (fresh) differs from the surface (coated). Oil-base and oil-emulsion muds coat the cuttings with oil, and care must be taken to distinguish such occurrences from formation oil. They are generally recognized because they coat all cuttings regardless of lithology, rather than being confined to one rock type. Such contamination can sometimes be removed by washing the samples with a detergent or with dilute HCl. Ligno-sulfate muds may present problems in samples used in palynological studies.
2-34
Miscellaneous Contaminants
Other lithic materials which may be present in cuttings samples and obscure their real nature, or might be logged as being in place, include rock fragments used as aggregate in casing shoes.
2-35
Fusing
Shales drilled by a diamond bit may be burned and fused, resulting in the formation of dark gray or black hard fragments that resemble igneous rock.
Spread
Spread is the separation of large from small cuttings by relative slippage (also called elutriation or differential settling) in the mud stream, so that the cuttings of a rock drilled up into fine chips may overtake the cuttings of a rock drilled up into coarse chips during their journey up the borehole. This results in the wrong sequence of rock types or very mixed sample being recovered.
2-36
2.14.1 Unconformities
Notation on a sample log of any data which suggest the presence of an unconformity is important, even though the evidence is inconclusive. Supporting evidence may be found in nearby wells. In cuttings, the following criteria may indicate the presence of an unconformity:
1. 2. 3. 4. 5. 6. 7. 8. 9. 10.
11.
12.
Concentrations of minerals, e.g., phosphate, pyrite, glauconite, calcite, manganese nodules. Abrupt changes in mineral assemblages, fauna, lithologic character, or cementing material. Iron oxide stains or manganese coatings. Corrosion surfaces, as developed on conglomerates (e.g., blackened limestone pebbles). Desert varnish, as polished surfaces on pebbles. Basal conglomerate - generally more heterogeneous and weathered than other conglomerates. Bone and tooth conglomerate - accumulated as a lag zone overlying an unconformity. Siliceous shells with beekite rings - small, bluish grey to white doughnut-like rings occurring on siliceous shells below some unconformities. Weathered chert - a definite indication of an unconformity, providing the chert is residual and not reworked. Asphaltic residues can be present at unconformities at which oil seeped out to the surface. In the case of cherts, the oil or asphaltic residue may be in the residual chert and not in the overlying reworked material. Porous zones in limestone, caused by solution by ground water, may be evidence of unconformities, but porous zones can develop for considerable distances below the surface. The porosity may not be in contact with the unconformity, but the erosional interval is the cause of it. Limestones that underlie unconformities should be more deeply leached than similar limestones which do not underlie conformities. Other porous zones may occur at unconformities in various types of lithology because of the occurrence there of coarser material and the effects of weathering. An unconformity so established may be traced from well to well by recognition of the porous zones. Caliche and vadose pisolites, may form in carbonate rocks exposed to surface weathering.
2-37
The presence of two or more associated criteria greatly increases the chances that an unconformity is present.
2.14.2 Environments
Environments of deposition may be interpreted from (1) geometry and distribution of depositional units, (2) sedimentary structures and lithologic associations, (3) fossil assemblages. Information from drill cuttings, excepting fossil assemblages, is often insufficient to allow interpretation of environments. When a number of control wells are available in a region and sedimentary units can be traced, it is often possible to interpret at least generalized environments on geometry and distribution of units, lithologic associations, and in some cases, electric log shapes. Sedimentary structures and fossils observed in slabbed cores are the principal physical basis for identifying specific sedimentary environments, and determining sediment genesis. Environments are classified with respect to sea level: continental, coastal, marine; and on the basis of physiography: shelf, slope, basin. Clastic sediments are controlled by the source of transported materials and the currents which disperse them; therefore, it is necessary to distinguish between coastal and continental environments in order to differentiate sand bodies which were formed by different processes and so have very different shapes and characters. The physiographic distinction between shelf and basin is important to the understanding of sandstones which may have been deposited in submarine fans and canyons. Carbonate sediments are generally best understood in terms of physiography. Tabular units may be expected to be present on the shelf, and lenticular units, such as mounds or reefs, form at the loci of major changes in slope; e.g., the shelf margin. The constituents of carbonate sediments are usually generated locally and not derived from external sources as are those of siliciclastics, so they may be found to change character abruptly in response to inherited or constructed topographic features anywhere in shallow marine environments. The distinction between continental, coastal, and marine is of lesser importance; most genetic units in carbonates are marine, although the landward limits of carbonate deposits may be within the coastal realm. Carbonates formed under subaerial conditions in a continental environment may not be volumetrically important but they demand particular attention as indicators of periods of exposure and thus of the intensive diagenesis which may occur under such conditions. Description and logging of drilled cuttings and cores is an essential step in providing data which will contribute to the interpretation of the environment of deposition and genesis of a sedimentary unit. Environmental interpretation from cuttings is extremely difficult and more often than not is impossible. However, in certain exploration areas even gross designations of basin, shelf, or continental is useful information. More specific
2-38
environmental interpretations can be very helpful in establishing local facies variations and sedimentologic sequences and should be recorded on the sample log along with qualifying data.
Acid, hydrochloric; 10% or 20% solution Alizarin Red S dye Beakers (50 ml) Bottles, dropper or wash. Use for acid, chlorothene, mineral oil, or glycerine Brush (paint brush) for cleaning sample trays Carborundum stone (paper or powder) or silicon carbide grit Cement: Lakeside 70 plastic cement, or Canada balsam Comparators for size, sorting, roundness, sphericity and percentage Chlorothene Ultraviolet box Glycerine Hot plate (electric) Lamp for microscope, with frosted and blue filters Magnet (pocket size) Manual (standard sample logging manual) Mineral oil (Nujol) Microscope, binocular, with polarizing attachment desirable Plate (glass) Probes Sieves (5 mm, 20 mm, 80 mm, 120 mm, and 200 mm) Slides (glass, 1 inch 3 inches or 1 inch 2 inches) Spot plates (porcelain) Streak plate
2-39
Trays, sample (metal) 25. Tweezers or forceps 26. Water (distilled) 27. Wetting agent (or household detergent) for wetting powdered or contaminated samples
24.
Binocular Microscope
The lens magnification should range from approximately 9 to 50 diameters. Magnification should be great enough to reveal the essential structure and texture of the sample. It should be low enough to reduce eyestrain and to provide a sufficiently wide field of view for estimating percentages of the rock samples. A magnification of approximately 9 to 12 diameters is best for routine sample examination, and 27 to 50 diameters for more detailed study. To reduce eyestrain, microscopes should be kept clean, properly focused, and in good condition. Lenses should be cleaned with lens paper; facial tissue, or a very soft, clean cloth may be used if lens paper is not available. A polarizing attachment for thin section study is a desirable accessory. Microscopes should be kept lightly lubricated.
Light
Natural or artificial light may be used; however, samples are usually examined under artificial light produced by one of several types of lamps. A Nikon, Bausch & Lomb, Zeiss, or American Optical illuminator is recommended, but any lamp with a blue bulb or blue filter plate may be used. A lamp that produces a rainbow of colors should not be used as it will tend to mask or distort the true colors of the sample and will cause eyestrain. Not only should sufficient light be applied to the sample itself, but the work area around the microscope should be well illuminated to prevent excessive eyestrain. A gooseneck desk lamp with a blue bulb produces an excellent light, if a conventional microscope lamp is not available.
2-40
large interval of the well poured into trays for study; in addition, it facilitates the picking of sample breaks, inasmuch as gradations and variations in lithology are often readily visible and may be recognized before the samples are subjected to detailed examination under the microscope.
Ultraviolet box
Surface data loggers should have access to an ultraviolet light box at all times while examining samples. All porous intervals should be thoroughly checked for hydrocarbons with this device.
Equipment
1. 2. 3. 4. 5. 6.
Hot plate Glass slides Lakeside #70 thermoplastic cement (broken to approximately 1/8 inch pieces) Glass grinding plate, 1/4 inch 10 inches 10 inches, carborundum stone or #300 or #400 grit (fine) silicon carbide paper Carborundum loose grain abrasive (#240, #400, #600 grit) Tweezers
Procedure
1.
Melt the cement on a glass slide on the hot plate and drop into it one or more cuttings.
2-41
2. 3.
4. 5. 6. 7.
Remove the slide from the hot plate and allow the cement to harden by cooling. Wet the grinding surface and hone a flat surface on the chips. Keep the grinding surface thoroughly wet by either dipping the slide in a water pan or by sprinkling additional water on the surface. The glass plate and loose abrasive method, the wet stone or the carbide paper may be used to make the thin section. Dry the slide and place it back on the hot plate. Using tweezers or a probe, turn over the honed surface of the chip when the cement melts. Remove the chip from the hot plate and press the chip (honed surface down) against the slide as the cement hardens. Hone the chip down to the desired thinness on the glass plate, wet stone, or carbide paper (as in Step 3).
Caution:
Take care in Step 7 not to grind away the mounted chip entirely. It is not necessary to achieve any critical thinness. All that is necessary is to make the rock reasonably transparent. Therefore, check frequently by examining under the microscope. With a new stone, and especially with silicon carbide paper, the small chips will grind down quickly.
This process is simple in practice, and reasonable proficiency can be achieved with very few attempts. Using a carborundum stone, the whole procedure takes less than 10 minutes; when using silicon carbide paper, even less time is required. Better quality sections result from using a glass plate with abrasives. Covering the finished section is unnecessary; simply wet the surface while examining.
2-42
Samples containing a high density of pores less than 1/2 mm in diameter do not stain well because the solution soaks in, coating the walls of the pore with stain, and pore differentiation is difficult. This type of rock is best stained if impregnated with plastic first. In this case, the matrix is stained and the pore space is void of stain. 4. If the sample does take a good stain with the first application, repeat the procedure for sample preparation.
3.
Procedure
Immerse the chips to be stained momentarily in acid, then briefly rinse in distilled water. (Dipping the chips, held in tweezers, in a spot dish depression filled with acid then one filled with distilled water is an effective way.) 2. Apply a drop of two of Alizarin Red S to the chip on a spot plate or impervious surface (or place the chip in the stain) for 45 seconds. 3. Remove the chip and wash off the excess staining solution with distilled water. Let stand until dry. 4. Examine the chip under the microscope. Calcite will stain red, dolomite and other minerals will not be stained.
1.
Note: The stain will come off if mineral oil, clove oil, or glycerine is applied to the samples. It may be removed from samples or thin sections by scrubbing with warm water and mild soap. The concentration of HCl in the staining solution is extremely critical. Variations of a few tenths of a percent will give different contrasts between stained and unstained areas. It is recommended that each time a new solution is mixed, it has exactly the same HCl concentration as the solution being replaced. 0.2% HCl gives a good contrast between stained and unstained areas.
2-43
more representative sample. However, since well cuttings are generally the only source of material available, it is from them that most residues are prepared. In the preparation of insoluble residues, a measured sample (approximately 10 grams by weight) is placed in a 200-ml beaker to which dilute (12% to 15%) HCl is added. The first application of acid should be done slowly, to prevent foaming and overflow caused by rapid effervescence. A few minutes after the initial application, additional acid may be added. After several hours of digestion, the samples should be washed once or twice to remove the spent acid, then a second application of acid should be made. If the sample has only a small percentage of carbonates, a single application of acid will suffice for total digestion. Evidence of incomplete digestion will be the presence in the samples of dolomite fragments with rough surfaces, dolomite rhombs, or rounded limestone fragments. Finally, the sample should be washed to remove all traces of acid and prevent scum and caking of the residues.
2-44
Testing procedure
Crush the sample and collect the portion that passes through a 150-mesh sieve. 2. Place approximately 0.02 grams of the sample in a 10-ml beaker and wet the sample with one drop of Aerosol solution. 3. Place the beaker under a mixer with a glass stirring rod and start the mixing motor. 4. Add 3 ml of the reagent solution to the beaker and start a stopwatch when the solution comes in contact with the sample.
1.
2-45
When the solution turns from blue to pink, stop the watch and record the time on the Versene Analysis data sheet in the Reaction Time column. 6. The calcite/dolomite ratio is obtained from the accompanying chart. Silica and argillaceous impurities of less than 30% have no effect on the reaction time, but the presence of +15% anhydrite decreases the reaction in both limestone and dolomite.
5. Table 2.1 % Calcite/Dolomite vs. Reaction Time Chart Calc/Dolo Ratio Time (sec) Calc/Dolo Ratio Time (sec) Calc/Dolo Ratio Time (sec)
100/0 99/1 95/5 90/10 89/11 87/13 84/16 80/20 79/21 77/23 76/24 75/25 72/28 70/30 69/31 68/32 67/33 66/34 64/36 63/37 62/38 61/39 60/40 59/41 58/42
29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53
57/43 56/44 55/45 54/46 53/47 52/48 51/49 50/50 49/51 48/52 47/53 46/54 45/55 44/56 43/57 42/58 41/59 40/60 39/61 38/62 37/63 36/64 35/65 34/66 33/67
/ /Dolo
54 55 56-57 58 59 60-61 62 63 64 65-66 67-68 69-70 71-72 73-74 75-76 77-78 79-80 81-82 83-85 86-88 89-90 91-92 93-96 97-99 100-103
Calc/Dolo Time
32/68 31/69 30/70 29/71 28/72 27/73 26/74 25/75 24/75 23/77 22/78 21/79 20/80 19/81 18/82 17/83 16/84 15/85 14/86 13/87 12/88 11/89 10/90
100-106 107-110 111-114 115-119 120-124 125-131 132-137 138-146 147-156 157-168 169-182 183-198 199-218 219-237 238-258 259-285 286-318 319-355 356-402 403-462 463-537 538-630 631-703
2-46
Result
If the rock turns purple, it is calcite or limestone. If the rock remains unchanged, it is dolomite.
Caution:
Equipment
1.
2. 3. 4. 5. 6. 7. 8. 9.
Plexiglass CO2 cell fitted with one of the following: Marsh Instrument Co. Type 11 pressure gauge (0-15 psi) Marsh Instrument Co. Type 28 pressure gauge (0-30 psi) Rustrack Model 2162 pressure recorder (0-15 psi or 0-30 psi) (Optional) Wrist-Action shaker or equivalent shaker capable of mild agitation with the complete CO2 cell. Analytical balance (o.1 mg precision for calibration) or a portable balance (10 mg precision for field use). 10 ml graduated pipette and aspirator bulb or equivalent Mortar and pestle Glazed paper or equivalent Vacuum grease or lubriseal Graph paper Small brush
2-47
Reagents
Dilute Hydrochloric acid solution. Dilute the concentrated reagent grade HCl with distilled water (20% HCl by volume). 2. Calcium carbonate - reagent grade (laboratory standard).
1.
Procedure
Construction of the calibration curve
1. 2. 3. 4.
5.
6. 7. 8. 9. 10.
11.
Weigh exactly.20,.40,.60,.80, 1.0, and 1.2 grams of reagent grade CaCO3 onto six pieces of glazed paper. Remove the top and acid cup from the CO2 cell. Inspect the CO2 cell to ensure that it is clean and dry. Slide the paper and sample to the bottom of the cell by holding the cell in a horizontal position. Raise the cell to vertical and dump the sample onto the cell bottom. Brush the paper with the small brush to remove traces of the sample. Measure and pour 10 ml of 20% dilute HCl into the acid cup and lower it into the cell. Be careful not to spill the acid and get any of it on the bottom of the cup. Be sure the cell O-ring seals and pressure connections are in good condition. Use a light coating of vacuum grease on the O-ring seals. Tighten the cell cap being careful not to splash acid onto the sample. Open the bleeder valve until a 0 pressure reading is obtained. Close the bleeder valve tightly. Tip the cell and allow the acid to run out of the cup onto the sample. Swirl the acid and sample gently. Keep the reactants in the lower part of the cell. If a mechanical shaker is available, agitate for 20 minutes. If not, swirl continuously by hand until a constant pressure is obtained. (Avoid getting acid into the pressure gauge.) Plot the grams of CaCO3 versus pressure on regular graph paper. Figure 2.2)
2-48
2-49
12.
Compute the average slope of the calibration curve by obtaining from the curve two high pressure readings (P2 - P1) and corresponding weights (W2 - W1). Compute the slope using the following formula:
2 1 Slope = -------------------
P P W2 W1
Repeat the slope calculation for two low pressure readings and corresponding weights. Finally, using the two slopes just computed, compute the average slope using the following formula:
1 2 Average Slope = ---------------------------------------
Slope + Slope 2
13.
1. 2. 3. 4. 5. 6.
7.
Dry the cuttings on a hot plate or under a heat lamp. Grind the cuttings with the mortar and pestle to at least 200 mesh. Sieve the cuttings through a 200-mesh sieve onto glazed paper. Weigh 1.0 gram of grounded cuttings (if available) to the nearest 10 mg or better, depending on the type of balance available. Follow Steps 2 through 11 in the above Construction of the Calibration Curve procedure. For interpretation of the pressure readings, refer to Figure 2.3 and Figure 2.4. These figures are representative of the CaCO3, dolomite and combined CaCO3/dolomite pressure versus time graphs as they would appear on the pressure recorder model. Pressure gauge models
Note: As soon as the cell is tipped to start the reaction, observe and record the rapidly rising pressure at its peak reading. Record this as the CaCO3 pressure. If dolomite is present there should be a slight pause, then a second rise in pressure. Keeping in mind that the dolomite reaction is slow, swirl the cell and allow sufficient time for the reaction to complete. The reaction is assumed to be complete when the pressure stops increasing and remains constant (approximately 15-20 minutes). This final pressure value is the total CaCO3 plus dolomite pressure.
8.
Note: Allow sufficient time for the reaction to complete (i.e., when the pressure stops increasing with time and remains constant, approximately 15-20 minutes). Refer to Figure 2.3 for interpretation of the recorded data. From Figure 2.3, the first pressure peak is due to CaCO3 and the final pressure is the total CaCO3 plus dolomite pressure. The CaCO3 pressure peak is not as well defined for higher dolomite concentrations (for dolomite
2-50
concentrations approaching 100% the Figure 2.3 curve begins to look like the Figure 2.4 (b) dolomite curve). Take care in determining CaCO3 pressure when dolomite concentrations are high. Use the following formulas for calculating the CaCO3 and dolomite percentages. % CaCO3 = Pressure reading in psi 100 Sample weight average slope % dolomite = (Total pressure - CaCO3 pressure) 100 .92 Sample weight average slope
2-51
Figure 2.4
Troubleshooting
Leaks in the pressure system are probably the greatest source of potential trouble. Periodically inspect and replace if necessary, the CaCO3 cell, O-rings, and seals. On models with pressure recorders, also check the fittings and tubing CaCO3 cell to the recorder. A serious leak will show up during a test as a decay in pressure after the normal pressure build up from the CaCO 3 reaction. To check for leaks, pressure the instrument up to 10 to 12 psi (use the CaCO3 and HCl reaction), and let the instrument stand at least one hour. Pressure should not decease unless leaks are present. The second source of potential trouble is lack of agreement with the Calibration Curve or degrade performance. Tests conducted in the laboratory on the CaCo3 cell with the recorder indicated an overall accuracy of +/= 5% even with a portable balance (10 mg precision). If there are no leaks in the system, but performance is questionable with the following: Check for a sticking pressure gauge or recorder malfunction, consult the Rustrack manual for the recorder. 2. Make sure the balance is clean and free of corrosion on the weighs and pan. Shield the balance from any air currents and vibrations as much as possible. 3. Check the CaCO3 cell for contaminants. Be sure the cell is clean and dry. 4. Check for impurities in the reagents. Moisture in the CaCO3 standard will introduce an error.
1.
2-52
2-53
expulsion through a sufficient permeability pipeline and the time for this escaping to take place. In this case, the shales will become compacted and firm, and the pore pressure and matrix stress will be normal. Except for abrupt changes in the sand/shale lithology, a plot of Shale Factor versus depth in the ideal situation show a continuous decline in montmorillonite with depth and the age of the formation. It is the departure from the normal water squeezing with depth and time trend that gives Shale Factor it potential. Wherever completion of this phase of diagenesis has been slowed or interrupted for whatever reason and is not still occurring, generally a higher than normal montmorillonite content is encountered and its corresponding higher than normal water content. This condition has its attendant drilling problems associated with the abnormal pore pressure or subnormal matrix stress. Rather than over-simplify the situation by alluding to an impermeable seal or cap, it seems more reasonable to picture this as a dynamic situation with diagenesis still occurring releasing new fluid but faster than the fluid can escape. This fundamentally, makes the Shale Factor technique an abnormal pressure detecting scheme. So, abnormal montmorillonite for a given depth must first be present for abnormal water content to exist. Further, to prevent rapid leak-off, there must be an absence of permeable lithology which is also verified by Shale Factor. (Other methods of pressure detection are all based on indirect measurements or secondary properties that occur as a result of this incomplete and presently occurring diagenesis; and these include increasing conductivity, decreasing resistivity, increasing sonic travel time, decreasing density, increasing drillability, increasing gas content, increasing temperature gradient, and decreasing salinity. These are not independent and are altered or masked by minor changes in lithology.) This is not meant to imply the other parameters are not useful. However, Shale Factor has its greatest value in correctly interpreting this other data in the light of its more positive identification of clay mineralogy and shale versus sand lithology. Also, Shale Factor is one of the few properties of the formation or cuttings that are not altered by exposure to the drilling fluids. Shale Factor logging experience indicates that there are few pure shales or clean sands in the formations that are drilled. There are instead sandy shales or shaly sands, and it is next to impossible for even an experienced geologist or surface data logger to correctly identify cuttings as one or the other by looking at grains under the microscope. Attempts to correlate drilling breaks with lithology by calling a drilling break a sandier section can be wrong about half the time as these drilling breaks also occur in wet shalier sections. Shale Factor provides a more precise number for this shaliness versus sandiness of a formation. A wet clay might have a Shale Factor as high as 15 to 20, whereas a relatively clean saltwater sand might have a Shale Factor as low as 5.
2-54
Equipment
Mortar and pedestal Methylene blue solution (3.74 per 1000 cc) Dilute sulfuric acid (approximately 5N) Hydrogen peroxide (3% solution) Distilled water Erlenmeyer flask, 250 ml 10-ml pipette 1-ml pipette glass stirring rod hot plate filter paper 200-mesh sieve small magnet portable balance or scale
Procedure
1. 2. 3. 4. 5. 6. 7. 8. 9. 10.
11.
Wash and screen sloughings from a shale sample from the shale shaker. Place the shale samples on the hot plate and cook off all the water from the sample. Grind the dry sample in the mortar and pedestal until it is a fine powder. Run the magnet through the sample to remove any metal particles which may be in the sample. Sift, do not force, the sample through a 200-mesh sieve. Measure 1 gram of sifted cuttings. Add the 1 gram of sifted cuttings, 10 ml of distilled water, 15 ml hydrogen peroxide, and 0.5 ml of sulfuric acid to the Erlenmeyer flask. Boil gently for 10 minutes. Dilute the solution to about 50 ml with distilled water. Add methylene blue solution to the flask in .5 ml applications. After each addition, swirl the contents of the flask for about 20 seconds, and while the solids are still suspended, remove one drop of fluid with the glass stirring rod and place the drop on the filter paper. The end point of the titration is reached when the dye spreads as a greenish-blue ring or tint around the spot of dyed solids. When the blue ring or tint is detected around the spot, shake the flask an additional 2 minutes and place another drop on the filter paper. If the blue ring is again evident, the end point has been reached. If the ring does not appear, continue as before until a drop taken after shaking for 2 minutes shows the blue tint.
2-55
12.
Report as follows: Methylene blue used (ml) 1.425 = Methylene blue capacity (FMBT)
Equipment
Sodium polytungstate (density 2.89 g/cc) Distilled water 350 ml graduated cylinder with stopper Calibration beads 2.14 g/cc white cylinder 2.22 g/cc clear bead 2.27 g/cc pale yellow bead 2.34 g/cc gray bead 2.41 g/cc gray/clear bead 2.47 g/cc blue bead 2.56 g/cc maroon bead 2.70 g/cc black/maroon bead 2.85 g/cc black bead low range calibration beads (optional) 1.80 g/cc green bead 1.90 g/cc orange bead 2.00 g/cc bright yellow bead cleaning basket stirring rod
2-56
graph paper (10 10 divisions per inch) wall mounting bracket containing a light source
4. 5.
6.
Pour 200 ml of sodium polytungstate, the heavy solution, into the graduated cylinder. Add the distilled water very slowly, decanting it onto the inside wall of the cylinder, until a total volume of approximately 330 ml is reached. Slowly lower the cleaning basket, containing the beads, until the sample retriever rests on the bottom of the cylinder. (Drop in the beads if the cleaning basket is already in the cylinder.) Slowly lower the stirring rod to a point about two inches below the beads. Lift the rod quickly about three inches. Wait for the beads to stop moving. Slowly lower the rod to a point no closer than three inches to the bottom of the cylinder. Lift the rod quickly to a point no closer than two inches from the top. Wait for the beads to stop moving. If the beads span less than three 20 ml graduations, mix again in the above fashion.
Note: Use only upward strokes to mix. Always wait for the beads to settle between strokes. Never mix the top and bottom ends of the column It is very easy to mix too much rather than too little. Be sure not to mix the bottom three inches or the top two inches of the column with the stirring rod. Be patient. Time is very effective at smoothing out an initially uneven density gradient. Add the extended low range beads now if desired. 8. Slowly add distilled water, decanting it against the cylinder wall as before to reach a total of 350 ml.
7.
Column Calibration
1.
Prepare a piece of graph paper with 10 10 divisions per inch for use as a calibration graph. a If using the regular set of calibration beads, along the long edge of the paper mark out a suitable span of 100 ml of column height, using a scale of one inch per 10 milliliters. Start with the 10 ml division below the lowest bead at the left margin of the graph. Along the width of the paper graph 2.15 - 2.85 g/cc, using a scale of one inch per 0.1 g/cc. See Figure 2.5.
2-57
If using the extended set of beads, prepare the graph paper as shown in Figure 2.5. Along the side, mark off 1.6 - 3.0 g/cc using a scale of one inch per 0.2 g/cc. Along the bottom edge mark from 20 to 340 ml as shown in Figure 2.6, using a scale of one inch per 40 milliliters.
2.
Draw guidelines, if desired, opposite the density of each bead to be used. This will aid in quick plotting of the calibration graph. See Figure 2.5. a b c Draw the guidelines for the lighter beads along the left margin. Draw the guidelines for the heavier beads along the right margin. The densities of the calibration beads are given on the envelope in which they are contained. They may vary slightly from those given in Section 6.20.2.
2-58
Plot the height of each bead in column milliliters using the markings on the cylinder. a b c d Read the height of the middle of the bead. Read the height with your eye at the same height as the bead to avoid parallax error. Plot the graph in pencil so that you can replot the same graph as the column shifts. Plot each point as a dot and circle it. X out any point improperly plotted. See Figure 2.6.
4.
Draw a smooth curve through the points. a b This curve need not be a straight line; it should, however, be a smooth curve. Your shale densities will be more accurate if you draw a smooth line instead of just connecting the points. This is because there is some error in the density of the calibration beads. Typically the line should flatten out at one or both ends and be fairly straight throughout the middle section.
c
5.
If the points do not make a suitable curve, slowly insert the stirring rod and make one or two strokes upward through the uneven areas. a Remember the curve will become smoother and more stable with time. Try to leave it alone.
2-59
b c
Remember the beads will usually not move upward and they will usually not move closer together. If a single bead is too high a short stroke downward, beginning above the bead and pushing down on it, will cause it to move downward in the column.
Replot the calibration graph whenever there have been noticeable changes in the positions of the beads. It is a good practice to replot the calibration graph every two days during continuous drilling. a b c Erase the penciled curve and plotted points, and reuse the same prepared graph paper. Do not plot beads resting at the bottom or floating at the top of the gradient mixture. Write the date and time on the new plot each time one is made.
2.
Remove the accumulated shale every two days of normal drilling, or whenever the equivalent amount of shale samples have accumulated. This prevents the premature contamination of the column by the absorption of the water contained in the shale. a b c d Lift the cleaning basket very slowly. Plan to take almost a minute to lift the basket from the bottom to the top of the column. Remove the shale and beads, clean the beads, and replace the beads in the cleaning basket. Just as slowly, lower the basket back to the bottom of the column. Replot the calibration curve.
2-60
The shale densities, for any reason, can no longer be discerned. The beads no longer form a smooth curve. The column is too dirty or too discolored to be useful. There are more than two beads on the bottom. There are more than three beads on the top.
Beads on the top or bottom of the column cannot be graphed. And, if shale floats or sinks, its density cannot be measured in the gradient column. The areas above the highest suspended bead and below the lowest suspended bead cannot be used. Remember that a new column is not stable so you should plan ahead. Mix a new column when first arriving on the site, before it is needed, and if it is necessary to replace an old column, plan to do it, whenever possible, while the bit is off bottom for at least a few hours.
Technical Tricks
Avoid unnecessary exposure of the column to ultraviolet radiation and vibration. Keep the column out of the sun and cushion it from vibrating machinery. In the logging unit, if light is a problem, shape a cardboard tube from a paper towel roll so that you can put it around the graduated cylinder when it is not in use. 2. Avoid inferior quality cuttings such as so-called gumbo shale. Not only will these give unreliable readings, but they will hydrate and clog up the column. Any time scum builds up in the column, the surface data logger should make sure, when making density readings, that the shale particles have freely found their own level without sticking to other material. 3. It is theoretically possible to raise the column by adding sodium polytungstate to the bottom of the column. To do this use at least 16 inches of Tygon tubing and a Sand Content Funnel to afford sufficient hydrostatic pressure to raise the column.
1.
For the most consistent density readings when the drilling mud system is using a water-based mud, follow these recommendations.
1.
2-61
Caution:
Do not handle the cuttings with your fingers. The oils in the skin will contaminate the density readings. Use tweezers to handle the individual cuttings.
2.
Under the microscope, select suitable bottom hole cuttings. See Figure 2.7. a b c d e f Select sharp, angular cuttings. Avoid smooth cuttings. Avoid cutting sloughed from the sides of the hole. These may appear smooth on one side or be flat rather than angular. Avoid cuttings which are cracked or have discernible crevices. These often contain trapped gas which would alter the density reading. Select bottom hole cuttings. They may appear slightly changed in color, if the bit is drilling a new formation. Pick cuttings of average size. Pick at least five good samples for each logging interval.
Note: Do not attempt to measure the density of very porous shale (so-called gumbo shale). Such shale absorbs the gradient fluid very quickly and thus no true density can be determined. The high water content of the shale also quickly contaminates the gradient solutions.
3.
Use the cuttings as soon as possible. a b c d Maintain a consistent routine. Pat the cuttings dry with a paper towel. Do not air dry, spin dry, or force dry the cuttings. If waiting time must be inserted at this stage, leave the cuttings submerged in water during the waiting period. Keep this period to a minimum. Select sharp, angular cuttings. Avoid smooth cuttings. Avoid cuttings sloughed from the sides of the hole. These may appear smooth on one side or be flat rather than angular. Avoid cuttings which are cracked or have discernable crevices.
Note 1: Note 2:
Note 3:
2-62
This procedure is basically the same as when using water-based drilling fluid, although oil-based mud allows more consistent readings. The extended low range calibration beads may be needed since samples become available at shallower depths. Use normally washed and screened cuttings 2. Select suitable cuttings as with water-based mud.
1.
2-63
3.
Note: If waiting time must be inserted, leave the cuttings submerged in the base oil of the mud during the waiting period.
Using Dry Cuttings
A good shale density gradient can be obtained using dry cuttings so long as dry cuttings are consistently used. It is much harder to select suitable cuttings once the shale has been dried. Cracks and crevices become a major difficulty. The density readings are harder to make since the cuttings absorb the gradient fluid after they have reached a first equilibrium. Hence, unless unavoidable, the use of dry cuttings is not recommended.
Using tweezers, release one cutting immediately above the surface of the mixture and wait for it to come to rest in fluid of the same density.
Note: The cuttings will normally fall quite rapidly at first, then begin to slow as it nears fluid of the same density. It will have carried a halo of less dense fluid with it, which will disperse in a few seconds. Read the level of the midpoint of the cutting about four seconds after it has slowed to near motionlessness. Immediately use the calibration graph to figure the density. 3. Do not record unreliable readings. The following are such indications:
2.
a b
4.
There are observable air bubbles released from the cutting. The cutting adheres to the column wall or another cutting.
After recording the densities of at least five cuttings in an interval (normally 30 feet), compute the average density. a b c Discard the highest reading and the lowest reading, and average the middle three. If you vary this procedure, note the variation in writing on the data sheet. Note any known changes during the interval which could affect shale density readings as comments on the data sheet. Include changes in lithology which affect density readings.
5.
Record these average densities on the standard Sperry-Sun Surface Logging Systems data sheet as shale density. Then enter the density in the computer.
If you vary the above procedures, do so consistently and note the variation in writing on the data sheet.
2-64
Replot the calibration graph whenever there have been noticeable changes in the positions of the beads. It is a good practice to replot the calibration graph every two days. Do not attempt to second-guess your results. Make predictions, but remain objective by following the established procedures. Note: The accuracy and reliability to the shale density graph is directly determined by the consistency and careful method of the surface data logger.
2-65
2-66
Chapter 3
Show Evaluation
When an oil or gas reservoir is drilled, a drilled interval of rock fragments and the formation fluids that it contains are carried to the surface. If it was possible to reconstruct these fragments and fluids at reservoir temperature and pressure, and if the fragments and fluids were the only source in the borehole, it would be possible to determine, the quantity, type and productivity of the fluids contained in the reservoir. This highly desirable hypothesis remains beyond practical achievement. A reservoir is a highly complex system whose stability is distorted even prior to being drilled. The addition of material from the borehole wall and radical changes in temperature and pressure in transit to the surface will change the composition of the sample. At the surface, variable efficiencies and characteristics of the surface extraction and analytical equipment will lead to further compositional and concentration changes as the sample proceeds to final analysis. Quantitative determination using Mud Log gas shows is, and will remain, an impossibility. On the other hand, when used as qualitative formation evaluation tool, the Mud Log is essential and irreplaceable. It is not the intent of this manual to propose changes in surface data logging technology or procedures, but to examine the physics and chemistry underlying the basic gas extraction and analysis process.
3-1
All that remains now is to convert this parameter to a meaningful representation of the character of the formation before being disturbed by the bit. This conversion is accomplished by three parts of the gas detector and related equipment. These parts are: The gas trap, which is the device for removing gases from the drilling mud. 2. The transporting equipment, consisting of a sample pump, pumping the gas-air mixture to the gas detector, the hoses, the plumbing, and flow regulation equipment. 3. The gas detectors proper (Total Hydrocarbon Analyzer and Gas Chromatograph). These detectors are the Flame Ionization Detectors, see the Sperry-Sun Drilling Services Gas Systems Manual for detailed information on these detectors and flow conditioners.
1.
3-2
Simultaneously, as the sample of mud is being pumped through the trap, the whirling action of the impeller whips air from the atmosphere inside the canister into the mud. These bubbles of air tend to become united with the tiny molecules of gas entrained in the mud and being much larger, develop a size having a surface tension sufficiently low to be released from bondage by the drilling mud. These bubbles of air thus serve as a carrier by going into the mud, uniting with the gas and carrying the gas out of the mud into the atmosphere of the canister.
2.
3. 4. 5.
The trap should be located as near as possible to the discharge of the flowline, or at least in a way so that it will have immediate access to the mud returning to the surface. See Figure 3.2. The trap should be located in an open atmosphere. The trap discharge must have immediate access to fresh air. This may not always be possible on some rigs. If difficulty is experienced with lost circulation material, removing the flange on the trap bottom sometimes helps maintain a steady flow of mud. Keep the locking screw on the adjustment jack tight after adjustments are made. Mud should not be discharging from the trap in an intermittent manner but should exit in a continuous flowing manner.
3-3
Agitator Motor Vacuum Line Air Port Return Flowline Ditch (Possum Belly) Gas Trap
Settling Pit
3-4
Vacuum Pump Input From Ditch Pressure Regulator Exhaust Detector Filter and Dryer Flowmeter
Milliameter
Chart Recorder
Figure 3.3 Gas Analysis System
In the logging unit, the flow is kept constant by the Total Hydrocarbon Conditioner, and the flow rate is read on the flowmeter. The flow is split either two or three ways, and the majority of the flow is discharged back to the atmosphere. The optimum flow of air from the trap is 6 to 8 cubic feet per hour (cfh). The total flow meter is a small flowrater, calibrated to read directly in cfh of 0 to 10 cfh. The flow is adjusted by an adjustment knob in the flowrater and should be kept at 6 to 8 cfh which will be maintained at this constant by the regulator. More volume of the air-gas mixture than is necessary for detection is drawn from the trap in order to minimize the lag time between the arrival of a show at the surface and its detection. It now becomes apparent that a reliable gas detecting technique demands that the number of influencing variables be kept in control. Ideally, only the amount of gas in the mud and its corresponding reading should be variable. For this reason, it is important that the trap consistently pump a constant volume of mud, that the amount of air drawn through the trap be maintained at a constant, that there be no leaks or restrictions in the flow system.
3-5
3-6
3.2.1 Definitions
Prior to considering these complex factors, it is necessary to define the usage of the term gas show as it is to be used in this manual. To do this it will be necessary to define also the terms background gas and circulation gas. Although apparently self-explanatory, these terms are commonly given multiple uses in the industry.
Circulation Gas
This is the value of gas seen by a gas detector when circulating under normal conditions, meaning a clean, balanced borehole with drill pipe in the hole and rotating but with the bit off bottom and with no vertical movement. Under such conditions some gas will be present in the sample drawn from the mud stream, but it will represent only contamination or recycled hydrocarbons in the mud.
Background Gas
When drilling through a consistent lithology, it is common for a consistent gas value to be recorded. Certain lithologies (for example, overpressured shales) may show considerable variation in the background gas. Background gas may be observed to vary with drilling, mud or surface conditions without any change in lithology or formation hydrocarbons. This should always be given consideration in both formation and hydrocarbon evaluation.
Gas Show
This any deviation in gas, amount or composition, from the established background. This may or may not accompany a change in lithology, may or may not indicate a significant or economic hydrocarbon accumulation. In other words, Gas Show is a term describing an observed response on the gas detector, having no causal or interpretive significance. It is common for the surface data logger to be asked, What is a good gas show? The answer to this is complex and relates to many factors beyond the simple number of gas units seen. To decide whether a gas show is good or poor, i.e., whether or not a significant hydrocarbon accumulation is indicated, requires a total evaluation of all mudlogging parameters plus consideration of the many variable system conditions. The manner and extent to which shows will manifest themselves varies so greatly among the many different regions that it is impossible to set even a general set of requirements which must be met to qualify a show for additional evaluation or to
3-7
determine to what that evaluation should be. To a large extent, evaluation of interesting zones becomes a matter of comparison of likes and differences. Parameters of the log through the sections in question are compared with the same parameters in adjacent known barren zones for perception of the degree to which they changed or did not change. Generally speaking, the key to interpretation lies not in the magnitude of the reading reached, but in the extent to which it did change.
3-8
3-9
Figure 3.5 Effects of Temperature, Pressure, and Salinity on the Solubility of Gas in Water
3-10
3-11
The imperfect gas law allows the computation of the volume occupied by a specific quantity of gas at varying temperatures and pressures (Figure 3.8). The converse of this would indicate that a fixed volume of pore space, as defined by porosity, hole diameter, and footage, would contain different quantities of gas at different depths, i.e., temperatures and pressures. The result will be that, with increasing depth, equal volumes of pore space, i.e., porosity and gas saturation, will produce greater volumes of gas into the mud stream. See Figure 3.9.
3-12
Although the reservoir will contain a percentage of bulk volume as void space, only a proportion of these voids will be interconnected in such a way as to provide a flowpath, i.e., effective porosity. Permeability is related, since a lithology with zero effective porosity would also have zero permeability. However, the two are not directly related since one, effective porosity, is a measure of the quantity of connected voids whereas the other, permeability, is a measure of the quality of the interconnections. See Figure 3.7. Nevertheless, where a formation lacks effective porosity and/or permeability, only a fraction of the pore fluids contained in it will be liberated to the mud stream during drilling, i.e., only those contained in pores exposed by the crushing process. If porosity is not effective, other pores within the cuttings will remain intact and retain their contained fluids. If permeability is poor, but the effective porosity is good, fluid will be lost from the pore spaces, but only slowly - some being retained even at the surface. In either case, analysis of the fluids retained in the cuttings is essential in estimating the total hydrocarbons in place and the ability of those hydrocarbons to flow out of the formation. Evaluation of cuttings gas is also essential in determining the gas content of non-reservoir rocks which are known to have little or no permeability or effective porosity, e.g., shales and mudstones. Although such rocks are rarely of interest in terms of petroleum accumulation, they are known to be the source of most formation fluids, both water and hydrocarbons. This is due to their high initial content of water and organic debris which is lost in diagenesis due to compaction and thermo-catalytic degradation. Analysis of hydrocarbon content in shale cuttings can be of value in determining the presence of both mobile and potential hydrocarbon sources. Enrichment of hydrocarbons will also occur in shales where rapid sedimentation has prevented dewatering and hence hydrocarbon displacement, resulting in abnormal formation pressures. It should also be remembered that in a potential reservoir where a mixture of fluids is present, (water and/or oil and/or gas), the reservoir will have an effective permeability to each fluid less than that to each if present alone. Some attempt to reconstruct the character of the fluids in place will be advantageous in estimating the nature of eventual production from the reservoir. Relative permeability can be defined as the effective permeability of a reservoir to a fluid (i.e., the actual rate at which that fluid will flow), when one or more other fluids are present, as a fraction of the permeability of the reservoir to that fluid when it is present alone under the same conditions. The effective permeability to each phase will be a function of the number of phases present, the proportion of pore volume occupied by each phase, the chemical composition and physical properties of each phase and the degree of continuity of each phase. See Figure 3.10.
3-13
3.3.5 Summary
It would appear that such complex interactive processes are at work as to hamper any estimate of quantity and type of fluids in place from a surface gas analysis.
3-14
While it is true that no quantitative determination of either fluid in place or eventual production is possible from a surface analysis, it is also true that similar reservoirs can be generalized sufficiently to allow good qualitative correlations. Thus it may be possible to discriminate between a high gravity and low gravity reservoir or even to estimate an oil-water contact by distinctive gas composition. Such interpretations are an essential guide prior to running logs, a tool in the interpretation of those logs, and where control is good, may even prove to be a better means than logs in selecting productive intervals. Most petroleum hydrocarbons originate from a similar organic source and proceed in their maturation via a similar temperature and pressure controlled physico-chemical process. For this reason, petroleum accumulations (although markedly different in composition) tend to show a spectral relationship to each other in terms of the type and amount of hydrocarbon species present. Therefore, though two crude oils may be extremely different in total composition, they will contain some similar components in similar compositional relationships to each other. Since petroleum maturation continues by the continuous cracking of complex branched molecules into simpler straight-chain molecules, these significant relationships are readily seen in the universal petroleum gases, methane through pentane. Thus, study of chromatographic analysis of these gases may often lead to a gross estimate of the type and quality of the reservoir. Nevertheless, all of the factors explained above plus those to follow will contribute to the inaccuracy of such estimates and to the likelihood of some exceptional reservoir giving false indications. In addition, the gas analysis is only one factor on the mud log. All others must be given consideration in evaluating the reservoir. Oil shows, water indications, lithology type, porosity and permeability are of equal or more importance in assessing potential productivity.
3.4.1 Flushing
It is well known that where the borehole pressure exceeds the formation pressure, and permeability exists, the drilling fluid will tend to flush into the formation. If the solids diameter is sufficiently high, filtration will result. Mud filtrate will flush
3-15
into the formation, leaving a clay filter cake which eventually becomes dense and thick enough to be sufficiently impermeable to prevent further filtration (Figure 3.11). Such flushing commonly causes little formation damage since invasion takes place only a short distance into the formation. However, where effective porosity is low, only a small volume of flushing may give a large diameter of invasion. Displacement of gas some distance from the borehole in this way may so reduce the reservoirs gas saturation and effective permeability to gas close to zero in the vicinity of the borehole. Thus a zone which gives good gas shows when drilled will appear water-bearing or recover only mud filtrate when logged or tested.
3-16
This effect can be accentuated when dispersed clay or clay cement is present in a reservoir. If the clays are diagenetically immature, i.e., so-called swelling or mixed layer clays, invasion of mud filtrate may result in swelling or recrystallization of clays within the pore space, drastically reducing both permeability and effective porosity within the reservoir adjacent to the borehole. Flushing will also take place at the bottom of the hole when an overbalance exists (Figure 3.11). In this circumstance no permanent mud filter cake can be formed due to the continuous action of the drill bit. Flushing below the drill bit will have most effect when the reservoir has high permeability and effective porosity. The differential pressure to the advantage of the borehole combined with high impact force due to the jet nozzle pressure drop will force mud filtrate into the formation ahead of the bit. Dependent upon the relative horizontal and vertical permeabilities of the formation, the contained formation fluid may be displaced ahead of or to the side of the bit's path. When the formation is eventually drilled, little or no gas will be liberated. At the surface a flat, unresponsive gas curve will be seen which may even indicate less gas than in nearby formations. Since permeability is high, the reservoir will return to its natural state soon after drilling, and an apparently water-bearing reservoir may later be logged or tested as productive. Common good drilling practice in minimizing mud weight and water loss will be advantageous in reducing flushing, but, to aid in the interpretation of such anomalies, the surface data logger should have sufficient information on the following:
1. 2. 3. 4. 5.
Pump pressure Jet nozzle sizes Mud rheology (Plastic Viscosity and Yield Point) Mud weight and ECD (Equivalent Circulating Density) Mud filtrate (water loss)
Evaluation of the flushing ability of the mud combined with the susceptibility of the formation to flushing should guide the surface data logger in the correct evaluation of the gas curves. Compared with mud gas, high cuttings gas may indicative of poor permeability or effective porosity. The lithological description should indicate which. Where a combination of mud gas and lithology description indicate good permeability, flushing is to be expected. This may result in simple fluid displacement or more permanent physical or chemical damage. (See Figure 3.12)
3-17
It is essential in assessing the possibility of formation permeability damage that the distribution of clays relative to porosity be determined and clay type identified on the log. For this reason clay ion exchange capacity (Shale Factor) should be determined and plotted on the log for the clay fraction of all samples. If this is not possible, the surface data logger should perform a wettability test on a fresh sample containing clay and report the water sensitivity of the included clays. In addition to this it has been suggested that formation permeability damage may result due to the flushing and flocculation of mud clays in the formation adjacent to the borehole. This is difficult to verify, but reference to the base exchange capacity of the drilling mud may be of assistance in ascertaining the possibility of damage of this type, which in any case is likely to occur only where porosity is extremely large, i.e., where pore throats are capable of passing mud solids. A formation indicating high porosity and permeability confirmed by a good rate of penetration, which shows little or no gas in either the mud or cuttings, should be strongly suspected of being flushed prior to drilling - especially where an overbalance exists. The occurrence of a clean, high porosity and permeability formation containing few or no hydrocarbons is in itself sufficient to lead to suspicion. However, the possibly does exist that the formation contains only water without even gas in solution. This possibility may be confirmed or rejected by monitoring mud salinity.
3-18
Like the Shale Factor test, differential salinity is a tool which can be used. So long as a sufficient salinity differential exists between the drilling mud and the formation fluids, the drilling of a high porosity formation containing water should be indicated by a noticeable, though not extreme, change in mud salinity and hence resistivity.
3-19
Casing
Casing
Gas in Sandstone
Oil in Sandstone
Gas in Sandstone
Figure 3.13 Comparison Between a Gas Kick and a Large Gas Show
3-20
Casing
Gas in Sandstone
Oil in Sandstone
3-21
Where permeability is effectively absent, e.g., in clays or shales, even a minor feed-in cannot take place. Fluid pressure in the rock will gain access to the borehole by the opening of preexistent microfractures and partings in the rock. The result will be the caving or sloughing of rock fragments into the borehole, accompanied by a small amount of gas (Figure 3.16). As above, a minimum gas background and, in this case, cavings recovery will exist even when circulating without drilling.
3-22
Permeable (Sandstone)
Bottomhole pressure, when circulating is higher than when the mud is static. This is due to annular pressure losses when circulating. It is, therefore, possible for a feed-in, caving or even a kick, to result due to a resultant underbalance when circulation is stopped. Furthermore, pressure if further reduced due to the so-called swabbing effect when the drill pipe is moved upward, e.g., when making a connection. The literal meaning of swabbing is the pulling of a full gauge tool from the hole, acting like a plunger in a syringe and initiating fluid flow into the borehole. Swabbing by moving the drill string does work in this way. When the drill pipe is pulled upward, the high viscosity gelled mud will attempt to move with the pipe, thus reducing the effective hydrostatic pressure acting on the borehole wall. Pressure reduction is a function of pulling speed, mud rheology and annular diameter. The important consideration is that pressure reduction takes places not just below the bit but at all points in the open hole. Downtime gas or connection gas is a gas show resulting from the momentary underbalance due to the pumps being shutdown and/or pipe movement. It can be recognized by the occurrence of discrete gas show appearance at, or slightly less than, the lag time after circulation recommences. This is gas actually being produced by the formation and the value should be reported on the log as it is indicative of formation permeability and fluid content. The surface data logger should also check a flowline mud sample when a connection gas occurs for evidence of produced oil or saltwater with the gas. The incidence of connection
3-23
gas should also be reported to the drilling foreman who may choose to increase the mud weight in response to the indicated underbalance. It is important to remember that the whole open hole section will be underbalanced by swabbing. The connection gas may not come from the bottom of the hole but from some horizon above. In fact, two or even more connection gases may result from a connection. For this reason, it is important that lag time and annular velocities should be accurately known by the surface data logger so that connection gases can be identified with the producing formation and the mud log annotated accordingly. Drilling into a permeable reservoir with an underbalance is potentially dangerous, as a kick may result. Even if a kick does not immediately occur, the hazardous situation will be marked by an increasing feed-in as more formation is drilled, accompanied by progressively larger connection gases. This condition should be reported by the surface data logger and noted on the mud log. If increases in mud weight alleviate or remove the effect, this should also be noted on the mud log in explanation of the consequent reduction in gas. Where a balanced or even overbalanced condition exists, it is still possible for formation fluids to move into the borehole. Normally, this will occur only where reasonable permeability exists in the formation. The mechanism involved is an interchange of fluids between the formation and the borehole, either by diffusion of lighter hydrocarbons or by the flushing of mud filtrate into the formation, displacing fluids out at some other point. Even though a permeable formation will be flushed with mud filtrate during and shortly after drilling, it is possible for some lighter gaseous hydrocarbons to move back into the flushed zone. This may be due to gaseous diffusion or to the mixing of mud filtrate with formation water containing dissolved gas. Where an impermeable mud filter cake has not yet formed or when the filter cake is removed by stabilizers, bit, or pipe movement, this gas-bearing mud filtrate may interchange with the mud in the annulus and be carried to the surface, resulting in a low, continuous gas background (Figure 3.17).
3-24
A more important incidence of fluid entering the borehole without an underbalance existing is in connection with flushing at the bottom as discussed above. It was shown how mud filtrate may flush into the formation ahead of the bit, displacing the original formation fluids before drilling takes place. Where a sufficient thickness of formation has been cut and vertical permeability exists, it is possible for these displaced formation fluids to be displaced back into the borehole at some point above the bit turbulence (Figure 3.18). The effect of this during normal drilling will be to effectively delay the appearance of a gas show until sometime after the formation is drilled. This may lead the inexperienced surface data logger to suspect some error in the lag time or even that the gas lag is somewhat longer than the cuttings lag.
3-25
It should be remembered that the fluid contained in this influx may not be representative of the total fluids in place, but will be, in effect, produced fluids under a highly energetic water drive. In fact, depending upon mud rheology, hydraulics, overbalance and formation permeability and effective porosity, more fluid may be flushed from the formation in this way than would be produced by drilling it. If drilling stops and circulation continues, this formation sweeping effect will continue and mud gas will be seen to remain relatively constant, declining only slowly as the formation becomes depleted progressively further from the borehole. The effect would appear to behave very much in the same way as a bleed-in. There are, however, dissimilarities which indicate the presence of a different mechanism. Once sufficient exposed formation exists to allow sweeping, only increases in formation porosity, permeability or mud hydraulics will increase the gas seen at the surface. The steady increase with drilling progress characteristic of a bleed-in will not be seen, nor will a connection result in large or increasing connection gases. Most significantly, if the bit is lifted off bottom to a point above the fluid invasion point to a position where normal flushing has produced an impermeable filter cake, the mud gas will show a rapid decline to normal with further circulation, i.e., sweeping no longer takes place. Trip gas is a complex phenomenon. At best it can be said that it is a combination of some or all of the above fluid invasion mechanisms. Nevertheless, regardless of the producing mechanism, it is true that one or more gas shows will be seen when breaking circulation after a trip, coming from the bottom of the hole or the open hole section above it. This gas has been produced from the formation and the surface data logger should report it on the mud log, and lag it back to the probable originating formation and test mud samples coincident with the gas shows for produced oil or water. Care should be taken in anticipating kelly cut and recirculated trip gas. Kelly cut is the result of air being pumped around from a partly empty drill string. This will appear at the surface after the in/out lag (bottoms up plus down pipe volume) has passed from the time circulation commenced. Air will be compressed under borehole hydrostatic pressure, resulting in a slug of aerated mud being pumped from the bit and up hole. The increased tranmissivity of this aerated fluid will assist in any sweeping of diffusion processes active in the open hole especially since immediately after a trip any mud cake stripped during the trip in will not yet be fully reformed. The aerated mud will thus be enriched with formation hydrocarbons and produce a gas show at the surface. The presence and magnitude of this indicate the presence of permeable hydrocarbon bearing formations downhole, but is of little diagnostic value beyond this. In a dry hole, i.e., where the kelly cut picks up few or no hydrocarbons, confusion may arise due to the arrival of the kelly cut, causing visible gas-cut mud at the flowline while the surface data logger reports no gas. An experienced surface data
3-26
logger will learn to anticipate arrival of the kelly cut and will report the fact to the driller. Recycled trip gas, connection gas, or any other gas show will behave in a similar manner to kelly cut, producing an extended gas show. Similarly, oil accompanying a trip gas may, if not checked on its first occurrence, return and be incorrectly logged as a deeper oil show. The experienced surface data logger should anticipate gas show recycling by knowledge of the total circulation time of the circulation system.
3-27
It would appear that the height of the cylinder of cut formation is not relevant since the mud log is a depth-controlled plot, and gas readings are plotted foot by foot as the well progresses. This is, in fact, a misapprehension and a formation identical in all ways will produce higher mud gas readings if drilled at a higher rate of penetration (Figure 3.20).
3-28
The reason for this is very simple and relates to the way in which the gas curve from the chart recorder is translated to a gas curve on the depth-based mud log. It is normal practice for the surface data logger to mark lagged depths onto the chart recorder; these depth marks will be more or less evenly spaced according to the uniformity of the ROP, i.e., the time/depth relationship. Slowly drilled intervals will be more spread out, quickly drilled intervals closely spaced. It is mathematically provable that the amount of gas passing through the gas detector over any depth interval is equivalent to the area defined by the gas curve. Only if the interval is very small can it be said that the quantity of gas is proportional to the height of the gas peak, since the duration is so short. Thus, if a fixed quantity of gas is introduced into the gas detector over two differing time intervals, the high gas reading of short duration, (e.g., high ROP) and the low gas reading of long duration (e.g., low ROP) will define similar areas. If only the height of the gas curve is considered, the longer duration reading would be interpreted as a lesser quantity of gas. For this reason, the magnitude of a gas show should never be considered alone as an indication of reservoir quality without, at least, some reference to and consideration of the rate of penetration. The second factor controlling the volume of the cylinder of formation cut is the hole diameter. Although this will not vary extremely over the commonly used bit sizes, it should be remembered that the volume of the cut cylinder will vary as the square of the hole diameter (Figure 3.21). It should also be remembered, in conjunction of the two factors, that in general, smaller diameter bits will drill less quickly than larger bits, thus reducing expected gas shows accordingly.
3-29
The cylinder of rock cut by the bit is not cut as a whole but as cuttings. The size of teeth on the bit, which is governed by both bit type and bit size, will control the size of cuttings produced. Where cuttings are smaller and more numerous, formation fluids will be more easily liberated from non-effective porosity and inferior permeability, giving improved gas shows.
As mud flow rate increases, the volume of gas and cuttings contained in a fixed volume of mud will decrease. Conversely, the volume of mud passing the gas trap will increase. The net effect should be zero. In fact, the complex geometries and variable efficiencies of the various parts of the system will introduce some variations, but the overall effect is probably not that great. Mud flow rate will not
3-30
vary greatly within any hole size or in relation to hole size within hole sections. This further removes the severity of this effect. Upon entering the drilling fluid at bottom under high pressure, high temperature and the influence of the bit turbulence, the gas and oil which are still retained within the cuttings will be finely dispersed or dissolved in the drilling fluid. As the mud travels up the annulus and the temperature and pressure fall, gas will come out of solution, expand and be released from the cuttings and form larger bubbles. Droplets of oil will release dissolved gas and accumulate into larger globules in the mud. This process has its major effect in the final stage of travel, i.e., the top 500 feet of the borehole. Here, the temperature and pressure are at their lowest and falling fastest. Also, the mud is traveling more slowly and uniformly at the maximum annular diameter. This is especially true offshore where most or all of the 500 feet consists of a large-diameter marine riser subjected to seawater cooling. As shown above, with increased depth, equivalent volumes of gas in place will result in increased volumes of gas to the surface. If hole size and mud flow rate are constant, the increased volume of gas on expansion will disperse through a greater length of borehole annulus. For this reason, although deeper reservoirs may liberate more gas into the mud stream, the net result may not be gas shows of higher magnitude but similar shows of longer duration. Where permeabilities are high, escape and accumulation of fluids in this way is rapid and requires little time. However, in medium to low permeabilities, high annular velocities due to high flow rates or low annular diameters, especially when coupled with high mud viscosities, will reduce the time spent in this critical section and increase the degree of turbulence in it. The result of this will be increased retention of gas and oil in the cuttings, i.e., high cuttings gas analyses relative to mud gas, and greater retention of gas and oil in solution and fine suspension in the drilling fluid, lowering mud gas extraction and hence detection.
3-31
A common source of gas contamination is the degradation of organic-based mud additives, e.g., lignosulphonate dispersant. These will degrade due to the effect of hydrostatic head and with the catalytic support of the clay ion exchanges sites on the mineral matrix. The common product of the degradation is methane, although more complex hydrocarbons may also be present. In order to determine only those hydrocarbons liberated by drilling, it is necessary to establish the amount of hydrocarbons detected from the gas trap when freshly circulated mud is present but drilling is not taking place. Ideally, to do this, it would necessary to circulate for several hours to establish complete consistency in the mud system, ensuring that a stable baseline above which all other readings could be taken. This is not only impractical and prohibitively expensive, but the system background established in this way would remain valid for only a short time, probably less time than is required to establish it.
3-32
3.6.1 Flowline
It is well known that a high degree of degassing takes place in the conductor and flowline. In a good gas show with extensive gas cutting of the mud, gas-produced foam can often be seen bubbling at the bell nipple. Loss of gas from the mud to the atmosphere will also occur extensively in the flowline, especially where (Figure 3.24).
Losses from the Bell Nipple Flowline Not Full Open Flowline Entry Above Mud Level Turbulence in Flowline
The flowline is not filled with mud. 2. Changes in the slope of the flowline cause turbulence. 3. Sections of the flowline are open to the atmosphere. 4. The flowline enters the possum belly above the mud level.
1.
Geometry of the ditch will be of considerable effect in the volume of mud and gas available to the gas trap. Location of the flowline entry, direction of major flow and the degree of turbulence will all affect the efficiency of the gas collection system.
3-33
maintenance and good operation. The trap and its immediate surrounding should be kept clear of cuttings, debris, settled debris or mud caking, all of which may restrict or modify the flow of mud and air through the trap (Figure 3.25).
Motor Malfunction
Weak Vacuum
Damaged Impellor
Mechanical efficiency of the trap is controlled by the rotational speed and surface area of the trap impeller, strength of the sample pump vacuum and the flow rate of air from the trap. Physical condition and submersion depth of the trap and impeller should checked regularly. The surface data logging unit should be equipped to monitor continuously the current drawn by the impeller motor, vacuum and sample flow rate. Even when installation and maintenance of the gas extraction system ensures maximum mechanical efficiency, there will be variations in the overall efficiency of extraction and the magnitude of gas shows. This will depend on the composition of gas present, distribution of gas in the mud, viscosity and gel strength of the mud, and the mud flow rate.
3-34
gas (all combustible fractions). However accurate the analysis, we are still dependent upon the same sample which, as has been seen, bears only indirect and unpredictable relationships with the volume of gas contained in the formation. The value of the gas unit as a quantitative measure is also reduced by limitations of the calibration technique commonly used. The Gas Chromatograph is designed to analyze only for a specific group of compounds (methane through pentane). Calibration of the GC is therefore easily achieved by introducing a mixture of these compounds of known composition into the chromatograph. The Total Hydrocarbon Analyzer, on the other hand, is required to analyze a mixture of unknown composition and constitution. It is commonly the practice to calibrate the THA with a single gas standard (i.e., methane) of known composition. Since only a small volume of gas actually passes into the detection systems, it is essential that this volume be maintained at a uniform rate to avoid fluctuations in the detected values. This is done by drawing a large volume of gas from the gas trap into the logging unit and diverting the excess needed by the detection systems.
3-35
suction pit sample when a drilling break occurs or during regular time intervals during the surface data loggers tour is the best that can be done. The sample must be taken from the suction pit, not the return pit since this pit will contain higher gas readings. It is a good idea to catch two samples each time. If the interval to be evaluated is thick enough that more than one complete circulation is completed before the bottom of the zone is encountered, then another set of background mud samples must be caught. After drilling any oil and/or gas zone, an additional amount of gas is entrained in the mud. Only time and dilution will lower these gas readings.
When catching the sample, the bottle must be submerged, mouth down, to a depth of approximately six to eight inches. Then with the bottle still mouth down in the mud, the top is then screwed on. The sample bottles should be completely full. Keep the samples, top down, at all times until there is time to run the sample through the Steam Still. If not, some of the gas will escape around the top of the sample bottle. The samples can be stored until there is time to run them with very little loss of gas from the bottles if they are handled correctly.
3-36
Boiler
Within the mixing chamber the mud is swept with steam that carries gases and heavier hydrocarbons into the reflux unit. The steam then condenses on the water jacketed surface of the reflux unit. All the condensing water and oil rolls back into the mixing chamber; air, and other gases are swept to the top of the reflux unit and captured there. The inner surface of the condensing jacket is covered with a thin film of condensing water that flows back into the mixing chamber. Droplets of oil will condense and collect on this film of water and also flow back into the mixing chamber. The portion of the reflux unit above this water jacket will remain hot and the walls essentially dry. This distinct separation of condensing water and oil with
3-37
the gases in the top of the reflux unit eliminates the possibility of the gases redissolving in the condensing water and oil. The hydrocarbons that are dissolved in the mud sample, are eluted by the steam and come off in the order of their vapor pressures. Only a small quantity of steam is necessary to strip methane from the mud sample; but, larger quantities are needed to strip heavier hydrocarbons. To strip heavier hydrocarbons the reflux unit condenses large amounts of steam which has passed through the mud sample. After reflux action has been carried out for at least four minutes essentially 100% of the normal butane and any lighter hydrocarbons will be stripped from the mud. During the time that the reflux action is taking place, that is water is flowing through the condensing jacket, all gases will be swept to the top of the reflux unit. When the flow of water in the water jacket is stopped and the refluxing action ceases, gases that are trapped in the top of the reflux unit will disperse throughout the unit as the unit and mixing chamber all come to the same temperature.
Operating Procedure
1. 2. 3.
4. 5. 6. 7. 8. 9.
Turn on the Steam Still and allow it to reach its working pressure (minimum of 10 psi, preferably 15 psi). Start the water flowing though the water jacket. Insert a glass syringe, without the plunger, through the top of the reflux unit. Allow all the excess air to be purged from the reflux unit. You should see water and air violently jetting from the syringe. This will take approximately 1 to 2 minutes. When air leaving the chamber is no longer audible, pull the syringe out. Then insert the syringe with the plunger, when air does not move the plunger, the chamber is properly purged. While the reflux unit is purging, agitate the mud sample, by shaking the bottle or by tapping it on the heal of your hand. Drain the excess water from the mixing chamber. Withdrawn 5 ml of mud from the sample bottle with a 5 ml syringe, ensure that a full 5 ml of mud is withdrawn. Slowly inject the 5 ml mud sample in the injection valve. Set the timer and allow the reflux action to continue for 4 minutes, and do not allow the boiler pressure to drop below 10 psi. Draw into a dry syringe, approximately 1 ml of air. Insert the syringe in the top of the reflux unit. Force the air into the reflux unit and allow the pressure in the reflux unit to return the air and whatever gases are trapped in the top of the unit. The plunger will rise until steam in the unit begins to condense in the needle of the syringe. At this point, if the plunger does not read 5 ml on the syringe, tapped the plunger with your finger until more gas enters the syringe. This may take several taps, but eventually the syringe will have 5 ml of sample. Once 5 ml of sample is in the syringe, grasp the plunger and barrel of the syringe and remove it from the top of the reflux unit.
3-38
Run the syringe under cool water for a minute to completely condense any water. With the needle down, tapped the syringe until all the water is at the bottom, then slowly eject all the water from the syringe. 11. Go to the LS-2000 command terminal and enter the volume factor and depth data (C C 10, 11, 12 and 13). 12. Inject the sample into the chromatograph. Start the chromatograph as soon as the sample is completely injected. 13. Drain the mixing chamber, purge the reflux unit and prepare the next sample.
10.
3-39
gas-in-mud when the well was drilled. A comparison of the plots led to two ideas: the first being that the magnitude of ratios of methane to each of the heavier hydrocarbons is essentially the same and the second is that the slope of the lines of these plotted ratios are nearly the same. The similarity seen in the comparison of the plots of the ppm logs and wellhead gas analysis data show that a simple correlation could be made between the Steam Still Reflux samples and interpretation of the in place formation content. A more detailed study of the correlations of the ratios has indicated the following principles: Productive dry gas zones may show only C1 but abnormally high shows of C1 are usually indicative of saltwater. 2. C1/C2 ratios between 2 and 15 indicate oil.
1.
Values in the range of 2 to 4 indicate low gravity oil. (10-15 on the API gravity comparison scale.) Values in the range 4 to 8 indicate medium gravity oil (15-35 on the API gravity comparison scale.) Values in the range of 8 to 15 indicate high gravity oil (35 + on the API gravity comparison scale.)
c
3.
C1/C2 ratios in the 15 to 65 range indicate gas. a Values in the 15 to 20 range indicates gas rich in heavies, and would most likely be associated with condensate. Values in the 20 to 50 range indicates gas rich in mid-ranged hydrocarbons. Values in the 50 to 65 range indicates gas rich in the lighter fractions.
c
4.
C1/C2 ratios less than 2 and greater than 65 indicate non-producible hydrocarbons. a A hydrocarbon with a value less than 2 indicates an oil with a gravity so low that it cannot be produced. A hydrocarbon with a value greater than 65 indicates a light gas, most C1, but the low permeability nature of the formation makes it non-producible.
b
5.
The slope of the line of all plotted ratios indicates if the zone will be produce hydrocarbons only or hydrocarbons and water. Line slope also provides some indication as to the formation permeability. a An all positive slope, up and to the right, indicates hydrocarbons only.
3-40
If any of the line segments have a negative slope, down and to the right, hydrocarbons and saltwater is indicated. A steep line slope indicates reduced permeability and the steeper the slope the lower the permeability. Steep is a relative term but the dashed lines separating the gas, oil, and non-producible areas of the graph are there for comparison. Oil plots whose line slope is somewhat steeper than the middle dashed line indicates low permeability. Oil plots whose line slope is equal to or less than the middle dashed line indicates good permeability. Gas plots whose line slope is somewhat steeper than the top dashed line indicates low permeability. Gas plots whose line slope is equal to or less than the top dashed line indicates good permeability.
Zone of Interest Report No - label this blank alphabetically if the report is only a zone of interest. a Show Report No. - label this blank numerically if this is a show report.
2. 3. 4. 5. 6. 7. 8. 9.
Depth Interval (MD)/(TVD) - enter the top and bottom depth values of the Zone of Interest or Show, whichever is the case. Well Name - enter the well name or OCS-G number if offshore. Location - enter the offshore block number Operator - enter the name of the operating oil company. Hole Diameter - enter the current bit diameter. Bit Type - enter the current bit type; X the appropriate box. Reported Prepared By - enter the logger's name who logged the Zone of Interest or Show. Report Delivered To - enter the person you delivered the report to. This would be the geologist even if you verbally gave him the report over the phone. Date - abbreviate or spell out the month rather than using a number value. Time - enter the hours and minutes using a 24 hour clock. New Orders - what the rig operation was after detecting the show. Max ROP @ _____ ft - depth where the maximum ROP occurred. Max Gas @ _____ ft - depth where the maximum gas occurred. Max Cl @ _____ ft - depth where the maximum chlorides occurred.
3-41
16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32.
MW in (before, during, after) - suction pit mud weight. MW out (before, during, after) - flowline mud weight, including any cut. Avg. WOB (before, during, after) - average weight-on-bit. Avg. ROP (before, during, after) - average rotary speed. Avg. GPM (before, during, after) - average mud flow rate. Avg. ROP (before, during, after) - average rate of penetration, mark the appropriate box for feet per hour or minutes per foot. Max ROP - the maximum rate of penetration during the interval. Avg Gas (before, during, after) - average gas reading, mark the appropriate box for either units or per cent gas. Max Gas - the maximum gas reading during the interval. Avg Cl (before, during, after) - the average chlorides (use K to indicate thousands of ppm's) Max Cl - the maximum chloride reading Avg. C1 (before, during, after) - the average C1 readings from the chromatograph. Avg C2 (before, during, after) - the average C2 readings from the chromatograph. Avg C3 (before, during, after) - the average C3 readings from the chromatograph. Avg C4 (before, during, after) - the average total C4 reading from the chromatograph. Avg C5 (before, during, after) - the average total C5 readings from the chromatograph. PPM at Max Gas - the parts per million of each gas component at the maximum gas reading. Note: If the gas readings do not contain a component, enter NA for that component, do not leave it blank.
33.
At Max Gas, the visual sample percentages were: ____ % ____ enter the cuttings percentages and types seen in the interval (up to four types and percentages). The reservoir rock was a ____ - enter the reservoir rock of the zone of interest or show. -colored ____ - enter the color or colors of the reservoir rock. The gain (crystal) size was ____ - the size or sizes of the reservoir rock. and the grain shape was ____ - the shape of the grains of the reservoir rock. Approximate visual porosity was ____% - estimate the visual porosity of the reservoir rock. and the visual permeability was ____ - estimate the visual permeability of the reservoir rock. Grain sorting was ____ - enter the sorting of the grains or crystal (i.e., poor, moderate, fair, well).
3-42
44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60.
and the rock cement was ____ - if the reservoir rock was cemented, enter the type of cement holding the fragments together. The porosity type was ____ - enter the type of porosity of the reservoir rock (i.e., intergranular, intraparticle). and the secondary components in the rock fragments were ____ enter only the secondary components in the rock fragments of the reservoir rock, not the secondary rock types in the entire sample. The rock hardness was ____ - the hardness of the reservoir rock. and the sample contamination was ____ any type of contamination of the sample (i.e., metal filings, pipe dope). Depth - enter the depth of any gas-in-air chromatograph readings that were taken during the interval. Gas - enter the mud gas reading for the above depth (mark the appropriate box for units or percent readings). C1 - enter the C1 ppm reading (in thousands) for the above depth. C2 - enter the C2 ppm reading for the above depth. C3 - enter the C3 ppm reading for the above depth. C4 - enter the total C4 ppm reading for the above depth. C5 - enter the total C5 ppm reading for the above depth. Oil Fluor y/n - mark the blank with a yes or no if any oil fluorescence was noted in the sample for the above depth. Cut Fluor y/n - mark the blank with a yes or no if any cut fluorescence was noted in the sample for the above depth. The liquid hydrocarbon was first detected at ____ ft - the top depth that any liquid hydrocarbons were detected. and continued through ____ ft - the bottom depth that any liquid hydrocarbons were detected. The liquid phase of the mud was ____. - enter the base of the drilling fluid (i.e., water, mineral oil). The liquid hydrocarbon occurred in the form of ____ enter the type of liquid hydrocarbon detected (i.e., gas condensate, oil). and was present in the ____. - was the liquid hydrocarbons detected in the mud, washed cuttings and/or unwashed cuttings. When the ____ (was/were) mixed with water and studied in the UV box, enter which of the three above samples were mixed with water and studied in the UV box. the liquid hydrocarbon covered ____% of the surface of the water - enter what percent of the water fluoresced. The oil was ____ in color, - enter the color of the oil as noted in unwashed cuttings mixed with water. exhibited a ____ fluorescence - enter the color of the oil fluorescence as noted in the unwashed cuttings mixed with water. and had an approximate API gravity of ____; - enter the API gravity closest to the color of the oil fluorescence as compared to the API Gravity Scale.
3-43
odor was ____ and staining (was/was not) present - enter any hydrocarbon odor that was noted (i.e., none, very faint, faint, strong). Cross out the appropriate response pertaining to any staining that may have occurred in the samples. 66. The cuttings exhibited a ____ cut - enter the type of solvent cut (i.e., streaming, blooming, crush). 67. that was ____ in color - enter the color of the solvent cut (i.e., clear, straw, light brown, brown). 68. with a ____ fluorescence. - enter the color of the cut (i.e., cream/white, yellow/green).
65.
Note: If no liquid hydrocarbons were detected in the interval, enter NA for each blank. Logger's Opinion of the Show Interval - the logger should enter any other pertinent information concerning the interval at this point. 70. 100% Gas-In-Air = ____ units - enter the gas scale for 100% gas, (i.e., 3000, 5000).
69.
3-44
3-45
8. 9. 10.
11.
Show Report No. - label this blank numerically Depth Interval (MD)/(TVD) - enter the top and bottom depth values of the show report. Well Name - enter the well name or OCS-G number if offshore. Location - enter the offshore block number. Operator - enter the name of the operating oil company. Report Prepared By - enter the logger's name who logged the show. Report Delivered To - enter the person you delivered the report to. This would be the geologist even if you verbally gave him the report over the phone. Date - abbreviate or spell out the month rather than using a number value. Time - enter the hours and minutes using a 24 hour clock. the production of this zone is deemed to be ____. based on your interpretation of all Steam Still hydrocarbon ratio plots of the zone enter either gas, oil, water, non-producible hydrocarbons or any combination. At approximately ____ feet there is - again referring to all the Steam Still hydrocarbon ratio plots in zone, enter the approximate measured depth where the fluid content of the zone changed. If the zone was 100% gas or oil, enter NA in all the blanks. a ____/____ contact - if the zone contains gas and/or, oil and/or water, enter the type of contact. (and a ____/____ contact - treat as above. at approximately ____ feet) - enter the measured depth of the second fluid content change of the reservoir. for a total of ____ feet - enter the total footage of the first fluid type of the reservoir. of ____ show - enter the first fluid type if the reservoir contains more than one fluid type. (and ____ feet - if the reservoir contains more than one fluid type, enter the total footage of the second fluid type. If the reservoir contains only one fluid type, enter NA. of ____ show.) - enter the second fluid type if the reservoir contains more than one fluid. Enter NA if the reservoir contains only one fluid type. DEPTH - enter the measured depth where the gas for this particular plot originated from. GAS - marked the appropriate box for either units or percent and enter the value in this blank. MUD CHLORIDES (ppm) - enter the mud chlorides for the plot measured depth. FLOWLINE ppm - enter the flowline ppm readings for the plot depth. If any components are not present, enter NA.
2001 Sperry-Sun, a Halliburton Company SDL Applied Fundamentals January 2001
3-46
SUCTION ppm - enter the suction pit ppm readings for the plot depth. If any components are not present, enter NA. 24. SHOW ppm - subtract the suction ppm's from the flowline ppm's and enter the results here. 25. HYDROCARBON RATIOS - enter the C1/C2, C1/C3, etc. ratios here. 26. PRODUCTION ANALYSIS - mark the appropriate box for the production type the ratios indicate.
23.
3-47
3-48
Allow the solvent to evaporate; if oil was cut from the sample a fluorescent ring will be left in the dish. A blank solvent spot may be used for comparison. Samples should be checked again for fluorescence after they have dried. Some oils are more easily seen under the UV light in dry samples. Examination of carbonate cuttings requires a somewhat different approach and additional considerations. Foremost is the fact that almost all limestones and dolomites exhibit some kind of mineral fluorescence. For this reason oil determination must depend almost entirely upon a fluorescent cut in a leaching agent rather than on fluorescence of untreated cuttings. The type of porosity usually found in carbonate rock is such that the cuttings themselves may be dense and have only one face that is part of the wall or a crystal-lined fracture that has been in contact with the oil. The cuttings in the sample tray should be studied for new and different types of fluorescence and these pieces picked out and placed in the spot plate. Under the microscope, pick out the cuttings indicating porosity and put them in the spot plate. Place the spot plate in the UV box and while observing, carefully spray chlorothene onto the cuttings. A strong cut should not be expected. In fact, an oil cut consists of no more than a flash or fluorescent leaching only immediately after application of the solvent. Add 10% hydrochloric acid to break down the cuttings and help liberate any oil present. Allow the solvent to evaporate and examine for a fluorescent ring. Do not use solvent from rubber stoppered bottles or medicine droppers as the rubber will add fluorescence to the solvent, making it unfit for oil determination. A hypodermic syringe is best for dispersing the chlorothene, a glass dispensing bottle may be used. Periodically check the solvent for contamination. Rig oils and greases must be examined for their fluorescence. Care must be exercised to determine that the fluorescence or cut is coming from within pieces of sample and not just from the overall surface. This will reduce chances of mistaking rig oil for crude oil. Before the oil in the cuttings evaluation entry is completed on the data sheet, the cuttings should also be examined under the microscope to obtain some idea of the sample porosity, staining and oil source.
3-49
Qualifying factors with which the above results should be tempered would be dispersion of the analyzed sample in the mud column which is in itself dependent on depth, mud characteristics, actual grain size, density, hole conditions and the degree of cementation of the individual particles of the cuttings. In spite of the mechanical factors, the quality of the show and massiveness of the member containing the show are the important reservoir characteristics to be gained from this analysis of the cuttings. Therefore, the evaluation of the show in the cuttings becomes a matter of first evaluating the show within individual cuttings which do exhibit a show and then interpreting that evaluation in the light of the factors affecting the percentage of these cuttings present in the sample. This allows the logger to arrive at the following table for evaluation or rating of the show:
Table 3.1 Quality of Show in Those Cuttings Which Exhibit a Show Percentage of show-containing cuttings in the sample Little or no porosity, little or no staining, weak fluorescence, cuts only after crushing Some visible porosity, fair visible porosity, fair fluorescence, cuts by discoloration Good porosity, good visible, staining, good fluorescence, streaming cut
Few pieces - 25% 25% - 50% 50% - 75% More than 75%
The Table 3.1 is a chart that applies considerations of both quality and quantity as a guide for placing a rating on the above described oil show in the cuttings. It is conceded that specific knowledge of and experience in a particular area should supersede the application of this chart. However, in the absence of such information or any other circumstance surrounding the occurrence of the show, the logger should use the chart to arrive at a final evaluation to place the show of a cuttings sample. Also, he should take into consideration any extraordinary circumstances that temper his rating. Tar or deal oil alone should never be give a value greater than 1/2, regardless of the amount present. A TRACE of live crude oil should never be given a value less than 1. Theoretically, if a 2-foot thick sand is drilled, the samples returning to the surface would contain 100% of this sand until the end of the interval passes over the shaker and then the samples would abruptly change to whatever follows the sand. Therefore, if the sand contained a GOOD SHOW it should be logged accordingly for two feet, though the formation might be barren above and below. In practice, however, zones are never this clearly defined even at a shallow depth. Actually, at a fairly shallow depth, the hypothetical sample above might indicate only 50% sand for the two foot interval with the balance of the sand from this interval
3-50
trailing off showing decreasing percentages for the next several samples. In a deeper well, the sample from this same sand might indicate only 25% of the sand the balance will trail off over an even longer interval. Thus, it is easy to see that evaluations based on percentages alone will lead to an erroneous representation. More correctly, the quality of the show should be logged, letting the depth intervals delineate the thickness or extent. Cuttings oil shows can be described using Table 3.2 below and the following guide:
1. 2. 3. 4. 5.
Percentage stain and distribution Color of the stain Intensity and distribution of fluorescence Color of fluorescence Type of cut
p fr g n vis stn
Very light Light Medium Dark Very dark Faint Dull Bright No fluorescence
Shade
Fluorescence intensity
v lt lt m dk v dk
3-51
Cuts fall into three categories: 1 Streaming cuts These are the quickest to develop and are the ones in which the oil fluorescence can be seen to stream out of the cuttings. imm fst slw lazy Streaming Cut strm cut
These are slower to develop than streaming cuts and are the ones where the solvent becomes discolored without any obvious source points for the producing the discoloration. v slw slw mod fst Milky Hazy mlky hzy
These are the cuts produced by crushing the cuttings, either prior to adding the solvent or in the case of a very slow discoloration cut/no cut after the addition of the solvent. A sample with a good crush cut and a very slow discoloration may be interpreted as having a low permeability. p fr g Crush crsh
3-52
million (ppm) oil-in-mud were detected with UV light; 15 ppm oil-in-mud were easily detected. The mud sample for study must be collected at the point of nearest access to the well head. Usually this point is at the flowline. This is necessary in order that the mud will have a minimum exposure to the atmosphere. The sample must never be taken from the discharge of the trap or after passing through the shaker screens. Fill a pyrex mud sample dish about one third full of mud. Add sufficient water (of no fluorescent impurities) to make the mud thin. Place the sample in the UV box immediately and with the door closed, look for distillate fluorescence which usually disappears almost as soon as it reaches the surface. Stir several seconds with a glass rod while examining the mud sample under the UV light. In stirring, be sure to stir around the edges of the dish near the surface of the mud as some crude oils tend to adhere to the dish. Be sure your eyes become accustomed to the darkness of the viewing box before stirring and examining the sample. This is especially important on bright, sunny days. The mud sample should be allowed to stand until the time for collecting the next sample and another examination of the sample made for any new fluorescence. In many cases oil appears to blossom and spread on the surface of the mud. High gravity crude oil shows usually appear as droplets which seem to pop out of the mud followed by and leaving a flower of fluorescence on the surface of the mud sample. An approximation of the gravity of the oil comprising the show may be made from the color and type of fluorescence. The color may range from white or very light blue for high gravity; dark yellow or almost brown fro that of low gravity crude. The fluorescence of paraffin base crude will usually fall more in the blue range while that of asphaltic crude will exhibit more of the yellow to brown cast. Since rig oil and greases will fluoresce under the UV light, a collection of these oils and greases on a small stick will help to identify fluorescence from these sources. A white light is provided in the UV box to assist in identifying such contamination as well as to assist in the complete examination. Each time the UV box is turned on, it should be checked to ensure that all four tubes are burning. The ultraviolet light in the UV box must be checked periodically and sufficient intensity maintained. Because of contamination by rig oils and grease and by oil recirculated in the mud, it is necessary to examine mud samples at regular intervals, even while coring, in order to locate any new fluorescence which might be caused by crude oil; otherwise, the new fluorescence might go unnoticed because of the contaminating fluorescence from oils and grease. The use of different types of oils in the making of oil emulsion muds, makes it imperative that the mud be examined at regular intervals. Only by doing so can the fluorescence from a new show be distinguished from the normal fluorescence
3-53
of the mud. Arrangements should be made with the mud engineer to notify the logger of any new additions of oil to the mud. Sometimes, cuttings samples are not readily available because of sand reaching the surface being finely divided, lost circulation material, hole conditions or fast drilling rates. In such cases, it is necessary to rely more on mud samples and in order to do so, the logger must have been examining the mud samples regularly. Accurate and logical evaluation of oil shows in the drilling mud is realized mainly by experience. Numerical evaluation of the oil shows in mud is normally a matter of visual comparison of a particular show with the known results of other shows that have been observed. The logger should have previously gained as much information from those already observed within the same or closely related areas. The oil shows are evaluated on the basis of the percentage of the mud covered by oil and tempered by such things as drilling rate, depth, gas readings, lithology, rate of mud circulation and mud properties. In the interest of maintaining as much uniformity as possible between various logging jobs and between individuals, the evaluation below should be used:
Table 3.3 Fluorescence Evaluation % Surface Fluorescence
A TRACE of live crude should never be given a value of less than T. A rating of a GOOD SHOW will be rare, and should only be given to a show accompanied by a good drilling break, increase in gas readings, and a good show of oil in the cuttings. No more than a TRACE should be given a show accompanied by only a small amount of gas increase or no increase al all. These suggestions are not intended as ironclad rules, since practices and greater knowledge of certain areas may dictate variations. They are intended to serve only as a guideline to standardization of the evaluation given to a show throughout all operating areas and as aids to the logger to serve in the gathering of information on remote wildcats where little previous information is known.
3-54
Chapter 4
Well Control
The control of formation pressure during drilling, completion and work over operations. The loss of well control is often a subtle and unnoticed event until it has suddenly become a threat to life, property and environment. Today, wells are being drilled at faster penetration rates, with lower mud weights, deeper, and in deeper water than ever before. Your ability to recognize the warning signs of kicks and potential blowouts, and to assist with the execution of correct and safe well control practices is important. Your life, as well as the success of the operation in which you are involved, may depend upon your knowledge of well control technology.
4-1
used to measure the amount of space in the rock that contains fluid. Sand generally has a greater potential for causing a kick than does a shale because, (1) the volume of rock occupied by fluid is greater than in a shale (porosity), and (2) the ability of the fluid to move through rock (permeability) is higher than it is for a shale. The third factor controlling kick severity is the amount of underbalance. The greater the amount of underbalance, the easier it is for formation fluids to flow from the formation into the wellbore. Kicks are commonly labeled by several different methods. For instance, the same kick may be labeled at various times as a 20-barrel kick (the amount of influx that occurred prior to the time the well was shut-in), a gas kick (the type of fluid influx), or a 1 ppg kick (the increase in mud weight necessary to control the kick).
4-2
Assume the driller wants to pull 5 stands (94 ft/stand) between hole fills. Can he do so safely? Solution: In order to find the answer to this problem, it is necessary to compute the fluid level drop in the hole after he has pulled 5 stands with filling the hole.
1.
Calculate the total fluid displaced by 5 stands of drill pipe. Drill pipe displacement, bbl/ft = (OD2 - ID2) .0009714 = (4.52 - 3.8262) .0009714 = .00545 bbl/ft
The total amount of pipe being pulled is 5 stands. 5 stands 94 ft/stand = 470 ft. So the total pipe length being pulled between fills is 470 feet. The total fluid displaced by 470 feet of drill pipe is: 470 ft .00545 bbl/ft = 2.562 bbl.
2.
How many feet does 2.562 barrels fill? Annular capacity (bbl/ft) = (Hole ID2 - Pipe OD2) .0009714 bbl/ft = (8.52 - 4.52) .0009714 = .0505 bbl/ft
Drill Pipe capacity (bbl/ft) = Pipe ID 2 .0009714 = 3.8622 .0009714 = .0145 bbl/ft Fluid level drop (ft) = 2.562 bbl 0.0144 bbl/ft +0.0505 bbl/ft = 39.5 feet
3.
What is the pressure reduction? The fluid level has dropped 79 feet. Therefore, the hydrostatic head reduction is:
4-3
Compute the hydrostatic head of 9,000 feet of 10.0 ppg mud. HH = .052 10.0 ppg 9,000 ft = 4680 psi
The formation pressure has been previously given as a 9.9 ppg equivalent. Calculate the formation pressure in psi. Formation pressure (psi) = 9.9 ppg .052 9,000 feet = 4633 psi
Since the driller pulled 5 stands without filling the hole, the hydrostatic pressure of the fluid in the hole was reduced by 21 psi. The hydrostatic head of the mud now in the hole is: 4,680 psi - 21 psi = 4,659 psi The difference between the new hydrostatic head (HH of mud with hole not full) and the formation pressure is: 4,659 psi - 4,633 psi = 26 psi Twenty-six psi is the safety factor, working for the driller while pulling 5 stands between hole fills. Each set of conditions requires separate calculations, so this analysis must be carried out on a case-by-case basis, considering each of the different pipe sizes and hole diameters to be used. Drill collars displace considerably more hole fluid than does drill pipe. Therefore, when drill collars (or Hevi-Wate) is being pulled, more frequent hole fills are required than for drill pipe. Assume the hole data being used in this example is the same as that used in the previous example. Assume the driller wants to pull 3 stands of drill collars between hole fills. Can he do so safely? Solution:
1.
Calculate the total fluid displaced by 3 stands (90 ft/stand) of drill collars DC displacement = (OD2 - ID2) .0009714 = (72 - 2.52) .0009714 = .0415 bbl/ft DC displacement
4-4
2.
Compute the feet filled (fluid drop) by 11.2 barrels. Annular capacity (bbl/ft) = (Hole ID 2 - DC OD2) .0009714 = (8.52 - 72) .0009714 = .0026 bbl/ft = ID2 .0009714 = 2.52 .0009714 = .006 (bbl/ft) Fluid displaced (bbl) Annular capacity (bbl/ft) + DC capacity (bbl/ft) = = 11.2 bbl .0285 bbl/ft 392 ft
DC capacity
3.
What is the hydrostatic pressure reduction as a result of the fluid level drop? The fluid level has dropped 293 feet. Hence, the hydrostatic pressure reduction is: 392 ft. .052 10.0 ppg = 204 psi The formation pressure is given as being a 9.9 ppg equivalent. FP = .052 9.9 ppg 9,000 feet = 4,633 psi The hydrostatic head of the mud in the hole is now equal to the hydrostatic head of the mud when the hole was full minus the pressure reduction caused by the pipe displacement. HH = .052 10.0 ppg 9,000 feet = 4,680 psi The hydrostatic head of the mud with the hole not full is: HH - pressure reduction (psi) 4,680 psi - 204 psi = 4,476 psi
4-5
Therefore, the hydrostatic head of mud in the hole after pulling 3 stands of drill collars is 4,476 psi, and our formation pressure is 4,633 psi. Thus, leading to an influx of formation fluids. The lesson to this story is two-fold: Frequent hole fills while pulling drill pipe is a good practice, especially when drilling close to balance. The MMS requires that the hole be filled with fluid before a bottom hole pressure reduction of 75 psi occurs or no more than 5 stands of drill pipe, whichever provides the least reduction. 2. Pipe which displaces a lot of mud due to its wall thickness (such as drill collars or Hevi-Wate) will demand more frequent hole fills, not only to comply with the law, but for personnel and rig safety as well. A good practice is to fill the hole every stand or continuously when pulling large drill collars or Hevi-Wate drill pipe.
1.
The MMS requires that the hole be filled with accurately measured volumes of mud. This implies that two things must be done: (1) the pipe displacement must be calculated, and (2) a method must be devised to accurately measure the mud volume used to fill the hole. The two most common methods for doing this are trip tanks and pump stroke counters. A trip tank is a small tank with an accurate calibration scale which can be used to measure the amount of mud entering and leaving the tank. On some trip tanks, the calibration scale is accurate enough to read mud volumes to the nearest tenth of a barrel. The trip tank may be placed at a height greater than that of the flowline to allow it to fill the hole by gravity feed. If the trip tank is not placed at the flowline height or higher, a centrifugal pump may be attached to the trip tank so that the pump can pump mud continuously through a fillup line. Another line is placed on the BOP stack at a height slightly greater than that of the fillup line. The trip tank pumps mud into the annulus continuously and any overflow is returned to the trip tank via the return line (Figure 4.1). The volume required to fill the hole can be easily seen by using the calibration scale and recording the drop in the fluid level of the trip tank as pipe is removed from the well. The second method commonly used is to periodically fill the well with mud using the rig pump and to count the number of strokes the pump runs while filling the hole. The number of strokes that will be required to fill the hole should be predetermined before a trip is begun by using the following formula: Strokes to fill = Pipe displacement volume (bbl) pump output (bbl/stk) Example: Suppose the driller wishes to pull 5 stands of drill pipe between hole fills. The pipe displacement volume for the 5 stands is 3 barrels, and
4-6
the rig pump output is .2 bbl/stk. How many pump strokes are required to fill the hole? Strokes to fill = .3 bbl .2 bbl/stk = 15 strokes
Figure 4.1
Trip Tank
If the hole fails to take the calculated amount of mud (either by strokes or by level decrease in a trip tank), formation fluids are entering the well and corrective action should be taken immediately.
Swab Pressure
Swab pressure is pressure created by pulling the drill string from the hole. Swab pressure acts like a negative hydrostatic pressure, thus causing reduced bottom hole pressures (BHP). If the BHP becomes less than the formation pressure, an influx of formation fluids can occur. There are a number of factors upon which swab pressures depend. Some of these are:
1.
4-7
Mud flow properties, especially the yield point and gel strength. 3. Hole geometry. 4. Balled-up equipment (bits, collars, stabilizers.
2.
Hole swabbing is very easy to recognize. First a hook load indicator may show more drill string weight than what the string actually weighs. Although drag may depend on any number of factors, it is an immediate tip-off that potential hole swabbing can be taking place. Hole swabbing can be detected by watching the fluid level as pipe is pulled from the hole. If the fluid level does not fall enough or if the fluid seems to be following the pipe as the pipe is pulled, swabbing is taking place. Again some remedial action must be taken immediately. Swab pressures can become severe enough at times so that a BHP reduction of several hundred psi may occur. This is a very common phenomenon in gumbo shale and happens because some of the drilling assembly has become caked with gumbo. As the pipe is removed from the hole, the balling that has taken place on the drilling assembly acts like a piston, which impedes the flow of mud required to fill the void created as the pipe was pulled. It is common practice to pull up into the casing, put the rig in high-high gear and gel that pipe on the floor in a hurry. Swab pressure is exerted at every point in the open hole below the bit. Pipe speed should only be increased if sufficient swab pressure versus pipe speed data is available with which to determine maximum safe pipe speeds. Remember bottom hole pressure is critical, if swab pressure reduces it below formation pressure, a kick is eminent.
Lost Circulation
Lost circulation may be a cause of kicks. If lost returns occur, the fluid level in the hole begins to fall. The length of the fluid column in the hole is decreased, thus causing the hydrostatic pressure of the mud in the hole to decrease. If the hydrostatic pressure of the mud decreases to the point where it becomes less than the formation pressure, a potential kick situation exists. If the lost circulation problem goes undetected, a large amount of fluid influx can occur in the wellbore. It is therefore a good practice to attempt to keep the annulus filled with measured volumes of water.
4-8
Most gas expansion takes place near the surface of the well. If a large diameter hole is being drilled at a high rate of penetration (as in a 171/2-inch surface hole offshore), the amount of gas cut up by the bit itself can be considerable. Expansion of the gas occurs as the gas nears the surface, so the hydrostatic head of the mud is reduced. If the hydrostatic head is reduced to a value lower than the formation pressure, a kick can occur. However, the hydrostatic head reduction caused on bottom by gas cut mud is generally not a significant value. The surface indication of gas cut mud often cause more concern than the situation warrants. The main concern when dealing with gas cut mud should be to make certain that the surface facilities are adequate to keep cut mud from being pumped back down the hole. An operational degasser is a necessity when dealing with gas cut mud. There are two major types of degassers. One is the atmospheric type, often called a gas separator or gas buster. Its construction is such that it can be constructed at the rig site and has a large mud handling capacity. It can handle the volumes of mud normally pumped in circulating the well or the volumes pumped during a well-kill operation. The atmospheric degasser is usually the first treatment which gas cut mud receives at the surface. In the atmospheric degasser, mud falls over baffles which tend to break the gas bubbles out of the mud. The gas is vented to the atmosphere through a 6-inch or larger gas outlet. The second common type of degasser is the vacuum type. Mud enters one side of the degasser and is pulled up into a chamber in which a partial vacuum exists. The gas is forced out of the mud by gravity segregation in the chamber under a partial vacuum. The mud flows back into the pits and the gas from the mud is vented to the atmosphere. The vacuum degasser is more effective than the atmospheric degasser, but its construction complexity and maintenance requirements are higher than those for the atmospheric degasser.
Water and Oil Cut Mud. Water and oil cut mud can cause a kick. The mud weight reduction at the surface caused by an influx or other additions of water and oil considerably decreases hydrostatic pressure in the wellbore. A .5 ppg cut at the surface implies that all the mud in the borehole is cut by at least .5 ppg. A .5 ppg cut on a 10,000-foot well reduces the bottom hole pressure by 260 psi.
4-9
abnormal formation pressures exist. Insufficient mud density and abnormal formation pressures often go hand-in-hand with one another. There are a number of techniques available to aid in the detection of abnormal pressure. Among those which may provide prior information that abnormal pressure exists are:
1. 2. 3. 4. 5. 6. 7.
Paleontology Offset well logs and analysis of offset well histories Temperature changes Gas readings Formation resistivity Cutting appearance Hole conditions
Each of the above may be an indication that abnormal pressure exists, but there are also a number of techniques available which can be used to determine closely what the magnitude of the abnormal pressure actually is. Among these are: d exponent 2. EWR resistivity plots 3. Shale Density 4. Shale Factor
1.
The theory of detection of abnormal pressure will be dealt with in a separate section. Since abnormal pressures and insufficient mud weights often go hand-in-hand, it would seem that a solution to this problem would be to drill with high mud weights. However, high mud weights may exceed the fracture gradient, causing stuck pipe, and lower the rate of penetration. The best practice is probably to maintain enough mud weight to keep a slight overbalance and set additional pipe when the mud weight nears the fracture gradient in the weakest part of the hole.
4-10
4-11
Drilling Breaks
It was stated previously that a rock must have sufficient permeability and porosity in order for a kick to occur. Increasing porosity will result in drilling rate increases because more of the rock itself is occupied by fluid and less space is occupied by the actual rock matrix. An abrupt change in the rate of penetration usually signals a formation change (as in going from a shale into a sand). The sand has a greater kick potential than a shale does, so it warrants stopping the pumps 3 feet to 5 feet into the sand and checking the well for flow. A gradual increase in the rate of penetration is not a drilling break, but it is an abnormal pressure detection device and may warrant flow checks if it persists.
4-12
decreases in the gas level at the time that gas is due at the surface. Connection and trip gases must be evaluated by the amount of change from previous readings rather than by the readings themselves. If the trip or connection gases keep increasing with each trip or connection, there is a good chance that the formation pressure is increasing. However, increased connection or trip gases may also be caused by deteriorating mud flow properties or balled up equipment. These two factors must be considered before evaluating connection gases or trip gases. Background gas is that gas which is encountered while actually drilling the well. If a gas bearing formation is being drilled, gas will breakout of the rock cuttings as it nears the surface. High gas readings do not automatically mean that an increase in mud weight is required, but simply that a gas bearing formation has been drilled. Of course, gas may enter the wellbore from the formation as a result of underbalance. In this case, the amount of gas entering the mud may not be sufficient to cause a kick or blowout because the volume of gas entering the mud stream is low (low rock permeability). However, if a zone of higher permeability is later drilled, a kick may occur. As a result, it often becomes necessary to determine whether the gas has been caused by core volume cutting or bleed-in from the formation. A common way of determining whether gas in the mud is caused by core volume cutting or formation gas influx is to circulate the cuttings from the wellbore. If the gas readings fall back to a low value (they will not generally reach 0 even if the hole is clean), the gas is probably a result of core volume cutting. If relatively high gas readings persist, the gas may be a result of bleed-in from the formation and an increase in mud weight may be necessary to correct this condition. Gas cutting increases are a better warning sign of impending abnormal pressure than they are of an impending kick. Caution must be used in evaluating gas readings, however, because of the variety of sources which may cause gas readings to change.
Water-Cut Mud
Water may enter the returning mud stream by the same mechanisms by which gas enters the mud stream. If the density of the formation water is less than the density of the drilling mud, a reduction in flowline mud weight may occur. Water does not expand as it travels up the annulus. As a result, any mud weight cuts caused by core volume cutting are generally insignificant because the ratio of the amount of water in a given formation compared to the volume of mud into which the water can disperse is quite small. An influx of formation water into the return mud stream can be detected by three methods which are (1) mud weight cut, (2) change in mud salinity, and (3) changes in mud viscosity. In the Gulf Coast area, most formation water is saltwater. The salinity of the formation saltwater is generally higher than that of the water in the drilling mud.
4-13
When this is true, formation water bleed-in can be detected by running a chloride ion test on the mud filtrate. If the chlorides keep increasing, the chances are good that formation fluid is probably bleeding into the return mud stream as a result of underbalanced conditions. The influx of saltwater in appreciable quantities will also cause viscosity increases in a freshwater mud system which uses bentonitic gel as a viscosifier. It is probably a good practice to check the chlorides in the mud anytime a sudden increase in viscosity is noted unless the exact cause of the mud thickening is known. If the mud in use at a particular time has a higher salinity than the formation water, the chlorides will decease. Freshwater mud cutting can occur in the same way as saltwater cutting, but an influx of freshwater due to formation pressure will cause a weight reduction, chloride decrease, and thinning of a freshwater gel mud system. Freshwater cutting of a mud is not a common occurrence on the Gulf Coast.
Conclusions
Flow rate increases, pit level increases, positive flow checks, decreases in pump pressure or increases in pump strokes, and improper hole fillups on trips are positive primary indicators which will generally manifest themselves at the time the well begins to kick. A drilling break may also be a good kick warning indicator and usually warrants a flow check. However, string weight changes, chloride changes, gas changes, and flowline mud weight changes generally take time to detect (usually the time to pump bottoms up). In most cases, the well would have kicked prior to any of these indications being manifested at the
4-14
surface. Cut mud is a good indicator of developing abnormal pressure, but is only of secondary importance as a kick warning signal.
A timely sequence of operations must be followed to achieve the main well control objectives. One of the most critical times in a well control operation is when the well is initially shut-in. After the well has been shut-in, the proper procedure for controlling the well must be used. A proper kill procedure is safer for the operation and personnel because it minimizes the stresses on both the wellbore and the surface equipment. This section will be comprised of three parts. The first part will deal with proper shut-in procedures for both surface and subsurface blowout preventer stacks. The second part will deal with U-tubes and basic well control theory, and the third part will deal with proper kill procedures using both surface and subsurface blowout prevention equipment.
If mud continues to flow from the wellbore or if doubt exists as to the results of the flow rate check, the well should be shut-in immediately and check for shut-in pressure.
4-15
Shut-in procedures are classified as either hard or soft. Each has primary objectives different from the other, and each is correctly applied in situations different from the other.
Hard Shut-In
The primary objective of a hard shut-in is to limit influx volume to a minimum. When a hard shut-in is to be performed, the adjustable choke is closed before taking a kick; and the wellbore is shut-in upon kick recognition by closing a blowout preventer. Influx is limited in volume to that which can flow into the wellbore during the time that the blowout preventer is closing. A hard shut-in does have disadvantages. Erosion of blowout preventer elements can lead to the immediate or premature failure of the blowout preventer. The minimized surface pressure and wellbore stress can exceed what the wellbore can withstand causing an underground or surface blowout.
Soft Shut-In
The primary objective of a soft shut-in is to control surface pressure during shut-in and to minimize erosion of blowout preventer elements. When a soft shut-in is to be performed, the adjustable choke is open before taking a kick; the wellbore is shut-in upon taking a kick by first allowing flow from the casing through the choke line; second, closing a blowout preventer; and third, closing the adjustable choke. Surface pressure is monitored and controlled if necessary while closing the adjustable choke. Erosion of blowout preventer elements is minimized by diverting flow through the choke line while the blowout preventer is closing. The soft shut-in does have disadvantages. Subsequent well control operations, such as circulating out the influx, are complicated by increased influx volume and surface pressure. Shut-in of the well may not be immediately possible. A dynamic kill procedure may be required. After a kick has been shut-in, and the shut-in pressure stabilized, (assuming an underground blowout has not occurred), no further formation fluid will flow into the wellbore; formation pressure is balanced with a combination of hydrostatic and shut- in surface pressure.
4-16
Figure 4.2
A correct shut-in procedure will safely control surface pressure and wellbore stress while minimizing influx volume. Four different shut-in procedures will be discussed; each involves what to do when a kick occurs and one of the following conditions exist. A hard shut-in will be assumed for each.
3. 4. 5. 6.
When a kick is detected or suspected, raise the kelly immediately until the uppermost tool joint in the drill string is above the rotary table. Shut down the rig pumps. Leaving the rig pumps on while the kelly is being raised helps to maintain extra mud pressure due to ECD and minimizes the possibility of additional swabbing. (Remember that the rate of fluid influx into the wellbore is a function of the amount of underbalance, and the permeability and porosity of the kicking formation. The circulating pressure helps to lessen the amount of underbalance.) Check for flow. Close the annular preventer. Open the HCR valve. Notify the operator personnel.
4-17
7.
Read and record the shut-in pressures and the amount of pit gain.
Note: Attempting to make up the valve when it is already closed can cause the installation of the valve to be difficult or impossible.
3. 4. 5. 6. 7. 8. 9.
Close the drill pipe safety valve. Close the annular preventer. Open the HCR valve. Notify the appropriate operator personnel. Pick up the kelly and make it up on the drill string. Open the safety valve. Read and record the pressures. Note the amount of pit gain.
Special consideration is given to placement of tool joints and drill string movement for floating vessels and subsurface blowout preventer stacks. To ensure tools joints are clear of closing blowout preventers, the drill string is picked up to a predetermined, space off, height before closing the blowout preventer. The correct space off height is determined from the pipe tally. Uncontrolled vertical movement in a closed blowout preventer can separate the drill string or damage the blowout preventer stack. Movement of the drill string is normally minimized with a motion compensator, but when the motion compensators are ineffective, the drill string is hung off on a pipe ram in what is called a hang off procedure. These considerations are included in the two following shut-in procedures.
4-18
3. 4. 5. 6. 7.
Check the well for flow. Close the annular preventer. Open the fail safe valve. Notify the operator personnel. Hang-off procedure: a Reduce the hydraulic pressure on the preventer until a slight preventer leak can be observed at the flowline. This will help save the rubber when the drill pipe is stripped through the preventer. Slack the kelly down and close the bottom kelly cock. Set the slips just below the uppermost tool joint. Break the kelly off above the kelly cock valve. Rig up pump-in lines and reopen the bottom kelly cock. Pick up the pipe to the space off height. Close the top pipe rams and lock in position. Slack off until the string weight is held by the pipe rams. This should occur when the tool joint reaches the closed ram element.
b c d e f g h
8.
Read and record the SIDPP, SICP, and the amount of pit gain.
It may not be necessary to hang off the drill string on the rams every time, but the procedure may be followed if a move off location is anticipated, the motion compensator is not working properly, or the weather is bad.
When a kick is detected, immediately set the slips just below the top tool joint. Make up a full opening, fully opened safety valve on the drill string. Close the safety valve. Close the annular preventer. Open the fail safe valve. Notify the operator personnel. Make the hang-off decision. Pick up the kelly and make the kelly up on the drill string. Open the safety valve. Read and record pressures and the amount of pit gain. The sequence of steps may again be altered to suit the operator preference. The hang-off procedure would probably not be followed if stripping back to bottom was anticipated.
4-19
Figure 4.3
Assume both sides are filled with a 10.0 ppg mud, and that the system is closed at the top on both sides. The bottom hole pressure (BHP) is equal to the sum of the hydrostatic pressure exerted by the mud (5200 psi at 10,000 feet) plus any pressures imposed at the surface (0). Both sides of this system exert the same hydrostatic pressure; both sides have 0 psi imposed surface pressure, so the system is balanced and is in equilibrium. The objective in well control is to balance the pressures on each side of the system so that the BHP is held equal to or greater than the formation pressure (Figure 4.3).
4-20
Assume the system is closed at the top on both sides; 6,000 psi of pressure is exerted at the bottom. The hydrostatic pressure of the mud is only equal to 5,200 psi. The 6,000 psi being exerted by the formation is enough to light the 10.0 ppg mud and blow it out of the U-tube, if the U-tube is left open. What must be done, then, is to balance the formation pressure by imposing pressures at the surface. Since the U-tube is filled by equal lengths of 10.0 ppg mud on both sides, the imposition of equal surface pressures will be necessary to keep the U-tube itself balanced (Figure 4.4).
Figure 4.4
Expressed mathematically, we want the bottom hole pressure to equal the formation pressure. BHP = HP + Surface Pressure Surface Pressure = BHP - HP Surface Pressure + (6000 - 5200) psi Surface Pressure = 800 psi The necessary surface pressure is 800 psi.
4-21
Figure 4.5
Unbalanced U-tube
Assume that the drill pipe side is filled with 10.0 ppg mud and annulus side is filled with 9.0 ppg mud. We wish to balance the BHP on both sides of the U-tube. Assume the U-tube is again a closed system. BHP9.0 ppg = .052 9.0 ppg 10,000 feet = 4680 psi = 5200 psi
In order to balance the U-tube, we need to impose surface pressure on the side which has the least amount of hydrostatic mud pressure. In this case, the annulus side has the lighter mud, so surface pressure must be imposed on that side of the U-tube. We need to impose enough surface pressure on the annulus side to bring the BHP on the annulus side up to the BHP on the drill pipe side. We want: HP(10.0 ppg) Drill pipe side = HP(9.0 ppg) + Surface Imposed Pressure Annulus Side
4-22
HP(9.0 ppg) The surface imposed pressure needs to be 520 psi on the annulus side to make the U-tube balance. BHP = 5200 psi HP(10.0 ppg) = 4680 psi HP(9.0 ppg) + 520 psi Surface Imposed pressure
Figure 4.6
The U-tube is closed on both sides. The drill pipe is filled with 10.0 ppg mud Total length of both the drill pipe side and annulus side = 10,000 ft. The formation pressure is 6,000 psi. The annulus is filled with 7,000 ft of 10.0 ppg mud and 3000 ft of gas. (Disregard the gradient of the gas.)
We want to compute the SIDPP and SICP which will be required to balance the U-tube with formation and which will balance both sides of the U-tube itself. We want to balance the BHP and FP on the drill pipe side, *BHP = FP = SIDPP + HP10.0 ppg @ 10000
4-23
and also to balance the BHP and FP on the annulus side, **BHP = FP = SICP + HP10.0 ppg @ 7000 and to balance the U-tube itself, SIDPP + HP10.0 ppg @ 10000 the HP10.0 ppg @ 10000 and HP10.0 ppg @ 7000 = SICP + HP10.0 ppg @ 7000 = 5,200 psi = 3,640 psi
FP = 6000 psi. We want the BHP = FP, so from * above, 6000 psi = SIDPP + HP10.0 ppg @ 10,000
if the SIDPP = 800 psi, the BHP = FP on the drill pipe side and from ** above, 6000 psi 6000 psi = SICP + HP10.0 ppg @ 7000 = SICP + 3,640 psi
The SICP required to balance the FP and BHP on the annulus side is 2,360 psi. With 800 psi imposed at the surface on the drill pipe side and 2,360 psi imposed at the surface on the annulus side, the U-tube is balanced with respect to the formation and both sides are in balance relative to each other.
4-24
Figure 4.7
The circulating pressure was 2,500 psi before the gas influx and with the well open. In this example assume the following conditions are known to exist:
1. 2. 3. 4. 5. 6.
The system is being circulated and back pressure is being imposed at the surface by varying the choke opening size. The total system pressure loss at the pump being used is 2,500 psi with well open. The annulus pressure loss at the pump rate in use is 100 psi. The drill pipe is filled with 10,000 feet of 10.0 ppg mud, and the annulus is filled with 3000 ft. of gas and 7,000 feet of 10.0 ppg mud. The formation pressure is 5,700 psi. Disregard the gradient of the gas.
We want to find out what drill pipe pressure reading is required and what casing pressure reading is required in order to balance both sides of the U-tube relative to each other and to equalize the BHP on both sides of the U-tube with the formation pressure. In other words, we want: BHPdrill pipe side = BHPannulus side = FP
4-25
The total system pressure loss was measured as being 2500 psi prior to the kick. Of this, 2500 psi, 2400 psi was lost as a result of mud friction in the surface equipment, drill pipe, and the bit jets; 100 psi was lost in the annulus. The annulus pressure loss exerts and additional 100 psi on bottom as a result of circulating friction in the annulus. The pressure losses in the surface equipment, drill pipe, and bit jets do not result in additional pressure being exerted on bottom. The annulus pressure loss plus the choke pressure loss plus the hydrostatic pressure of the mud and gas on the annulus side are equal to the BHP exerted on the annulus side. BHPannulus side Loss HPmud HPgas = HPmud + HPgas + Choke Pressure + Annulus Pressure = 10.0 ppg .052 7000' = 3460 psi = 0 psi
Annulus Pressure Loss (APL) = 100 psi BHP needs to be equal to FP, so 5700 psi 5700 psi 5700 psi = FP + BHPannulus = 3640 psi + 0 psi + Choke Pressure + 100 psi = 3740 psi + Choke Pressure
A choke pressure of 1,960 psi will balance the annulus side with the formation pressure. We now need to compute the drill pipe pressure which satisfies the specified balances. The system pressure loss was measured as being 2,500 psi with the annulus side open prior to the kick. Under static conditions, FP 5,700 psi SIDPP = SIDPP + HP10.0ppg @ 10,000 + BHP = 5,200 psi + SIDPP = 500 psi
500 psi must be exerted at the surface on the drill pipe side before any mud movement can take place. We already know that an additional 2,500 psi of pressure will be required because of friction to pump the mud at our desired pump rate. Therefore, a pump pressure reading of 3,000 psi will be indicated when the pump is brought up to the necessary circulating rate.
4-26
In a closed U-tube system, a balance between the two sides of the U is automatically established. The surface indicated pressures will automatically be those necessary to maintain balance in the closed system. The imposition of additional pressure anywhere in a closed, balance U-tube system will result in that additional imposed pressure being felt equally at all points in the U-tube system. Assume for example that we have a closed and balanced U-tube system with 100 psi pressure gauge readings on both sides. If we pressure up the annulus side to make it read 200 psi at the gauge, the drill pipe gauge will also read 200 psi and the BHP will rise by 100 psi. The 100 psi increase will also be felt at all other points in the closed system. The drill pipe side of the U-tube is normally uncontaminated with formation fluid, so the hydrostatic pressure exerted by the mud in the drill pipe is effective over the total vertical depth and this pressure is known. The system pressure loss is known by prior measurement. The SIDPP is known also. This means that the BHP on the drill pipe side is known. If we pump at a constant rate and restrict the choke openings correctly, we can calculate the BHP on the drill pipe side. Since the system is balanced, we can hold the bottom hole pressure on the drill pipe side equal to the formation pressure. The annulus side will also be balanced relative to the formation pressure since the drill pipe side and annulus side are balanced relative to each other. Thus, we can hold the bottom hole pressure constant by manipulating the drill pipe pressure. Given the system pressure loss, hydrostatic pressure of the mud in the drill pipe, and the SIDPP, we can compute the formation pressure and the necessary drill pipe pressure at all times. The annulus side of the U is contaminated with an unknown weight and volume of formation influx. Due to this, the casing pressure cannot be used to compute BHP; but any change in casing pressure from a balanced condition will result in a like change in drill pipe pressure. The system is balanced and closed, so any change in the imposed pressure at the choke will be felt equally at all points in the system. Thus, it is possible to hold the bottom hole pressure constant at the desired level. This is the underlying principle of well control.
4-27
The pumping rates at which system pressure losses are taken are normally from 1 to 3 barrels per minute. The system pressure loss is generally not taken at the normal circulating rate for a variety of reasons. Among these reasons are: The system pressure loss at normal circulating rate plus and shut-in pressures may be excessive for the pump and surface conditions in use. 2. Any change in the choke opening size at normal pump rates may cause drastic pressure loss changes across the choke, which makes choke control difficult. 3. The mud being displaced in the annulus during a kill operation needs to be weighted before it is pumped into the well. It may also be contaminated and need treatment. Most rigs do not have the mud handling facilities to treat and weight the mud rapidly enough to keep up at normal pumping rates. 4. Faster pump rates result in higher annulus pressure losses. High annulus pressure losses increase the possibility of lost circulation.
1.
The pump rate at which the system pressure loss is recorded for purposes of well control is called the slow pump rate, reduced pump rate, slow pump pressure, kill rate, or reduced circulating pressure. All these terms may be used interchangeably. A slow pump rate of 1 to 2 barrels per minute is considered optimum in many cases because the system pressure losses are relatively low at this rate, the mud can be weighted and treated, and the mechanical stresses on the pumps are not too great. Slower rates are often impractical because the pump motors may not have enough power to drive the pump with low RPMs. If the rig pump cannot be run at the desired output and pressure, an efficiency reduction can be effected by pulling various discharge and suction valves on the pump. The manufacturers recommendations should be followed concerning which valves to pull to obtain a desired reduction in pump efficiency. A high pressure - low volume pump (e.g., a cement pump) can also be used if the desired pump rate cannot be reached by the rig pumps. If the kill rate or pressure is not known before the kick, a kill rate pressure can be obtained using the following procedure. Shut in the well and record the SIDPP and the SICP. 2. Hold the SICP constant with the choke and bring the pump up to the desired pumping rate. 3. Note the circulating pressure obtained after the pump has been brought up to the kill rate speed. 4. The kill rate pressure is equal to the observed circulating pressure minus the SIDPP.
1.
4-28
There are instances in which a float valve is installed in the drill string. A float valve is a simple one way valve that prevents the movement of fluids and pressures up the drill pipe. When a float valve is present in the string it naturally prevents the kick pressure from being recorded on the drill pipe. In the cases, the equivalent shut-in drill pipe pressure must be found another way. The SIDPP can be found by either of two methods if a float valve is in use. One of the suggested methods involves the case in which the kill rate is known prior to
4-29
taking the kick. The second suggested method involves a procedure which can be followed if the system pressure loss was not known prior to taking the kick. Case 1: Obtaining a shut-in drill pipe pressure if the kill rate pressure is known, float valve in use. Shut-in the well, record the shut-in casing pressure and obtain the prerecorded kill rate from the drillers log. 2. Hold the casing pressure constant with the choke and bring the pumps up to the prerecorded kill speed. 3. Note the circulating pressure obtained with the pump at the kill rate. 4. Shut down the pump and close the choke. The circulating pressure obtained with the pump at the kill rate minus the prerecorded circulating pressure at the same pump rate is the shut-in drill pipe pressure.
1.
Case 2: Obtaining a shut-in drill pipe pressure if the kill rate is unknown - float valve in use. Shut in the well and line up a low volume high pressure pump on the standpipe. 2. Start the pump and fill up all the lines with mud. Any air left in the lines will cause false pressure readings. 3. Increase the pressure on the pump. Note the pressure obtained when the fluid first begins to move. Fluid is incompressible, so no movement can take place until the pressure exerted on the bottom side of the float valve is overcome. Note: The fluid has moved when a small volume of mud has been pumped without a change in pressure. 4. The pressure obtained when the fluid first begins to move is the shut-in drill pipe pressure.
1.
4-30
The influx gradient valve can be divided by.052 to obtain the density of the influx in ppg. As a general rule, an influx with an equivalent mud weight of 1 to 3 ppg is assumed to be gas, 3 to 5 ppg is assumed to be a mixture of gas and water or gas and oil, and 5 to 7 ppg is assumed to be either oil, water, or an oil-water mixture. Example: Assume the following wellbore conditions: TVD Mud weight in drill pipe Hole ID Drill Pipe OD SIDPP SICP Pit gain Determine the nature of the influx. Solution: Mud gradient = 12.0 ppg .052 = .624 psi/ft 2. Annular volume (bbl/ft) = (9.8752 - 52) .0009714 = .0700 bbl/ft 3. Length of influx (ft) = bbl gained (bbl) annular volume (bbl/ft)
1.
= = = = = = =
10,000 feet 12.0 ppg 9.875 inches 5 inches 520 psi 650 psi 40 bbl
=
4.
= 571 ft 40 bbl .0700 bbl/ft = = = = = mud gradient - (SICP - SIDPP) length of influx .624 psi/ft - (650 psi - 520 psi) 571 ft .624 psi/ft - 130 psi 571 ft 624 psi/ft -.228 psi/ft .396 psi/ft
Change the influx gradient to an equivalent mud weight .396 psi/ft .052 = 7.61 ppg. 6. An flux with a 7.61-ppg mud weight equivalent would probably be oil or water. The comparatively high weight means that not much gas is associated with the influx.
5.
4-31
= Kill weight mud (weight necessary to balance the well) = True Vertical Depth of the well = Shut-in drill pipe pressure = Original mud weight
4-32
and the mud weight is raised while the kick is being circulated out. The use of this method may require several circulations before the well is fully killed.
Final Circulating Pressure = Kill Rate Pressure KWM OMW where: KWM OMW = Kill weight mud = Original mud weight
4-33
The final circulating pressure is higher than the kill rate pressure because the higher mud weight used in the kill operation has a greater system pressure loss than the lower weight mud. The factor KWM/OMW is a way of compensating for this difference in system pressure loss. Plot the amount of mud pumped (either barrels or strokes) along the horizontal axis. 5. Plot the drill pipe pressure along the vertical axis. The drill pipe pressure will fall at a constant rate as the drill pipe is displaced with the kill weight mud, at which point the drill pipe pressure will be held constant at the final circulating pressure. Therefore, the plot can be completed by plotting only two points. Plot initial circulating pressure at 0 strokes. Plot the final circulating pressure at the strokes required to displace the drill pipe with kill weight mud (Figure 4.8).
4.
Connect the two points with a straight line. The straight line indicates the pressure which should be held on the drill pipe at all times while the kill weight mud is being pumped down the drill pipe. The choke opening is varied as required to keep the drill pipe pressure regulated.
Figure 4.8
Fill in the drill pipe pressure schedule: Assume SIDPP Kill rate pressure Original mud weight = = = 500 psi 1000 psi 10.0 ppg
4-34
Initial Circulating Pressure = Kill rate pressure + SIDPP = 1,000 psi + 500 psi = 1,500 psi (at 0 strokes)
Final circulating pressure = Kill rate pressure KWM OMW = 1000 psi 11.0 ppg 10.0 ppg = 1100 psi (after 500 strokes) Plot 1,500 psi at 0 strokes and plot 1,100 psi at 500 strokes. Connect the two points with a straight line. When the wait and weight method of well control is being used, the following steps should be followed:
1. 2. 3.
4.
5. 6.
7.
8.
Shut-in the well using the preferred shut-in sequence. Record the SIDPP, SICP, and the amount of pit gain. Check for trapped pressure. Compute the mud weight necessary to kill the well and weigh up the surface mud system to this weight. Fill out the kill sheet while the mud is being weighted up. Hold the casing pressure constant by controlling the choke and bring the pump up to the prerecorded kill rate. Once the pump has been brought up to the kill rate, be certain that the drill pipe pressure is equal to the sum of the SIDPP and the prerecorded circulating pressure at the kill rate. Follow the drill pipe pressure schedule while the light mud in the drill pipe is being displaced with kill weight mud. Once the drill pipe has been filled with heavy mud, the well may be shut-in. The SIDPP should be 0 at this point. This step is optional and is simply a check to make certain that the kill weight mud calculation is correct. If Step 6 has been followed, hold the casing pressure constant by controlling the choke and bring the pump up to the kill rate once more. Be certain the drill pipe pressure is equal to the final circulating pressure. Hold the drill pipe pressure constant (while maintaining the pump rate at the kill rate) at the final circulating pressure until the kill weight mud reaches the surface.
4-35
9.
When the kill weight mud reaches the surface, shut down the pump and shut-in the well. All shut-in pressures should be zero. Open the choke and check for flow. Then open the preventer stack, close the choke, and re-check to see that the well is dead.
Some people prefer a mathematical method of doing drill pipe pressure schedules. The reason is that a table listing the drill pipe pressure values versus amount of mud pumped is easier for some people to read than the graph. Consider the data used in the previous drill pipe pressure schedule: SIDPP Kill rate pressure Original mud weight Kill weight mud Stroke to displace drill pipe = = = = = 500 psi 1000 psi 10.0 ppg 11.0 ppg 500 strokes
The initial circulating pressure is 1,500 psi. The final circulating pressure is 1,100 psi. We know that the drill pipe pressure should fall at a constant rate while kill weight mud is being pumped down the drill pipe. In this case it falls from 1,500 psi to 1,100 psi in 500 strokes. Therefore, the drill pipe pressure decrease per stroke is: DP pressure decrease per stroke = Initial circulating pressure - Final circulating pressure Strokes required to displace DP with KWM = 1500 psi - 1100 psi 500 strokes = .8 psi/stroke A .8-psi/stroke decrease is impossible to read with present rig instrumentation, but if we construct a table of drill pipe pressure values, we can use the drill pipe pressure decrease per 100 strokes pumped which is: .8 psi/stroke 100 strokes = 80 psi per 100 strokes The table is constructed as follows:
Table 4.1 Drill Pipe Pressure Values Strokes Pumped Drill Pipe Pressure
0 100 200
1500 psi 1500 psi - 80 psi = 1420 psi 1420 psi - 80 psi = 1340 psi
4-36
Table 4.1
1340 psi - 80 psi = 1250 psi 1250 psi - 80 psi = 1180 psi 1180 psi - 80 psi = 1100 psi
The table states that the drill pipe pressure should be 1,420 psi after pumping 100 strokes, 1,340 psi after pumping 200 strokes, etc. This method eliminates any errors which may occur as a result of reading the drill pipe pressure graph incorrectly.
6. 7. 8.
9.
Shut-in the well using the preferred shut-in procedure. Record the SIDPP, SICP, and the amount of pit gain. Be sure to check for trapped pressure. Compute the mud weight necessary to kill the well. Hole the choke pressure constant by controlling the choke and bring the pump up to the kill rate. Hole the drill pipe pressure constant at the initial circulating pressure (SIDPP + kill rate pressure) by controlling the choke and pump at the kill rate until the influx is out of the hole. After the influx is out of the hole, shut-in the well and raise the mud weight in the pits to the kill weight. Hold the casing pressure constant and bring the pump up to the kill rate. While pumping at the kill rate, hold the casing pressure constant by controlling the choke and displace the mud in the drill pipe with kill weight mud. Once the drill pipe has been displaced with kill weight mud, observe the final circulating pressure on the drill pipe pressure gauge.
4-37
Keep pumping at the kill rate and hold the drill pipe pressure constant at the observed final circulating pressure by controlling the choke until the kill weight mud reaches the surface. 11. When the kill weight mud reaches the surface, shut down the pump, close the choke, and verify that the SIDPP and SICP are both zero. If so, then open the choke to be sure the well does not flow. If no flow is observed, open the preventers and check again to be sure the well is completely dead.
10.
The drill pipe pressure graph may be eliminated when the drillers method is used. Since the kill weight mud is not circulated until after the influx is out of the hole, the choke can be used to hold the drill pipe pressure constant at the initial circulating pressure (kill rate pressure + SIDPP). This action is enough to ensure a constant bottom hole pressure equal to the formation pressure. Once the influx has been removed from the wellbore, the annulus is no longer contaminated, so the SICP is now the same as the SIDPP. Holding the casing pressure at a constant value equal to the original SIDPP now assures a constant bottom hole pressure equal to the formation pressure. The casing pressure is held constant until kill weight mud reaches the bit. Once kill weight mud reaches the annulus, the hydrostatic forces in the annulus begin to change, but the hydrostatic forces in the drill pipe are no longer subject to change.
4-38
It has been decided to kill the well using the Concurrent Method. Since it is necessary to begin killing the well as soon as possible, ad decision has been made to increase the surface mud weights in .2 ppg increments. Fill out a drill pipe pressure schedule to be used during the kill operation. Solution: Kill weight mud = 520 psi 19.23 + 10.0 ppg = 11.0 ppg 10,000 feet Initial circulating pressure = 1,000 psi + 520 psi = 1520 psi Final circulating pressure = 1,000 psi 11.0 ppg = 110 psi w/KWM 10.0 ppg The mud density needs to be increased by 1.0 ppg to kill the well. Since the density is going to be increased by .2 ppg at a time, 5 drill pipe mud displacements will be made. The initial and final circulating pressures are exactly the same as those which would be used in the Wait and Weight Method. This method is different only in that it takes 5 times as long to kill the well, but the principles involved are exactly the same as those used in the Wait and Weight Method. To fill out a drill pipe pressure schedule the following procedure may be followed: Compute the time necessary to pump from the surface to the bit at the kill rate, the kill weight mud, the initial circulating pressure and the final circulating pressure. 2. Decide the density increments in which the mud weight will be raised (.1 ppg per circulation, .2 ppg per circulation, etc.) 3. Calculate the number of drill pipe displacements which will be necessary to raise the mud weight and calculate the total time or number of strokes involved. In this case, 5 displacements of the mud in the drill pipe are necessary and 30 minutes are required for each displacement at 30 strokes/min. 4. The mud weights being pumped down the drill pipe and the strokes (or times) involved are plotted along the horizontal axis. The drill pipe pressures are plotted along the vertical axis as follows (Figure 4.9):
1.
4-39
2000
Figure 4.9
A procedure which can be followed when the Concurrent Method is being used is as follows:
1. 2. 3. 4.
5. 6.
7.
8.
Shut-in the well using the preferred shut-in procedure. Allow the SIDPP and SICP to stabilize. Check for trapped pressures. Record the SIDPP, SICP, and the amount of pit gain. Fill out the kill sheet. Weight up the pits quickly to the desired weight. While holding the choke pressure constant, bring the pump speed up to the kill rate. The initial circulating pressure should be equal to the sum of the SIDPP and the kill rate circulating pressure. Pump at the kill rate and follow the drill pipe pressure schedule until the kill weight mud reaches the bit. When the kill weight mud reaches the bit, the pump may be shut down, the well shut-in, and the shut-in pressures checked. The SIDPP should be equal to zero. Restart the pump and maintain its speed at the kill rate. Hold the drill pipe pressure constant at the final circulating pressure by partially/gradually opening or closing the choke until the kill weight mud reaches the surface. When the kill weight mud reaches the surface, stop by partially/gradually opening or closing the choke, close the choke, and check to see that the SIDPP and SICP are both equal to zero. If they are, open the choke and check for flow. The preventer stack may be opened and the well again checked for flow.
4-40
4-41
Figure 4.10 Pit Level Changes that Result from Heave and Roll of a Floating Drilling Vessel - No Change in the Amount of Mud in the Pits
A single float mounted in the pit is insufficient to monitor the actual pit volume. However, a pit level monitor can be mounted at both ends of the pit. If the two (or more) floats in each pit are tied together in such a way that the total volume is read out on a recorder, the pit volume can be monitored accurately. This principle is illustrated in Figure 4.11.
4-42
Figure 4.11
Compensating for Pit Level Changes that Result from Heave and Roll on a Floating Rig - Constant Volume of Mud in the Pits
In the illustration above, a rise in mud level on one side of the pit results in a decrease in mud level on the other side of the pit. Since the mud volume remains constant and both pit level monitors are tied together, the total volume can be read as a constant because the rise in level of one pit level monitor is accompanied by an equal decrease in level by the other pit monitor. The sum of their readings is indicated on the recorder. Pit volume totalizers have been developed which can accommodate eight or more pit floats, sum their individual readings and display the total volume and total change in volume on a recorder. In order to monitor the return flow rate with any accuracy, a sensor must be placed in a position where level changes do not take place. The vessel motion at the surface does not affect the rate at which mud passes a point on the ocean floor. Electronic sensors have been developed which are able to monitor return mud flow at or near the ocean floor. This approach allows the accurate measurement of flow rate from the well. Once a kick has been detected on a floating drilling rig, the well needs to be shut-in. The hang-off procedure discussed earlier may be necessary to prevent damage to the drill string or a stationary preventer. If the weather is severe, the motion compensator will not prevent all drill string motion. The weight of the string may be placed on the rams and flexible pump-in lines can be rigged up. Hanging-off will result in the drill string remaining motionless relative to the preventer elements, and the flexible pump-in lines will allow vessel motion without placing tension or compression forces on the drill string, which would
4-43
result if a kelly or other fixed-position pump-in line was used. Hanging-off may not be necessary or desirable in every instance, but it should be used if any of the following conditions develop or are anticipated: Bad weather 2. Motion compensation failure 3. Moving off location
1.
Two chokes lines and additional preventers are generally included to serve as backups in case of component failure on a subsea stack. Two annular preventers, several sets of rams (pipe rams, blind rams, and shear rams) are included. The bottom set of pipe rams is generally placed below the choke lines. This allows the well to be closed in if repairs need to be made to any BOP components or choke lines. The shear rams are also generally located beneath all the choke lines so that the drill string can be severed and the rig moved off quickly if an impending blowout develops. Choke lines generally run from the ocean floor to the surface when a subsea preventer stack is in use. These choke lines have relatively small inner diameters, and in deepwater applications where long choke lines are in use, considerable pressure loss resulting from mud flow through the choke line can develop. The pressure loss in a choke line results in additional bottom hole pressure. If the bottom hole pressure is to be held constant at a known value, the amount of pressure loss in the choke line must be known. The slow pump rate should be taken through both the riser (preventers open, choke closed) and through the choke line with a preventer closed. If both readings are taken at the same pump rate and mud properties, the difference in the two readings will be equal to the pressure loss in the choke line. Example: After making a connection, the driller obtained a reduced circulating pressure through the riser (normal drilling conditions) of 800 psi at 30 stk/min. He then closed the upper annular preventer, opened the choke and choke lines, and brought the pump up to 30 stk/min. The pressure obtained through the choke line was 1,00 psi at 30 stk/min: The pressure loss in the choke line was: 1,100 psi Circulating pressure obtained through choke line 800 psi Circulating pressure obtained through riser = 300 psi Circulating loss through choke line
The 300 psi pressure loss in the choke line results in an additional 300 psi of bottom hole pressure. In fact, an additional 300 psi is applied to all points in the wellbore that are below the base of the choke line. At shallower depths, the
4-44
300 psi can cause considerable increases in the equivalent mud weights seen by the formation. The reduced circulating pressure should be obtained at the slowest possible pump rate. If a gas kick occurs, the gas will begin to expand as it is pumped from the wellbore. As the bubble expands, the amount of mud returning from the well will be greater than the amount of mud being pumped into the well. This increased return flow rate results in an increase in choke line pressure. (Pressure loss through a conduit is dependent upon the flow properties and velocity of the fluid traveling through that conduit.) If the choke line pressure loss is of concern prior to the time a kick occurs, the pump rate should be slowed down enough to hold this pressure loss to a minimum. It may be necessary to obtain reduced circulating pressures with a cement pump or other high pressure low volume pump in order to meet this requirement. The reduced circulating pressure which was obtained through the riser is the pressure which should be used in making the kill sheet calculations. The initial circulating pressure is equal to the sum of the reduced circulating pressure obtained through the riser and the shut-in drill pipe pressure. The final circulating pressure is obtained by multiplying the ratio of the kill weight mud to original mud weight by the reduced circulating pressure obtained through the riser. The other procedures used in killing a well from a floating drilling vessel are very similar to those used in killing a well which employs a surface-mounted preventer stack. The main consideration, therefore, is obtaining a reduced circulating pressure. Failure to obtain the reduced circulating pressure will necessitate calculating the pressure loss through the choke line which, as with most pressure loss calculations, is an approximation. The approximation that results from calculating the pressure loss through a choke line can result in significant errors in kill sheet calculations. Knowledge in advance of the choke line pressure loss is simpler and eliminates errors. The choke line pressure loss may be approximated by using the following formula:
0.000077 p Q ( PV ) L Pressure loss through choke line = ----------------------------------------------------------------4.8 (D )
0.8 1.8 2
where:
p q
PV = plastic viscosity (centipoise) of the mud D = inner diameter of the choke line L Example: = length of the choke line
4-45
We need to determine the choke line pressure loss, give the following conditions: Choke line length = 2000 feet
Choke line inner diameter = 2.5 inches Pump rate PV Mud weight = 100 gpm = 35 centipoise (cp) = 12.8 ppg
0.8 1.8 0.2
0.000077 [ 12.80 ( 100 ) ( 35 ) ] 2000 Pressure loss through choke line = -----------------------------------------------------------------------------------------------------4.8 ( 2.5 )
= 118 psi The calculated pressure loss through the choke line is 118 psi. If a reduced circulating pressure is not known, it may be determined by using the following procedure:
1. 2. 3. 4.
5.
6.
7.
Decide what the kill rate is to be (either in barrels/min or strokes/min). Calculate the pressure loss through the choke line at the desired pump rate. Subtract the calculated choke line pressure loss from the shut-in casing pressure. As the pump is brought up to speed gradually, reduce the casing pressure until it is reduced by an amount equal to the answer obtained in Step 3 and hold it at this value until the pump is brought up to speed. Once the pump is at the desired kill speed and the casing pressure is at the desired value, record the standpipe pressure reading. This reading is equal to the initial circulating pressure. The initial circulating pressure minus the shut-in drill pipe pressure corresponds to the reduced circulating pressure, which is used in making the final circulating pressure calculation. Using the answers obtained in steps 5 and 6, the kill sheet can be completed.
Example: We need to obtain a reduced circulating pressure. The well has been shut-in and following conditions exist: SIDPP SICP Choke line length Choke line ID PV = = = = = 500 psi 800 psi 2000 feet 21/2 inches 35 cp
4-46
= =
The operator has decided to use 100 gal/min as the kill rate. The choke line pressure loss has been calculated as 118 psi. The operator brings the pump up to 100 gal/min, and he allows the casing pressure to decrease by 118 psi as he brings the pump rate up. The new casing pressure is 800 psi - 188 psi = 682 psi (700 on gauge). He notes the standpipe pressure is 1,000 psi with the pump displacing 100 gal/min, and the casing pressure is 682 psi. The initial circulating pressure (which corresponds to what the system pressure through the riser would have been at 100 gal/min) is equal to the initial circulating pressure minus the shut-in drill pipe pressure. In this case, the reduced circulating pressure is 1,000 psi - 500 psi = 500 psi. As a result we have: SIDPP Initial circulating pressure Reduced circulating pressure = = = 500 psi 1000 psi 500 psi
The pressure loss through the choke line is often called the hidden choke effect. It must be taken into account each time a well using a subsea stack is killed. A sequence which can be followed during well control operations using subsea blowout preventers can now be developed. A recommended sequence of events is as follows:
1. 2. 3. 4.
5.
Upon detection of a kick, pick up the kelly, shut down the pumps, and shut-in the well. Read and record the SIDPP, SICP, and the amount of pit gain. Notify the operator personnel. Fill out the kill sheet if the reduced circulating pressure and choke line pressure drops are known. If a reduced circulating pressure was not previously obtained, calculate the choke line pressure loss by formula or by hydraulic slide rule. Bring the pump up to the desired kill rate and allow the casing pressure to fall below its initial shut-in value by an amount equal to the calculated choke line pressure loss. (Controlling the choke in such a manner that allows the casing pressure to fall at the proper rate as the pump is brought up to speed is awkward. It may be more simple to open the choke first, bring the pump up to speed and then gradually close the choke opening until the casing pressure reads the desired value.) Observe the initial circulating pressure and then shut down the pump and close in the well once again. Fill out the kill sheet and begin weighting up the surface mud system if the Wait and weight or concurrent method is to be used. If the drillers method is to be used, the kill operation can begin as soon as the necessary calculations are completed. Bring the pump up to speed. Regulate the casing pressure as the pump is brought up to speed. Once the pump is at the desired kill rate, fine tune the choke opening until the drill pipe pressure gauge indicates the proper value.
4-47
Kill the well by following the appropriate pressure schedule and kill procedure. 7. Once kill weight mud reaches the surface, shut down the pump, close the choke, and verify that the well is dead. Open the choke and check for flow through the choke line. 8. Since the preventers are located on the ocean floor, the riser is still full of original mud. This light mud must be displaced prior to opening the preventer stack. To do this, close the bottom set of pipe rams and circulate kill weight mud down the choke line. Continue this reverse circulation procedure until the riser has been displaced with kill weight mud. The rams can then be opened without the danger of allowing additional influx to enter the wellbore.
6.
The fracture gradients are low in deepwater drilling operations, because the fracture gradient of a formation is directly related to the overburden stresses applied to it. A larger overburden generally implies a higher fracture gradient (this is the same as saying that fracture gradients increase with increasing depth, which they generally do). Water exerts less stress on an underlying formation than the same amount of overlying rock would exert on that same formation. For example, a formation at 10,000 feet relative to the kelly bushing in 2,000 feet of water may have only 8,900 psi of overburden earth acting upon it. This formation would have a fracture gradient similar to an 8,900-foot formation on land (remember overburden stress is 1.0 psi/ft). In 1,000 feet of water, a formation at 4,000 feet (relative to the kelly bushing) will have a .3-to .5-ppg lower fracture gradient than a formation at 4,000 feet on land. This can make a considerable difference in what happens during a well control operation. Lost returns which occur during a well control operation present severe danger to a floating drilling vessel. The formation may allow the gas to be expelled to the sea directly underneath the drilling vessel. This will result in aerated water.
4-48
Aerated water is incapable of sustaining a floating vessel, and the rig may sink if the aeration is severe and uniform at all points directly beneath the hull of the vessel. There has been some discussion that water aeration will only occur beneath certain portions of a drilling vessel at any given time. If this is indeed true, the vessel would probably not sink, but it could dip at a precarious angle in the area where water aeration is taking place due to the decreased buoyancy in that particular area.
4-49
borehole stresses encountered are lower than those which would be expected in a gas kick. Example: Hole diameter Drill pipe OD Drill pipe ID Total depth Mud weight in use Pump output Kill rate pressure Pit gain SIDPP SICP Assume the following wellsite conditions exist: = = = = = = = = = = 81/2 inches 41/2 inches 3.826 inches 15000 feet 15.0 ppg .15 bbl/stk 1000 psi at 20 stk/min 25 bbl 780 psi 1100 psi
Surface stack is in use. For purposes of simplicity, a constant hole and pipe diameter are assumed. The following calculations have been made: Drill pipe capacity Kelly to bit Capacity of annulus Strokes, bit to surface Total strokes required to displace mud in the well Barrels required to displace mud in the well Kill weight mud Initial circulating pressure Length of influx Influx gradient = = = = = = = = = = 213 bbl 1420 stks 758 bbl (0.0505 bbl/ft) 5053 stks 6473 stks 971 bbl 16.0 ppg 1780 psi 25 bbl .0505 bbl/ft = 495 ft
A .13-psi/ft gradient for the wellbore fluid influx means that the kick is gas.
4-50
Hydrostatic pressure exerted by the gas = .13 psi/ft 495 feet = 64 psi Formation pressure = 15.0 ppg .052 15,000 feet + 780 psi = 11,700 psi + 780 psi = 12,480 psi Using the data above, the casing pressure expected while killing this kick were calculated. The casing pressure versus barrels of mud pumped were calculated for each of the three previously discussed well control methods. The Wait and Weight Method assumes that the kill will be made in one circulation. The drillers method would require two circulations to complete the kill. The concurrent method assumes five displacements of the mud in the drill pipe plus two to three displacements in the annulus to complete the kill. The concurrent method does not require any set number of circulations to complete a kill operation. The number of circulation required and the rate of mud weight increase depend upon rig capabilities and operator preference. In addition, the table includes a listing of the casing pressures which occur when the well is overkilled with 17.0 ppg mud (Table 4.2).
Table 4.2 Casing Pressures Expected (psi) Wait and Weight 16.0 ppg Kill Mud Drillers Method 16.0 ppg Kill Mud Concurrent Method 16.0 ppg Kill Mud Wait and Weight 17.0 ppg Kill Mud
0 100 213 413 513 680 758 858 971 1071 1184 1284 1384 1584 1684 1729 1942
1110 1131 1152 1197 1253 1678 780 780 780 678 561 458 355 150 46 0 0
1110 1131 1152 1163 1199 1619 660 639 624 624 624 603 582 562 535 468 312
4-51
Table 4.2
Casing Pressures Expected (psi) Wait and Weight 16.0 ppg Kill Mud Drillers Method 16.0 ppg Kill Mud Concurrent Method 16.0 ppg Kill Mud Wait and Weight 17.0 ppg Kill Mud
2913 3884
0 0
0 0
156 0
0 0
4-52
Of the three methods used, the wait and weight method using the proper kill weight mud results in the least amount of casing pressure and the least borehole stresses. The casing pressures are lower using the wait and weight method with 16.0 ppg kill weight mud than they are if the drillers method and 16.0 ppg kill weight mud are used. The reason is that the kill weight mud reaches the annulus while the influx is still in the hole, and this heavier kill weight mud helps to exert additional hydrostatic pressure on bottom. The additional hydrostatic pressure exerted by the kill weight mud means that the choke pressure (surface imposed pressure) can be held at a lower value and still maintain the bottom hole pressure at the desired value. Using the concurrent method allows lower casing pressure values than the drillers method. This is because some of the heavier mud reaches the annulus before the gas bubble is out of the hole. Since this heavier mud (15.2 ppg) is less than the kill weight mud (16.0 pg), the casing pressures are higher than those encountered when using the wait and weight method. This brings up an additional interesting point. Since heavier mud in the annulus seems to allow the use of lower casing pressures, why not kill the well with a greater-than-necessary mud weight? Most engineers agree that overkilling a well (using a greater-than-necessary mud weight to kill the well) has no tangible benefit, and may, in fact, be detrimental to the kill operation. The reason is that overkilling a well causes a higher bottom hole pressure to be exerted on the drill pipe side due to the extra hydrostatic pressure exerted by the mud in the drill pipe. The bottom hole pressure is held constant on both the drill pipe and annulus sides of the U. The two bottom hole pressures are also held equal to one another during the well kill. Since the bottom hole pressure exerted by the mud in the drill pipe is higher than necessary, higher choke pressures must be held on the casing side in order to keep the system in balance. In fact, overkilling a well can cause a considerable increase in borehole stresses. Excessive safety factors should not be included in the kill mud. Only enough safety factors should be included to overcome any errors in pressure gauge readings on the rig. If a trip margin is desired, it can be added while the mud is conditioned following a kill operation. Overkilling a well is the same as having a slug in the drill pipe. When the drill pipe is slugged, mud flows from the drill pipe into the annulus. The mud level falls in the drill pipe until it is low enough for the bottom hole pressure to be the same as the annulus side bottom hole pressure. In the case of overkill, the use of the choke precludes any hydrostatic pressure decrease in the drill pipe. Therefore, more choke pressure must be used to keep the bottom hole pressure on both sides equal. The conclusions reached in this section are as follows: The wait and weight method, using only the mud weight necessary to balance the formation pressure, allows lower casing pressures to be imposed. 2. Overkilling a well has no tangible benefit and may be detrimental. 3. If a trip margin is to be added, it should be added after the well is dead.
1.
4-53
Using the same well and casing pressure data, a table of equivalent mud weights at various depths was compiled.
Table 4.3 Equivalent Mud Weights at Various Depths vs. Barrels of Mud Pumped - Wait and Weight Method 16.0 ppg Kill Mud - Gas Kick 1000 ft 5000 ft 12,000 ft 15,000ft
It should be evident that the choke pressure imposed at the surface has a pronounced effect on the equivalent mud weights at various depths. The equivalent mud weights at the shallower well depths are particularly affected by the choke pressure. In order to kill any kicks safely, the equivalent mud weights at any depth in an open hole must not be allowed to exceed the fracture gradients at that particular depth. A 5000-foot string of casing with a fracture gradient at the shoe of 17.0 ppg would not have been sufficient to allow the well to be killed, because the equivalent mud weight at 5,000 feet reached 19.4 ppg. The formation would have broken down at initial shut-in producing a down-hole blowout. A casing string must be set at a depth which has a high enough fracture gradient to exceed the maximum anticipated mud weight at that depth. Often, field personnel assume that the difference between the existing mud weight and fracture gradient in weakest exposed part of the well is equal to the amount of kick which can be tolerated. This is not true, particularly with only surface pipe set in the well. Example: Mud weight Last casing set = 10.0 ppg
Fracture gradient at the weakest point (3,500 feet) = Total depth Question: =
4-54
Can we tolerate a 3.0 ppg kick? Solution: A 3.0-ppg increase in pore pressure at 10,000 feet corresponds to a shut-in drill pipe pressure of 1,560 psi. The 1,560 psi is felt equally at all points in the system, because the system is a closed, balanced U. Therefore, the pressure felt by the formation at 3,500 feet is equal to the hydrostatic pressure exerted by the mud at 3,500 feet plus the 1,560 psi of shut-in pressure. Pressure at 3500 ft = (10.0 ppg .052 3500) ft + 1560 psi = 3380 psi 3380 psi at 3500 feet is equivalent to a .966-psi/ft pressure gradient or an 18.6-ppg equivalent mud weight. Obviously, we cannot stand a 3.0-ppg kick even though the present mud weight is 3.0 ppg less than the fracture gradient. Question: If we are unable to tolerate a 3.0 ppg kick, what size kick are we able to tolerate? Solution: The fracture gradient at 3,500 feet is 13.0 ppg. The mud weight in use is 10.0 ppg. Therefore, the increase in pressure which can be tolerated is the difference between the hydrostatic pressure exerted by a 10.0 ppg mud at 3,500 feet and the hydrostatic pressure exerted by a 13.0 ppg mud at 3,500 feet. (.052 13.0 ppg 3500 feet) - (.052 10.0 ppg 3500 feet) = 2366 psi - 1820 psi = 546 psi The 546 psi is the maximum casing pressure which can be withstood at 3500 feet with 10.0 ppg mud in the well. An increase in mud density will result in a lower casing pressure tolerance. If no influx occurs (i.e., if the kick is caught instantly and the well is shut-in instantly) the shut-in casing pressure and the shut-in drill pipe pressure are numerically equal. A 546-psi SIDPP is the maximum SIDPP that can be tolerated with 10.0 ppg mud in the well. The size of the kick which can be tolerated prior to circulation is: 546 psi 19.23 TVD
4-55
where TVD = true vertical depth of the well 546 psi 19.23 = 1.05 ppg 10,000 feet A 1.05 ppg kick is the maximum kick which can be shut-in safely under the given well conditions. Circulation adds limits as described in the following section.
Weakest point in the well (last casing shoe) has a fracture gradient of 15.0 ppg at 7,000 feet. Question: What is the kick tolerance factor and its equivalent ppg?
1.
Equivalent fracture gradient at 7,000 feet is 15.0 ppg. Psi required to break the formation down is: Fracture pressure = 15.0 ppg .052 7,000 feet = 5,460 psi
2.
4-56
Subtract the HP from the fracture pressure to obtain the maximum allowable pressure (kick tolerance factor). Kick tolerance factor = 5460 psi - 4,732 psi = 728 psi
4.
Convert 728 psi kick tolerance factor into equivalent ppg. 728 psi .052 12,000 feet = 1.17 ppg
A 1.17- ppg kick is the maximum size kick which can be handled. If any influx has occurred in the annulus, this size kick cannot be handled without breaking the formation down.
4-57
Drill pipe ID Total depth Mud weight in use Pump output Kill rate pressure Pit gain SIDPP SICP
= 3.826 inches = = = = = = = 15,000 feet 15.0 ppg .15 bbl/stk 1,000 psi at 20 stks/min 25 bbl 780 psi 960 psi
Surface stack in use. For purposes of simplicity, the hole and pipe diameters are assumed to be constant. The following calculation were made: Drill pipe capacity Capacity of the annulus Total mud in well Kill weight mud Length of influx Influx gradient = = = = = = = = = 213 bbl 758 bbl 971 bbl 16.0 ppg 495 feet Mud gradient - (SICP - SIDPP) Length of influx .78 psi/ft - (960 psi - 780 psi) 495 feet .78 psi/ft - .36 psi/ft .42 psi/ft influx gradient
The influx gradient indicates that the kick is probably a mixture of saltwater and oil. Therefore, no influx fluid expansion should occur. The following table of expected casing pressures versus amount of kill mud pump was constructed using the data above. The calculations were based on using the Wait and Weight Method.
Table 4.4 16.0 ppg Kill Weight Mud Casing Pressures Expected with Oil/Saltwater Kick Barrels of Mud Pumped Casing Pressure Expected (psi)
4-58
Table 4.4
16.0 ppg Kill Weight Mud Casing Pressures Expected with Oil/Saltwater Kick Barrels of Mud Pumped Casing Pressure Expected (psi)
The well data is exactly the same as that mud in Table 4.2 (Wait and Weight column) with the exception of changing the type of wellbore fluid influx. Note: No expansion occurs and that the casing pressures are constant at the initial shut-in value until the kill weight mud reaches the annulus. Once kill weight mud reaches the annulus, the casing pressure begins a continuous decline which lasts until the well is dead (kill weight mud at the surface). Using the same well data used to compile Table 4.4 and Table 4.5, which gives the equivalent mud weights at various depths in the well versus the amount of mud pumped was compiled.
Table 4.5
Equivalent Mud Weights at Various Depths versus Barrels of Mud Pumped - Wait and Weight Method - 16.0 ppg Kill Weight Mud - Combination Oil and Saltwater Kick 1000 ft 5000 ft 12,000 ft 15,000 ft
Note: The equivalent mud weight is constant at the required level at 15,000 feet. A comparison of Table 4.3 with Table 4.5 shows lesser borehole stresses at most points in the well. In almost all cases, an oil or saltwater kick will result in lower stresses on the wellbore than would occur with a gas kick.
4-59
Lost returns Plugged drill pipe or plugged bit Washed out or plugged choke Hole in the drill pipe Excessive casing pressure Drill pipe off bottom or out of the hole Gas bubble migration and expansion
If a problem does arise during the kill operation, knowledge of all the alternative solutions to that problem will help the operation go more smoothly and will help to realize the stated objectives of the well control operation.
The detection of lost returns may depend upon a combination of the above given factors. Oftentimes, especially if returns are only partially lost, the detection of lost returns may be difficult.
4-60
In order to make any decision regarding remedies to a lost returns problem, the location of the zone which is taking mud must be known. Gulf Coast operators generally assume the loss zone is at or near the last casing shoe. However, this may not always be true. A number of methods of loss zone detection exist. A radioactive tracer may be pumped down the drill pipe, through the bit, and up the annulus. It should enter the loss zone when it reaches that zone in the annulus. The loss zone can then be detected by wireline radioactivity detection devices. The disadvantage to this method is that the wellbore may be contaminated with radioactivity, which makes pinpointing the exact location of the loss zone difficult. Temperature surveys may also be run on a wireline and used to locate a loss zone. The theory behind a temperature survey is that fluid from deeper, hotter zones is entering a shallower, cooler zone. As the wireline probe is lowered, the temperature increases at a fairly steady rate until the loss zone is encountered. As the loss zone is encountered the temperature increases sharply.
If lost returns occur, fluid flow above the loss zone will normally cease. In an underground blowout situation, the fluid flow is taking place between the kicking zone and the loss zone. This fluid flow naturally creates some noise. A noise log which is lowered on a wireline will indicate very little noise above the loss zone. Once the probe is opposite the loss zone, the noise intensity increases. A disadvantage to this type of survey is the sensitivity of the noise probe. Any extraneous surface noise can cause severe distortion in the noise log. If no pipe is in the wellbore, a loss zone may be detected mechanically by running a spinner survey. The spinner will detect fluid motion mechanically.
4-61
Killing a Well with Lost Circulation Problems by Curing the Lost Circulation First
Once a loss zone location has been pinpointed, the operator is faced with a decision of whether to control the well first and worry about the lost returns later, or whether it will be necessary to cure the lost returns problem prior to killing the well. If an operator elects to cure the lost circulation problem first, he has a number of alternatives available to him. Among the alternatives are: lost circulation material, squeezes, chemical or cement squeezes, and time. Once the bottom of the influx is above a section of the wellbore, the stresses imposed against that section decrease. Note: This is true only for the wait and weight method of kick control. If the returns loss is partial, use of a light hand on the choke and continuing ahead as planned with the well control procedure may be sufficient to effect a cure for the problem. This type of cure will only be effective in areas where the rocks exhibit some plasticity. Fractures in many of the types of formation on the Gulf Coast can be healed with the simple application of time and the relief of stress. In hard rock areas the formations are not generally as plastic as Gulf Coast formations. Time and the relief of stress against a loss zone will probably not be sufficient to cure a lost circulation problem in this type of formation. In hard rock country, lost circulation material may be useful in combatting a lost circulation problem. These types of rocks generally require plugging of the fracture or isolation of the loss zone before returns can be regained. Lost circulation material in sufficient quantities may be effective in plugging the fracture. Pills consisting of lost circulation material can be pumped down the drill string and spotted in the loss zone, or they can simply be bullheaded down the annulus and forced into the lost circulation zone. The amount of lost circulation material used to form a pill varies from about 5 ppb to about 60 ppb, depending upon the characteristics of the loss zone and lost circulation material. Another method of curing a lost circulation problem prior to killing a well is to spot a high solids squeeze or a gunk squeeze in the loss zone. When a mud which exhibits a high filtration rate characteristics is subjected to pressure, liquid will be squeezed from the drilling fluid providing there is a filter medium of sufficient permeability to allow the passage of the liquid phase of the drilling mud. The loss zone, in this case, provides the permeable medium for liquid to be squeezed out of the mud, a filter cake is deposited on the permeable medium itself. This filter cake deposit consists of the solids which were originally associated with the liquid in the mud. When high filtration rates are in evidence, a great deal of liquid is squeezed out of a mud, and a large amount of solids are deposited in the form of a filter cake (Figure 4.14 and Figure 4.15).
4-62
In Figure 4.15 (after filtration) the filter cake is deposited on the face of the sand. The sand presents enough of a permeable medium to allow the passage of any liquid squeezed from the mud as a result of differential pressure (Pm - Pf0). A cure for lost circulation which uses a high filtration rate requires, spotting this pill opposite the loss zone, and applying differential pressure to effect the filtrate loss. The solids which are deposited as a result of filtration plug the fractures in the loss zone and enable normal circulation to be restored. Pressure should be
4-63
allowed to build up and should be held on the loss zone for several hours. If squeeze pressure does not build up, this technique will not be successful. The volume of the pill normally varies from 50 to 200 barrels, depending upon the amount of uncased hole in the well and the extent of fracturing in the loss zone.
Choose the desired slurry weight. Decide on the volume of slurry to be used. 500 barrels is a good guideline to use for a plug length of 150 feet to 200 feet in the hole. Calculate the barrels of slurry required and add 10+ barrels extra to compensate for hole washout. Mix the barite through the cement pump hopper directly into the drill string. Pump the barite plug slurry down the drill pipe with the cement pumps at 6 to 12 barrels per minute and bleed and equal volume of mud through the choke on the annulus side. (On technique is simply to hold the casing pressure constant at its value prior to beginning the displacement.)
4-64
6.
7.
8. 9. 10. 11.
Note the initial drill pipe pressure after the pump has been brought up to the desired displacement rate, keeping the pump rate constant. If a gas bubble reaches the surface prior to completing the placement of the barite plug, the choke can be adjusted to hold the drill pipe pressure constant at whatever value it had when the displacement started. Follow the barite plug slurry with a mud weight of calculated kill weight or higher. This will allow the drill pipe to be kept under control following the spotting of the plug. Underdisplace and leave about 2 barrels of plug slurry in the drill pipe. Pull up out of the plug. Hold back pressure at the surface and allow about 8 hours for the plug to bridge and solidify. Once a plug is formed, work on the lost circulation problem. An operator should be prepared to keep accurate volume counts during the displacement operation. Provision should also be made to switch immediately to the rig pumps if the cement pump malfunctions. Any delay in switching pumps may allow the plug to solidify in the drill pipe.
Another possible solution to a lost circulation problem involves the spotting of a heavy slug of mud between the kicking zone and the loss zone. If the distance between the two formations is great enough, the hydrostatic pressure exerted by a shortened column of heavy mud on bottom may be sufficient to balance the formation pressure. Once the kick is controlled, the lost circulation problem can be worked on (Figure 4.16).
4-65
The hydrostatic pressure which must be exerted by the mud in order to control the well is 6,760 psi. The is no mud above the loss zone, which is at 3,500 feet. This leave the bottom 6,500 feet of the well to work with. In this case, 6,500 feet of mud must be spotted below the loss zone, and this mud must exert 6,760 psi of hydrostatic pressure. This corresponds to a 1.04 psi/ft hydrostatic pressure gradient or a 20.0 ppg mud weight. The surface system should be weighted to 20.0 ppg and then pumped down the drill pipe. Continue pumping until the first of the 20.0 ppg mud is calculated to enter the loss zone. When this occurs, the well should be dead. The lost circulation problem can then be attacked without danger of a blowout. In order for this method to work, there must be enough height difference between the kicking zone and the loss zone for a shortened column of mud to exert enough hydrostatic pressure to balance the formation pressure in the kicking zone. If the distance between the two respective formations is relatively short, unrealistically high mud densities will be required, and this method will not work. The kicking zone will either have to isolated beneath a plug or the lost circulation cured.
A decision can be made to live with the problem until the well is killed. a The pump can be slowed down while holding the casing pressure at a constant value. Once a new circulating pressure is established at the reduced rate, use the newly obtained drill pipe pressure in subsequent kill sheet calculation and kill operations. This will probably necessitate redoing the drill pipe pressure schedule and new pump rate.
4-66
If the increase in drill pipe pressure which resulted because of the plugging can be tolerated at the old pump rate, note the new value of drill pipe pressure and redo the kill sheet to accommodate the pressure changes.
2.
If the plugging is complete or nearly so and the pressures cannot be tolerated, then either the jets must be blown out of the bit (if the jets are what is plugged), the drill pipe backed off above the area that is plugged, or the drill pipe perforated. If perforations or backing-off are anticipated, care must be taken to insure that the perforations or back-off points are deep enough that mud of reasonable density can be used to kill the well. A back-off or perforations that are significantly above the bit will require a higher kill weight mud because of the shortened length of the mud column.
4-67
a sharp reduction in drill pipe pressure with no commensurate reduction in casing pressure. Allowing the problem to go uncorrected will eventually result in a washout which will make it impossible to circulate kill mud to the bit and may even result in parting of the drill string and resultant fish in the hole. The most conventional technique used to correct this problem is to pump some soft line down the drill pipe. The soft line follows the fluid flow and will flow into a hole in the drill pipe. When the soft line reaches the hole in the drill pipe and seals it, the drill pipe pressure should increase immediately. Counting the strokes it takes to pump the soft line down gives a good indication of the depth of the hole in drill pipe. If the use of soft line is not successful, it may be necessary to run a string of tubing down into the original drill string. A packer is affixed to the end of the tubing. Once the packer is positioned beneath the hole in the drill pipe, the packer can be set and the kill operation can be continued through the tubing. If absolutely necessary, the drill pipe can be stripped out of the hole, the faulty joint replaced, and the pipe stripped back in the hole. If a back pressure valve is not already in the string, a wireline bridge plug can be run.
4-68
procedures and checks to be sure that the hole is not swabbing can eliminate the problem of kicks with the bit off bottom. If a bubble enters the wellbore and is below the bit, the bubble cannot be circulated out until the bit is below the level of the bubble, and it may, in fact, be impossible to kill the well without first working the bit down to a point where mud can be circulated beneath the bubble. One method of working the pipe to bottom is the volumetric method. As pipe enters the wellbore, it displaces mud which is already present in the wellbore. If the mud which the pipe displaces is not released from the wellbore, higher pressures will develop in the wellbore. The amount of mud released should be equal to the amount displaced by pipe entering the wellbore. The problem with this method is that relatively small volumes are involved, and it is very difficult to bleed the exact amount necessary through the choke. More mud than necessary is generally bled through the choke, which results in additional fluid influx into the wellbore. The second method of allowing displaced mud to leave the wellbore is called the pressure method. A cementing pump is rigged up to the choke line and pressure is brought up to a level slightly greater than the level which is imposed on the choke by the wellbore. The return line is rigged up to the cement mixing tank. The mast valve to the choke line is then opened. Since the cement pump is holding a slightly higher pressure on the choke than that imposed by the wellbore, no additional influx occurs. As pipe is moved (stripped/snubbed) into the wellbore, the mud that the pipe displaces results in increased wellbore pressure being imposed on the choke. When the wellbore pressures exceed the cement pump pressure, fluid will escape the wellbore and enter the return line running to the cement suction tank. The mud volume in the cement suction tank can be accurately monitored to insure that the correct volume of mud is leaving the wellbore. As the pipe enters the influx bubble, a large volume increase will be noted because of the increase in vertical height of the bubble. Once the pipe is through the bubble, the influx can be circulated from the well using a standard kill procedure.
4-69
Shallow gas reservoirs are usually high pressure, low volume accumulations. The combination of low fracture gradients and difficulty in detecting a shallow pressured gas zone further enhances the hazards involved. Seismic data and offset well histories have aided considerably in the detection of those shallow pressure gas zones, but the operator still needs contingency plans in case one of these reservoirs is penetrated without prior warning. Contingency plans for encountering shallow pressured gas zones should include at least the following considerations: Can the well be shut-in or should it be diverted? The fracture gradient must be known or closely estimated in order to make this decision correctly. 2. If the drilling operation is to be carried out from a floating rig, should a marine riser be used in a diverter operation or should diversion be carried out subsea.
1.
4-70
In those cases in which a floating rig is being used, and the location can be abandoned quickly, an operator may on occasion attempt to shut-in a kick which occurs when drilling shallow hole. The operators reasoning is that diversion is a risky procedure. If shut-in can be accomplished without fractures broaching to the surface, the chances are good that the well can be killed using conventional techniques, thus minimizing the risk of dealing with a potentially disastrous surface blowout. If broaching does occur, the operator simply moves off location. The decision to attempt shut-in of a shallow kick from a highly mobile floating rig is often a value judgement, and all factors affecting this decision need to be considered most carefully. In most cases, the risks involved in diversion are less than those involved in the attempted shut-in of a shallow well kick.
4-71
The operator also has sufficient time to attempt to control the well by spotting a shortened column of heavy mud beneath the mudline. The primary disadvantage involved with drilling ahead blind in shallow hole without a marine riser is that a gas kick may result in reduced rig buoyancy due to the presence of gas bubbles in the water beneath the rig. Those who advocate dispensing with the riser in shallow portions of the well, however, agree that reduced buoyancy does not present severe danger to a drilling vessel because ocean current is usually sufficiently strong to carry all of the gas cut water safely away.
Since a diverter operation does not result in the imposition of surface pressure to counteract formation pressure, the operator has only three (3) chances of successfully controlling the well. He may be able to spot enough mud in the annulus above the kicking zone to balance the formation pressure. 2. The shallow reservoir may deplete itself before too much time has passed. 3. The hole may bridge, thus sealing off the kicking zone.
1.
Possibility number one generally involves pumping some heavy mud down the drill pipe and into the annulus in hopes enough of this heavy mud can be spotted to stop the well flow. Some operators have adopted the practice of mixing several hundred barrels of heavy reserve mud prior to spud-in and keep this mud on board for possible use during a diverter operation. Possibilities two and three are natural events over which the operator has little control. If the operator is unable to balance the formation pressure with mud hydrostatic pressure, the operators only
4-72
objective is to divert the kick safely downwind until the hole bridges over or the reservoir depletes. A diverter operation should be carried out in the following manner: If a kick is suspected, pick up the kelly high enough to clear any tool joints which may be opposite the sealing element on the diverter bag type preventer. 2. If the well is flowing, open the downwind diverter vent line. 3. Close the bag type preventer. (You must be sure that at least one of the vent lines is completely open before the bag preventer closes on the drill pipe. Failure to do this may result in an underground blowout.) 4. Pump mud own the drill pipe at an increased pump rate. Many operators keep a pit of heavy reserve kill mud for use in diverter operations. Keep pumping mud down the drill pipe until the well stops flowing. If you pump away the reserve kill mud, and the well continues to flow, switch immediately to the active pit and keep pumping. If the well is still flowing after the entire system has been pumped away, switch to sea water. Pumping should then be continued at an increased rate until the flow stops.
1.
Under no circumstances should both diverter lines be closed on a kicking well. Diverter systems should be designed so that one or both vent lines are completely open prior to the time the bag type preventer closes. Diverter systems should also be tested at frequent intervals for proper functioning. Since diverters are not designed as high pressure assemblies, no maximum test pressures are specified.
Carrying this logic a bit further, we can state that: PiVi = PfVf where: Pi Vi Pf = = = initial pressure of the gas initial volume occupied by the gas final pressure of the gas
4-73
Vf
As the pressure confining a certain amount of gas is decreased, its volume increases. For example, consider a one barrel volume of gas which has a pressure of 5,200 psi downhole. According to the formula, this gas would have a volume of 347 barrels when it reaches the surface because the only pressure confining the gas at that point is atmospheric pressure (15 psi). Under oilfield conditions, the volume occupied by a given gas also depends on the temperature of the gas and the compressibility of the particular gas. The formula used to account for these two additional variables results in modification to the original PV = K gas law formula. The modified formula is then:
Pf Vf Pi Vi --------- = --------Z i Ti Z f Tf
where:
Pi = Vi = Zi = Ti = Pf = Vf = Zf = Tf =
initial pressure of gas initial volume of gas initial compressibility of gas initial gas temperature final pressure of gas final volume of gas final gas compressibility final temperature of gas
This particular formula is only generally used to predict maximum casing pressures. Since gas is lighter than mud, gas will tend to rise or percolate up through a column of mud if given sufficient time. The rate at which gas migrates depends heavily on mud properties and composition of the gas itself. The maximum rate at which a gas bubble will normally rise through a column of mud on a shut-in well is generally assumed to be about 1,000 ft/hr. However, this figure may vary greatly under different circumstances. If a kicking well can be shut-in without lost returns or equipment failure, it is often assumed that all the conditions will remain stable indefinitely and that the well can be killed almost at leisure. Gas migration on a shut-in well can change this assumed stable condition to an unstable and dangerous situation. In order to understand the effects of gas migration on a shut-in well, consider the following example. Example: Well depth = 10,000 ft
4-74
= =
At initial shut-in, assume the following conditions: SIDPP SICP Influx is gas
1.
= =
The pressure in the gas bubble at initial shut-in is equal to the formation pressure (Figure 4.17).
Figure 4.17 Pressure in the Well with a Gas Bubble at Initial Shut-In Equal to the Formation Pressure
At this time, the formation pressure is balanced by the pressure of the gas which is in the wellbore. The gas itself is confined by the combination of mud hydrostatic pressure and surface imposed pressure.
2.
Five hours later, the gas has migrated 5000 feet up the hole. The operators representative has spent all this time waiting on orders, and has done nothing. Since no mud has been pumped or bled off, the volume of mud in the annulus has not changed. The volume of gas has not changed either. The gas law states that PV = constant. Therefore, if the volume of gas is not changed, then the pressure is also unchanged. At this point, the pressure in the gas bubble is still 5200 psi. The formation sees the following (Figure 4.18).
4-75
Figure 4.18 Pressures After the Gas Bubble has Migrated 5000 Feet
The pressure in the gas bubble acts uniformly in all directions. Since the gas pressure is the same as it was initially, the pressure confining the gas must also be the same. The pressure confining the gas results from a combination of surface imposed pressured and hydrostatic pressure exerted by the mud above the bubble. Since the hydrostatic pressure exerted by the mud above the bubble has decreased due to gas migration, the surface imposed pressure must increase to compensate. The formation on bottom now feels the pressure in the gas bubble plus the hydrostatic pressure exerted by the mud below the bubble. Thus, the bottom hole pressure on the annulus side is not equal to 7,540 psi. Since the system is a closed, balanced U-tube, the drill pipe pressure also increases. Note that the shut-in pressures on both sides have increased over initial values by equal amounts. The increased surface imposed pressure results in dramatic increases in the equivalent mud weights at various points in the wellbore.
3.
If the bubble is allowed to migrate all the way to the surface without allowing any expansion, the following wellbore conditions will result (Figure 4.19).
4-76
Figure 4.19 Pressures After the Gas Bubble has Migrated to the Surface Without any Expansion
At this point, the formation feels the pressure of the gas bubble plus the hydrostatic pressure of the 9,829 feet of mud now beneath the bubble. The drill pipe side of the U-tube also reflects an equal amount of increased surface imposed pressure. Gas migration is recognized by noting the increase in shut-in pressures on both the drill pipe and casing sides. During gas migration, both pressures will continue to increase equally until the bubble reaches the surface, the formation breaks down, casing burst or equipment failure occurs. In most cases the bubble will not reach the surface before one of the latter possibilities becomes reality. If the bit is on bottom when a kick occurs, handling gas migration is a relatively simple matter. As the shut-in drill pipe pressure increases to values above that necessary to balance formation pressure, small amounts of mud should be bled from the annulus side to relieve the excess drill pipe pressure. Mud should be bled in amounts of less than one barrel. Bleeding mud through a manual choke and directing this mud into a calibrated tank is one way of accurately gauging the amounts bled off. Mud should be bled whenever the shut-in drill pipe pressure rises significantly above its initial shut-in value. Bleeding small amounts of mud from the annulus each time the well pressure is bled off helps to insure that no additional influx occurs as a result of the bleeding operation. In those instances in which the drill pipe pressure gauge cannot be used as a direct indicator of bottom hole pressure (i.e., pipe off bottom as out of the hole, drill pipe or bit plugged, float value in use), the procedure for handling gas migration becomes more complicated. Gas migration can be recognized by noting increases
4-77
in casing pressures to values higher than those at initial shut-in. The position of the gas bubble in the wellbore must be estimated and the number of barrels of mud required to exert a given hydrostatic pressure must be known before a procedure can be implemented.
4-78
Appendix A
Appendix A
above absent abundant acicular after agglomerate aggregate algac, algal allochem altered alternating amber ammonite amorphus amount amphipora and angular anhedral anhydrite(ic) anthracite aphanitic apparent appears approximate aragonite arenaceous argillaceous argillite arkose(ic) as above asphalt(ic) assemblage associated at authigenic average band(ed)
ab abs abd acic aft Aglm Agg Alg, alg Allo alt altg amb Amm amor amt Amph & ang ahd Anhy, anhy Anthr aph apr ap apprx arg aren arg argl Ark, ark a.a. Asph, asph Assem assoc @ authg Av, av Bnd, bnd
barite(ic) basalt(ic) basement become(ing) bed(-ed) bedding belemnites bentonite(ic) bioclastic biothem(-al) biomicrite biostrom(-al) biotite bioturbated birdseye bitumen(-inous) black(-ish) blade(-ed) blocky blue(-ish) bored(-ing) botryoid(-al) bottom boudinage boulder boundstone brachiopod brackish branching break breccia(-ted) bright brittle brown bryozoa bubble buff bulbous
bar Bas, bas Bm bcm Bd, bd Bdg Belm Bent, bent biocl Bioh, bioh Biomi Biost, biost Biot bioturb Bdeye Bit, bit blk, blksh Bld, bld blky bl, blsh Bor, bor Bot, bot Btm boudg Bld Bdst Brach brak brhg Brk, brk Brec, brec brt brit brn Bru Bubl bu bulb
A-1
Appendix A
Word
Abbreviation
Word
Abbreviation
burrow(-ed) calcarentie calcareous calcilutite calcirudite calcisilitite calcisphaera calcisphere calcite(-ic) caliche carbonaceous carbonate carbonized cavem(-ous) caving cement(-ed, -ing) center(-ed) cephalopod chaetetes chalcedony(-ic) chalk(-y) charophyte chert(-y) chitin(-ous) chitinozoa chlorite(-ic) chocolate circulate(-ion) clastic clay(-ey) claystone clean clear cleavage cluster coal coarse coated(-ing) coated grains cobble colonial
Bur, bur Clcar calc Clclt Clcrd Clslt Casph Clcsp Calc, calctc cche carb crbnt cb Cav, cav Cvg Cmt, cmt Cntr, cntr Ceph Chaet Chal, chal Chk, chky Char Cht, cht Chit, chit Chtz Chlor, chlor choc Circ, circ clas Cl, cl Clst cln clr Clvg Clus C crs cotd, cotg cotd gn Cbl coin
color common compact compare concentric conchoidal concretion(-ary) conglomerate(-ic) conodont conquina conquina(-iod) considerable consolidated conspicuous contact contamination(-ed) content contorted coral, coralline core covered cream crenulated crinkled crinoid(-al) cross cross-bedded cross-laminated cross-stratified crumpled crystal(-line) crystocrystalline cube(-ic) cuttings cypridopsis dark dead debris decrease(-ing) dendrite(-ic) dense
col com cpct cp cncn conch Conc, conc Cgl, cgl Cono Coq Coqid cons consol conspic Ctc Contam Cont cntrt Cor, corin c cov crm cren crnlk Crin, crinal x x-bd x-lam x-strat crpld Xl, xln crpxln Cub, cub ctgs Cyp dk, drk dd Deb Decr, decr dend dns
A-2
Appendix A
Word
Abbreviation
Word
Abbreviation
depauperate description desication detrital devitrified diabase diagenesis(-etic) diameter disseminated distillate ditto dolomite(-ic) dolostone dominant(-ly) drill stem test drilling drusy earthy East echnoid elevation elongate embedded endothyra equant equivalent euhedral euryamphipora exinic evaporite(-itic) excellent exposed extraclast(-ic) extremely extrusive facet(-ed) faint fair fault(-ed) fauna favosites
depau Descr dess detr devit Db Diagn, diagn Dia dissem Dist " or do Dol, dol dolst dom DST drlg dru ea E Ech Elev elong embd Endo eqnt Equiv euhd Euryamph eux Evap, evap ex exp Exclas, exclas extr exv Fac, fac fnt fr Flt, flt Fau Fvst
feet feldspar(-athic) fenestra(-al) ferro-magnesian ferruginous fibrous fill(-ed) fine(-ly) firm fissile fleggy flake(-ky) flat flesh floating flora fluorescene(-ent) foliated foot foraminifer foraminiferal formation fossil(-iferous) fracture(-d) fragment(-al) framework frequent fresh friable fringe(-ing) frosted frosted quartz fucoid(-al) fusulinid gabbro galeolaria gas impression in part geopetal gilsonite
ft Fspr, fspr Fen, fen Fe-mag ferr fibr fl f, fnly frm fis flg Flk, flk fl fls fltg Flo Fluor, fluor fol Ft Foram foram Fm Foss, foss Frac, frac Frag, frag frmwk freq frs fri Frg, frg fros F.O.G. Fuc, fuc Fus Gab Gal G imp I.P. gept Gil
A-3
Appendix A
Word
Abbreviation
Word
Abbreviation
girvanella glass(-y) glauconite(-itic) globigerina(-inal) gloss(-y) gneiss(-ic) good grading grain(-s, -ed) grainstone grainte grainte wash granule(-ar) grapestone graptolite gravel gray, grey(-ish) graywacke greasy green(-ish) grit(-ty) gypsum(-iferous) hackly halite(-iferous) hard heavy hematite(-ic) heterogeneous heterosteqina hexagonal high(-ly) homogeneous horizontal hornblende hydrocarbon igneous rock imbedded inch inclusion(-ded) increasing indistinct
Girv Glas, glas Glauc, glauc Glob, glob Glos, glos Gns, gns gd grad Gr, gr Grst Grt G.W. Gran, gran grapst Grap Grv Gry, grysh Gwke gsy gn, gnsh Gt, gt Gyp, gyp hkl Hal, hal hd hvy Hem, hem hetr Het hex hi hom hor hornbd Hydc Ig, ig imbd in Incl, incl incr indst
indurated inoceramus insoluble interbedded intercalated intercrystalline interfragmental intergranular intergrown interlaminated interparticle interpretation intersticies interval intraclast(-ic) intraparticle intrusive invertebrate iridescent ironstone irregular(-ly) isopachous ivanovia jasper joint(-d, -ing) kaolin(-itic) lacustrine lamina(-itions, ated) large laterite(-itic) lavender layer leached lens, lenticular lentil(-cular) light lignite(-itic) limestone limonite(-itic) limy lithic
ind Inoc insl intbd intercal intxin intfrag intgran intgn intrlam intpar intpt instst Intvl Intclas, intclas intrapar intr Invtb irid Fe-st irr iso Ivan Jasp Jt, jt Kao, kao lac Lam, lam lge Lat, lat lav Lyr lchd Len, lent len lt Lig, lig Ls Lim, lim lmy lit
A-4
Appendix A
Word
Abbreviation
Word
Abbreviation
lithographic lithology(-ic) little littoral local long loose lower lumpy lustre lutite macrofossil magnetite, magnetic manganese marble marine marl(-y) marlstone maroon massive material matrix maximum medium member meniscus metamorphic(-osed) metamorphic rock metasomatic mica(-iceous) micrite(-ic) micro micro-oolite microcrystalline microfossil micrograined micropore(-osity) microspar microstylolite middle miliolid
lithgr Lith, lith Ltl litt loc lg lse l lmpy Lstr Lut Macrofos Mag, mag Mn Mbl marn Mrl, mrl Mrlst mar mass Mat Mtrx max m or med Mbr men meta, metaph Meta msm Mic, mic Micr, micr mic Microol microxin microfos micgr micropor Microspr Microstl Mid Milid
milky mineral(-lized) minor minute moderate mold(-ic) mollusc mosaic mottled mud(-dy) mudstone muscovite nacreous no sample no show no visible porosity nodules(-ar) North novaculite numerous occasional ochre odor oil oil source rock olive olivine oncolite(-oidal) ooid(-al) oolicast(-ic) oolite(-itic) oomold(-ic) opaque orange(-ish) orbitolina organic orthoclase orthouartzite ostrcod overgrowth oxidized
mky Min, min mnr mnut mod Mol, mol Moll mos mott md, mdy Mdst Musc nac n.s. n/s n.v.p. Nod, nod N Novac num occ och od O O.S.R. olv olvn Onc, onc Oo, oo Ooc, ooc Ool, ool Oomol, oomol op or, orsh Orbit org Orth O-Qtz Ostr ovgth ox
A-5
Appendix A
Word
Abbreviation
Word
Abbreviation
oyster packstone paper(-y) paraparchites part(-ly) particle parting parts per million patch(-y) pearly pebble pelecypod pellet(-al) pelletoid(-al) pendular(-ous) pentamerus permeability(-able) petroleum phlogopite phosphate(-atic) phreatic phyllite (-ic) pin-point (porosity) pink(ish) pisoid(-al) pisolite(-ic) pitted plagioclase plant plastic platy polish(-ed) pollen polygonal poor(-ly) porcelaneous porosity (porous) porphyry possible(-ly) predominant(-ly) preserved
Oyst Pkst Pap, pap Para Pt, pt Par Ptg PPM Pch, pch prly Pbl Pelec Pel, pel Peld, peld Pend, pend Pent Perm, perm pet Phlog Phos, phos phr Phyl, phyl p.p. pk, pksh Piso, piso Pisol, pisol pit Plag Plt plas plty Pol, pol Poln poly P porcel Por, por prphy poss pred pres
primary prism(-atic) probable(-ly) production prominent pseudo oolite(-ic) pseudopumicestone purple pyrite(-itized, itic) pyrobitumen pyroclastic pyroxene quartz(-ose) quartzite(-ic) radial(-ating) radiaxial range rare recemented recovery(-ered) recrystallized red(-ish) reef(-old) remains renalcis replaced(-ment) residue(-ual) resinous rhomb(-ic) ripple rndd,frosted,pitted rock round(-ed) rubble(-bly) rubist rugose saccharodial salt(-ly) salt and pepper salt cast(-ic)
prim pris prob Prod prom Psool, psool ps Pst purp Pyr, pyr Pyrbit pyrcl pyrxn Qtz, qtz Qtzt, qtzt Rad, rad Radax rng r recem Rec, rec rexizd rd, rdsh Rf, rf Rem Ren Rep, rep Res, res rsns Rhb, rhb Rpl r.f.p. Rk rnd, rndd Rbl, rbl Rub rug sacc Sa, sa s&p sa-c
A-6
Appendix A
Word
Abbreviation
Word
Abbreviation
salt water same as above sand(-y) sandstone saturation(-ated) scales sacphopod scarce scattered schist(-ose) scolecodont secondary sediment(-ary) selenite septate shadow shale(-ly) shell shelter porosity show siderite(-itic) sidewall core silica(-iceous) size sheletal slabby slate(-y) slickenside(-d) slight(-ly) small smooth soft solenopora solitary solution, soluble somewhat sorted(-ing) South spar(-ly) spare(-ly) speck(-ed)
S.W. a.a. Sd, sdy Sst Sat, sat sc Scaph scs scat Sch, sch Scol sec Sed, sed Sel sept shad Sh, sh Shl Shlt por shw Sid, sid S.W.C. Sil, sil sz skel slb Sl, sl Slick, slick sli, slily sml sm sft Solen sol Sol, sol smwt srt, srtg S Spr, spr sps, spsly Spk, spkld
sphaerocodium sphalerite shperule(-itic) spiclue splintery sponge spore spotted(-y) stachyode stain(-ed, -ing) stalactitic strata(-ified) streaming striae(-ted) stringer stromatolite stromatoporoid structure styliolina stylolite(-itic) sub subangular sublithic subrounded sucrosic sugary sulphur(-ous) superficial oolite surface syntaxial syringopora vermillon tabular(-ate) tan tasmanites tension tentaculties ternginous texture(-d) thamnopora thick
Sphaer Sphal Spher, spher Spic Splin Spg Spo sptd, spty Stach Stn, stn stal Strat, strat stmg Stri, stri strgr Stromilt Strom Str Stylio Styl, styl sb sbang sblit sbrndd suc sug Su, su Spfool Surf syn syring verm tab tn Tas tns Tent ter Tex, tex Tham thk
A-7
Appendix A
Word
Abbreviation
Word
Abbreviation
thin thin section thin-bedded throughout tight top tough trace translucent transparent trilobite tripoli(-itic) tube(-ular) tuff(-aceous) type(-ical) unconformity unconsolidated underclay underlying unidentifiable uniform upper vadose variation(-able) varicolored variegated vared vein(-ing, -ed) veinlet
thn T.S. t.b. thru ti Tp tgh Tr trnsl trnsp Tril Trip, trip Tub, tub Tf, tf Typ, typ Unconf uncons Uc undly unident uni u Vad Var, var varic vgt vrvd Vn, vn Vnlet
vertebrate vertical very very poor sample vesicular violet visible vitreous(-ified) volatile volcanic rock vug(-gy) wackestone washed residue water wavy waxy weak weathered well West white with without wood yellow(-ish) zeolite zircon zone
vrtb vert v V.P.S. ves vi vis vit volat Volc Vug, vug Wkst W.R. Wtr wvy wxy wk wthd Wl W wh w/ w/o Wd yel, yelsh zeo Zr Zn
A-8
Appendix B
Appendix B
Table B.1 Rock Type Color
Classification Tables
Crystal or Grain Size Major and Minor Characteristics Hardness Porosity Oil Show
Siltstone
White Gray Green Brown Gray green Gray brown Clear Yellow Pink Clear White Gray Green Brown Gray green Gray brown White Gray Amber Brown Spotted
Quartz Variety Pyritic Glauconitic Coal Inclusions Carbonaceous Argillaceous Angularity Sorting Angularity Sorting
Common Variety Pyritic Glauconitic Coal Inclusions Calcareous Dolomitic Minerals Clay Matrix Minerals Clay Matrix
Cementation or Induration
Intergranular
Possible
Sand
Grain size
Unconsolidated
Intergranular
Possible
Sandstone
Grain size
Cementation
Intergranular
Possible
Chert
Microcrystalline by definition
none
none
B-1
Appendix B
Table B.2
Classification of Argillaceous Sediments Color Crystal or Grain Size Major and Minor Characteristics Hardness Porosity Oil Show
Rock Type
Shale
Green Gray Brown Gray green Gray brown Red Yellow Black Blue Green Gray Brown Gray green Gray brown Green Gray Brown Gray green Gray brown Yellow Blue Black
See Definition
Calcareous Pyritic Glauconitic Silty Sandy Carbonaceous Micaceous Fossiliferous Coal Inclusions
Not applicable
Clay
See Definition
Not applicable
None
Marl
See Definition
Not applicable
None
B-2
Appendix B
Table B.3
Classification of Limestone Color Crystal or Grain Size Major and Minor Characteristics Hardness Porosity Oil Show
Rock Type
Calcirudte
Argillaceous Carbonaceous Pyritic Glauconitic Silty or sandy Fossil Fragments Dolomitic Micrite Sparite Argillaceous Carbonaceous Pyritic Glauconitic Silty or sandy Fossil fragments Dolomitic Micrite Sparite Argillaceous Carbonaceous Pyritic Glauconitic Silty or sandy Dolomitic Micrite Sparite
Possible
Calcarenite
Possible
Calcilutite
Possible
B-3
Appendix B
Table B.4
Classification of Dolomite Color Crystal or Grain Size Major and Minor Characteristics Hardness Porosity Oil Show
Rock Type
Dolomite
Possible
Table B.5
Classification of Limestone and Limestone Matrices Color Crystal or Grain Size Major and Minor Characteristics Hardness Porosity Oil Show
Rock Type
Micrite
Clear White Buff Gray Tan Brown Clear White Buff Gray Tan Brown
Argillaceous Carbonaceous Pyritic Glauconitic Silty or sandy Fossil fragments Dolomitic Argillaceous Carbonaceous Pyritic Glauconitic Silty or sandy Fossil fragments Dolomitic
Possible
Sparite
Possible
B-4
Appendix B
Table B.6
Classification of Sulfates Color Crystal or Grain Size Major and Minor Characteristics Hardness Porosity Oil Show
Rock Type
Crystalline Amorphous
Transparency Salt inclusions Platey Anhydritic Clay (not found below approx. 3,000 ft.)
Not applicable
Probably none
Anhydrite CaSO4
Crystalline Amorphous
Pearly luster Dolomite inclusions Gypsum inclusions Limestone inclusions Salt inclusions Pyritic Glauconitic Argillaceous Carbonaceous
Indurated
Possible if fractured
B-5
Appendix B
B-6
Appendix C
Appendix C
lb/gal Buoyancy Factor
Buoyancy Factors
lb/gal Buoyancy Factor lb/gal Buoyancy Factor lb/gal Buoyancy Factor
6.0 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 7.0 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 8.0 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 9.0 9.1 9.2 9.3 9.4 9.5 9.6 9.7
.9083 .9068 .9053 .9037 .9022 .9007 .8991 .8976 .8961 .8946 .8930 .8915 .8900 .8884 .8869 .8854 .8839 .8823 .8808 .8793 .8778 .8762 .8747 .8732 .8716 .8701 .8686 .8671 .8655 .8640 .8625 .8609 .8594 .8579 .8564 .8548 .8533 .8518
9.8 9.9 10.0 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 10.9 11.0 11.1 11.2 11.3 11.4 11.5 11.6 11.7 11.8 11.9 12.0 12.1 12.2 12.3 12.4 12.5 12.6 12.7 12.8 12.9 13.0 13.1 13.2 13.3 13.4 13.5
.8502 .8487 .8472 .8457 .8441 .8426 .8411 .8395 .8380 .8365 .8350 .8334 .8319 .8304 .8289 .8273 .8258 .8243 .8227 .8212 .8197 .8182 .8166 .8151 .8136 .8120 .8105 .8090 .8075 .8059 .8044 .8029 .8013 .7998 .7983 .7968 .7952 .7937
13.6 13.7 13.8 13.9 14.0 14.1 14.2 14.3 14.4 14.5 14.6 14.7 14.8 14.9 15.0 15.1 15.2 15.3 15.4 15.5 15.6 15.7 15.8 15.9 16.0 16.1 16.2 16.3 16.4 16.5 16.6 16.7 16.8 16.9 17.0 17.1 17.2 17.3
.7922 .7906 .7891 .7876 .7861 .7845 .7830 .7815 .7800 .7784 .7769 .7754 .7738 .7723 .7708 .7693 .7677 .7662 .7647 .7631 .7616 .7601 .7586 .7570 .7555 .7540 .7524 .7509 .7494 .7479 .7463 .7448 .7433 .7417 .7402 .7387 .7372 .7356
17.4 17.5 17.6 17.7 17.8 17.9 18.0 18.1 18.2 18.3 18.4 18.5 18.6 18.7 18.8 18.9 19.0 19.1 19.2 19.3 19.4 19.5 19.6 19.7 19.8 19.9 20.0
.7341 .7326 .7311 .7295 .7280 .7265 .7249 .7234 .7219 .7204 .7188 .7173 .7158 .7142 .7127 .7112 .7097 .7081 .7066 .7051 .7035 .7020 .7005 .6990 .6974 .6960 .6944
C-1
Appendix C
C-2
Appendix D
Appendix D
D-1
Appendix D
Method
Ordinarily, when formation cuttings are drilled they retain much of the formation pore fluid. This fluid is released to the mud column as the cuttings travel up the annulus. Most of the formation fluid in the cuttings will be produced into the drilling mud during the top 500 feet of hole travel. Conventionally, a mud sample is diverted to a mechanically operated gas trap to obtain a sample of the gas in the mud. The efficiency of this trap is from 15 to 70%, depending upon the gel strength of the mud, the amount of mud flowing through the trap and the rotation speed of the trap impeller. The magnitude of the conventional gas show is, therefore, quantitative only to the air-gas sample obtained. The sample is accurately analyzed by the gas chromatograph; but, because the sample furnished by the conventional gas trap represents only a fraction of the gas present in the mud, and because that fraction is not representative of the total gases in the mud, the results are still only qualitative. When the Steam Still Reflux Unit is used to obtain the gas sample, the gas sample will represent almost 100 percent of the hydrocarbon fractions C1 through C5 that were in the mud sample. This enables the chromatograph analysis to be related quantitatively to the mud, and the readings to be reported as parts per million of each hydrocarbon vapor (C1 through C5) to mud volume. Because the cuttings from a particular formation produce the gas they contain into the drilling mud, it was reasonable to assume that this same formation, if completed, would produce gases of similar composition. This assumption led to a comparison of PPM logs of hydrocarbon vapors with similar data from producing wells. Plots were made of the ratio of methane to each of the heavier hydrocarbons from many analyses of wellhead samples. These plots were compared with plots, made from PPM logs, of gas in mud. Both groups of plots showed definite patterns between (1) the magnitude of the ratios of methane to each of the heavier hydrocarbons, and (2) the slope of the lines of the plotted ratios. These in turn, indicate productive potential and reservoir permeability. The Steam Still Reflux Unit consists of a small steam boiler, mud-injection port, mud-steam mixing chamber, Reflux-Condensing Unit and a gas-extraction port. Five (5) milliliters (ml) of mud are injected into the purged mud-steam mixing chamber. The mud is rolled with 2,000 to 4,000 volumes of steam. The hydrocarbons (C1 through C5) extracted from the mud are collected at the Reflux-Condensing Unit, withdrawn with a syringe, diluted to the standard chromatograph sample size and injected into the chromatograph for analysis. The Reflux-Condensing Unit removes only the lighter paraffin series hydrocarbons from the mud sample tested. For example, if the mud contains diesel oil, the more complex hydrocarbons - C6 and above - condense and drop back into the
D-2
Appendix D
mud-steam mixing chamber. Therefore, regardless of whether the fluid phase of the mud is oil or water, the gas sample analyzed contains only the light fractions through C5, and the analysis is representative of the formation gas. The full importance of determining formation gas composition has not always been apparent. At first it was observed that if the magnitude of butane in the mud was greater than the magnitude of either propane or ethane, the zone in question would produce water and hydrocarbons. Later, the ratios of methane to each of the heavier hydrocarbons components were plotted on semilog paper. Hydrocarbon ratio plots obtained from PPM logs and available data from wellhead gas sample analyses were compared. The comparison of the plots from PPM logs and wellhead gas analysis showed a striking correlation. The correlation demonstrated that PPM logs made with Steam Still-Reflux samples could be interpreted The magnitude of the methane-to-ethane ratio determines if the reservoir contains gas or oil or if it is nonproductive. The slope of the line of the ratio plot of C1/C2, C1/C3, C1/C4 and C1/C5 indicates whether the reservoir will produce hydrocarbons or hydrocarbons and water. Positive line slopes indicate production; negative slope lines indicate water-gearing formations. An undersaturated reservoir may show a negative slope, but such occurrences are rare. The ratio plots may not be definitive for low permeability zones, but unusually steep plots indicate tight zones. A ratio of C1/C2 between 2 and 15 indicates oil. A ratio of C1/C2 between 15 and 65 indicates gas. The lower the C1/C2 ratio, the richer the gas or the lower the oil gravity. If the ratio of C1/C2 is below about 2 or above about 65 the zone is nonproductive.
Figure 1
D-3
Appendix D
Figure 2
Hydrocarbon ratio plots, productive reservoirs, South Texas reservoirs, Rocky Mountain area
Field Examples
Figure 1 shows average hydrocarbon ratio plots from limestone reservoirs in the Rocky Mountain area. Plot 1 is derived from analyses of gases from Mississippian oil-producing reservoirs. The C1/C2 ratio is 3.5. The slope of the line is again positive and not steep. Plot 2 was obtained from analyses of gases from wells producing gas-condensate from the Silurian. The C1/C2 ratio is 12; the line slope is again positive and not steep. Plot 3 is from gas-condensate wells producing from the Ordovician. The C1/C2 is 15 and, again, the slope of the line is not steep; all three plots show slopes favorable to production. Plot 4 shows ratios obtained from an analysis of gas from the Lower Ordovician, which produced gas and water. The plot shows a negative slope of the section from the C1/C4 ratio to the C1/C5 ratio. Many tests have verified the fact that if a ratio plot shows a negative slope, the zone in question is water-bearing. Figure 2 shows plotted hydrocarbon ratios for productive reservoirs in South Texas. Plot 1 was made from an analysis of a wellhead sample of gas-condensate produced from a Frio sand, Hidalgo County. The production is rich in liquid hydrocarbons as indicated by the low C1/C2 ratio. Plot 2 is from an analysis of a wellhead gas sample from a 11,000 ft oil reservoir, North Lindsey field. The pentane was not reported, but the low C1/C2 ratio indicates oil production. Plot 3 was obtained from a gas show at 12,690 ft on the PPM log of the State Tract 49 No.1 Well, Nueces County, Texas. Formation tests resulted in gas production.
D-4
Appendix D
Experience shows that if the C1/C2 ratio is above 65 the zone is too tight for commercial production. Figure 3 shows the ratio plots obtained from PPM logs on the Texas Gulf Coast wells that were nonproductive in the zones of interest. Plot 1 is from the PPM log of the R.A. Tally No.1 well, Victoria County, Texas. The C1/C2 ratio was 470. The zone was tested extensively but it was a low permeability reservoir that could not be commercially completed. Plots 2 and 3 are from the State Tract 49 No.1 well, Nueces County, Texas. Plot 2 was from a sand encountered at about 8,060 ft. The relatively high ratios of C 1/C2, C1/C3, C1/C4, and C1/C5 indicated that the zone was nonproductive because of the low permeability. This was subsequently verified by testing. Plot 3 was obtained from a sand at 9,130 ft. The negative slope of the ratio plot, C1/C2 to C1/C3, indicated that the zone was water-bearing. Subsequent formation tests showed water and non-commercial amounts of gas.
D-5
Appendix D
Figure 3
Plot 4 was obtained from the PPM log of the Kovar No.1 well, Victoria County, Texas. The sand encountered from which the plot was made is at 10,120 ft. The gas show appeared to be good, but a negative slope of the C 1/C3 ratio to the C1/C4 ratio was positive identification of a water-bearing formation.
Evaluation Technique
It is apparent that with this evaluation system, potential production can be accurately predicted. The only significant time lapse between penetration of the formation and evaluation of its productive possibilities is the time required to pump the mud from the bottom of the hole to the surface and analyze it by the Steam Still-Reflux and chromatograph method. Figure 4 shows the evaluation technique, which may be described as follows. First, record the net increase of each gas component over the background gas; next, plot the ratios C1/C2, C1/C3, C1/C4, C1/C5 on the ratio lines as indicated. Then evaluate, within the following limits, the section in question for probable production as indicated by the plotted curve: Productive dry gas zones may show only C1, but abnormally high shows of C1 only are usually indicative of salt water. 2. If the C1/C2 ratio is low in the oil section and the C1/C4 ratio is high in the gas section the zone is probably nonproductive. 3. If any ratio (C1/C5 excepted if oil is used in the mud) is lower than a receding ration, the zone is probably nonproductive. For example, if C1/C4 is less than C1/C3, the zone is probably water-bearing. 4. The ratios may not be definitive for low permeability zones; however, steep ratio plots may indicate tight zones.
1.
D-6
Appendix D
Figure 4
PPM log and report form for analyses from gas shows on PPM logs
Application
The PPM log is only one of many tools that are ordinarily used for formation evaluation. But in many instances, the PPM log has furnished the vital information necessary to make the final decision on a well. One well drilled in inland waters of Louisiana had what appeared on the PPM log to be a good sand body, but the PPM log showed only a nominal increase in gas. After the sand was penetrated and the well deepened, hole trouble was encountered. No other information of interest was available on the sand. The cost of the sidetracking to investigate the sand was sizable. Tight hole conditions and the low magnitude of the gas show indicated that the sand had good permeability and that possibly formation hydrocarbons had been flushed ahead of the bit. A plot of hydrocarbon ratios indicated oil production. Therefore, at considerable expense, the sand was investigated and a new oil field was found. An interesting well drilled in St. Martin Parish, LA, was the No.1 St. Martin Bank and Trust located on the southeast flank of the Anse La Butte Dome. A good sand was encountered at about 8,000 ft showed oil, but the negative slope of the ratio plot indicated that the sand was water-bearing. The well was deepened to approximately 9,600 ft. One of the partners, a successful independent with a talent for finding oil by feel and by prudent use of the latest technology, decided that the formations in which the well was being drilled were tilted to almost vertical.
D-7
Appendix D
On his recommendation, the well was plugged back to about 7,000 ft and sidetracked. The sand that was drilled at 8,000 ft in the first hole was encountered in the directional hole at approximately 7,000 ft and the entire sand was hydrocarbon saturated. The PPM log and the ratio plots from the sand in the sidetrack hole are shown in Figure 5. Table 1 shows the mud gas components related to percent of total gas. In actual practice, the ppm gas shows obtained from the PPM log are not converted to percent of total gas; but note the general decrease in percent methane in the lower section of the sand compared with that in the upper section. The magnitude of the gas show in the straight hole and in the sidetrack hole was significant. An accurate determination, however, of the composition of the gas in both cases led to correct conclusions on the potential productivity of the sand at the different depths in each hole. Note that the ratio Plot 1 at the top of the sand indicates a gas cap. As shown in Table 1, the gas was 93.1 percent methane. Subsequent plots indicated that production would be oil. In each of these cases the C1/C2 ratio was less than 9. The lowest ratio, 4.5, is shown in Plot 6, which was made from the show at the bottom of the sand.
Figure 5
PPM log and hydrocarbon ratio plots, No.1 St. Martin Bank and Trust Well, St. Martin Parish, LA.
D-8
Appendix D
Another example of the application of the PPM log is No.1 State Tract 198 well, Aransas County, Texas. Many sands were encountered showing the present of hydrocarbons. The logging crew submitted more than 60 ratio plots to the operator during the drilling of the well. In almost all instances subsequent information verified the logging engineers predictions of probable productivity based on the ratio plots. Figure 6 shows a section of the PPM log and ratio plots for this well. The gas composition relating the percent of each gas component to the total gas is shown in Table 2. Gas condensate production is indicated by the PPM log and ratio plots as shown. The zones are tight marine deposits - especially the 10,520 ft zone. Plot 4 has the steepest line slope; pentane was not present. The slope of Plot 3 is steep. Plots 2 and 5 show more favorable (less steep) line slopes. The electric log and subsequent formation tests made of each zone indicated probable production. The well was completed as a gas condensate producer in the 11,300 ft section, which is the section plotted as No.5.
D-9
Appendix D
Figure 6
PPM log and hydrocarbon ratio plots, No.1 State Tract 198 well, Aransas County, Texas
D-10
Appendix D
Conclusions
Only qualitative shows of hydrocarbons in the mud can be derived from conventual mud logs. If chromatography is used, only a general indication of in-place gas composition is obtainable. Such hydrocarbon shows may be reported as units of gas or percent hydrocarbons or parts per million (PPM) as in the air-gas mixture tested. Only the presence in relative amounts, not the actual quantity, of hydrocarbons in mud is indicated, and other supplemental information may be necessary to evaluate the formation in terms of potential productivity. However, if the composition of the gas sample obtained from the mud is representative of the in-place formation gas, then the gas analysis is accurate. The use of the Steam Still-Reflux Unit makes possible a report of formation gas composition on the PPM log. Meaningful ratio plots of gas composition can then be made. Even though many factors affect the amount of reservoir fluid released to the drilling mud, reservoir potential productive capabilities can be determined by a study of the ratio of methane to each of the heavier hydrocarbon components. The hydrocarbon ratio plot is a unique technique and provides the operator with new information for evaluating productive possibilities of exploratory wells. Computer programs involving percent gas in mud (PPM log) and gas composition are being used in special cases to determine reservoir potential production. The use of computers in mud log interpretation, although new, will contribute significantly towards a better application of the data shown on the PPM log.
D-11
Appendix D
D-12
Appendix E
Appendix E
Abstract
To be able to evaluate formation potentials and possible productivity, using the hydrocarbon ratios minutes after the drilling fluid attains the surface. Thereby decreasing the need for running expensive drilling operations (Drill Stem Test, cores, log runs) when formation potential is questionable. That is to say, the use of Drill Stem Tests, cores and log analysis can be minimized through this method.
Theory
That a knowledge of hydrocarbon ratios as they relate to formation fluids, enables a logging technician, on wellsite, to determine possible hydrocarbon shows: and that the relationship between quantitative amounts of methane (C1), ethane (C2), propane (C3), butane (C4) and pentane (C5) to in-place reservoir production and potential. Therefore, it is the presence in relative amounts, not the actual quantity of hydrocarbons in drilling fluids, that determine reservoir potential.
Introduction
In any discussion of the validity of different methods of hydrocarbon detection and evaluation, it is perhaps a good idea to try and envisage exactly what is taking place in the process of drilling a hole in the ground.
E-1
Appendix E
The original rock strata-reservoir or not - lies undisturbed before the advent of the bit, and our objective is to attempt to reconstruct the exact physical properties of this strata without interrupting the drilling operation to run a DST, core, etc. As the bit approaches the strata that we are interested in, a phenomenon known as flushing occurs. This, in short, is simply a replacement of some of the pore fluids of the rock by some of the mud, and is caused by hydrostatic pressure of the mud column usually being in excess of that exerted by the formation. The amount of flushing that takes place, depends on a number of factors, e.g., (depth, ROP, hole size, volume of mud being circulated, physical properties of the formation and the mud, etc.) and in fact, is very difficult to estimate. However, despite the fact that some portion of the interstitial fluids are flushed away from the wellbore, the formation will still contain some of these fluids when the bit finally does arrive. At this point, the bit will mechanically break up the solid formation into small cuttings and, depending mainly on the porosity of the formation (but also on other factors) some oil and gas, if present, will be released into the mud. Most of the interstitial fluids remain contained in the cuttings, however, being released as the cuttings travel up the hole; the pressure on them being reduced from that of the hydrostatic head of the mud to atmospheric. Most of these fluids will be produced into the mud stream in the final 200 meters of travel, especially if effective permeability is present. Some of the interstitial fluids, however, will remain in the cuttings, although greater importance is usually placed on the returning mud stream as a source of hydrocarbon information. We have, anyway, two places in which to look for evidence of hydrocarbons in economic quantities - the mud and the cuttings.
Discussion
It has been noted that it is the presence and magnitude of the show of methane that is the most crucial factor in surface data logging interpretations. However, the magnitude of this show has been the cause of some misunderstandings in the past. Both the hot wire log of gas hydrocarbons and the percent-of-gas log obtained with the conventional gas trap and the gas chromatograph indicate only that the reservoir in question contains hydrocarbons. There is no accurate method to determine the quantitative amounts of the various hydrocarbons in the formation. We are of the opinion that, using a Steam Still, chromatograph, and Show Evaluation Report, a more complete and valid picture of the commercial possibilities of any formation can be gleaned from the gas in the mud, rather than any residual fluid in the cuttings. The method, briefly, is as follows:
E-2
Appendix E
A 5-ml sample of mud is taken from as deep in the possum belly as possible close to the flow line - and injected into the mixing chamber of the Steam Still, where it is swept by steam vapor about 2000 times its own volume. The mud is completely stripped of its hydrocarbon content, which collects in the condenser, whence it is extracted by another syringe and injected into the chromatograph where the various components are accurately measured. The aeration gas trap used in most logging operations is an efficient machine for removing most of the methane from the drilling fluid. However, the heavier the hydrocarbon, the more inefficient the gas trap becomes. Variables such as, type of drilling mud, viscosity and density of mud, unstable power supplies, drop-out and absorption in the trap sampling line, and effects of temperature, humidity, etc., all have an adverse affect upon the gas trap. The efficiency of this system, under these conditions is anywhere from 15 to 70% Therefore, the aeration trap method is useful for establishing trends but not evaluating shows effectively. The Steam Still and chromatograph will give a very accurate qualitative and quantitative reading of the volatile hydrocarbons in the drilling mud - methane, ethane, propane, iso - and normal butane, and iso- and normal pentane. The Steam Still is a very efficient method of doing this, is easy and simple to operate, and requires only basic maintenance. Because the cuttings of the formation produce virtually all the producible gases they contain into the mud, it is a reasonable assumption that this same formation, if completed, would produce gases of similar composition. Now, it has been observed that it is the composition, rather than the magnitude of the gases in a show that is the important factor in terms of formation productivity. The relationship of methane to the heavier hydrocarbon gases is indicative of gas, oil or water productive potential, and in some cases, the reservoir permeability. A comparison of the ratios of methane to the heavier hydrocarbons from producing wellheads and PPM logs produce a striking similarity. In this similarity lies the evidence that the hydrocarbon ratio pattern can be used to predict the productive potential of a reservoir. The Show Evaluation Report is a simple, easy-to-interpret graphical method of outlining these ratios and is usually done for every significant gas show. So far as the cuttings gas goes, however, the situation is somewhat different. Although they undergo a normal production cycle during their travel up the hole in the drilling mud, the cuttings, on arrival at the surface, still contain a certain amount of the original oil and/or gas. These interstitial fluids will be contained within the pore-spaces of the cuttings, and will be released by the chopping and beating action of the blender.
E-3
Appendix E
It is obvious that most of the volatile hydrocarbons will already have been produced from the cuttings by the time they reach the logging unit, unless the rock has very low permeability. However, much of the heavier liquid hydrocarbons will remain and, in fact, most Cuttings Gas Detectors can detect up to about C18. It is thus primarily an oil detector, and its readings are representations of the oil content of the cuttings obtained instrumentally - independent, that is, of human judgement.
We believe, however, that this is not that important - that a visual estimation by a qualified technician of any oil show in the cuttings can be done more easily, and to sufficient accuracy. Fluorescence under ultra-violet light with or without the aid of a leaching agent, the colour, nature and speed of the cut, leaching, staining, etc., are all properties that can be easily investigated, and, put together, plotted on the log as a qualitative estimation of the amount of oil in the cuttings. A quantitative estimation has little value in that it is difficult to relate to any production potential, and its accuracy, in any case, is doubtful. Cuttings gas estimation, on the other hand, is useful to a certain extent, but then only as a negative consideration. If you have negligible amounts of volatile gases in the mud, but considerably more in the cuttings, you are looking at a very tight impermeable formation - not exactly a potential reservoir. To sum up then, the oil in the mud is given a qualitative examination by the logger, as is the oil in the cuttings. The gas in the mud is given a qualitative and quantitative examination, resulting in the case of a gas show, in an estimation of some of the economic and productive capabilities of the formation. The remaining gas in the cuttings is, for the above reasons, ignored.
Evaluation Method
We find then that we have available to us a viable method of determining potential production almost immediately upon penetration of the zone of interest. The only significant time lapse is that interval taken for the drilling mud to travel to the surface, in order to be analyzed by the Steam Still and chromatograph. Figure 3 shows the evaluation technique which can be described as follows:
C1 C2
C1 C3
C1 C4
C1 C5
To being with, record only the net increase of each gas over background gas; then calculate and plot the ratios on the ratio lines as indicated. Finally, the show can
E-4
Appendix E
then be evaluated, within bounds for probable production and potential, as indicated by the plot. Productive dry gas zones may show only C1 but abnormally high C1 only shows are usually indicative of salt water. 2. If the C1/C2 ratio is low in the oil section and the C1/C4 ratio is high in the gas section, the zone is probably non-productive. 3. If any ratio (C1/C5 excepted, when oil is used in the mud) is lower than a preceding ratio, the zone is probably non-productive. For example, if C1/C4 is less than C1/C3, the zone is probably water bearing. 4. The ratios may not be definitive for low permeability zones; however, steep ratio plots may indicate tight zones.
1.
Field Examples
Figure 4 shows the hydrocarbon ratio plots from a proven reservoir in the Fox Creek area from a well producing gas from the Lower Cretaceous. The C1/C2 ratio is over 15; the slope of the line is positive and moderately steep, but not overly. This sample of gas was taken from the flareline during a DST. The well tested at 202 mcf/d with gas to surface in 7 minutes. Plot #1 was obtained from analysis of fluid from the Upper Devonian. This well had produced through perforations in the production casing. The C1/C2, for Plot #2a, is slightly under 6. This is the ratio plot of fluid recovered from 3285m - 3304m and is indicative of gas condensate; the slope being moderately steep. The C1/C2, for Plot 2b, is slightly under 5; this is the ratio plot of fluid recovered from 3304m - 3311m and is indicative of light crude. The slight decrease in the latter portion of Plot 2b indicates the zone is possibly wet and may be down-dip. Figure 5 shows plotted hydrocarbon ratios from a well drilled in the Karr Lake area. Plot #1 was made from an analysis of gases from the Lower Cretaceous. The C1/C2 ratio is slightly under 15; the line slope is even, positive and fairly steep. This may indicate poor permeability in this particular formation. Plot #2 was made from analysis of fluid from the Jurassic. The C1/C2 ratio is slightly under 3; the line is fairly flat and low in the oil section and indicates a more viscous crude, which may be water bearing. Both zones, however, appear to be producible with the Lower Cretaceous producing gas-condensate while the Jurassic produces oil and water.
Practical experience shows that if the C1/C2 ratio is above 65, the zone is too tight for commercial production or may only indicate very dry gas (mostly methane).
Figure 6 shows the ratio plots obtained from Alberta and B.C. wells that were non-productive in the zones of interest. The C 1/C2 ratio for Plot 1 is slightly under
E-5
Appendix E
7; this is the ratio plot taken from the PPM log, of a well in B.C. The slope is obviously negative from C1/C3 ratio to C1/C4 ratio and is indicative of a water bearing formation and this turned out to be the case upon logging and testing. The C1/C2 ratio for Plot 2 is slightly over 70; this ratio plot was taken from the PPM log, of a well in B.C. The slope is again negative from C1/C3 to C1/C4 and did, in fact, indicate a water bearing formation. In both of these cases, however, the hot wire gas detector did show a substantial increase over background gas, and the formation would probably have been tested, had not the recommendation from the logging technician, been heeded. At first glance they appear to be reasonable shows, because of the magnitude of the hydrocarbons present, and the relatively good porosity of the sandstone indicated by the wellsite geologist. However, after show reports were accomplished, it is quite evident that if tested, the results would have indicated water bearing formations.
Applications
The PPM log is only one of several tools that can be used for formation evaluation. However, often times it can furnish the vital information necessary to make the final decision on a well. The author was on the wells indicated by Figure 1 and Figure 2 when totally unexpected shows of gas were encountered in the Lower Cretaceous. The total gas readings showed no great change in the amount of gas present, and the drill rate was not affected except for an anomaly at the top of the formation. When the Steam Still analysis was performed, the amount of heavy hydrocarbons increased greatly while the methane content decreased. Drilling continued until the total gas readings returned to normal, (as shown in Figure 2). Ratio plots were made as drilling progressed through the section noted and the results were relayed to the wellsite geologist. A Drill Stem Test was deemed necessary and one was performed with the results shown on the PPM log in Figure 2. A ratio plot was made later of the fluid recovered from the DST and confirmed the ratio plot made slightly after penetration of the formation. The result is noted by Figure 1. Plot 1 shows the ppm ratio performed by the logger minutes after the formation was penetrated, while Plot 2 shows the percentage of gases, realized by analysis from the DST. The interesting thing to note here is that without the PPM log this formation could have been passed by until the electric log analysis were available, increasing the chances of damaging the formation, etc. Another example of the PPM log and ratio plot is taken from a well drilled in Western Canada. Many sands and carbonates were noted showing the presence of hydrocarbons. The logging crew submitted over several ratio plots to the operator during the drilling of the well. In most instances it turned out that DST and electric log analysis substantiated their findings. Figure 7 shows a plot of the PPM log and subsequent ratio plots with a corresponding table.
E-6
Appendix E
Conclusions
As we have observed by the preceding examples, only qualitative trends can be established by using the conventional aeration gas trap and hot wire or hydrogen flame gas detector. If chromatography is used, we are then able to get an idea of the rough composition, in a gas to air ratio, and this is unsatisfactory for formation evaluation. Such hydrocarbons are usually reported as units of gas or parts per million of gas in the gas-air mixture. We are only seeing the presence, in relative amounts, not the actual quantity of the hydrocarbons in the mud. In this case, it is necessary to have other forms of formation evaluation in order to deal with productive formation potential. However, if an accurate account of the in-place hydrocarbons is made available, then more meaningful ratio plots of gases can be made. Even though several factors affect the reservoir potential and its productive capabilities; these factors can be lessened by studying the ratio of methane to each of the heavier hydrocarbons. This plot is unique and provides the operator with new and relevant information for evaluating productive possibilities of exploratory and frontier wells. The ratio plot can also be used to aide the operator in his pre-well planning. When a play is to be drilled, information can be gained from the study of wells in the area to be looked at. Ratio plots from the gas analysis records can be drawn up and evaluated beforehand enabling the operator to see what they should be running into before the formation has been drilled. When the formation is drilled, the ratio plots made by the logger can be compared to those made previously, and evaluation conclusions can be drawn from this. Advances are being made daily in the exploration industry and it is hoped that this method, though not new, will aide the operator in effectively completing his well.
E-7
Appendix E
Figure 1
Example 1 illustrates the relationship between a chromatographic analysis taken from two different sources. Plot 1 is the ratio taken from the mud stream on wellsite, while drilling is in operation. Plot 2 is the ratio, taken from the formation, of the gas sampled from the DST. The two plots are taken from the lower section of the Basa-Gething. The interesting point to note is that Plot 1 was obtained while drilling was in operation. Within reason, the quality of gas, oil, salt water, etc., can be determined before an expensive testing operation is run. Lithological data is determined, at the same time, by the wellsite geologist; therefore, if money or time is a factor, in well completion, this method can be a viable alternative to expensive testing or logging runs, while drilling.
E-8
Appendix E
Figure 2
The PPM log is obtained through a quantitative analysis of the hydrocarbons in the drilling fluids, specifically parts per million of the lighter hydrocarbons in the fluid. The total PPM log is the summation of the following factors: a drilling rate column which plots penetration rate inversely to the gas, on a linear scale, a visual porosity column, a lithology column which shows either an interpretive or percentage lithology; five
E-9
Appendix E
columns showing C1 through C5 in ppm; a remarks column; and finally a column detailing lithological descriptions.
Figure 3
E-10
Appendix E
Figure 4
E-11
Appendix E
Figure 5
Hydrocarbon ratio plots, productive reservoirs, obtained from wellhead analysis data, Karr Lake area.
E-12
Appendix E
Figure 6
Hydrocarbon ratio plots, non-productive reservoirs, Alberta, B.C. - analysis from gas shows on PPM logs.
E-13
Appendix E
Figure 7
E-14
Appendix F
Appendix F
Factors Affecting the Surface Expression of Hydrocarbon Shows (T. George Paper)
Introduction
The surface expression of hydrocarbon shows as detected by surface data logging techniques is greatly influenced by both surface and downhole factors. These factors are readily understandable on their own, but are rarely applied to surface data logging shows. This lack of application of common occurrences often leads to a misinterpretation of surface data logging show evaluation at best, and at worst to a total disregard for valid show evaluation data. This paper explains how these common surface and subsurface factors influence surface data logging shows and why there are often discrepancies in show evaluation data collected by other means, wireline, cores, sidewall cores, drill stem test, and retrievable formation test devices. Topics include:
1. 2. 3. 4. 5.
How hydrocarbons enter the wellbore Formation flushing The source of detectable hydrocarbons Factors affecting hydrocarbon measurements Background gas
In the petroleum industry there seems to exist wide differences of opinion and varying degrees of understanding about how to interpret surface data logging hydrocarbon shows. This conclusion has been reached after conducting numerous employee surface data logging programs and customer seminars on the subject during the past six years. This paper deals with the surface expression of hydrocarbon shows as detected by surface data logging equipment. It contains some explanation as to the reasons for some of the mud log show interpretation problems encountered by geologists and mud loggers.
Basic Definitions
equivalent circulating density (ECD) - the density (ppg) equivalent of pressure exerted by a column of fluid in motion.
F-1
Appendix F
filter cake - the discontinuous solids phase of a drilling mud that is deposited on the wall of a wellbore as a result of filtrate entering permeable formations. The cake consists of primarily of gel, barite, and fine drill solids; also called wall cake. filtrate - the continuous, liquid phase of a drilling mud; can be oil (diesel, crude, mineral, etc.) or water (salt or fresh). flushing - the process where the filtrate or liquid phase of the drilling mud enters permeable formations and displaces formation pore fluids back into the formation. gas unit - an arbitrary ratio of gaseous hydrocarbons to air as measured by a hot-wire or other style surface gas detector. hydrostatic head - the pressure exerted by a column of fluid at rest. overbalanced - when the effective pressure of the drilling mud column exceeds the pore pressure of the formation. permeability - the ability of a porous medium to allow fluids to pass through it. pore - the void space in a rock, filled with either gas, oil or water (salt or fresh). pore pressure - pressure of the fluid (gas, oil, or water) in a pore. show - with regard to surface data logging, an exhibition of either gaseous or liquid hydrocarbons at the surface in either the cuttings or drilling mud. underbalanced - when the effective pressure of the drilling mud column is less than the pore pressure of the formation.
Primary
Generally, a well is drilled in an overbalanced condition, usually in the range of .3 to .5 ppg. This is done primarily to prevent the uncontrolled influx of formation pore fluids into the mud column and secondarily to prevent the walls of the wellbore from caving in. As the drill bit penetrates the formation, rock chips or
F-2
Appendix F
cuttings are dislodged from the bottom of the hole. As cuttings are generated by the bit, some formation pore fluids enter the drilling mud from exposed pores. While in this overbalanced condition, the only other method by which pore fluids can enter the mud column is when the cuttings reach a point in their travel up the annulus where the hydrostatic pressure exerted on them by the mud column is less than the pore pressure of the fluid inside them. When the cuttings reach this point, any gaseous fluids still inside them will begin to expand and enter the surrounding drilling mud. Liquids in the form of oil or water have a very low coefficient of expansion and essentially remain inside the cuttings unless they are pushed out by expanding gas. (Gas readings may also be generated from formation fragments that slough off the wall of the hole, but they are discounted in this discussion.)
Secondary
Oftentimes, small amounts of hydrocarbons or formation water may enter the well directly from the formation during a connection or a trip. This only occurs, however, when the effective hydrostatic pressure of the mud is temporarily reduced (swab pressure is created and ECD is removed) such that a momentary underbalanced condition is artificially created.
Formation Flushing
Also of importance in understanding surface data logging shows is the process of flushing or filtration. This process is controlled by a number of parameters.
1. 2. 3. 4. 5.
The amount of pressure overbalance The type of mud filtrate (oil or water). Whether the mud has been chemically treated to control the water loss. The wall cake building capabilities of the mud (polymer or gel). The type of pore fluids (gas, oil, or water).
Formation Permeability
Of primary consideration with regard to flushing is the formation permeability.
F-3
Appendix F
Impermeable Formations
An impermeable formation will not allow filtrate to penetrate and thereby displace formation pore fluids. When this occurs, generally there is no filter cake on the wall of the hole or only a very thin one. This results from the fact that no filtrate moved into the rock to leave the solids behind as filter cake.
Permeable Formations
If the formation is permeable, mud filtrate will enter it and displace pore fluids away from the wellbore and further back into the formation. This process continues until an impermeable filter cake is deposited. (It is significant to note the wall cake can be 10,000 to 50,000 times as impermeable as the formation on which it is deposited.) This may take only a few minutes or a few hours, or the flushing process may remain essentially dynamic since the wall cake may never become completely impermeable. In brief, when drilling an impermeable formation, the rock itself controls flushing, and when drilling a permeable formation the filter cake becomes the controlling factor. Therefore, when penetrating a permeable formation, wall cake is deposited on the side and bottom of the hole. However, because the drill bit is continually rotating and cutting on the bottom of the hole, the filter cake there is constantly being removed. This causes the rate of filtration and flushing action directly below the bit to remain at an elevated level. (See Figure 1.) On the side of the hole, where the wall cake is relatively free from erosion, the flushing action is controlled by the wall cake itself. The area directly below the bit is where the greatest amount of flushing occurs. This is significant in that it also happens to be the area from which most of the hydrocarbon readings associated with surface data logging come. This is in contrast to the fact that all of the hydrocarbon information from wireline logs, formation tests, sidewall cores or drill stem test, is derived from the area of formation to the side of the hole behind the protective layer of filter cake. Hence, it is one of the reasons there is not always good correlation between mud log shows and non-mud log type shows.
F-4
Appendix F
Figure 1
F-5
Appendix F
Conversely, if a formation does contain hydrocarbons and is characterized by reduced permeability, the likelihood of a surface expression of hydrocarbons is increased since flushing was kept to a minimum. Consequently, a very poor wireline hydrocarbon show, as categorized by a tight section, may exhibit excellent oil and gas show characteristics as detected by the mud logger on the surface. Additionally, the technique of cuttings gas analysis, where a measured quantity of cuttings and water is placed in a blender, agitated and the released hydrocarbons measured by means of a hot wire or equivalent, could also lead one into the trap of thinking that high means good and low means bad. In reality, a high cuttings gas reading could indicate the existence of a very tight hydrocarbon-bearing formation. On the other hand, a low reading could imply that an extremely permeable, hydrocarbon-rich zone had been flushed or that the zone was truly low in hydrocarbon content.
F-6
Appendix F
few parts per million of oil can be detected using U.V. but not if the sun is bright, your eyes are not adequately adjusted to the darkness of the box, the oil is low gravity (minimal U.V. reflectance) and youre in a hurry.
Background Gas
A final area of importance is that of background gas. Usually, background gas is defined as the normal flowline gas reading under average ROP conditions. The problem is how the background gas is used in regards to describing a show. Oftentimes, the logger is directed to subtract the average background gas reading from the gas reading measured during a show. Once again recalling the circumstances under which hydrocarbons enter the drilling mud, it is quickly recognizable why this is a faulty practice. To further the point, consider the following example.
F-7
Appendix F
Example
Assume the bit is penetrating a shale section and the surface background gas reading is 30 gas units. Drilling proceeds until the bit is just at the boundary of a show zone; the driller picks the bit up off bottom but continues to circulate the shale cuttings to the surface. Upon bottoms-up, the gas reading goes to zero since no more pores are being opened on the bottom of the hole and no more cuttings are releasing their pore fluids to the drilling mud. Usually, the only circumstances where the gas reading would not go to zero would be if the gas was being recirculated or if the well was being drilled in an underbalanced condition. Obviously, drilling a permeable hydrocarbon-bearing formation in an underbalanced condition would allow the formation in situ to bleed hydrocarbons into the well regardless of the lack of drilling and cuttings. (This in fact is an excellent method to determine the balance condition.) Once the surface gas reading has dropped to zero, the driller is instructed to begin making hole again and the bit immediately penetrates the show zone. Upon bottoms-up this time the gas reading shows 200 units. In no way does the 30 units of background gas from the shale have anything to do with the 200 unit reading in the show zone and should therefore not be subtracted.
Gas in the suction pit could also cause a misinterpretation of the balance condition if it were determined using the gas versus cuttings method described above.
F-8
Appendix F
Conclusion
In conclusion, there are many factors to consider when interpreting shows, many of which have not been discussed in this article. However, experience shows the following factors to be of paramount importance:
1. 2. 3. 4. 5.
How hydrocarbons enter the wellbore. Formation flushing. The source of detectable hydrocarbons. Factors affecting hydrocarbon measurement. Background gas.
Hopefully, this information will provide answers to some of the questions and bothersome problems encountered when reading and interpreting mud log shows.
F-9
Appendix F
F-10
Appendix G
Appendix G
A
Glossary
abandon v: to cease producing oil and gas from a well when it becomes unprofitable. A wildcat well may be abandoned after it has proven nonproductive. Several steps are involved in abandoning a well: part of the casing is removed and salvaged; one or more cement plugs are placed in the borehole to prevent migration of fluids between the different formations penetrated by the borehole; and the well is abandoned. In many states, it is necessary to secure permission from official agencies before a well may be abandoned. absolute permeability n: a measure of the ability of a single fluid (as water, gas, or oil) to flow through a rock formation when the formation is totally filled (saturated) with the single fluid. The permeability measure of a rock filled with a single fluid is different from the permeability measure of the same rock filled with two or more fluids. Compare effective permeability. acid fracture v: to part or open fractures in productive, hard-limestone formations by using a combination of oil and acid or water and acid under high pressure. See formation fracturing. acidize v: to treat oil-bearing limestone or other formations, using a chemical reaction with acid, to increase production. Hydrochloric or other acid is injected into the formation under pressure. The acid etches the rock, enlarging the pore spaces and passages through which the reservoir fluids flow. The acid is held under pressure for a period of time and then pumped out, and the well is swabbed and put back into production. Chemical inhibitors combined with the acid prevent corrosion of the pipe. adjustable choke n: a choke in which a conical needle and seat vary the rate of flow. See choke. air-actuated adj: powered by compressed air, as the clutch and brake system in drilling equipment. air drilling n: a method of rotary drilling that uses compressed air as the circulation medium. The conventional method of removing cuttings from the wellbore is to use a flow of water or drilling mud. Compressed air removes the cuttings with equal or greater efficiency. The rate of penetration is usually increased considerably when air drilling is used. However, a principal problem in air drilling is the penetration of formations containing water, since the entry of water into the system reduces the ability of the air to remove the cuttings. American Petroleum n: 1. founded in 1920, this national oil trade organization is the leading Institute standardizing organization on oil field drilling and producing equipment. It maintains departments of transportation, refining, and marketing in Washington, D.C., and a department of production in Dallas. 2. (slang) indicative of a job being properly or thoroughly done (as, His work is strictly API). 3. degrees API, used to designate API gravity. See API gravity. angle of deflection n: in directional drilling, the angle, expressed in degrees, at which a well is deflected from the vertical by a whipstock or other deflecting tool. See whipstock. annular blowout n: a large valve, usually installed above the ram preventers, that forms a seal in the preventer annular space between the pipe and wellbore or, if no pipe is present, on the wellbore itself. Compare ram blowout preventer.
G-1
Appendix G
annular space n: 1. the space surrounding a cylindrical object within a cylinder. 2. the space around a pipe in a wellbore, the outer wall of which may be the wall of either the borehole or the casing; sometimes termed the annulus. anticline n: an arched, inverted-trough configuration of folded and stratified rock layers. Compare syncline. API abbr: American Petroleum Institute. API gravity n: the measure of the density or gravity of liquid petroleum products in the United States, derived from specific gravity in accordance with the following equation: API gravity = 141.5 - 131.5 specific gravity API gravity is expressed in degrees, a specific gravity of 1.0 being equivalent to 10o API.
B
back off v: to unscrew one threaded piece (as a section of pipe) from another. back up v: to hold one section of an object (as pipe) while another is being screwed into or out of it. bail n: a cylindrical steel bar (similar to the handle or bail of a bucket, only much larger) that supports the swivel and connects it to the hook. Sometimes, the two cylindrical bars that support the elevators and attach them to the hook are called bails. v: to recover bottomhole fluids, samples, or drill cuttings by lowering a cylindrical vessel called a bailer to the bottom of a well, filling it and retrieving it. See bailer. bailer n: a long cylindrical container, fitted with a valve at its lower end, used to remove water, sand, mud or oil from a well. bailing line n: cable attached to the bailer, passed over a sheave at the top of the derrick, and spooled on a reel. See sheave. barge n: any one of many types of flat-decked, shallow draft vessels, usually towed by a boat. A complete drilling rig may be assembled on a drilling barge, which usually is submersible; that is, it has a submersible hull or base that is flooded with water at the drilling site. Drilling equipment, crew quarters, and so forth are mounted on a superstructure above the water level. barite or baryte n: barium sulfate BaSO4; a mineral used to increase the weight of drilling mud. Its specific gravity is 4.2 (i.e., it is 4.2 times heavier than water). See barium sulfate and mud. barium sulfate n: 1. a chemical combination of barium, sulfur, and oxygen. Also called barite. See barite. 2. a tenacious scale that is very difficult to remove. barrel n: a measure of volume for petroleum products. One barrel is the equivalent of 42 U.S. gallons or 0.15899 cubic meters. One cubic meter equals 6.2897 barrels. basket sub n: a fishing accessory run above a bit or mill to recover small pieces of metal or junk in a well. bed n: a specific layer of earth or rock in contrast to other layers of different material lying above, below, or adjacent to it.
G-2
Appendix G
belt n: a flexible band or cord connecting and passing about each of two or more pulleys to transmit power or impart motion. bit n: the cutting or boring element used in drilling oil and gas wells. The bit consists of the cutting element and the circulating element. The circulating element permits the passage of drilling fluid and utilizes the hydraulic force of the fluid stream to improve drilling rates. In rotary drilling, several drill collars are joined at the bottom end of the drill string. The bit is attached to the end of the drill collar. Most bits used in rotary drilling are roller cone bits. bit breaker n: a heavy plate that fits in the rotary table and hold the drill bit while it is being made up in or broken out of the drill string. See bit. bit record n: a report on each bit used in a drilling operation that lists the bit type, the amount of footage the bit has drilled, and the nature of the formation penetrated. blind ram n: an integral part of a blowout preventer that serves as the closing element. Its ends do not fit around the drill pipe but seal against each other and shut off the space below completely. block n: any assembly of pulleys on a common framework; in mechanics one or more pulleys, or sheaves, mounted to rotate on a common axis. The crown block is an assembly of sheaves mounted on beams at the top of the derrick. The drill line is reeved over the sheaves of the crown block alternately with the sheaves of the traveling block, which is hoisted and lowered in the derrick by the drill line. When the elevators are attached to a hook on the traveling block, and when drill pipe is latched in the elevators, the pipe can be raised or lowered in the derrick. See crown block, elevator, hook, reeve, sheave, and traveling block; also see drilling block. blooey line n: the discharge pipe from a well being drilled by air drilling. The blooey line is used to conduct the air or gas used for circulation away from the rig to reduce the fire hazard as well as to transport the cuttings a suitable distance from the well. See air drilling. blowout n: an uncontrolled flow of gas, oil, or other well fluids into the atmosphere. A blowout, or gusher, occurs when formation pressure exceeds the pressure applied to it by the column of drilling fluid. A kick warns of an impending blowout. See formation pressure, gusher, and kick. blowout preventer n: one of several valves installed at the wellhead to prevent the escape of pressure either in the annular space between the casing and drill pipe or in open hole (i.e., hole with no drill pipe) during drilling or completion operations. Blowout preventers on land rigs are located beneath the rig at the lands surface; on jackup or platform rigs, they are located at the waters surface; and on floating offshore rig, on the seafloor. See annular blowout preventer, inside blowout preventer, and ram blowout preventer. boll-weevil n: (slang) an inexperienced rig or oil-field worker, sometimes shortened to weevil. bomb n: a thick-walled container, usually steel, used to hold sample of oil or gas under pressure. See bottom-hole pressure. bond n: the state of one material adhering or being joined to another material (as cement to formation). v: to adhere or be joined to another material. BOP abbr: blowout preventer.
G-3
Appendix G
borehole n: the wellbore; the hole made by drilling or boring. See wellbore. bottomhole n: the lowest or deepest part of a well. adj: pertaining to the bottom of the wellbore. bottomhole choke n: a device with a restricted opening placed in the lower end of the tubing to control the rate of flow. See choke. bottomhole pressure n: 1. the pressure at the bottom of a borehole. It is caused by the hydrostatic pressure of the drilling fluid in the hole and, sometimes, any back-pressure held at the surface as when the well is shut in with blowout preventers. When mud is being circulated, bottomhole pressure is the hydrostatic pressure plus the remaining circulating pressure required to move the mud up the annulus. 2. the pressure in a well at a point opposite the producing formation, as recorded by a bottomhole pressure bomb. box n: the female section of a tool joint. See tool joint. brake n: a device for arresting the motion of a mechanism, usually by means of friction, as in the draw works brake. Compare electrodynamic brake and hydromatic brake. break out v: 1. to unscrew one section of pipe from another section, especially drill pipe while it is being withdrawn from the wellbore. During this operation, the tongs are used to start the unscrewing operation. See tongs. 2. to separate, as gas from liquid. breakout cat-head n: a device attached to the shaft of the draw works that is used as a power source for unscrewing drill pipe; usually located opposite the drillers side of the draw works. See cat-head. breakout tongs n: tongs that are used to start unscrewing one section of pipe from another section, especially drill pipe coming out of the hole. Also called lead tongs. See tongs. bring in a well v: to complete a well and put it in producing status. buck up v: to tighten up a threaded connection (as two joints of drill pipe). bullet perforator n: a tubular device that, when lowered to a selected depth within a well, fires bullets through the casing to provide hole through which the well fluids may enter.
C
cable n: a rope of wire, hemp, or other strong fibers. See wire-rope. cable-tool drilling n: a drilling method in which the hole is drilled by dropping a sharply pointed bit on the bottom of the hole. The bit is attached to a cable, and the cable is picked up and dropped, picked up and dropped, over and over, as the hole is drilled. cap rock n: 1. impermeable rock overlying an oil or gas reservoir that tends to prevent migration of oil or gas out of the reservoir. 2. the porous and permeable strata overlying salt domes that may serve as the reservoir rock. cased adj: pertaining to a wellbore in which casing is run and cemented. See casing. cased hole n: a wellbore in which casing has been run. casing n: steel pipe placed in an oil or gas well as drilling progresses to prevent the wall of the hole from caving in during drilling and to provide means of extracting petroleum if the well is productive.
G-4
Appendix G
casing centralizer n: a device secured around the casing at regular intervals to center it in the hole. Casing that is centralized allows a more uniform cement sheath to form around the pipe. casing coupling n: a tubular section of pipe that is threaded inside and used to connect two joints of casing. casing elevator n: See elevator. casing head n: a heavy steel, flanged fitting that connects to the first string of casing and provides a housing for the slips and packing assemblies by which intermediate strings of casing are suspended and the annulus sealed off. Also called a spool. See annular space. casing shoe n: also called a guide shoe. See guide shoe. casing string n: the entire length of all the joints of casing run in a well. Casing is manufactured in lengths of about 30 feet, each length or joint being joined to another as casing is run in a well. See combination string. catch samples v: to obtain cuttings for geological information as formation are penetrated by the bit. The samples are obtained from drilling fluid as it emerges from the wellbore or, in cable-tool drilling, from the bailer. Cuttings are carefully washed until they are free of foreign matter, dried, and labeled to indicate the depth at which they were obtained. See bailer, cable-tool drilling, and cuttings. cat-head n: a spool-shaped attachment on a winch around which rope for hoisting and pulling is wound. See breakout cat-head and makeup cat-head. cat-line n: a hoisting or pulling line powered by the cat-head and used to lift heavy equipment on the rig. See cat-head. caving n: collapse of the walls of the wellbore, also called sloughing. cellar n: a pit in the ground to provide additional height between the rig floor and the wellhead to accommodate the installation of blowout preventers, rat hole, mouse hole, and so forth. It also collects drainage water and other fluids for subsequent disposal. cement casing v: to fill the annulus between the casing and hole with cement to support the casing and prevent fluid migration between permeable zones. cement channeling n: an undesirable phenomenon that can occur when casing is being cemented in a borehole. The cement slurry fails to rise uniformly between the casing and borehole wall, leaving spaces void of cement. Ideally, the cement should completely and uniformly surround the casing and form a strong bond to the borehole wall. cementing n: the application of a liquid slurry of cement and water to various points inside and outside the casing. See primary cementing, secondary cementing, and squeeze cementing. chain drive n: a drive system using a chain and chain gears to transmit power. Power transmissions use a roller chain, in which each link is made of side bars, transverse pins, and rollers on the pins. A double roller chain is made of two connected rows of links, a triple roller chain of three, and so forth. chain tongs n: a tool consisting of a handle and releasable chain used for turning pipe or fittings of a diameter larger than that which a pipe wrench would fit. The chain is looped and tightened around the pipe or fitting, and the handle is used to turn the
G-5
Appendix G
tool so that the pipe of fitting can be tightened or loosened. check valve n: a valve that permits flow in one direction only. choke n: an orifice installed in a line to restrict the flow and control the rate of production. Surface chokes are part of the Christmas tree and contain a choke nipple, or bean, with a small-diameter bore that serves to restrict the flow. Chokes are also used to control the rate of flow of the drilling mud out of the hole when the well is closed in with the blowout preventer and a kick is being circulated out of the hole See adjustable choke, blowout preventer, bottomhole choke, Christmas tree, kick, nipple, and positive choke. choke line n: an extension of pipe from the blowout preventer assembly used to direct well fluids from the annulus to the choke manifold. choke manifold n: the arrangement of piping and special valves, called chokes, through which drilling mud is circulated when the blowout preventers are closed to control the pressures encountered during a kick. See choke and blowout preventer. Christmas tree n: the control valves, pressure gauges, and chokes assembled at the top of a well to control the flow of oil and gas after the well has been drilled and completed. circulate v: to pass from one point throughout a system and back to the starting point. For example, drilling fluid is circulated out of the suction pit, down the drill pipe and drill collars, out the bit, up the annulus, and back to the pits. circulation n: the movement of drilling fluid out of the mud pits, down the drill string, up the annulus, and back to the mud pits. combination string n: a casing string that has joints of various collapse resistance, internal yield strength, and tensile strength designed for various depths in a specific well to best withstand the conditions of that well. In deep wells, high tensile strength is required in the top casing joints to carry the load, whereas high collapse resistance and internal yield strength are needed for the bottom joints. In the middle of the casing, average qualities are usually sufficient. The most suitable combination of types and weights of pipe helps to ensure efficient production at a minimum cost. come out of the hole v: to pull the drill string out of the wellbore. This withdrawal is necessary to change the bit, change from a core barrel to the bit, run electric logs, prepare for a drill stem test, run casing, and so on. company man n: also called company representative, See company representative. company representative n: an employee of an operating company whose job is to represent the companys interest at the drilling location. complete a well v: to finish work on a well and bring it to productive status. See well completion. compound n: a mechanism used to transmit power from the engines to the pump, draw works, and other machinery on a drilling rig. It is composed of clutches, chains and sprockets, belts and pulleys, and a number of shafts, both driven and driving. v: to connect two or more power-producing devices (as engines) to run one piece of driven equipment (as the draw works). conductor pipe n: a short string of large-diameter casing used to keep the top of the wellbore open and to provide a means of conveying the up-flowing drilling fluid from the wellbore to the mud pit. contract depth n: the depth of the wellbore at which the drilling contract is fulfilled.
G-6
Appendix G
core n: a cylindrical sample taken from a formation for geological analysis. Usually a conventional core barrel is substituted for the bit and procures a sample as it penetrates the formation. See also sidewall coring. v: to obtain a formation sample for analysis. core analysis n: laboratory analysis of a core sample to determine porosity, permeability, lithology, fluid content, angle of dip, geological age, and probable productivity of the formation. core barrel n: a tubular device from 25 to 60 feet long run at the bottom of the drill pipe in place of a bit to cut a core sample. core catcher n: the part of the core barrel that holds the formation sample. core cutterhead n: the cutting element of the core barrel assembly. In design it corresponds to one of the three main types of bits: drag bits with blades for cutting soft formations; roller bits with rotating cutting for cutting medium formations; and diamond bits for cutting very hard formations. coupling n: 1. in piping, a metal collar with internal threads used to join two sections of thread pipe. 2. in power transmission, a connection extending longitudinally between a driving shaft and a driven shaft. Most such couplings are flexible and compensate for minor misalignment of the two shafts. crooked hole n: a wellbore that has deviated from the vertical. It usually occurs in areas where the subsurface formations are difficult to drill, such as a section of alternating hard and soft strata steeply inclined from the horizontal. crown block n: an assembly of sheaves or pulleys mounted on beams at the top of the derrick over which the drill line is reeved. See block, reeve, and sheave. cuttings n pl: the fragments of rock dislodged by the bit and brought to the surface in the drilling mud. Washed and dried samples of the cuttings are analyzed by geologists to obtain information about the formations drilled.
D
daylight tour n: (pronounced tower) the shift of duty on a drilling rig that starts at or about daylight; also called morning tour. Compare evening tour and graveyard tour. deadline n: the drill line from the crown block sheave to the anchor, so called because it does not move. Compare fastline. deadline tie-down n: a device to which the deadline is attached, securely fastened to the mast or anchor derrick substructure. Also called a deadline anchor. degasser n: the equipment used to remove unwanted gas from a liquid, especially from a drilling fluid. density n: the mass or weight of a substance; often expressed in weight per unit volume. For instance, the density of a drilling mud may be 10 pounds per gallon (ppg), 74.8 pounds per cubic foot (lb/ft3), or 1,198.2 kilograms per cubic meter (kg/m3). Specific gravity and API gravity are other units of density. See API gravity and specific gravity. derrick n: a large load-bearing structure, usually of bolted construction. In drilling, the standard derrick has four legs standing at the corners of the substructure and reaching to the crown block. The substructure is an assembly of heavy beams used to elevate the derrick and provide space to install blowout preventers, casing
G-7
Appendix G
heads, and so forth. Because the standard derrick must be assembled piece by piece, it has largely been replaced by the mast, which can be lowered and raised without disassembly. See crown block, mast, and substructure. derrickman n: the crew member who handles the upper end of the drill string as it is being hoisted out of or lowered into the hole. He is also responsible for the conditioning of the drilling fluid and circulating machinery. de-sander n: a centrifugal device for removing sand from drilling fluid to prevent abrasion of the pumps. It may be operated mechanically or by a fast-moving stream of fluid inside a special cone-shaped vessel, in which case it is sometimes called a hydrocyclone. See de-silter. de-silter n: a centrifugal device for removing very fine particles, or silt, from drilling fluid to keep the amount of solids in the fluid to the lowest possible point. Usually, the lower the solids content of mud, the faster the rate of penetration. It works on the same principle as a de-sander. Compare de-sander. development well n: 1. a well drilled in proven territory in a field to complete a pattern of production. 2. an exploitation well. See exploitation well. deviation n: the inclination of the wellbore from the vertical. The angle of deviation, angle of drift, or drift angle is the angle in degrees that shows the variation from the vertical as revealed by a deviation survey. See deviation survey. deviation survey n: an operation made to determine the angle from which a bit has deviated from the vertical during drilling. There are two basic deviation survey, or drift survey, instruments: one reveals the angle of deviation only; the other indicates both the angle and direction of deviation. diamond bit n: a drilling bit that has a steel body surfaced with industrial diamonds. Cutting is performed by the rotation of the very hard diamonds over the rock surface. diesel-electric power n: the power supplied to a drilling rig by diesel engines driving electric generators, used widely offshore and gaining popularity onshore. diesel engine n: a high-compression, internal-combustion engine used extensively for powering drilling rigs. In a diesel engine, air is drawn into the cylinders and compressed to very high pressures; ignition occurs as fuel is injected into the compressed and heated air. Combustion takes place within the cylinder above the piston, and expansion of the combustion products imparts power to the piston. directional drilling n: intentional deviation of a wellbore from the vertical. Although wellbores are normally drilled vertically, it is sometimes necessary or advantageous to drill at an angle from the vertical. Controlled directional drilling makes it possible to reach subsurface areas laterally remote from the point where the bit enters the earth. It involves the use of turbo-drills, Dyna-Drills, whipstocks, or other deflecting tools. See Dyna-Drill, turbo-drill, and whipstock. discovery well n: the first oil or gas well drilled in a new field; the well that reveals the presence of a petroleum-bearing reservoir. Subsequent wells are development wells. Compare development well. displacement fluid n: in oil well cementing, the fluid, usually drilling mud or salt water, that is pumped into the well after the cement to force the cement out of the casing and into the annulus. doghouse n: 1. a small enclosure on the rig floor used as an office for the driller or as a storehouse for small objects. 2. any small building used as an office or for storage.
G-8
Appendix G
double n: a length of drill pipe, casing, or tubing consisting of two joints screwed together. Compare thribble and fourble. See joint. double-board n: the name used for the working platform of the derrickman, or monkey board, when it is located at a height in the derrick or mast equal to two lengths of pipe joined together. Compare fourble board and thribble board. See monkey board. draw works n: the hoisting mechanism on a drilling rig. It is essentially a large winch that spools off or takes in the drill line and thus raises or lowers the drill string and bit. drill bit n: the cutting or boring element used for drilling. See bit. drill collar n: a heavy, thick-walled tube, usually steel, used between the drill pipe and the bit in the drill string. Drill collars are used to put weight on the bit so that the bit can drill. driller n: the employee directly in charge of a drilling rig and crew. His main duty is operation of the drilling and hoisting equipment, but he is also responsible for the operation of downhole tools, and pipe measurement. drilling block n: a lease or a number of leases of adjoining tracts of land that constitute a unit of acreage sufficient to justify the expense of drilling a wildcat. drilling contractor n: an individual or group of individuals that own a drilling rig or rigs and contract their services for drilling wells to a certain depth. drilling crew n: a driller, derrickman, and two or more helpers who operate a drilling rig for one tour each day. See derrickman, driller, and tour. drilling fluid n: circulating fluid, one function of which is to force cuttings out of the wellbore and to the surface. While a mixture of clay, water, and other chemical additives is the most common drilling fluid, wells can also be drilled using air, gas, or water as the drilling fluid. Also called circulating fluid. See mud. drilling foreman n: the supervisor of drilling operations on a rig; also the tool pusher or superintendent. drill line n: a wire rope used to support the drilling tools. drilling rate n: the speed with which the bit drills the formation; usually called the rate of penetration. drilling rig n: See rig. drill pipe n: the heavy seamless tubing used to rotate the bit and circulate the drilling fluid. Joints of pipe 30 feet long are coupled together by means of tool joints. drill ship n: a ship constructed to permit a well to be drilled from it at an offshore location. While not as stable as other floating structures (as a semisubmersible), drill ships, or ship shapes, are capable of drilling exploratory wells in relatively deep waters. They may have a ship hull, a catamaran hull, or a trimaran hull. See semisubmersible drilling rig. drill stem n: all members in the assembly used for drilling by the rotary method from the swivel to the bit, including the kelly, drill pipe and tools joints, drill collars, stabilizers, and various subsequent items. Compare drill string. drill-stem test n: a method of gathering data on the potential productivity of a formation before installing casing in a well. See formation testing.
G-9
Appendix G
drill string n: the column, or string, of drill pipe with attached tool joints that transmits fluid and rotation power from the kelly to the drill collars and bit. Often, especially in the oil field, term is loosely applied to include both drill pipe and drill collars. Compare drill stem. drum n: 1. a cylinder around which wire rope is wound in the draw works. The draw works drum is that part of the hoist upon which the drill line is wound. 2. a steel container of general cylindrical form. Refined products are shipped in steel drums with capacities of about 50 to 55 U.S. gallons (about 200 liters). DST abbr: drill-stem test Dyna-Drill n: a downhole motor driven by drilling fluid that imparts rotary motion to a drilling bit connected to the tool, thus eliminating the need to turn the entire drill string to make hole. The Dyna-Drill, a trade name, is used in straight and directional drilling. dynamic positioning n: a method by which a floating offshore drilling rig is maintained in position over an offshore well location. Generally, several motors called thrusters are located on the hull(s) of the structure and are actuated by a sensing system. A computer to which the system feeds signals then directs the thrusters to maintain the rig on location.
E
effective permeability n: a measure of the ability of a single fluid to flow through a rock when the pore spaces of the rock are not completely filled or saturated with the fluid. Compare absolute permeability and relative permeability. electric well log n: a record of certain electrical characteristics of formation traversed by the borehole, made to identify the formation, determine the nature and amount of fluids they contain, and estimate their depth. Also called an electric log or electric survey. electrodynamic brake n: a device mounted on the end of the draw works shaft of a drilling rig. The electrodynamic brake (sometimes called a magnetic brake) serves as an auxiliary to the mechanical brake when pipe is lowered into a well. The braking effect in an electrodynamic brake is achieved by means of the interaction of electric currents with magnets, with other currents, or with themselves. elevator n: a set of clamps that grips a stand, or column of casing, tubing, or drill pipe so that the stand can be raised or lowered into the hole. evening tour n: (pronounced tower) the shift of duty on a drilling rig that starts in the afternoon and runs through the evening. Compare daylight tour and graveyard tour. exploitation well n: a well drilled to permit more effective extraction of oil from a reservoir. It is sometimes called a development well. See development well. exploration well n: a wildcat. See wildcat.
F
fastline n: the end of the drill line that is affixed to the drum or reel of the draw works, so called because it travels with greater velocity than any other portion of the line. Compare deadline.
G-10
Appendix G
fault n: a break in subsurface strata. Often strata on one side of the fault line has been displaced (upward, downward, or laterally) relative to its original positions. field n: a geographical area in which a number of oil and gas wells produce from a continuous reservoir. A field may refer to surface area only or to underground productive formations as well. In a single field, there may be several separate reservoirs at varying depths. fill the hole v: to pump drilling fluid into the wellbore while the pipe is being withdrawn in order to ensure that the wellbore remains full of fluid even though the pipe is withdrawn. Filling the hole lessens the danger of blowout or of caving of the wall of the wellbore. filter cake n: 1. compacted solid or semisolid material remaining on a filter after pressure filtration of mud with the standard filter press. Thickness of the cake is reported in thirty-seconds of an inch or in millimeters. 2. the layer of concentrated solids from the drilling mud that forms on the walls of the borehole opposite permeable formations; also called wall cake or mud cake. fingerboard n: a rack that supports the tops of the stands of pipe being stacked in the derrick or mast. It has several steel finger-like projections that form a series of slots into which the derrickman can set a stand of drill pipe as it is pulled out of the hole. fish n: an object left in the wellbore during drilling operations that must be recovered or drilled around before work can proceed. It can be anything from a piece of scrap metal to a part of the drill string. v: 1. to recover from a well any equipment left there during drilling operations, such as a lost bit or drill collar or part of the drill string. 2. to remove from an older well certain pieces of equipment, such as packers, liners, or screen pipe, to allow reconditioning of the well. fishing tool n: a tool designed to recover equipment lost in the well. float collar n: a special coupling device, inserted one or two joints above the bottom of the casing string, that contains a check valve to permit fluid to pass downward but not upward through the casing. The float collar prevents drilling mud from entering the casing while it is being lowered, allowing the casing to float during its descent, which decreases the load on the derrick. The float collar also prevents a back flow of cement during the cementing operation. floorman n: a drilling crew member whose work station is on the derrick floor. On rotary drilling rigs, there are at least two and usually three or more floormen on each crew. Also called rotary helper and roughneck. fluid n: a substance that flows and yields to any force tending to change it shape. Liquids and gases are fluids. formation n: a bed or deposit composed throughout of substantially the same kinds of rock; a lithologic unit. Each different formation is given a name, frequently as a result of the study of the formation outcrop at the surface and sometimes based on fossils found in the formation. formation fracturing n: a method of stimulating production by increasing the permeability of the producing formation. Under extremely high hydraulic pressure, a fluid (as water, oil, alcohol, dilute hydrochloric acid, liquefied petroleum gas, or foam) is pumped downward through tubing or drill pipe and forced into the perforations in the casing. The fluid enters the formation and parts or fractures it. Sand grains, aluminum pellets, glass beads, or similar materials are carried in suspension by the fluid into the fractures. These are called propping agents or proppants. When
G-11
Appendix G
the pressure is released at the surface, the fracturing fluid returns to the well, and the fractures partially close on the proppants, leaving channels for oil to flow through them to the well. This process is often called a frac job. See propping agent. formation pressure n: the force exerted by fluids in a formation, recorded in the hole at the level of the formation with the well shut in. It is also called reservoir pressure or shut-in bottom-hole pressure. See reservoir pressure and shut-in bottom-hole pressure. formation testing n: the gathering of data on a formation to determine its potential productivity before installing casing in a well. The conventional method is the drill stem test. Incorporated in the drill stem testing tool are a packer, valves or ports that may be opened and closed from the surface, and a pressure-recording device. The tool is lowered to bottom on a string of drill pipe and the packer set, isolating the formation to be tested from the formations above and supporting the fluid column above the packer. A port on the tool is opened to allow the trapped pressure below the packer to bleed off into the drill pipe, gradually exposing the formation to atmospheric pressure and allowing the well to produce to the surface, where the well fluids may be sampled and inspected. From a record of the pressure readings, a number of facts about the formation may be inferred. fourable n: a section of drill pipe, casing or tubing consisting of four joints screwed together. Compare double and thribble. See joint. fourable board n: the name used for the working platform of the derrickman, or monkey board, when it is located at a height in the derrick equal to approximately four lengths of pipe joined together. Compare double board and thribble board. See monkey board. fracturing n: shortened form of formation fracturing. See formation fracturing.
G
gas-cut mud n: a drilling mud that has entrained formation gas giving the mud a characteristically fluffy texture. When entrained gas is not released before the fluid returns to the well, the weight or density of the fluid column is reduced. Because a large amount of gas in mud lowers its density, gas-cut mud must be treated to lessen the chance of a blowout. gas sand n: a stratum of sand or porous sandstone from which natural gas is obtained. gas show n: the gas that appears in drilling fluid returns, indicating the presence of a gas zone. geologist n: a scientist who gathers and interprets data pertaining to the strata of the earths crust. geology n: the science that relates to the study of the structure, origin, history, and development of the earth and its inhabitants as revealed in the study of rocks, formations, and fossils. graveyard tour n: (pronounced tower) the shift of duty on a drilling rig that starts at or about midnight. Compare daylight tour and evening tour. gravity n: the attraction exerted by the earths mass on objects at its surface; the weight of a body. See API gravity and specific gravity. guide shoe n: a short, heavy, cylindrical section of steel filled with concrete and rounded at the bottom, which is placed at the end of the casing string. It prevents the casing
G-12
Appendix G
from snagging on irregularities in the borehole as it is lowered. A passage through the center of the shoe allows drilling fluid to pass up into the casing while it is being lowered and cement to pass out during cementing operations. Also called casing shoe. gun-perforate v: to create holes in casing and cement set through a productive formation. A common method of completing a well is to set casing through the oil-bearing formation and cement it. A perforating gun is then lowered into the hole and fired to detonate high-powered jets or shoot steel projectiles (bullets) through the casing and cement and into the pay zone. The formation fluids flow out of the reservoir through the perforations and into the wellbore. See jet-perforate and perforating gun. gusher n: an oil well that has come in with such great pressure that the oil jets out of the well like a geyser. In reality, a gusher is a blowout and is extremely wasteful of reservoir fluids and drive energy. In the early days of the oil industry, gushers were common and many times were the only indication that a large reservoir of oil and gas had been struck. See blowout.
H
hoist n: an arrangement of pulleys and wire rope or chain used for lifting heavy objects; a winch or similar device; the draw works. See draw works. hoisting drum n: the large, flanged spooled in the draw works on which the hoisting cable is wound. See draw works. hook n: a large hook-shaped device from which the swivel is suspended. It is designed to carry maximum loads ranging from 100 to 650 tons and turns on bearings in its supporting housing. A strong spring within the assembly cushions the weight of a stand (90 feet) of drill pipe, thus permitting the pipe to be made up and broken out with less damage to the tool joint threads. Smaller hooks without the spring are used for handling tubing and sucker rods. See stand and swivel. hopper n: a large funnel- or cone-shaped device into which dry components (as powdered clay or cement) can be poured in order to uniformly mix the components with water (or other liquids). The liquid is injected through a nozzle at the bottom of the hopper. The resulting mixture of dry material and liquid may be drilling mud to be used as the circulating fluid in a rotary drilling operation or may be cement slurry used to bond casing to the borehole. hydraulic fracturing n: an operation in which a specially blended liquid is pumped down a well and into a formation under pressure high enough to cause the formation to crack open. The resulting cracks or fractures serve as passages through which oil can flow into the wellbore. See formation fracturing. hydrocarbons n: organic compounds of hydrogen and carbon, whose densities, boiling points, and freezing points increase as their molecular weights increase. Although composed only of two elements, hydrocarbons exist in a variety of compounds, because of the strong affinity of the carbon atom for other atoms and for itself. The smallest molecules of hydrocarbons are gaseous; the largest are solids. Petroleum is a mixture of many different hydrocarbons. hydromatic brake n: a device mounted on the end of the draw works shaft of a drilling rig. The hydromatic brake (often simply called the hydromatic) serves as an auxiliary to the mechanical brake when pipe is lowered into the well. The braking effect in a
G-13
Appendix G
hydromatic brake is achieved by means of a runner or impeller turning in a housing filled with water.
I
impermeable adj: preventing the passage of fluid. A formation may be porous yet impermeable if there is an absence of connecting passages between the voids within it. See permeability. inland barge rig n: a drilling structure consisting of a barge upon which the drilling equipment is constructed. When moved from one location to another, the barge floats, but when stationed on the drill site, the barge is submerged to rest on the bottom. Typically, inland barge rigs are used to drill wells in marshes, shallow inland bays, and in areas where the water covering the drill site is not too deep. instrumentation n: a device or assembly of devices designed for one or more of the following functions: to measure operating variables (as pressure, temperature, rate of flow, speed of rotation, etc.); to indicate these phenomena with visible or audible signals; to record them, to control them within a predetermined range; and to stop operations if the control fails. Simple instrumentation might consist of an indicating pressure gauge only. In a completely automatic system, the desired range of pressure, temperature, and so on is predetermined and preset. intermediate casing string n: the string of casing set in a well after the surface casing, but before the production casing, to keep the hole from caving and to seal off troublesome formations. The string is sometimes called protection casing.
J
jackup drilling rig n: an offshore drilling structure with tubular or derrick legs that support the deck and hull. When positioned over the drilling site, the bottoms of the legs rest on the seafloor. A jackup rig is towed or propelled to a location with its legs up. Once the legs are firmly positioned on the bottom, the deck and hull height are adjusted and leveled. jet bit n: a drilling bit having replaceable nozzles though which the drilling fluid is directed in a high-velocity stream to the bottom of the hole to improve efficiency of the bit. See bit. jet gun n: an assembly, including a carrier and shaped charges, that is used in jet perforating. jet-perforate v: to create a hole through the casing with a shaped charge of high explosives instead of a gun that fires projectiles. The loaded charge is lowered into the hole to the desired depth. Once detonated, the charges emit short, penetrating jets of high-velocity gases that cut holes in the casing and cement and some distance into the formation. Formation fluids then flow into the wellbore through these perforations. See bullet perforator and gun-perforate. joint n: a single length (about 30 feet) of drill pipe or of drill collar, casing, or tubing, that has threaded connections at both ends. Several joints screwed together constitute a stand of pipe. See stand, single, double, thribble, and fourble. junk n: metal debris lost in a hole. Junk may be a lost bit, pieces of a bit, milled pieces of pipe, wrenches, or any relatively small object that impedes drilling and must be fished out of the hole. v: to abandon (as a nonproductive well).
G-14
Appendix G
K
kelly n: the heavy steel member, four- or six-sided, suspended from the swivel through the rotary table and connected to the topmost joint of drill pipe to turn the drill string as the rotary table turns. It has a bored passageway that permits fluid to be circulated into the drill string and up the annulus, or vice versa. See drill stem, rotary table, and swivel. kelly bushing n: a special device that, when fitted into the master bushing, transmits torque to the kelly and simultaneously permits vertical movement of the kelly to make hole. It may be shaped to fit the rotary opening or have pins for transmitting torque. Also called the drive bushing. See kelly and master bushing. kelly spinner n: a pneumatically operated device mounted on top of the kelly that, when actuated, causes the kelly to turn or spin. it is useful when the kelly or a joint of pipe attached to it must be spun up; that is, rotated rapidly in order to make it up. kick n: an entry of water, gas, oil, or other formation fluid into the wellbore. It occurs because the pressure exerted by the column of drilling fluid is not great enough to overcome the pressure exerted by the fluids in the formation drilled. If prompt action is not taken to control the kick or kill the well, a blowout will occur. See blowout.
L
LACT unit n: an automated system for measuring and transferring oil from a lease gathering system into a pipeline. See lease automatic custody transfer. latch on v: to attach elevators to a section of pipe to pull it out of or run it into the hole. lead tongs n: (pronounced leed) the pipe tongs suspended in the derrick or mast and operated by a wireline connected to the breakout cat-head. Also called breakout tongs. lease n: 1. a legal document executed between a land-owner, as lessor, and a company or individual, as lessee, that grants the right to exploit the premises for minerals or other products. 2. the area where production wells, stock tanks, separators, LACT units, and other production equipment are located. See LACT unit and lease automatic custody transfer. lease automatic custody n: the measurement and transfer of oil from the producers tanks to the transfer connected pipeline on an automatic basis without a representative of either having to be present. See LACT unit. location n: the place where a well is drilled. log n: a systematic recording of data, as from the drillers log, mud log, electrical well log, or radioactivity log. Many different logs are run in wells being produced or drilled to obtain various characteristics of downhole formations.
M
magnetic brake n: also called an electrodynamic brake. See electrodynamic brake. make a connection v: to attach a joint of drill pipe onto the drill string suspended in the wellbore to permit deepening of the wellbore.
G-15
Appendix G
make a trip v: to hoist the drill string out of the wellbore to perform one of a number of operations such as changing bits, taking a core, and so forth, and then to return the drill string to the wellbore. make hole v: to deepen the hole made by the bit; to drill ahead. make up v: 1. to assemble and join parts to form a complete unit (as to make up a string of casing). 2. to screw together two threaded pieces. 3. to mix or prepare (as to make up a tank of mud). 4. to compensate for (as to make up for lost time). make up a joint v: to screw a length of pipe into another length of pipe. makeup cat-head n: a device attached to the shaft of the draw works that is used as a power source for screwing together joints of pipe; usually located on the drillers side of the draw works. See cat-head. mast n: a portable derrick capable of being erected as a unit, as distinguished from a standard derrick that cannot be raised to a working position as a unit. For transporting by land, the mast can be divided into two or more sections to avoid excessive length extending from truck beds on the highway. Compare derrick. master bushing n: a device that fits into the rotary table. It accommodates the slips and drives the kelly bushing so that the rotating motion of the rotary table can be transmitted to the kelly. Also called rotary bushing. See slips and kelly bushing. mechanical rig n: a drilling rig in which the source of power is one or more internal-combustion engines and in which the power is distributed to rig components through mechanical devices (as chains, sprockets, clutches, and shafts). It is also called a power rig. mill n: a downhole tool with rough, sharp, extremely hard cutting surfaces for removing metal by grinding or cutting. Mills are run on drill pipe or tubing to grind up debris in the hole, remove stuck portions of drill stem or sections of casing for sidetracking, and ream out tight spots in the casing. They are also called junk mills, reaming mills, and so forth, depending on what use they have. v: to use a mill to cut or grind metal objects that must be removed from a well. mix mud v: to prepare drilling fluids from a mixture of water or other fluids and one or more of the various dry mud-making materials (as clay, weighting materials, chemicals, etc.). monkey board n: the derrickmans working platform. As pipe or tubing is run into or out of the hole, the derrickman must handle the top end of the pipe, which may be as high as 90 feet in the derrick or mast. The monkey board provides a small platform to raise him to the proper height to be able to handle the top of the pipe. See double board, fourable board, and thribble board. morning tour n: (pronounced tower) also called daylight tour. See daylight tour. motorman n: the crew member on a rotary drilling rig responsible for the care and operation of drilling engines. mouse hole n: an opening through the rig floor, usually lined with pipe, into which a length of drill pipe is placed temporarily for later connection to the drill string. mouse hole connection n: the procedure of adding a length of drill pipe or tubing to the active string in which the length to be added is placed in the mouse hole, made up to the kelly, then pulled out of the mouse hole, and subsequently made up into the string.
G-16
Appendix G
mud n: the liquid circulated through the wellbore during rotary drilling operations. In addition to its function of bringing up cuttings to the surface, drilling mud cools and lubricates the bit and drill stem, protects against blowouts by holding back subsurface pressures, and deposits a mud cake on the wall of the borehole to prevent loss of fluids to the formation. Although it originally was a suspension of earth solids (especially clays) in water, the mud used in modern drilling operations is a more complex, three-phase mixture of liquids, reactive solids, and inert solids. The liquid phase may be fresh water, seawater, and may contain one or more conditioners. See drilling fluid. mud analysis n: examination and testing of the drilling mud to determine its physical and chemical properties. mud cake n: the sheath of mud solids that forms on the wall of the hole when the liquid from the mud filters into the formation; also called wall cake or filter cake. mud circulation n: the act of pumping mud downward to the bit and back up to the surface by normal circulation or reverse circulation. See normal circulation and reverse circulation. mud conditioning n: the treatment and control of drilling mud to ensure that it has the correct properties. Conditioning may include the use of additives, the removal of sand or other solids, the removal of gas, the addition of water, and other measures to prepare the mud for conditions encountered in a specific well. mud engineer n: a person whose duty is to test and maintain the properties of the drilling mud that are specified by the operator. mud gun n: a pipe that shoots a jet of drilling mud under high pressure into the mud pit to mix additives with the mud. mud man n: also called a mud engineer. See mud engineer. mud pit n: a series of open tanks, usually made of steel plates, through which the drilling mud is cycled to allow sand and sediments to settle out. Additives are mixed with the mud in the pit, and the fluid is temporarily stored there before being pumped back into the well. Modern rotary drilling rigs are generally provided with three or more pits, usually fabricated steel tanks fitted with built-in piping, valves and mud agitators. Mud pits are also called shaker pits, settling pits, and suction pits, depending of their main purpose. See shaker pit, settling pit and suction pit. mud pump n: a large, reciprocating pump used to circulate the mud on a drilling rig. A typical mud pump is a single- or double-acting, two- or three-cylinder piston pump whose pistons travel in replaceable liners and are driven by a crankshaft actuated by an engine or motor. Also called a slush pump. mud-return line n: a trough or pipe placed between the surface connections at the wellbore and the shale shaker, through which drilling mud flows upon its return to the surface from the hole. mud screen n: also called a shale shaker. See shale shaker.
N
natural gas n: a highly compressible, highly expandable mixture of hydrocarbons having a low specific gravity and occurring naturally in a gaseous form. Besides hydrocarbon gases, natural gas may contain appreciable quantities of nitrogen, helium, carbon dioxide, and contaminants (as hydrogen sulfide and water vapor).
G-17
Appendix G
Although gaseous at normal temperatures and pressures, certain of the gases comprising the mixture that is natural gas are variable in form and may be found either as gases or as liquids under suitable conditions of temperature and pressure. needle valve n: a globe valve that incorporates a needle-point disk to produce extremely fine regulation of flow. nipple n: a tubular pipe fitting threaded on both ends and less than 12 inches long. nipple up v: in drilling, to assemble the blowout-preventer stack on the wellhead at the surface. normal circulation n: the smooth, uninterrupted circulation of drilling fluid down the drill stem, out the bit, up the annular space between the pipe and the hole, and back to the surface. See mud circulation and reverse circulation.
O
offshore drilling n: drilling for oil in an ocean, gulf, or sea, usually on the continental shelf. A drilling unit for offshore operations may be a mobile floating vessel with a ship or barge hull, a semisubmersible or submersible base, a self-propelled or towed structure with jacking legs (jackup drilling rig), or a permanent structure used as a production platform when drilling is completed. In general, wildcat wells are drilled from mobile floating vessels (as semisubmersible rigs and drill ships) or from jack-ups, while development wells are drilled from platforms. See drill ship, jackup drilling rig, platform, semisubmersible drilling rig and wildcat. oil field n: the surface area overlying an oil reservoir or reservoirs. Commonly, the term includes not only the surface area, but may include the reservoir, the wells, and production equipment as well. oil sand n: 1. a sandstone that yields oil. 2. (by extension) any reservoir that yields oil, whether or not it is sandstone. oil zone n: a formation or horizon of a well from which oil may be produced. The oil zone is usually immediately under the gas zone and on top of the water zone if all three fluids are present and segregated. open adj: 1. of a wellbore, having no casing. 2. of a hole, having no drill pipe or tubing suspended in it. open hole n: 1. any wellbore in which casing has not been set. 2. open or cased hole in which no drill pipe or tubing is suspended. operator n: the person or company, either proprietor or lessee, actually operating an oilwell or lease. Compare unit operator. overshot n: a fishing tool that is attached to tubing or drill pipe and lowered over the outside wall of pipe lost or stuck in the wellbore. A friction device in the overshot, usually either a basket or a spiral grapple, firmly grips the pipe, allowing the lost fish to be pulled from the hole.
P
P&A abbr: plug and abandon. pay sand n: the producing formation, often one that is not even sandstone. It is also called pay, pay zone, and producing zone.
G-18
Appendix G
perforate v: to pierce the casing wall and cement to provide holes through which formation fluids may enter or to provide holes in the casing so that materials may be introduced into the annulus between the casing and the wall of the borehole. Perforating is accomplished by lowering into the well a perforating gun, or perforator, that fires electrically detonated bullets or shaped charges from the surface. See perforating gun. perforating gun n: a device fitted with shaped charges or bullets that is lowered to the desired depth in a well and fired to create penetrating holes in casing, cement and formation. See gun-perforate. permeability n: 1. a measure of the ability of fluids to flow through a porous rock. 2. fluid conductivity of a porous medium. 3. the ability of a fluid to flow within the interconnected pore network of a porous medium. See absolute permeability, effective permeability, and relative permeability. petroleum n: oil or gas obtained from the rocks of the earth. See hydrocarbons. pin n: the male section of the tool joint. See tool joint. pipe ram n: a sealing component for a blowout preventer that closes the annular space between the pipe and the blowout preventer or wellhead. See annular space and blowout preventer. platform n: an immobile, offshore structure constructed on pilings from which wells are drilled, produced, or both. plug and abandon v: to place a cement plug into a dry hole and abandon it. pore n: an opening or space within a rock or mass of rocks, usually small and often filled with some fluid (as water, oil, gas, or all three). Compare vug. porosity n: the condition of something that contains pores (as a rock formation). See pore. positive choke n: a choke in which the orifice size must be changed to change the rate of flow through the choke. See choke and orifice. pressure n: the force that a fluid (liquid or gas) exerts when it is in some way confined within a vessel, pipe, hole in the ground, and so forth, such as that exerted against the inner wall of a tank or that exerted on the bottom of the wellbore by drilling mud. Pressure is often expressed in terms of force per unit of area, as pounds per square inch (psi). pressure gauge n: an instrument for measuring fluid pressure that usually registers the difference between atmospheric pressure and the pressure of the fluid by indicating the effect of such pressures on a measuring element (as a column of liquid, a weighted piston, a diaphragm, or other pressure-sensitive device). pressure gradient n: a scale of pressure differences in which there is a uniform variation of pressure from point to point. For example, the pressure gradient of a column of water is about 0.433 psi/ft of vertical elevation (9.794 kPa/m). The normal pressure gradient in a well is equivalent to the pressure exerted at any given depth by a column of 10 percent salt water extending from that depth to the surface (i.e., 0.465 psi/ft or 10.518 kPa/m). pressure relief valve n: a valve that opens at a preset pressure to relieve excessive pressures within a vessel or line; also called a relief valve, safety valve, or pop valve. preventer n: shortened form of blowout preventer. See blowout preventer.
G-19
Appendix G
primary cementing n: the cementing operation that takes place immediately after the casing has been run into the hole; used to provide a protective sheath around the casing, to segregate the producing formation, and to prevent the migration of undesirable fluids. See secondary cementing and squeeze cementing. prime mover n: an internal-combustion engine that is the source of power for a drilling rig in oilwell drilling. production n: 1. the phase of the petroleum industry that deals with bringing the well fluids to the surface and separating them and with storing, gauging, and otherwise preparing the product for the pipeline. 2. the amount of oil or gas produced in a given period. proppant n: also called propping agent. See propping agent. propping agent n: a granular substance (as sand grains, aluminum pellets, or other material) carried in suspension by the fracturing fluid that serves to keep the cracks open when the fracturing fluid is withdrawn after a fracture treatment. psi abbr: pounds per square inch. See pressure. pump n: a device that increases the pressure on a fluid or raises it to a higher level. Various types of pumps include the reciprocating pump, centrifugal pump, rotary pump, jet pump, sucker rod pump, hydraulic pump, mud pump, submersible pump, and bottomhole pump. pump pressure n: fluid pressure from the action of the pump.
R
radioactivity well n: the recording of the natural or induced radioactive characteristics of subsurface logging formations. A radioactivity log, also known as a radiation log, normally consists of two recorded curves: a gamma ray curve and a neutron curve. Both indicate the types of rock in the formation and the types of fluids contained in the rocks. The two logs may be run simultaneously in conjunction with a collar locator in a cased or uncased hole. ram n: the closing and sealing component on a blowout preventer. One of three types blind, pipe, or shear - may be installed in several preventers mounted in a stack on top of the wellbore. Blind rams, when closed, form a seal on a hole that has no drill pipe in it; pipe rams, when closed, seal around the pipe; shear rams cut through drill pipe and then form a seal. See blind ram, pipe ram, and shear ram. ram blowout preventer n: a blowout preventer that uses rams to seal off pressure on a hole that is with or without pipe. It is also called a ram preventer. See blowout preventer and ram. rat hole n: 1. a hole in the rig floor 30 to 35 feet deep, lined with casing that projects above the floor, into which the kelly and swivel are placed when hoisting operations are in progress. 2. a hole of a diameter smaller than the main hole that is drilled in the bottom of the main hole. v: to reduce the size of the wellbore and drill ahead. reeve v: to pass (as the end of a rope) through a hole or opening in a block or similar device. reeve the line v: to string wire-rope drill line through the sheaves of the traveling and crown blocks to the hoisting drum. relative permeability n: a measure of the ability of two or more fluids (as water, gas, and oil) to flow through a rock formation when the formation is totally filled with several fluids.
G-20
Appendix G
The permeability measure of a rock filled with two or more fluids is different from the permeability measure of the same rock filled with only one fluid. Compare absolute permeability. reserve pit n: 1. (obsolete) a mud pit in which a supply of drilling fluid was stored. 2. a waste pit, usually excavated, earthen-walled pit. It may be lined with plastic to prevent contamination of the soil. reservoir n: a subsurface, porous, permeable rock body in which oil and/or gas is stored. Most reservoir rocks are limestones, dolomites, sandstones, or a combination of these. The three basic types of hydrocarbon reservoirs are oil, gas, and condensate. An oil reservoir generally contains three fluids - gas, oil, and water with oil the dominant product. In the typical reservoir, these fluids occur in different phases because of the variance in their gravities. Gas, the lightest, occupies the upper part of the reservoir rocks; water, the lower part; and oil, the intermediate section. In addition to occurring as a cap or in solution, gas may accumulate independently of the oil; if so, the reservoir is called a gas reservoir. Associated with the gas, in most instances, are salt water and some oil. In a condensate reservoir, the hydrocarbons may exist as a gas, but when brought to the surface, some of the heavier ones condense to a liquid or condensate. reservoir pressure n: the pressure in a reservoir under normal conditions. reverse circulation n: the return of drilling fluid through the drill stem. The normal course of drilling fluid circulation is downward through the drill stem and upward through the annular space surrounding the drill stem. For special problems, normal circulation is sometimes reversed, and the fluid returns to the surface through the drill stem, or tubing, after being pumped down the annulus. rig n: the derrick or mast, draw works, and attendant surface equipment of a drilling unit. rig down v: to dismantle the drilling rig and auxiliary equipment following the completion of drilling operations; also called tear down. rig up v: to prepare the drilling rig for making hole; to install tools and machinery before drilling is started. roller cone bit n: a drilling bit made of two, three, or four cones, or cutters, that are mounted on extremely rugged bearings. Also called rock bits. The surface of each cone is made up of rows of steel teeth or rows of tungsten carbide inserts. See bit. rotary bushing n: also called master bushing. See master bushing. rotary drilling n: a drilling method in which a hole is drilled by a rotating bit to which downward force is applied. The bit is fastened to and rotated by the drill stem, which also provides a passageway through which the drilling fluid is circulated. Additional joints of drill pipe are added as drilling progresses. rotary helper n: a worker on a drilling rig, subordinate to the driller, sometimes called a roughneck, floorman, or rig crewman. rotary hose n: a reinforced, flexible tube on a rotary drilling rig that conducts the drilling fluid from the mud pump and standpipe to the swivel and kelly; also called the mud hose or kelly hose. See kelly, mud pump, standpipe, and swivel. rotary table n: the principal component of a rotary or rotary machine, used to turn the drill stem and support the drilling assembly. It has a beveled gear arrangement to create
G-21
Appendix G
the rotational motion and an opening into which bushings are fitted to drive and support the drilling assembly. roughneck n: also called a rotary helper. See rotary helper. round trip n: the action of pulling out and subsequently running back into the hole a string of drill pipe or tubing. It is also called tripping. roustabout n: 1. a worker on an offshore rig who handles the equipment and supplies that are sent to the rig from the shore base. The head roustabout is very often the crane operator. 2. a worker who assists the foreman in the general work around a producing oil well, usually on the property of the oil company. 3. a helper on a well-servicing unit. run in v: to go into the hole with tubing, drill pipe, and so forth.
S
samples n pl: 1. the well cuttings obtained at designated footage intervals during drilling. From an examination of these cuttings, the geologist determines the type of rock and formation being drilled and estimates oil and gas content. 2. small quantities of well fluids obtained for analysis. sand n: 1. an abrasive material composed of small quartz grains formed from the disintegration of preexisting rocks. Sand consists of particles less than 2 millimeters and greater than 1/16 of a millimeter in diameter. 2. sandstone. scratcher n: a device fastened to the outside of casing that removes the mud cake from the wall of the hole to condition the hole for cementing. By rotating or moving the casing string up and down as it is being run into the hole, the scratcher, formed of stiff wire, removes the cake so that the cement can bond solidly to the formation. secondary cementing n: any cementing operation after the primary cementing operation. Secondary cementing includes a plug-back job, in which a plug of cement is positioned at a specific point in the well and allowed to set. Wells are plugged to shut off bottom water or to reduce the depth of the well for other reasons. See primary cementing and squeeze cementing. seismograph n: a device that detects reflections of vibrations in the earth, used in prospecting for probable oil-bearing structures. Vibrations are created by discharging explosives in shallow boreholes, by striking the surface with a heavy blow, or by generating low-frequency sound waves. The type and velocity of the vibrations as recorded by the seismograph indicate the general characteristics of the section of earth through which the vibration pass. semi-submersible n: a floating, offshore drilling structure that has hulls submerged in the water drilling rig but not resting on the seafloor. Living quarters, storage space, and so forth are assembled on the deck. Semisubmersible rigs are either self-propelled or towed to a drilling site and either anchored or dynamically positioned over the site or both. Semi-submersibles are more stable than drill ships and are used extensively to drill wildcat wells in rough water such as the North Sea. See dynamic positioning. set casing v: to run and cement casing at a certain depth in the wellbore. Sometimes, the term set pipe is used when referring to setting casing. settling pit n: the mud pit into which mud flows and in which heavy solids are allowed to settle out. Often auxiliary equipment (as de-sanders) must be installed to speed up this process.
G-22
Appendix G
shaker n: shortened form of shale shaker. See shale shaker. shaker pit n: the mud pit adjacent to the shale shaker, usually the first pit into which the mud flows after returning from the hole. shale n: a fine-grained sedimentary rock composed of consolidated silt and clay or mud. Shale is the most frequently occurring sedimentary rock. shale shaker n: a series of trays with sieves that vibrate to remove cuttings from the circulating fluid in rotary drilling operations. The size of the openings in the sieve is carefully selected to match the size of the solids in the drilling fluid and the anticipated size of the cuttings. Also called a shaker. shaped charge n: a relatively small container of high explosive that is loaded into a perforating gun. Upon detonation, the charge releases a small, high-velocity stream of particles (a jet) that penetrates the casing, cement, and formation. See gun-perforate. shear ram n: the components in a blowout preventer that cut, or shear, through drill pipe and form a seal against well pressure. Shear rams are used in mobile offshore drilling operations to provide a quick method of moving the rig away from the hole when there is no time to trip the drill stem of the hole. sheave n: (pronounced shiv) a grooved pulley. show n: the appearance of oil or gas in cuttings, samples, cores, and so forth of drilling mud. shut down v: to stop work temporarily or to stop a machine or operation. shut-in bottomhole n: the pressure at the bottom of a well when the surface valves on the well are pressure completely closed. The pressure caused by fluids that exist in the formation at the bottom of the well. sidetrack v: to drill around broken drill pipe or casing that has become lodged permanently in the hole, using a whipstock, turbo-drill, or other mud motor. See directional drilling, turbo-drill, and whipstock. sidewall coring n: a coring technique in which core samples are obtained from a zone that has already been drilled. A hollow bullet is fired into the formation wall to capture the core and then retrieved on a flexible steel cable. Core samples of this type usually range from 3/4 to 13/16 inches in diameter and from 3/4 to 1 inch in length. This method is especially useful in soft rock areas. single n: a joint of drill pipe. Compare double, thribble, and fourable. slips n pl: wedge-shaped pieces of metal with teeth or other gripping elements that are used to prevent pipe from slipping down into the hole or to hold pipe in place. Rotary slips fit around the drill pipe and wedge against the master bushing to support the pipe. Power slips are pneumatically or hydraulically actuated devices that allow the crew to dispense with the manual handling of slips when making a connection. Packers and other downhole equipment are secured in position by slips that engage the pipe by action directed at the surface. slurry n: a plastic mixture of cement and water that is pumped into a well to harden; there it supports the casing and provides a seal in the wellbore to prevent migration of underground fluids. sonic logging n: the recording of the time required for a sound wave to travel a specific distance through a formation. Difference in observed travel times is largely caused by
G-23
Appendix G
variations in porosities of the medium, an important determination. The sonic log, which may be run simultaneously with a spontaneous potential log or a gamma ray log, is useful for correlation and often is used in conjunction with other logging services for substantiation of porosities. It is run in an uncased hole. spear n: a fishing tool used to retrieve pipe lost in a well. The spear is lowered down the hole and into the lost pipe, and when weight, torque, or both are applied to the string to which the spear is attached, the slips in the spear expand and tightly grip the inside of the wall of the lost pipe. Then the string, spear, and lost pipe are pulled to the surface. specific gravity n: the ratio of the weight of a given volume of a substance at a given temperature to the weight of an equal volume of a standard substance at the same temperature. For example, if 1 cubic inch of water at 39oF weighs 1 unit and 1 cubic inch of another solid or liquid at 39oF weighs 0.95 unit, then the specific gravity of the substance is 0.95. In determining the specific gravity of gases, the comparison is made with the standard of air or hydrogen. See gravity. spinning cat-head n: a spooling attachment on the makeup cat-head to permit use of a spinning chain to spin up or make up drill pipe. See spinning chain. spinning chain n: a Y-shaped chain used to spin up (tighten) one joint of drill pipe to another. In use, one end of the chain is attached to the tongs, another end to the spinning cat-head, and the third end is free. The free end is wrapped around the tool joint, and the cat-head pulls the chain off the joint, causing the to spin (turn) rapidly and tighten up. After the chain is pulled off the joint, the tongs are secured in the same spot, and continued pull on the chain (and thus on the tongs) by the cat-head makes up the joint to final tightness. spud v: to move the drill stem up and down in the hole over a short distance without rotation. Careless execution of this operation creates pressure surges that can cause a formation to break down, which results in lost circulation. See spud in. spud in v: to being drilling, to start the hole. squeeze cementing n: the forcing of cement slurry by pressure to specified points in a well to cause seals at the points of squeeze. It is a secondary cementing method that is used to isolate a producing formation, seal off water, repair casing leaks, and so forth. See cementing. stab v: to guide the end of a pipe into a coupling or tool joint when making a connection. See coupling and tool joint. stabbing board n: a temporary platform erected in the derrick or mast some 20 to 40 feet above the derrick floor. The derrickman or another crew member works on the board while casing is being run in a well. The board may be wooden or fabricated of steel girders floored with anti-skid material and powered electrically to raise or lower it to the desired level. A stabbing board serves the same purpose as a monkey board but is temporary instead of permanent. stake a well v: to locate precisely on the surface of the ground the point at which a well is to be drilled. After exploration techniques have revealed the possibility of the existence of a subsurface hydrocarbon-bearing formation, a certified and registered land surveyor drives a stake into the ground to mark the spot where the well is to be drilled.
G-24
Appendix G
stand n: the connected joints of pipe racked in the derrick or mast when making a trip. On a rig, the usual stand is 90 feet long (three lengths of pipe screwed together) or a thribble. Compare double and fourble. standpipe n: a vertical pipe rising along the side of the derrick or mast, which joins the discharge line leading from the mud pump to the rotary hose and through which mud is pumped going into the hole. See mud pump and rotary hose. stimulation n: any process undertaken to enlarge old channels or create new ones in the producing formation of a well (e.g., acidizing or formation fracturing) See acidize.
stratification n: the natural layering or lamination characteristic of sediments and sedimentary rocks. stratigraphic trap n: a petroleum trap that occurs when the top of the reservoir bed is terminated by other beds or by a change of porosity or permeability within the reservoir itself. Compare structural trap. See trap. string n: the entire length of casing, tubing, or drill pipe run into a hole; the casing string. Compare drill string and drill stem. string up v: to thread the drill line through the sheaves of the crown block and traveling block. One end of the line is secured to the hoisting drum and the other to the derrick substructure. See sheave. structural trap n: a petroleum trap that is formed because of deformation (as folding or faulting) of the rock layer that contains petroleum. Compare stratigraphic trap. See trap. stuck pipe n: drill pipe, drill collars, casing, or tubing that has inadvertently become immobile in the hole. It may occur when drilling is in progress, when casing is being run in the hole, or when the drill pipe is being hoisted. sub n: a short, threaded piece of pipe used to adapt parts of drill string that cannot otherwise be screwed together because of differences in thread size or design. A sub may also perform a special function. Lifting subs are used with drill collars to provide a shoulder to fit the drill pipe elevators. A kelly saver sub is placed between the drill pipe and kelly to prevent excessive thread wear of the kelly and drill pipe threads. A bent sub is used when drilling a directional hole. Sub is a sort expression for substitute. submersible drilling rig n: an offshore drilling structure with several compartments that are flooded to cause the structure to submerge and rest on the seafloor. Most submersible rigs are used only in shallow water. substructure n: the foundation on which the derrick or mast and usually the draw works sit; contains space for storage and well control equipment. suction pit n: the mud pit from which mud is picked up by the suction of the mud pumps; also called a sump pit and mud suction pit. surface casing n: also called surface pipe. See surface pipe. surface data logging n: the recording of information derived from examination and analysis of formation cuttings made by the bit and mud circulated out of the hole. A portion of the mud is diverted through a gas-detecting device. Cuttings brought up by the mud are examined under ultraviolet light to detect the presence of oil and gas. Surface data logging is often carried out in a portable laboratory set up at the well. surface pipe n: the first string of casing (after the conductor pipe) that is set in a well, varying in length from a few hundred to several thousand feet. Some states require a
G-25
Appendix G
minimum length to protect freshwater sands. Compare conductor pipe. swivel n: a rotary tool that is hung from the rotary hook and traveling block to suspend and permit free rotation of the drill stem. It also provides a connections for the rotary hose and passageway for the flow of drilling fluid into the drill stem. syncline n: a down-warped, trough-shaped configuration of folded, stratified rocks. Compare anticline.
T
TD abbr: total depth. thread protector n: a device that is screwed onto or into pipe threads to protect the threads from damage when the pipe is not in use. Protectors may be metal or plastic. thribble n: a stand of pipe made up of three joints and handled as a unit. See stand. Compare single, double, and fourble. thribble board n: the name used for the working platform of the derrickman, or monkey board, when it is located at a height in the derrick equal to three lengths of pipe joined together. Compare double board and fourble board. See monkey board. throw the chain n: to flip the spinning chain up from a tool joint box so that the chain wraps around the tool joint pin after it is stabbed into the box. The stand or joint of drill pipe is turned or spun by a pull on the spinning chain from the cat-head or draw works. tight formation n: a petroleum- or water-bearing formation of relatively low porosity and permeability. See porosity and permeability. tight hole n: a well about which information is restricted and passed only to those authorized for security or competitive reasons. tongs n pl: the large wrenches used for turning when making up or breaking out drill pipe, casing, tubing, or other pipe; variously called casing tongs, rotary tongs, and so forth according to the specific use. Power tongs are pneumatically or hydraulically operated tools that serve to spin the pipe up tight and, in some instances, to apply the final makeup torque. See also chain tongs. tool joint n: a heavy coupling element for drill pipe made of special alloy steel. Tool joints have coarse, tapered threads and seating shoulders designed to sustain the weight of the drill stem, withstand the strain of frequent coupling and uncoupling, and provide a leakproof seal. The male section of the joint, or the pin, is attached to one end of a length of drill pipe, and the female section, or box, is attached to the other end. The tool joint may be welded to the end of the pipe or screwed on or both. A hard metal facing is often applied in a band around the outside of the tool joint to enable it to resist abrasion from the walls of the borehole. tool pusher n: an employee of a drilling contractor who is in charge of the entire drilling crew and the drilling rig. Also called a drilling rig foreman, manager, supervisor, or rig superintendent. See drilling foreman. torque n: the turning force that is applied to a shaft or other rotary mechanism to cause it to rotate or tend to do so. Torque is measured in foot-pounds, joules, meter-kilograms, and so forth. torque converter n: a connecting device between a prime mover and the machine actuated by it. The elements that pump the fluid in the torque converter automatically increase
G-26
Appendix G
the output torque of the engine to which the torque is applied, with an increase of load on the output shaft. Torque converters are used extensively on mechanical rigs that have a compound. See mechanical rig. total depth n: the maximum depth reached in a well. tour n: (pronounced tower) an 8- or 12-hour shift worked by a drilling crew or other oil field workers. The most common divisions of tours are daylight, evening, and graveyard, if 8-hour tours are employed. transmission n: the gear or chain arrangement by which power is transmitted from the prime mover to the draw works, mud pump, or rotary table of a drilling rig. See prime mover. trap n: layers of buried rock strata that are arranged so that petroleum accumulates in them. traveling block n: an arrangement of pulleys, or sheaves, through which drilling cable is reeved and that moves up and down in the derrick or mast. See block, crown block, and sheave. tricone bit n: a type of bit in which three cone-shaped cutting devices are mounted in such a way that they inter-mesh and rotate together as the bit drills. The bit body may be fitted with nozzles, or jets, through which the drilling fluid is discharged. A one-eyed bit is used in soft formations to drill a deviated hole. See directional drilling and bit. trip n: the operation of hoisting the drill stem from and returning it to the wellbore. See make a trip. turbo-drill n: a drilling tool that rotates a bit attached to it by the action of the drilling mud on the turbine blades built into the tool. When a turbo-drill is used, rotary motion is imparted only at the bit; therefore, it is unnecessary to rotate the drill stem. Although straight holes can be drilled with the tool, it is used most often in directional drilling.
U
unit operator n: the oil company in charge of development and producing in an oil field in which several companies have joined together to produce the field.
V
valve n: a device used to control the rate of flow in a line, to open or shut off a line completely, or to serve as an automatic or semiautomatic safety device. Those with extensive usage include the gate valve, plug valve, globe valve, needle valve, check valve, and pressure relief valve. See check valve, needle valve and pressure relief valve. V-belt n: a belt with a trapezoidal cross-section that is made to run in sheaves or pulleys, with grooves of corresponding shape. See belt. vug n: a cavity in a rock.
G-27
Appendix G
W
waiting on cement adj: pertaining to or during the time when drilling or completion operations are suspended so the cement in a well can harden sufficiently. wall cake n: also called filter cake and mud cake. See filter cake and mud cake. weevil n: shortened form of boll weevil. See boll weevil. weight indicator n: an instrument near the drillers position on a drilling rig. It shows both the weight of the drill stem that is hanging from the hook (hook load) and the weight that is placed on the bit by the drill collars (weight on bit). weighting material n: a material that has high specific gravity and is used to increase the density of drilling fluids or cement slurries. wellbore n: a borehole; the hole drilled by the bit. A wellbore may have casing in it or may be open (i.e., uncased), or a portion of it may be cased and a portion of it may be open. Also called borehole or hole. See cased and open. well completion n: the activities and methods necessary to prepare a well for the production of oil and gas; the method by which a flow line for hydrocarbons is established between the reservoir and the surface. The method of well completion used by the operator depends on the individual characteristics of the producing formation or formations. These techniques include open-hole completions, sand exclusion completions, tubingless completions, multiple completions, and miniaturized completions. wellhead n: the equipment installed at the surface of the wellbore. A wellhead includes such equipment as the casing head and tubing head. adj pertaining to the wellhead (as wellhead pressure). well logging n: the recording of information about subsurface geologic formations. Logging methods include records kept by the driller, mud and cutting analyses, core analysis, drill stem tests, and electric and radioactivity procedures. See electric well log, mud logging, radioactivity well logging, and sonic logging. well stimulation n: any of several operations used to increase the production of a well. See acidize and formation fracturing. whipstock n: a long, steel casing that uses an inclined plane to cause the bit to deflect from the original borehole at a slight angle. Whipstocks are sometimes used in controlled directional drilling, to straighten crooked boreholes, and to sidetrack to avoid unretrieved fish. See directional drilling, fish, and sidetrack. wildcat n: 1. a well drilled in an area where no oil or gas production exists. With present-day exploration methods and equipment, about one wildcat out of every nine proves to be productive although not necessary profitable. 2. (nautical) a geared sheave of a windlass used to pull anchor chain. v: to drill wildcat wells. wireline n: a slender, rodlike or threadlike piece of metal, usually small in diameter, this is used for lowering special tools (such as logging sondes, perforating guns, and so forth) into the well. Compare wire rope. wire rope n: a cable composed of steel wires twisted around a central core of hemp or other fiber to create a rope of great strength and considerable flexibly. Wire rope is used as drill line (in rotary and cable-tool rigs), coring line, servicing line, winch line, and so on. It is often called cable or wireline; however, wireline is a single, slender metal rod, usually very flexible. Compare wireline.
G-28
Appendix G
WOC abbr: waiting on cement. worm n: a new and inexperienced worker on a drilling rig.
G-29
Appendix G
G-30