Replacement Interstage Seals for Steam Turbines

SED
N
A L
LICE

R I

M AT E
Technical Report

Effective December 6, 2006, this report has been made publicly available in accordance
with Section 734.3(b)(3) and published in accordance with Section 734.7 of the U.S. Export
Administration Regulations. As a result of this publication, this report is subject to only
copyright protection and does not require any license agreement from EPRI. This notice
supersedes the export control restrictions and any proprietary licensed material notices
embedded in the document prior to publication.
Replacement Interstage Seals for
Steam Turbines

1010214

Final Report, December 2005

EPRI Project Manager

S. Hesler

ELECTRIC POWER RESEARCH INSTITUTE

3420 Hillview Avenue, Palo Alto, California 94304-1395 ▪ PO Box 10412, Palo Alto, California 94303-0813 ▪ USA
800.313.3774 ▪ 650.855.2121 ▪ [email protected] ▪ www.epri.com
DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES
THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN
ACCOUNT OF WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH
INSTITUTE, INC. (EPRI). NEITHER EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE
ORGANIZATION(S) BELOW, NOR ANY PERSON ACTING ON BEHALF OF ANY OF THEM:

(A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I)

WITH RESPECT TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR
SIMILAR ITEM DISCLOSED IN THIS DOCUMENT, INCLUDING MERCHANTABILITY AND FITNESS
FOR A PARTICULAR PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR
INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY PARTY'S INTELLECTUAL
PROPERTY, OR (III) THAT THIS DOCUMENT IS SUITABLE TO ANY PARTICULAR USER'S
CIRCUMSTANCE; OR

(B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER

(INCLUDING ANY CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE
HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR
SELECTION OR USE OF THIS DOCUMENT OR ANY INFORMATION, APPARATUS, METHOD,
PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT.

ORGANIZATION(S) THAT PREPARED THIS DOCUMENT

Turbo-Technic Services

Electric Power Research Institute (EPRI)

NOTICE: THIS REPORT CONTAINS PROPRIETARY INFORMATION THAT IS THE

INTELLECTUAL PROPERTY OF EPRI. ACCORDINGLY, IT IS AVAILABLE
ONLY UNDER LICENSE FROM EPRI AND MAY NOT BE REPRODUCED
OR DISCLOSED, WHOLLY OR IN PART, BY ANY LICENSEE TO ANY
OTHER PERSON OR ORGANIZATION.

NOTE
For further information about EPRI, call the EPRI Customer Assistance Center at 800.313.3774 or
e-mail [email protected].

Electric Power Research Institute and EPRI are registered service marks of the Electric Power
Research Institute, Inc.

Copyright © 2005 Electric Power Research Institute, Inc. All rights reserved.

THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN

ACCOUNT OF WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH
CITATIONS

This report was prepared by

Turbo-Technic Services
36 Country Heights Drive
Richmond Hill, Ontario L4E 3M8

Principal Investigator
B. Sanders

Electric Power Research Institute (EPRI)

1300 W. T. Harris Blvd.
Charlotte, NC 28262

Principal Investigator
S. Hesler

This report describes research sponsored by EPRI.

The report is a corporate document that should be cited in the literature in the following manner:

Replacement Interstage Seals for Steam Turbines. EPRI, Palo Alto, CA: 2005. 1010214.

iii
PRODUCT DESCRIPTION

The installation of replacement steam turbine seals is a cost-effective means for improving steam
turbine efficiency and power output. Seals are often damaged by shaft rubs that increase leakage
flow and reduce efficiency. Frequent turbine overhauls were common practice in the past and
provided the opportunity to restore these damaged seals. The current industry strategy toward
longer maintenance intervals, coupled with increased fuel costs and the focus on capacity, has
increased interest in advanced replacement seals that are more tolerant of shaft vibration.

Results and Findings

This report presents a process for estimating the performance penalty associated with increased
interstage leakage that is caused by damaged seals. The input parameters of fuel cost and time
intervals between repairs can be varied to determine the cost-effectiveness of advanced seal
designs. An overview of the designs offered for advanced seals is also included in the report.
Many of these seals are specifically designed to withstand shaft vibration experienced during
startups, which greatly reduces the heat rate increase that is typically experienced between
turbine overhauls.
Challenges and Objectives
Plant owners are challenged to achieve high unit availability and to reduce operating costs. This
report describes commercially available options for retrofit seal designs. Some designs have been
used in the field for decades, whereas more recent advanced designs have less history in large
steam turbines. Plant operators are seeking to make effective decisions on options for seal
repair/replacement options that will affect their plant’s future reliability and performance.

Applications, Value, and Use

Members can use this report as guidance in their initial evaluation of seal replacement options.
Details on upgrade costs needed to make an economic assessment will be customer specific and
should be addressed individually. This report is intended to increase the awareness of options
and to prompt plant owners to seek details on issues related to the possible application of specific
designs. EPRI is not endorsing any specific product or commercial supplier.

EPRI Perspective
Steam leakage losses are a major component of controllable losses in large commercial steam
turbines. A demonstrated improvement in turbine section efficiencies of 5% can be achieved for
full section seal upgrades. Advanced retrofit seal designs address the most common cause of seal
degradation—shaft rubs that occur during unit startup/shutdown. The field experience with
advanced seal designs, first introduced 10 years ago, is growing rapidly. EPRI expects that
confidence in the use of these designs will increase with additional success in field applications.

v
Approach
Commercial providers of replacement seals for large steam turbines provided information on
various designs that were available for replacement or retrofit. A calculation procedure is
introduced and demonstrated for estimating the performance loss and increased operating costs
associated with increased seal clearance.

Keywords
Steam turbine
Seals
Labyrinth
Steam leakage
Packing
Heat rate

vi
 EPRI Proprietary Licensed Material

ACKNOWLEDGMENTS

EPRI acknowledges the information and material supplied by several commercial providers of
steam turbine seals. Contributions from the following individuals are appreciated:
• Bill Sanders (Turbo-Technic Services) for providing the material used in Section 2 of this
report covering seal economics
• Jerry Johnson (Turbo Parts LLC) for the photos and diagrams of the traditional labyrinth seal
provided in this report and all information pertaining to the Guardian seal and Vortex
Shedder design
• Mary Foley and Sandye Simmons (TurboCare) for the information and figures provided on
conventional retractable seal and retractable brush seal designs
• Jim McGregor for the information and figures provided on Brandon Engineering seal designs
• Donna Forsyth (Steam Turbine Alternative Resources [STAR]) for providing comments on
the importance of proper installation
• Al Deamer for the information provided on PerkinElmer seal designs
• James Feltz for the information and figures provided on Dresser-Rand seal designs

vii
 EPRI Proprietary Licensed Material

CONTENTS

1 INTRODUCTION ....................................................................................................................1-1
1.1 Report Objective ..........................................................................................................1-1
1.2 Report Scope ...............................................................................................................1-1
1.3 Description of Steam Turbine Seal Function ...............................................................1-2
1.4 Damage and Degradation of Seals..............................................................................1-4
1.5 Maintaining Desired Seal Performance .......................................................................1-4
1.5.1 During Maintenance Outages .............................................................................1-5
1.5.2 During Normal Operation ....................................................................................1-5
1.6 Nomenclature ..............................................................................................................1-5

2 ECONOMICS OF SEAL OPERATION ...................................................................................2-1

2.1 Introduction ..................................................................................................................2-1
2.1.1 Seal Maintenance Decision Analysis ..................................................................2-1
2.1.2 Factors Influencing Analysis ...............................................................................2-1
2.2 Determination of Annual Fuel Cost Penalty .................................................................2-2
2.2.1 Elements Affecting Overall Steam Path Efficiency Degradation .........................2-2
2.2.2 Determination of the Financial Penalty of Losses ...............................................2-4
2.3 Calculation of Steam Leakage Flow ..........................................................................2-11
2.3.1 Presentation of Martin’s Equation .....................................................................2-11
2.3.2 Application of Martin’s Equation........................................................................2-14
2.3.3 The Sensitivity of Losses to Cost ......................................................................2-21
2.4 Advanced Seal Designs and Influence on Operating Costs ......................................2-25
2.4.1. Unit Startup .......................................................................................................2-25
2.4.2 Unit Shutdown...................................................................................................2-27
2.4.3 Unit Operating Transients and Trips .................................................................2-27

3 COMMERCIALLY AVAILABLE STEAM TURBINE SEALS..................................................3-1

3.1 Traditional Spring-Backed Labyrinth Packing Rings....................................................3-1

ix
 EPRI Proprietary Licensed Material

3.2 Retractable Packing.....................................................................................................3-2

3.3 Brandon Sensitized Packing ........................................................................................3-4
3.4 Retractable Brush Seal Packing ..................................................................................3-5
3.5 Guardian Seal Design..................................................................................................3-8
3.6 Seals Used on Rotating Blade Tips .............................................................................3-9
3.6.1 TurboCare Brush Spill Strip ................................................................................3-9
3.6.2 HELP-R Spill Strip.............................................................................................3-10
3.6.3 Vortex Shedder Spill Strip .................................................................................3-11
3.7 Summary of Commercially Available Replacement Seals .........................................3-12

4 DISCUSSION..........................................................................................................................4-1
4.1 Reliability Versus Performance....................................................................................4-1
4.2 Installation Issues ........................................................................................................4-1
4.3 Springs on Variable-Clearance Packing ......................................................................4-3
4.4 Operational Issues .......................................................................................................4-3
4.5 Industry Experience .....................................................................................................4-3

5 REFERENCES .......................................................................................................................5-1

x
 EPRI Proprietary Licensed Material

LIST OF FIGURES

Figure 1-1 Diagram of a Stepped Labyrinth Seal Assembly ......................................................1-3

Figure 1-2 Simulation of Flow Patterns in a Stepped Labyrinth Seal.........................................1-3
Figure 2-1 Financial Penalty of a kW of Lost Output as a Function of Fuel Cost and Load
Factor for an SHR of 10,000 Btu/kW-hr (10,551 kJ/kW-hr)................................................2-5
Figure 2-2 Section Steam Conditions and Flows.......................................................................2-8
Figure 2-3 Steam Conditions and Flows Around a Three-Section Unit ...................................2-10
Figure 2-4 Critical Pressure Ratio............................................................................................2-12
Figure 2-5 Series Arrangement of Labyrinth Seals ..................................................................2-13
Figure 2-6 a) Series Arrangement of Seals Causing Successive Throttling of Leaking
Steam from Pressure P1–P5, Past Seals A–D and b) Steam Leakage of Figure 2-
6a on the Mollier Diagram ................................................................................................2-14
Figure 2-7 Basic Flow Coefficients ..........................................................................................2-16
Figure 2-8 Leakage Discharge Coefficient for a Labyrinth Seal in Terms of the Seal
Geometry .........................................................................................................................2-17
Figure 2-9 Details of a 50% Reaction Stage for Example 6.....................................................2-17
Figure 2-10 Details of an Impulse Stage for Example 7 ..........................................................2-19
Figure 2-11 Financial Penalty for Excess Clearance at the Diaphragm Inner Seals in an
Impulse Type Stage .........................................................................................................2-24
Figure 2-12 Financial Penalty for Excess Clearance at the Blade Tip Seals in an Impulse
Type Stage.......................................................................................................................2-24
Figure 2-13 Steam Pressure Distribution Around a Gland Ring During Both Normal
Operation (Light Shading) and at Startup (Dark Shading) ...............................................2-26
Figure 3-1 Diagram and Photograph of Spring-Backed Labyrinth Seal .....................................3-2
Figure 3-2 Diagram of Brandon Retractable Packing ................................................................3-3
Figure 3-3 Photograph of Brandon-Type Retractable Packing Segments .................................3-3
Figure 3-4 Photograph of TurboCare Retractable Packing Segment – Similar to Brandon
Design ................................................................................................................................3-4
Figure 3-5 Photograph of Conventional Spring-Backed Seals (Left) and Variable-
Clearance Seals (Right) with the Typical Clearance Provided at Startup ..........................3-4
Figure 3-6 Diagram of Sensitized Packing.................................................................................3-5
Figure 3-7 Diagram of the Brush Bristle Pack that Replaces a Standard High Knife in a
Retractable Brush Seal Packing Assembly ........................................................................3-6
Figure 3-8 Brush Bristle Pack Installed in a Retractable Seal Packing Element .......................3-6

xi
 EPRI Proprietary Licensed Material

Figure 3-9 Photograph Showing the Condition of a Brush Seal After Seven Years of
Operation in a Fossil Steam Turbine..................................................................................3-7
Figure 3-10 Photograph of a PerkinElmer Centurion Variable-Clearance Brush Seal
Assembly............................................................................................................................3-8
Figure 3-11 Photograph of Guardian Seal Showing the Post Elements Designed to
Protect Adjacent Labyrinth Knives .....................................................................................3-9
Figure 3-12 Photograph of Brush Spill Strip ............................................................................3-10
Figure 3-13 TurboCare HELP-R Spill Strip ..............................................................................3-11
Figure 3-14 Photograph of Vortex Shedder Spill Strip for Steam Turbine Application.............3-12
Figure 3-15 Diagram of Vortices Created In Leakage Jet Above Rotating Blades ..................3-12
Figure 4-1 Diagram Showing Neck Clearance on a Variable-Clearance Packing .....................4-2

xii
 EPRI Proprietary Licensed Material

LIST OF TABLES

Table 2-1 Table of Annual Fuel Cost Savings ...........................................................................2-7

Table 2-2 Fuel Cost Penalty for the 450,000-kW Unit in Example 4..........................................2-8
Table 2-3a Leakage Losses as a Function of Radial Clearance (in English Units) .................2-21
Table 2-3b Leakage Losses as a Function of Radial Clearance (in SI Units)..........................2-22
Table 2-4a Annual Fuel Cost Penalty Resulting from Excess Clearances (in English
Units)................................................................................................................................2-23
Table 2-4b Annual Fuel Cost Penalty Resulting from Excess Clearances (in SI Units)...........2-23
Table 3-1 Information on Commercially Available Seals for Steam Turbine Applications .......3-13

xiii
 EPRI Proprietary Licensed Material

1
INTRODUCTION

1.1 Report Objective

The continued focus by plant owners on increasing production capacity and the efficiency of
existing generating units has led to an interest in low-cost upgrades to turbine steam paths.
Replacement interstage seal assemblies, designed as retrofits to the existing steam path
components, can be part of a viable and cost-effective steam turbine upgrade strategy. These seal
improvements can be particularly beneficial in the high-pressure (HP) and intermediate-pressure
(IP) turbine sections, where stage pressure drops are high, and the steam leakage is a relatively
greater portion of the overall flow. Both original equipment manufacturer (OEM) and non-OEM
suppliers now offer improved seal design features and materials including advanced labyrinth,
brush, and retractable seals. The current industry emphasis is on maintaining increased efficiency
over an extended time between maintenance overhauls. Therefore, both the initial
thermodynamic performance and durability are important criteria in deciding which replacement
seal alternatives to employ. This report will address commercially available replacement seals
from an end-user perspective and will provide information that can assist in decisions on
replacement seal alternatives including the expected performance benefits, a basic understanding
of how the seal assembly operates, and the experience of current users of these replacement seal
designs. Because of the extensive use of labyrinth-type seals in large steam turbines, the focus of
this report will be on seals that can be retrofitted into labyrinth assemblies.

It is emphasized that EPRI does not endorse any specific replacement seal type, design, or
manufacturer. The purpose of this report is to provide objective information on the value of any
potential seal replacement project, the basic types of designs offered, and field experience where
available.

1.2 Report Scope

Sections 2–4 of this report cover the following topics:

• The performance effects and economics associated with seal degradation, increased leakage
flow, and reduced clearance through maintenance or upgrade
• A general description of several replacement seal design concepts
• Information provided by plants on operating experiences with various seal types

1-1
 EPRI Proprietary Licensed Material

Introduction

The basic seal types described in this report include labyrinth, retractable packing, retractable
brush packing, brush tip, Guardian 1 , and Vortex Shedder 2 . In addition to these seal concepts,
ongoing research is exploring improvements in the basic labyrinth seal design, including
optimizing knife shape and spacing [1, 2].

In preparing this report, both OEM and non-OEM steam turbine seal providers were invited to
contribute information on their current design concepts and products for both new steam paths
and upgrades. This report contains information from all of the companies that responded to
EPRI’s invitation to provide information.

1.3 Description of Steam Turbine Seal Function

Many technical resources are available that describes the role of interstage seals in maintaining
large steam turbine thermal performance [3–5].

The function of the turbine seal or packing is to reduce the quantity of steam leakage at two
locations:
• From the steam path at the shaft end of the turbine section
• Around the stage blade elements internal to the steam path at the clearances between the
shaft and diaphragms, and between the casing and blade/bucket tips

In high-efficiency turbines, a substantial effort is made to confine steam flow to the steam path
defined by the nozzle and blade cascades. Stepped labyrinth seals (high-low packing) with
multiple knives are the most common method used in order to reduce leakage flow by providing
a tortuous flow area (see Figure 1-1). The leakage jet in a stepped labyrinth seal is bounded by
counter-rotating vortices that restrict the leakage flow (see Figure 1-2). Leakage steam that
bypasses the fixed blades is not properly directed toward the downstream rotor blades, reducing
rotor performance. Likewise, steam that bypasses the rotating blade tips does not produce work
within the turbine stage. When all leakage flow streams re-enter the steam path, there are losses
associated with its interaction and entrainment with the main flow. The overall resulting
reduction in work performed reduces turbine efficiency and output and increases unit heat rate.

1
Guardian is a registered trademark of Turbo Parts LLC, an MD&A Company.
2
Vortex Shedder is a registered trademark of Turbo Parts LLC, an MD&A Company.

1-2
 EPRI Proprietary Licensed Material

Introduction

Figure 1-1
Diagram of a Stepped Labyrinth Seal Assembly [1]

Figure 1-2
Simulation of Flow Patterns in a Stepped Labyrinth Seal [1]

The amount of steam leakage is strongly related to the operating clearance between the rotating
and non-rotating elements that are separated by the seal. A second major influence on steam
leakage is the pressure difference that separates the inlet and exit of the seal assembly. Seals
located in regions of high pressure drop experience greater leakage flow. Steam turbines that
have opposed, single-flow HP and IP blading sections in a single casing must maintain a very
high pressure drop between main steam inlet pressure and the reheat turbine, thus requiring a
multistage seal assembly in order to prevent leakage flow. Often referred to as the N2 packing,
this seal is at the mid-span of the HP-IP rotor and is therefore subject to an increased risk of
damage due to shaft rubs at startup. Several other factors that influence seal performance are
discussed in Section 2 of this report.

1-3
 EPRI Proprietary Licensed Material

Introduction

1.4 Damage and Degradation of Seals

The stationary and rotating surfaces of turbine seal elements can be subjected to high-speed rubs,
large pressure differentials, and high temperatures of surrounding steam and metal. If rubbing
occurs with high-contact forces, the seal knives and shaft will heat up to excessive levels, raising
the temperature of the affected components beyond the austenitizing temperature. The
surrounding steam creates a quenching effect that can cause the seal material to become brittle
and susceptible to failure during future rub events. Similarly, there is a potential for quenching
damage to the shaft, resulting in areas of increased hardness and possible crack initiation sites.
The resulting deformation of the shaft surface due to wear will also increase the seal clearance
and leakage flow.

Labyrinth seal knives have tapered cross-sections so that any accidental rub will limit frictional
heating by reducing the contact area. A major part of the effort in reducing the steam leakage
involves minimizing the incidence and severity of these rubs during unit startup, shutdown, and
system transients. Shaft rubs can occur during startup when the rotor passes through critical
speed and the resulting mid-span lateral shaft deflections are high enough to exceed the
clearance. The permanent deformation of knife shape as a result of these rubs increases the
leakage area and leakage flow. In many steam turbines, backing springs are used to hold seal
segments in place but also allow them to be compliant in the radial direction in the event of a
hard rub between the shaft and the seal. This compliance reduces the frictional heating associated
with rubs.

Seal degradation and the need to periodically restore damaged elements to their original
condition represent a barrier to the current industry desire to extend scheduled maintenance
intervals on large steam turbines. It is clear that design improvements that allow seal integrity
and performance to be sustained over a longer period of time would be valuable and must be
achieved if the continued need for increased capacity and availability is to be met.

Some additional seal damage mechanisms not associated with rubs include:
• Water droplet erosion, most likely to occur in a nuclear unit because the steam admitted to
the HP turbine is below saturation
• Solid particle erosion, due to boiler scale carryover in the areas of the turbine that receives
steam directly from the boiler (occurs primarily in the initial stages of the HP and IP
turbines), a condition that can cause an increase in leakage area in the seal strips (the outer
diameters of the stages)
• The potential effect of mechanical impact damage caused by foreign objects

1.5 Maintaining Desired Seal Performance

Replacing and repairing steam turbine seals returns the clearances to design values, but the goal
is also to maintain the turbine efficiency as long as possible until the next scheduled turbine
outage. The following items are important steps that should be taken in order to maintain seal
performance.

1-4
 EPRI Proprietary Licensed Material

Introduction

1.5.1 During Maintenance Outages

During maintenance outages:

• Ensure that the total alignment of the unit is set at the design-specified conditions.
• Check and adjust the rotor balance if any corrective action has been taken on the rotating
elements.
• Check the inner casing and diaphragm ovality.
• Clean the steam path components of any deposits.

1.5.2 During Normal Operation

During normal operation:

• Maintain the control of water quality by avoiding condenser leaks and implementing proper
water treatment procedures.
• Maintain steam conditions at the design values.
• Follow specified operating procedures at startup and shutdown by observing dwell times and
the requirements for passing through rotor critical speeds.
• Ensure that the unit remains on turning gear until the rotor and casings are cool.

1.6 Nomenclature

HP high pressure
IP intermediate pressure
LP low pressure
OEM original equipment manufacturer
psi pounds per square inch pressure
MPa megapascals pressure
FOD foreign object damage
SHR station heat rate
Btu British thermal unit
kW kilowatts
hr hour
lbs pounds
kg kilogram
mm millimeter

1-5
 EPRI Proprietary Licensed Material

Introduction

kJ kilojoule
FC fuel cost
m mass
H enthalpy
HR heat rate
LF leakage flow
F loss factor
Ae leakage area
Ds diameter of leakage annulus
Cl clearance
Φ flow coefficient
X pressure ratio
P pressure
Vsi specific volume
N number of flow restrictions

1-6
 EPRI Proprietary Licensed Material

2
ECONOMICS OF SEAL OPERATION

2.1 Introduction

2.1.1 Seal Maintenance Decision Analysis

The function of steam path seals within the turbine is to minimize any leakage that allows steam
to bypass the blade rows. This leakage flow develops no power and thereby reduces the
mechanical power output of each stage. A stage loss reduces the amount of power produced for a
fixed level of steam available energy and mass flow and, as a consequence, lowers the unit
efficiency. Today’s elevated fuel costs and reduced capacity margins on many systems have
increased the importance of steam turbine performance. Therefore, it is essential that any means
available to maintain leakage at or below original design levels will have an overall beneficial
effect on system economics. In this section of the report, economics of seal performance will be
considered at the blade row, turbine stage, section, and unit levels. A method for evaluating the
effect of increased leakage flows on output and generating costs will be presented.

In preparing and conducting a steam turbine maintenance outage, decisions regarding the
continued use or refurbishment of seal elements need to be examined and justified on an
economic basis. The decisions should be based on the expected return on investment (ROI),
factoring in the cost of repair and the effect of future operating efficiency. This is a logical
approach and makes economic sense. The intent of this section of the report is to provide turbine
engineers responsible for making economic decisions on seal maintenance with an analysis
methodology that will allow the calculation of ROI, either in terms of saved fuel costs or
increased power output.

2.1.2 Factors Influencing Analysis

ROI analyses must recognize that the power plant and the steam turbine in particular are
sensitive to a number of external influences whose presence or effects are difficult to predict. In
conducting the analyses required to justify maintenance on steam turbine seals, it is necessary for
the engineers responsible to make certain assumptions and projections regarding future operating
characteristics and requirements. The required projections include analysis input factors such as
station heat rate, unit load patterns, and fuel cost changes over the anticipated operating period. It
is acknowledged that some of this input is based on projections, but results based on these
assumptions and subsequent sensitivity analyses will provide better decision support than non-
rigorous decision methods. Perhaps the most appropriate analysis method is to examine the
previous operating history of the unit and use it as a basis for anticipating future events and

2-1
 EPRI Proprietary Licensed Material

Economics of Seal Operation

operating characteristics. This is an important consideration when a unit is being returned to

service for an extended period of 8–10 years.

Historical data on wear and damage versus operating time for specific units can be a basis for
projecting the scope and interval of future seal refurbishment. It is the philosophy of many
turbine owners to refurbish all turbine seals that are accessible at every maintenance outage.
With today’s escalating fuel costs, shrinking generating reserves, and the cost of new power
plants, this approach is prudent and may be easily justified on large base-load units. In order to
decide on the most cost-effective course of action regarding seal refurbishment, replacement, or
possible design upgrade, an analysis is undertaken based on the projected losses associated with
the damage that is expected to occur.

2.2 Determination of Annual Fuel Cost Penalty

2.2.1 Elements Affecting Overall Steam Path Efficiency Degradation

During operation, there are a number of mechanisms that will reduce the expansion efficiency of
the steam path and therefore reduce the output from the unit. Most of the losses are recoverable
because the steam path can be restored through maintenance during overhaul. Steam path audits
performed at the start of an overhaul can identify the cost and benefit of undertaking each
maintenance activity related to improving the condition of the steam path.

To monitor the degradation during operation, a reduction in section state-line efficiency can be
trended by conducting periodic enthalpy drop tests. Enthalpy drop tests are straightforward for
turbine sections in which the steam remains superheated throughout the cylinder. For those
sections, the actual enthalpy drop from inlet to exhaust (based on steam temperature and
pressure) is compared to the ideal (isentropic) expansion to the same exit pressure level. A
gradual reduction in the actual enthalpy drop indicates that less work is extracted from the steam
as a result of steam path degradation and reduced internal efficiency. Turbine sections in which
the steam undergoes a phase change (such as LP turbine sections) cannot easily be monitored
using enthalpy drop tests because the steam exit temperature is no longer a parameter that
determines the exit enthalpy. In these cases, it would be necessary to measure the steam quality
of the exhaust flow [6].

The most common types of steam path losses are identified in the following paragraphs:
• Steam path deposits – Steam path deposits include chloride deposits in the LP turbine and
caustic deposits, silica, and copper in the HP turbines of units with an inlet pressure above
2400 psi (16.55 MPa). Copper deposition has the greatest effect on unit output because it
influences the mass flow admitted to the entire turbine. All deposits, including those
downstream of the control stage, increase surface roughness and reduce stage efficiency
accordingly.
• Erosion/corrosion – Erosion/corrosion at the inlet stages of the HP and IP turbine can occur
due to a carryover of exfoliated oxides from the inside surfaces of superheat and reheat tubes.
This erosion affects the nozzle geometry and throat discharge area. Blade surface erosion can
also occur due to the impact of water droplets entrained in the flow after the phase transition

2-2
 EPRI Proprietary Licensed Material

Economics of Seal Operation

zone. The performance impact of moisture erosion is typically not significant. Corrosion
pitting of LP turbine elements can also increase surface roughness and reduce stage
efficiency.
• Mechanical damage – Mechanical damage to steam path elements can occur as a
consequence of impact by loose objects within the steam path. This is referred to as foreign
object damage (FOD). The foreign objects could be material inadvertently left in the steam
path following maintenance or produced by damaged components such as blade coverbands,
tiewires, or sections of blade tips that experience failure and become loose inside the turbine.
• Nozzle degradation – Proper steam path performance is dependent upon maintaining the
individual stage pressure drops intended by the designer. This pressure drop is primarily
determined by the discharge area of each row of nozzles and is selected by the designer for
each stage to optimize overall turbine efficiency. Any damage that produces a change to the
geometry of the nozzle throat area will influence the pressure drop and affect the efficiency
of that stage and those stages downstream.
• Increased seal clearances – Seal rubbing will damage the labyrinth knife elements and open
the clearances, thus increasing the leakage area and steam leakage flow. This results in
reduced unit output.

The traditional approach to controlling leakage loss has been to replace worn seals at each
outage, if justified by the extent of wear. However, unless the root cause of repeated seal rubs is
identified and corrected, rubs will reoccur, and seal clearances will again increase. This damage
typically occurs shortly after the unit is returned to service, thus eliminating much of the
anticipated economic gain on which the replacement project was based. The root cause may be
operational (poor startup procedure) or maintenance-related (improper rotor balance or closing
clearances that are incorrect). Hard rub contact produces deformation of the knife edges of the
labyrinth, causing the radial clearance and leakage area to increase. The knife deformation
damage has two detrimental effects on sealing efficiency. First, it will open the clearance,
causing an increase in leakage area. Second, the rub will modify the shape of the knife profile,
changing the sharp knife corner to a rounded corner and increasing the flow coefficient. Both of
these effects will increase the leakage flow rate for a given pressure drop.

It is important for the turbine operator to know what level of seal clearance area increase can be
tolerated, and what is the consequential financial penalty that is associated with additional
leakage resulting from seal rubs. Such information will allow a quantitative judgment to be made
at outages, improve operator awareness of maintaining unit performance, and support decisions
to replace the existing seal elements. Turbine manufacturers often provide plant operators
information on the rate of expected power loss per unit increase in seal radial clearance in order
to provide guidance in repair decisions. These data can be used with reasonable accuracy in order
to estimate losses and establish the need for the replacement of seals.

2-3
 EPRI Proprietary Licensed Material

Economics of Seal Operation

2.2.2 Determination of the Financial Penalty of Losses

If a steam turbine section is opened for planned maintenance on four- to eight-year intervals, a
calculation procedure should be available in order to allow an assessment of the losses and
financial consequences associated with continued operation in the as-found condition. The
methodology that allows loss and penalty to be established can then be used to anticipate losses
during the next operating period.

The seven examples presented in this section show how the operator can predict additional fuel
costs associated with an increase in seal clearances and leakage. An appropriate means of
making such economic evaluation data available is through a parametric analysis relating values
assumed for principal governing factors to resulting operating costs (primarily increased fuel
costs). The following examples should help establish a methodology that plant personnel can use
to make more informed decisions.

Example 1: Consider a unit with an annual load factor of 80% having a station heat rate (SHR)
of 10,000 Btu/kW-hr (10,551 kJ/kW-hr). The plant uses fuel that costs $2.00/million Btu
($1.90/million kJ). It is desired to predict the annual additional fuel cost per lost kW as the result
of steam path deterioration. (Note that there are 8,760 hours in a year).

For this unit, fuel cost per kW-hr generation = fuel cost × SHR
2.0
= × 10,000
1,000,000
= $0.02/kW-hr

That is, it will cost $2.00 to generate 100 kW for 1 hour.

The annual fuel savings per kW for the unit discussed above is equal to:
Cost savings = 1 year × 8,760 hours × 0.8 × $0.020/kW-hr = $140.16/kW per year

2-4
 EPRI Proprietary Licensed Material

Economics of Seal Operation

Figure 2-1
Financial Penalty of a kW of Lost Output as a Function of Fuel Cost and Load Factor for an
SHR of 10,000 Btu/kW-hr (10,551 kJ/kW-hr)

It should be noted that this loss is independent of the unit rating. That is, a kW lost from a
30,000-kW unit is just as expensive as a kW lost from a 750,000-kW machine if the fuel costs,
load factor, and station heat rate are the same.

Using the method above, the chart in Figure 2-1 has been generated for fuel costs ranging from
$0.75 per million Btu ($0.71 per million kJ) to $3.00 per million Btu ($2.84 per million kJ). This
chart also covers a range of load factors from 30–100% for a unit with a station heat rate of
10,000 Btu/kW-hr (10,551 kJ/kW-hr). For other station heat rates, the cost per kW lost
generation can be determined by the ratio of the actual SHR to the reference value of 10,000
Btu/kW-hr (10,551 kJ/kW-hr).

Example 2: Consider a generating unit that burns fuel that costs $2.00 per million Btu ($1.90 per
million kJ), has a load factor of 80%, and a station heat rate of 12,650 Btu/kW-hr (13,252
kJ/kW-hr). What are the costs per kW of output?

From Figure 2-1, the loss for a unit with a SHR of 10,000 Btu/kW-hr (10,551 kJ/kW-hr) is:

$140.16/kW per year, (see Example 1)

The loss per kW for a SHR of 12,650 Btu/kW-hr (13,252 kJ/kW-hr) per year is:

2-5
 EPRI Proprietary Licensed Material

Economics of Seal Operation

$ 12,650 $177.44
= $140.16 × =
kW 10,000 kW

Example 3: This example demonstrates the use of curves in Figure 2-1 and the determination of
losses over an extended operating period for a 500,000-kW unit. This unit has an annual load
factor of 70%, an SHR of 12,500 Btu/kW-hr (13,188 kJ/kW-hr), and a fuel cost at $2.50 per
million Btu ($2.37 per million kJ). Assume that a steam path audit performed on the unit
removed from service showed that leakage losses had increased by a total of 2,350 kW beyond
the design values. Also assume that these additional losses were corrected by restoring the seal
clearances before the unit was returned to service. The following calculation estimates the annual
fuel cost savings derived from the restored seals.

Referring to Figure 2-1 and using a capacity factor of 70%, a fuel cost of $2.50 per million Btu
($2.37 per million kJ), and adjusting for SHR difference, the annual fuel cost per kW production
is:

$ 12,500 $191.63
= $153.30 × =
kW 10,000 kW

Therefore, as returned to service, the annual fuel cost saving is:

= 2,350kW × $191.63/kW
= $450,318.75

This amount represents the savings in the first year after the return to service. For subsequent
years, it is possible that the fuel costs will increase and that the initial improvement will not be
sustained throughout the operating period to the next maintenance outage. Based on the
assumptions of constant fuel costs and a five-year maintenance cycle, with the initial kW
improvement being reduced due to seal degradation, the projected cost savings shown in
Table 2-1 can be anticipated, assuming the load factor remains constant at 70%. Note that the
assumed projected kW improvement over the five-year period is reduced nonlinearly from the
initial 2,350-kW improvement.

2-6
 EPRI Proprietary Licensed Material

Economics of Seal Operation

Table 2-1
Table of Annual Fuel Cost Savings

Years After kW Improvement Predicted Fuel Cost in Annual Fuel Cost

Return to Service Relative to End of Prior $/million Btu Savings ($)
Operating Cycle ($/million kJ)

1 2,350 2.50 (2.37) 450,319

2 2,010 2.50 (2.37) 385,166

3 1,860 2.50 (2.37) 356,423

4 1,760 2.50 (2.37) 337,260

5 1,670 2.50 (2.37) 320,014

Predicted Total $1,849,182

Fuel Cost Savings:

Therefore, as the unit is returned to service, there is a five-year savings of about $1.85M,
assuming that there are no transient operating conditions that would cause a seal rub and an
excessive opening of the clearances.

An expression that allows the annual operating fuel cost to be determined for any unit is as
follows:

kW × FC × SHR × ΔHR × 8,760 × LF

annual fuel cost penalty = Eq. 2-1
1.0E10

Where:
kW = Unit rating in kilowatts
FC = Fuel cost in $/million Btu (multiply by 0.9478 to obtain
$/million kJ)
SHR = Station heat rate
ΔHR = Percent change in station heat rate
8,760 = Number of hours in a year
LF = Unit load factor (%)
Example 4: As an example of the application of Equation 2-1, consider a unit rated at
450,000 kW that has a station heat rate of 11,000 Btu/kW-hr (11,606 kJ/kW-hr). It is assumed
that for the next 10 years the unit will operate with a load factor of 85% and that the increase in
heat rate over this time period will eventually total 3.05%. A nonlinear time variation in heat rate
degradation is assumed in accordance with the data in column two of Table 2-2, with the
maximum rate of increase occurring during the early years of operation after return to service. In
this example, it has been assumed that the fuel cost will increase by $0.10 per year over the 10-
year operating cycle. The annual and cumulative fuel cost penalties are calculated using Equation
2-1 and presented in Table 2-2.

2-7
 EPRI Proprietary Licensed Material

Economics of Seal Operation

Table 2-2
Fuel Cost Penalty for the 450,000-kW Unit in Example 4

Year HR Increase (%) FC in $ per Million Btu Annual FC Annual

($/million kJ) Penalty ($) Cumulative FC ($)
0–1 0.60 2.50 (2.37) 552,900 552,900
1–2 1.25 2.60 (2.46) 931,680 1,484,580
2–3 1.70 2.70 (2.56) 1,315,820 2,800,360
3–4 2.00 2.80 (2.65) 1,605,360 4,405,720
4–5 2.25 2.90 (2.75) 1,870,530 6,276,250
5–6 2.45 3.00 (2.84) 2,107,030 8,383,280
6–7 2.60 3.10 (2.94) 2,310,570 10,693,850
7–8 2.78 3.20 (3.03) 2,550,230 13,244,080
8–9 2.85 3.30 (3.13) 2,696,140 15,940,220
9–10 3.05 3.40 (3.22) 2,972,780 18,913,000

The analysis presented in Table 2-2 shows that during a 10-year cycle the cumulative additional
fuel costs due to heat rate deterioration are nearly $19,000,000.

Equation 2-1 requires that a projected time/history be assigned for the heat rate increase due to
seal degradation. This projection can be established from past unit performance using station
instruments or calculated from information gathered in enthalpy drop tests as described in
Section 2.2.1.

The influence of efficiency degradation of individual turbine sections on overall cycle

performance can be evaluated. Using the rated heat balance and the steam parameters from
Figure 2-2, an assessment can be made using Equations 2-2 through 2-4.

Figure 2-2
Section Steam Conditions and Flows

2-8
 EPRI Proprietary Licensed Material

Economics of Seal Operation

If the loss factor in the HP section is designated as Fh, the loss factor in the IP section as Fr, and
the loss factor in the LP section as Fl, these factors can be defined as:

Fh =
(M hi ) (
+ Mh o × Hhi − Hh o )
2 × ∑ MH

Fr =
(M ri ) (
+ M ro × H ri − H ro ) Eq. 2-2
2 × ∑ MH

Fl =
(M li ) (
+ Mlo × Hli − Hlo )
2 × ∑ MH

These terms are defined below. In these expressions:

( ) ( )
∑ MH = M h H h i − H h o + M r H ri − H ro + M l H l i − H l o ( ) Eq. 2-3

where M = the mean mass flow from inlet (i) to discharge (o) of each cylinder.

=
1
(M i + M o ) Eq. 2-4
2

- Subscript h refers to the HP section.

- Subscript r refers to the IP section.
- Subscript l refers to the LP section.

Example 5: As an example, consider a 500,000-kW unit in which the HP section efficiency has
a measured deterioration of 1.5% in state-line efficiency and a further deterioration of 1% in the
IP section. Also assume that there are sufficient indications from previous inspections, in order to
estimate a 0.75% performance loss for the LP section. Using these data, the total heat rate
degradation can be found. This unit has an SHR of 10,500 Btu/kW-hr (11,078 kJ/kW-hr), and the
fuel costs are $2.50 per million Btu ($2.37 per million kJ).

2-9
 EPRI Proprietary Licensed Material

Economics of Seal Operation

Using the steam cycle parameters shown in Figure 2-3 (to convert from Btu to kJ, multiply by
1.055:

( )
M h H h i − H h o = 1 / 2 (3.428 + 3.111) × (1461.2 − 1305.9)

( )
M r H ri − H ro = 507.8 x 106 Btu
Mr (H ri − H ro ) = 1/ 2 (3.150 + 3.023) × (1520.2 − 1262.0)
( )
M l H l i − H l o = 798.6 x 106 Btu
(H ) = 1/ 2 (2.844 + 2.227) × (1262.0 − 1011.4)
Eq. 2-5
Ml li − Hlo

∑ MH = 635.4 × 106 Btu

∑ MH = (507.8 + 798.6 + 635.4 ) × 106 = 1941.8 × 106 Btu

Figure 2-3
Steam Conditions and Flows Around a Three-Section Unit

Therefore, according to Equation 2-2:

Fh = 507.8/1941.8 = 0.2615
Fr = 798.6/1941.8 = 0.4113
Fl = 635.4/1941.8 = 0.3272

Therefore, the total change in cycle heat rate is calculated as the product of the observed cylinder
efficiency degradation values and the respective loss factors:

HP section: 1.51% × 0.2615 = 0.3923%

IP section: 1.00% × 0.4113 = 0.4113%
LP section: 0.75% × 0.3272 = 0.2454%
Total change in unit heat rate: = 1.0490%

2-10
 EPRI Proprietary Licensed Material

Economics of Seal Operation

This change in cycle heat rate requires that extra fuel be consumed in order to develop the same
unit output. The additional fuel cost (or fuel cost penalty) per year can be calculated from
Equation 2-1 as:

kW × FC × SHR × ΔHR × 8,760 × LF

annual fuel cost penalty =
1.0E10

500,000 × 2.50 × 10,600 × 1.0490 × 8,760 × 80

annual fuel cost penalty =
1.0E10

annual fuel cost penalty = $974,059

2.3 Calculation of Steam Leakage Flow

2.3.1 Presentation of Martin’s Equation

Section 2.2 presented a method for calculating the fuel cost penalty associated with an assumed
lost generating capacity or heat rate increase. Section 2.3.1 describes the procedure for
estimating the actual leakage flow and the resulting impact on efficiency and turbine power
output. The mathematical basis for calculating the quantity of steam leaking past a labyrinth seal
was first proposed by H.M Martin in 1913 [7]. Martin’s equation has been used for many years
by the majority of manufacturers in estimating stage losses and establishing turbine performance
guarantees.

Martin’s equation states that the quantity of flow past a series of labyrinth seals can be
determined from Equation 2-6:

⎛ 1 ⎞
Pi ⎜1 − 2 ⎟
m = k × φ × Ae ⎝ x ⎠
Vsi (N + ln [X ])
Eq. 2-6

Where: Ae = The leakage area = π × Ds × Cl

Ds = The mean diameter of the leakage annulus (in.)
Cl = The radial clearance (in.)
Φ = The flow coefficient (dimensionless)
X = The pressure ratio across the series seals Pi/Pd (dimensionless)
Pi = The inlet pressure (psi)
Pd = The pressure at discharge (psi)
Vsi = The specific volume corresponding to Pi (ft3/lb)
N = The number of series constrictions, see Figure 2-4
k = A numerical conversion constant
= 0.472 for the flow in lbs/second
= 1,700 for the flow in lbs/hour

2-11
 EPRI Proprietary Licensed Material

Economics of Seal Operation

For a series of N labyrinth strips, Martin’s equation assumes that pressure ratio X is constant
across each of these constrictions of the total seal arrangement. The equation also assumes that
the steam thermodynamic conditions at the inlet and discharge of the seal assembly are in either
the superheat or saturated regions for the entire expansion. The leakage flow will be at a
maximum when the value of pressure ratio X reaches the critical value. Because the flow cannot
exceed that associated with the critical pressure ratio, if the value of X exceeds the critical value,
the critical value from Figure 2-4 should be used.

Figure 2-4
Critical Pressure Ratio

Variations of Martin’s equation are used in the determination of the steam leakage past labyrinth
seals, including both shaft seals that are typically installed on fixed blades and spill strips found
on rotating blade tips.

Consider the labyrinth system shown in Figure 2-5. In Figure 2-5(a), the geometric layout of the
labyrinth seal arrangement is shown, and in Figure 2-5(b), the steam properties and enthalpy
levels are presented on a Mollier diagram showing state points at the inlet and discharge
locations of the labyrinth assembly. The leakage flow quantity past this series of labyrinth seals
can now be determined by Equation 2-6.

2-12
 EPRI Proprietary Licensed Material

Economics of Seal Operation

Figure 2-5
Series Arrangement of Labyrinth Seals

To determine the loss in kW, first determine ΔkW, the leakage flow, m, and the enthalpy drop
ΔH across the seals. The kW loss is then calculated using Equation 2-6:

m × ΔH m × (H i − H d )
ΔkW = = Eq. 2-7
3412.14 3412.14

In most steam turbine applications, labyrinth seal groups are employed in series with the pressure
falling successively through each element. At each throttling constriction, a portion of the
thermal potential energy across the entire seal assembly is converted to kinetic energy, and is
subsequently destroyed in the steam chamber that is formed between the labyrinth strips (see
Figure 1-2). This kinetic energy is partially reconverted to pressure energy as its velocity reduces
in the chamber. Most of the remaining energy is converted to heat or thermal energy.

Consider this series’ throttling effect for the four throttling strips in a single-seal assembly as
shown in Figure 2-6(a). The steam has an initial pressure P1 at the entry to the assembly. This
condition is represented on the Mollier diagram in Figure 2-6(b) by point A. After expanding
past this first constriction, the pressure will have been reduced to condition Xo, with pressure P2.
In the chamber formed between the first and second seal strips, the kinetic energy of the steam is
destroyed and reconverted at constant pressure P2 to condition X. From point X, there is then a
further expansion of the steam past the second constriction, with the pressure falling to P3 at
condition Yo. The kinetic energy is again reconverted in the chamber between the second and
third seal strips, raising the thermal energy level from condition Yo to Y at constant pressure P3.
This process of expansion and kinetic energy reconversion is continued throughout the series of
seal strips until the final expansion takes the steam to condition Qo at pressure P5. The locus of
the points Xo.......Qo is called the Fanno curve.

2-13
 EPRI Proprietary Licensed Material

Economics of Seal Operation

Figure 2-6
(a) Series Arrangement of Seals Causing Successive Throttling of Leaking Steam from
Pressure P1–P5, Past Seals A–D and (b) Steam Leakage of Figure 2-6(a) on the Mollier
Diagram

Labyrinth seals installed in operating steam turbines have inlet and discharge pressures that are
determined by the overall cycle parameters, as well as the internal arrangement of the steam path
expansion stages. The two types of flow or pressure distributions that should be considered are:
• Seal assemblies in which the number of constrictions is sufficiently large so that the pressure
drop across each constriction, including the last, is less than critical
• Seal assemblies in which there are insufficient constrictions for the energy range so that flow
through the final constriction has a pressure ratio, X, that exceeds the critical value

If the pressure ratio indicates that the flow is critical, then the value of X from the curve in
Figure 2-4 should be used.

2.3.2 Application of Martin’s Equation

In calculating leakage flow using Martin’s equation (Equation 2-6), the unknown in any seal
arrangement is the flow coefficient, Φ. The flow coefficient is dependent upon a number of
physical characteristics of the labyrinth knives, including tip geometry and the arrangement of
the seals.

2-14
 EPRI Proprietary Licensed Material

Economics of Seal Operation

The difficulty in establishing a flow coefficient is that once a rub has occurred that damages the
knife edges, the seal is no longer in a condition for which accurate flow coefficient data are
available. Rubbed labyrinth seals could be blunted, bent, or rolled or have a mushroom surface at
the sealing point. All of these are conditions that invalidate standard flow coefficient data
obtained from seals that are tested in a new or undamaged condition. However, an estimate of the
flow coefficient for damaged labyrinth knives can be established taking into account the
following characteristics [8]:

Φ=Φ1×Φ2 Eq. 2-8

Where Φ = The flow factor for Martin’s equation

Φ1 = The basic form factor (see Figure 2-7) [8]
Φ2 = Sharpness factor:
= 1.0 for 100% circumferential rub
= 0.9 for 50% circumferential rub
= 0.8 for sharp teeth

A correction term for the flow coefficient allowing for the effect of a rub is also provided in
Figure 2-8, indicating the flow coefficient for a labyrinth seal in terms of the ratio of clearance to
knife edge thickness.

2-15
 EPRI Proprietary Licensed Material

Economics of Seal Operation

Figure 2-7
Basic Flow Coefficients

2-16
 EPRI Proprietary Licensed Material

Economics of Seal Operation

Figure 2-8
Leakage Discharge Coefficient for a Labyrinth Seal in Terms of the Seal Geometry [8]

Example 6: As an example of the application of Martin’s equation, consider the 50% reaction
stage shown as Figure 2-9. In Figure 2-9(a), the geometry of the stage is presented and the
nominal clearance is 0.025 in. (0.635 mm). In Figure 2-9(b), the steam properties at the stage
ends and row transition points are defined.

Figure 2-9
Details of a 50% Reaction Stage for Example 6

2-17
 EPRI Proprietary Licensed Material

Economics of Seal Operation

In making these loss calculations, it can be assumed that the flow coefficient estimated, using the
factors shown in Equation 2-8 for both the stationary and rotating blade rows, is determined to be
0.82.

For the stationary blade row:

Leakage area: Ae = (π)(20.40)(0.025) = 1.602 in.2 (1,034 mm2 )
Pressure ratio: X = 1970/1871 = 1.0529

Therefore, applying Martin’s equation:

⎛ 1 ⎞
1970 ⎜1 − ⎟
m = 0.472 × 0.82 × 1.602 ⎝ 1.05292 ⎠
0.3852 × (2 + ln [1.0529])

Resulting in: m = 9.688 lbs/second (4.394 kg/second) = 34,878 lbs/hour (15,820 kg/hr)

The resulting loss of kW, using Equation 2-8 is estimated as:

34,878 × (1450.4 − 1443.7 )

ΔkW = = 68.49 kW
3412.14

Therefore, the loss rate per 0.001 in. (0.025 mm) of clearance for this stationary blade row would
be:

68.49 kW
Loss per mil of clearance = = 2.74
25 mil

For the rotating blade row:

Leakage area: Ae = (π)(21.50)(0.025) = 1.689 in.2 (1,090 mm2)
Pressure ratio: X = 1871/1772 = 1.0559

Therefore, applying Martin’s equation:

⎛ 1 ⎞
1871 ⎜1 − ⎟
m = 0.472 × 0.82 × 1.602 ⎝ 1.05592 ⎠
0.4013 × (2 + ln [1.0559])

Resulting in: m = 9.996 lbs/second (4.534 kg/second) = 35,984 lbs/hr (16,322 kg/hr)

The loss of kW, using Equation 2-7 is:

35,984 × (1443.7 − 1437.1)

ΔkW = = 69.60 kW
3412.14

2-18
 EPRI Proprietary Licensed Material

Economics of Seal Operation

Therefore, the loss rate per 0.001 in. (0.025 mm) for this stationary blade row would be (to
convert kW/mil to kW/mm, multiply by 39.37):

69.60 kW
Loss per mil of clearance = = 2.78
25 mil

Example 7: A similar evaluation of an impulse (low reaction) type unit can also be made. Figure
2-10 shows the geometric layout of the stage and the corresponding steam conditions provided at
four locations.

In making these calculations, it is noted that the steam properties at the location between fixed
and rotating rows vary from hub to tip. This variation is a result of the radial pressure gradient
between the stages. In general, this variation is small enough that it can be ignored for stages
with a ratio of blade height to mean diameter smaller than 0.15. Again, assume that the flow
coefficient is 0.85 for both rows.

Figure 2-10
Details of an Impulse Stage for Example 7

For the stationary blade (diaphragm):

Leakage area: Ae = (π)(33.60)(0.025) = 2.639 in.2 (1,703 mm2)
Pressure ratio: X = 1945/1690 = 1.1509

2-19
 EPRI Proprietary Licensed Material

Economics of Seal Operation

Therefore, applying Martin’s equation:

⎛ 1 ⎞
1945 ⎜1 − ⎟
m = 0.472 × 0.85 × 2.639 ⎝ 1.15092 ⎠
0.390 × (12 + ln [1.1509])

Resulting in: m = 12.504 lbs/second (5.672 kg/second) = 45,016 lbs/hr (20,419 kg/hr)

The loss of kW, using Equation 2-8 is:

45,016 × (1442 − 1426)

ΔkW = = 211.09 kW
3412.14

Therefore, the loss rate per 0.001 in. (0.025 mm) for this stationary blade row would be (to
convert kW/mil to kW/mm, multiply by 39.37):

211.09 kW
Loss per mil of clearance = = 8.44
25 mil

For the rotating blade row:

Leakage area: Ae = (π)(41.60)(0.060) = 7.841 in.2 (5,059 mm2)
Pressure ratio: X = 1725/1662 = 1.0379

Therefore, applying Martin’s equation:

⎛ 1 ⎞
1725 ⎜1 − 2 ⎟
m = 0.472 × 0.85 × 7.841 ⎝ 1.0379 ⎠
0.426 × (2 + ln [1.0379])

Resulting in: m = 37.554 lbs/second (17.034 kg/second) = 135,194 lbs/hr (61,323 kg/hr)

The loss of kW, using Equation 2-7 is:

135,194 × (1428 − 1442)

ΔkW = = 158.49 kW
3412.14

Therefore, the loss rate per 0.001 in. (0.025 mm) for this stationary blade row would be (to
convert kW/mil to kW/mm, multiply by 39.37):

158.49 kW
Loss per mil of clearance = = 2.64
60 mil

2-20
 EPRI Proprietary Licensed Material

Economics of Seal Operation

2.3.3 The Sensitivity of Losses to Cost

For the impulse stage used in Example 7, the effects of changes to clearance on losses is
tabulated for both the stationary and rotating rows in Table 2-3. A flow coefficient of 0.85 is
again used in all calculations. In Table 2-3, the design losses are established as 211.1 kW for the
diaphragm and 158.5 kW for the rotating blade tip.

The losses that are significant in terms of establishing financial impact are shown as ΔkW and
are a consequence of clearance changes, ΔCl. Clearance changes are relative to the design value.

Table 2-3a
Leakage Losses as a Function of Radial Clearance (in English Units)

Diaphragm
Clearance (in.) 0.025 0.035 0.045 0.055 0.065 0.075
Leakageflow (lbs/hr) 45,016 63,022 81,029 99,035 117,042 135,048

ΔH 1428 - 1424 = 16.0 Btu/lbs

ΔkW 211.1 295.5 380.0 464.4 548.8 633.3

ΔClearance (mils) 0 10 20 30 40 50

ΔkW 0 84.4 168.9 253.3 337.7 422.2

Rotating Blade Row
Clearance (in.) 0.055 0.060 0.065 0.075 0.085 0.095
Leakageflow (lbs/hr) 123,928 135,194 146,460 168,993 191,525 214,057

ΔH 1428 - 1424 = 4.0 Btu/lbs

ΔkW 145.3 158.5 171.7 198.1 224.5 250.9

ΔClearance (mils) -5 0 5 15 25 35

ΔkW -13.2 0 13.2 39.6 66.0 92.4

2-21
 EPRI Proprietary Licensed Material

Economics of Seal Operation

Table 2-3b
Leakage Losses as a Function of Radial Clearance (in SI Units)

Diaphragm
Clearance (mm) 0.635 0.889 1.143 1.397 1.651 1.905
Leakage flow (kg/hr) 20,419 28,586 36,754 44,921 53,089 61,257

ΔH 1428 - 1424 = 16.0 Btu/lbs (37.216 kJ/kg)

ΔkW 211.1 295.5 380.0 464.4 548.8 633.3

ΔClearance (mm) 0 0.254 0.508 0.762 1.016 1.270

ΔkW 0 84.4 168.9 253.3 337.7 422.2

Rotating Blade Row
Clearance (mm) 1.397 1.524 1.651 1.905 2.159 2.413
Leakage flow (kg/hr) 56,213 61,323 66,433 76,654 86,874 97,095

ΔH 1428 - 1424 = 4.0 Btu/lbs (9.304 kJ/kg)

ΔkW 145.3 158.5 171.7 198.1 224.5 250.9

ΔClearance (mm) -0.127 0 0.127 0.381 0.635 0.889

ΔkW -13.2 0 13.2 39.6 66.0 92.4

If the fuel costs for this unit are $2.50 per million Btu ($2.37 per million kJ) and the load factor
is assumed to be 80%, Figure 2-1 shows an annual fuel cost penalty of $174 per year. However,
if the station heat rate is 12,500 Btu/kW-hour (13,188 kJ/kW-hr), the actual fuel cost penalty is:

$ 12,500 $217.50
= $174 × =
kW − hr 10,000 kW − hr

Table 2-4 shows the annual fuel cost penalty for the diaphragm and blade tip leakage resulting
from clearances that are above the design specified values. For the blade tip, an additional data
point is shown that represents a clearance set below the design value of 0.055 in. (1.397 mm).
The design clearances are typically set to provide a sufficient radial space in order to prevent
rubs when the unit is started and shut down in accordance with design specifications. When seal
clearances are used less than design, there is a greater potential for rubs. The occurrence of these
rubs would result in increased clearances.

The financial penalty associated with these losses is shown in Table 2-4 for leakage under the
diaphragm and for the blade tip leakage. Figure 2-11 shows a diagram of the diaphragm losses,
and Figure 2-12 shows the same losses for the rotating blade tip.

2-22
 EPRI Proprietary Licensed Material

Economics of Seal Operation

Table 2-4a
Annual Fuel Cost Penalty Resulting from Excess Clearances (in English Units)

Diaphragm
Clearance (in.) 0.025 0.035 0.045 0.055 0.065 0.075

ΔClearance (mils) 0 10 20 30 40 50

ΔkW 0 84.4 168.9 253.3 337.7 422.2

Δ$ 0 18,357 36,735 55,092 73,450 91,830

Rotating Blade Row

Clearance (in.) 0.055 0.060 0.065 0.075 0.085 0.095

ΔClearance (mils) -5 0 5 15 25 35

ΔkW -13.2 0 13.2 39.6 66.0 92.4

Δ$ -2,870 0 2,870 8,613 14,355 20,097

Table 2-4b
Annual Fuel Cost Penalty Resulting from Excess Clearances (in SI Units)

Diaphragm
Clearance (mm) 0.635 0.889 1.143 1.397 1.651 1.905

ΔClearance (mm) 0 0.254 0.508 0.762 1.016 1.270

ΔkW 0 84.4 168.9 253.3 337.7 422.2

Δ$ 0 18,357 36,735 55,092 73,450 91,830

Rotating Blade Row

Clearance (mm) 1.397 1.524 1.651 1.905 2.159 2.413

ΔClearance (mm) -0.127 0 0.127 0.381 0.635 0.889

ΔkW -13.2 0 13.2 39.6 66.0 92.4

Δ$ -2,870 0 2,870 8,613 14,355 20,097

2-23
 EPRI Proprietary Licensed Material

Economics of Seal Operation

Figure 2-11
Financial Penalty for Excess Clearance at the Diaphragm Inner Seals in an Impulse Type Stage

Figure 2-12
Financial Penalty for Excess Clearance at the Blade Tip Seals in an Impulse Type Stage

2-24
 EPRI Proprietary Licensed Material

Economics of Seal Operation

2.4 Advanced Seal Designs and Influence on Operating Costs

Replacement steam turbine seal designs, currently available as retrofit packages, offer
advantages over the original equipment installed in turbines 20–30 years ago. The main emphasis
with these new designs is to maintain the desired design clearances of seals over an increasingly
longer period of time between scheduled section overhauls. To accomplish this goal, it is
necessary to introduce design provisions that minimize the seal damage that typically occurs
during brief excursions in shaft lateral vibration levels. Excursions that lead to damaging rubs
can take place during startup, while running at critical speeds, or during normal operation if
unbalance forces increase over time. Maintaining seal clearances is one aspect of keeping the
state-line efficiency of stages and entire sections as close as possible to the value corresponding
to the unit’s return to service. These new sealing devices influence only the leakage losses and
will have no effect on losses due to deposits or mechanical damage.

Turbine operating conditions that can cause rubs and seal damage are summarized in the
following paragraphs.

2.4.1. Unit Startup

At startup, the unit will pass through critical speeds that cause the rotor to vibrate due to residual
unbalance. The critical speed vibration amplitude is typically highest at the mid-span of the
rotors, and it can damage the shaft seals on the diaphragms if the vibration amplitude exceeds the
clearances. Conventional seal designs use stiff backing springs to hold the labyrinth seal rings
radially inward. These springs permit some level of protection by allowing the labyrinth
assembly to move outward in the event of a hard rub. Depending on the spring stiffness, the
extent and consequences of startup rubs remain serious.

Retractable seal designs allow the labyrinth seal assemblies to displace radially away from the
rotor surface when steam pressures are low (typically associated with unit startup and shutdown).
Figure 2-13 shows the force diagram around a labyrinth ring assembly of a retractable design
during both normal operation and at startup/shutdown when the steam forces are significantly
lower. It can be seen that there is a radially inward force, which is the sum of the spring force
and the steam force. In the high- and reheat-pressure sections of a turbine, the steam force is
considerably larger than the spring force, forcing the labyrinth assemblies to move inward
toward the shaft.

2-25
 EPRI Proprietary Licensed Material

Economics of Seal Operation

Figure 2-13
Steam pressure Distribution Around a Gland Ring During Both Normal Operation (Light
Shading) and at Startup (Dark Shading)

Other design concepts feature a rub-tolerant strip placed among the labyrinth knives. This rub-
tolerant strip is designed slightly taller than the standard knives surrounding it. This strip is the
first to come in contact with the shaft in rub conditions and thus protects the remaining labyrinth
knives. This design is used in combination with a soft backing spring so that the protective strip
pushes the seal assembly away. The backing spring stiffness is sufficient to ensure that the ring is
held radially inward at all times under normal operation.

For those seals incorporating rub-tolerant strips, the clearances between the conventional knife-
edged seal strip and rotating surfaces are set at normal design levels. The rub-tolerant strip has a
marginally smaller clearance, but it is large enough so that during normal unit startup, contact
will not occur. At startup, the pressures in the steam path are considerably reduced; therefore, the

2-26
 EPRI Proprietary Licensed Material

Economics of Seal Operation

steam forces on the gland rings, shown in Figure 2-13, are not large. Should contact occur
between the shaft and the protective strip, relatively low contact pressures are developed at the
rub surfaces, and the controlling force is determined by the soft backing spring.

2.4.2 Unit Shutdown

When a unit is shut down either under a controlled condition or emergency condition, the main
and intermediate stop and/or control valves close, and the steam pressure levels within the flow
path decay as the rotor slows. For a normal shutdown, the steam pressure is controlled, but the
rotor still passes through its critical speeds with the potential to initiate rubs if procedures are not
followed. So, under controlled shutdown conditions, the radial inward force on the labyrinth seal
assemblies when the rotor passes through critical speed is governed by the spring device behind
the labyrinth rings. If the steam pressure is not reduced as required during shutdown, the steam
forces add to the backing spring forces, hold the labyrinth assembly in place with tight
clearances, and thus risk damaging the seals during coast down.

2.4.3 Unit Operating Transients and Trips

Unit trips represent a risk of seal damage during an uncontrolled shutdown. Under these
conditions, the rotor can accelerate momentarily to overspeed and vibrate either in going to
overspeed or coasting down through its critical speeds. Immediately upon disconnection from the
grid, the stop and control valves will assume a control status, but there is a finite period when
steam pressures are high, and heavy shaft-seal rubs can occur at critical speeds. The retrofit seal
designs described in this report could potentially provide protection to the knife edges of the
conventional seals in this situation.

2-27
 EPRI Proprietary Licensed Material

3
COMMERCIALLY AVAILABLE STEAM TURBINE SEALS

This section includes a description of retrofit seal designs applicable to large steam turbines. The
cost benefit of these design offerings for specific applications can be evaluated using the
techniques described in Section 2. As steam turbine overhaul intervals are lengthened, plant
owners are seeking seal designs that provide the best performance over an extended time, not just
during the time immediately following initial return to service. To achieve this goal, the
replacement seal designs must allow periodic rotor vibration excursions to occur without resulting
in significant damage to the labyrinth knives. The economic analyses previously presented clearly
demonstrate that maintaining proper seal clearances over many years will provide a greater
benefit than restoring operation with initially tight clearances that quickly degrade.

The discussions included in this section describe various design options for retrofit turbine seals.
The first description is of the traditional spring-backed labyrinth packing rings that were used as
original equipment in most large steam turbines that are in use today. A discussion of various
alternatives follows. EPRI does not endorse any particular design concept or supplier. The
purpose of this report is to provide information for plant operators in order to help them make
better-informed decisions.

3.1 Traditional Spring-Backed Labyrinth Packing Rings

The spring-backed labyrinth seal, shown in Figure 3-1, is the most common design employed as
original equipment in hundreds of large steam turbines. This seal design is available as
replacement hardware from a range of suppliers, including third-party component vendors and
steam turbine OEMs. The following are characteristics of this type of seal:
• Fluid leakage is minimized by the use of tortuous flow paths (that is, labyrinths). It is
typically employed as high-low packing in order to improve performance, but where
differential expansion is large, designers use a smooth rotor surface.
• Backing springs located behind the multiple packing ring segments push the seals radially
inwards toward the turbine shaft. The compliance of these springs permit some radial
deflection in the event of a heavy rub condition.
• Fluid pressure distribution (from surrounding steam) around the packing segments produces
a radially inward force that moves the seal assemblies to the minimum clearance position
relative to the shaft (see Figure 2-13). This additional force is normally proportional to
turbine flow and is insignificant under conditions of zero steam flow.
• Rubbing between the shaft and the packing can cause nonuniform shaft heating that leads to
a bowed rotor condition and shaft vibration.

3-1
 EPRI Proprietary Licensed Material

Commercially Available Steam Turbine Seals

Figure 3-1
Diagram and Photograph of Spring-Backed Labyrinth Seal
(Courtesy of Turbo Parts LLC)

3.2 Retractable Packing

Retractable packing was introduced in the early 1980s by Ronald E. Brandon and is now
installed in over 600 steam turbines worldwide [9]. The goal of the Brandon design is to avoid
packing rubs that can occur during unit startup when the shaft passes through critical speeds and
the resulting increase in shaft vibration exceeds the seal clearance. Brandon designed the
retractable packing to provide a very large shaft clearance during unit startup, when turbine
steam flow and local pressure are very low. During normal operation when the flow and pressure
are at design levels, the seal elements close, or move radially inward (closer to the shaft). The
resulting reduced clearance lowers leakage losses.

Unlike conventional spring-backed packing, in which the flat springs press the labyrinth
segments inward toward the shaft, the springs in retractable packing assemblies hold the packing
at a large radial clearance to the shaft (typically 0.150 in. [3.81 mm]). This concept is shown in

3-2
 EPRI Proprietary Licensed Material

Commercially Available Steam Turbine Seals

Figure 3-2. The photographs in Figures 3-3 through 3-5 show the packing elements with the
circumferential coil spring. As the turbine steam flow increases, the increasing steam pressure
closes the packing, causing the springs to compress to mechanical stops that establish the design
clearance specified by the manufacturer (see Figure 3-2). By operating in this manner, retractable
packing eliminates startup problems caused by the shaft rubbing and bowed rotors. Over the life
span of the equipment, this results in higher sustained efficiency and lower maintenance costs in
addition to the reduced probability of vibration at startup. Retractable packing can be installed
throughout the steam path with the exception of the three outer rings in the shaft end packing
boxes. Commercial suppliers of variable clearance packing include Brandon Engineering,
TurboCare, and Dresser-Rand (see Table 3-1). Several steam turbine OEMs also incorporate
these same commercial seal products in their retrofits or new designs.

Figure 3-2
Diagram of Brandon Retractable Packing
(Courtesy of Brandon Engineering)

Figure 3-3
Photograph of Brandon-Type Retractable Packing Segments
(Courtesy of Brandon Engineering)

3-3
 EPRI Proprietary Licensed Material

Commercially Available Steam Turbine Seals

Figure 3-4
Photograph of TurboCare Retractable Packing Segment – Similar to Brandon Design
(Courtesy of TurboCare)

Figure 3-5
Photograph of Conventional Spring-Backed Seals (Left) and Variable-Clearance Seals with
the Typical Clearance Provided at Startup
(Courtesy of Dresser-Rand)

3.3 Brandon Sensitized Packing

Sensitized packing does not rely on steam pressure to close, so it can be applied in all stages as
well as in the end seals, or in conjunction with retractable or conventional packing. The key
feature of sensitized packing is that the entire packing ring assembly is supported on soft springs
on the stationary elements. The soft springs do not strongly resist movement of the ring in the

3-4
 EPRI Proprietary Licensed Material

Commercially Available Steam Turbine Seals

event of a shaft rub. The heat generated in the seals and shafts during rubs is proportional to the
contact force, which is in part determined by these support springs. It is therefore advantageous
to minimize this spring force selected to keep the segments in their most optimum position
possible relative to the shaft.

Sensitized packing always operates in the closed position, with coil springs holding the segments
in place as shown in Figure 3-6. The difference is that conventional packing requires as much as
300–400 lbs (1,334–1,779 N) of force in order to deflect and move the packing segments in a rub
situation, whereas the contact force necessary to move sensitized packing segments is less than
10 lbs (44 N). The packing weight in this design must be balanced with the spring force in order
to avoid rubs due to sagging of the upper packing and increased clearance of the lower packing.
If the rotor touches a packing segment, the springs easily deflect, allowing the segment to move
so that heavy rubs at startup can be avoided.

Figure 3-6
Diagram of Sensitized Packing (Courtesy of Brandon Engineering)

3.4 Retractable Brush Seal Packing

Retractable brush seal packing has experienced significant use in gas turbine applications. It is a
modified spring-back labyrinth packing in which one of the high knives is replaced with an
element consisting of a fine wire bristle material sandwiched between two plates as shown in
Figure 3-7. This assembly is inserted into an enlarged slot in the packing assembly designed to
accommodate the greater width of the brush bristle pack compared to a standard knife (see
Figure 3-8).

3-5
 EPRI Proprietary Licensed Material

Commercially Available Steam Turbine Seals

Figure 3-7
Diagram of the Brush Bristle Pack that Replaces a Standard High Knife in a Retractable
Brush Seal Packing Assembly
(Courtesy of TurboCare)

Figure 3-8
Brush Bristle Pack Installed in a Retractable Seal Packing Element
(Courtesy of TurboCare)

3-6
 EPRI Proprietary Licensed Material

Commercially Available Steam Turbine Seals

The key design feature of the brush element, as shown in Figure 3-7, is the bristle pack in which
the brush elements are slanted in the direction of the shaft rotation. The brush is designed to be
compliant if it comes in contact with the shaft, which is why the slanted orientation is used. The
function of the back plate is to support and prevent axial deflection of the relatively flexible
bristles against the pressure loading across the seal. The back plate does not extend to the tip of
the bristle pack, and its diameter is set to prevent contact with the shaft during the maximum
expected lateral excursions in shaft vibration. Brush seal elements are designed such that the
bristle tips are not fully in contact with the shaft, but maintain a small nominal clearance
(significantly less than a conventional knife element). It is this reduced clearance, relative to
conventional labyrinths, that provides the performance characteristics of this seal design.

Incorporating the brush seal element into a spring-back packing assembly provides additional
compliance in the event of a shaft rub. This feature is expected to extend the life of the bristle
pack by reducing the likelihood of a hard rub that can damage the bristles due to excessive bristle
deflection, overheating, or wear.

Retractable brush seals have been used in both nuclear and fossil steam turbines. Suppliers of the
seals have indicated that the brushes can maintain their integrity even after several years of
operation (see Figure 3-9).

Figure 3-9
Photograph Showing the Condition of a Brush Seal After Seven Years of Operation in a
Fossil Steam Turbine
(Courtesy of TurboCare)

3-7
 EPRI Proprietary Licensed Material

Commercially Available Steam Turbine Seals

Commercial suppliers of brush seal packing assemblies include TurboCare (with over 100
installations), Dresser-Rand, and PerkinElmer.

PerkinElmer’s Centurion brush seal (see Figure 3-10) uses an engineered bristle technology in a
standard labyrinth design that permits a wear-in to an optimum, minimal clearance. PerkinElmer
claims that the seal reduces initial leakage by as much as 50% compared to a standard labyrinth
design, thus improving performance through a typical maintenance cycle.

Figure 3-10
Photograph of a PerkinElmer Centurion Variable-Clearance Brush Seal Assembly

3.5 Guardian Seal Design

The Guardian seal design concept (essentially an adaptation of the spring-backed labyrinth seal)
has seen steadily increasing application since it was first introduced for commercial use in June
of 1997. The unique feature of this design is that two of the high knives in each segment are
replaced by a slightly higher sacrificial post designed to contact the shaft in the event of a rub
(see Figure 3-11). This post is essentially sacrificial and protects the remaining adjacent
conventional knives. In the event of contact between the shaft and the packing, this higher post
pushes the entire spring-backed assembly outward, preventing the shaft from contacting the
adjacent knives and causing rub damage. The coil springs supporting the Guardian packing rings
are softer than those used in retractable packing design in order to permit easier radial movement
and reduced contact force in the event of a rub. The post is nominally 0.005 in. (0.127 mm)
closer to the shaft than the adjacent high knives, which still provides significant clearance under
typical operating conditions when shaft vibration is normal. The clearance between the shaft and
the knives in the Guardian packing is typical of that used in conventional labyrinth designs.

3-8
 EPRI Proprietary Licensed Material

Commercially Available Steam Turbine Seals

Figure 3-11
Photograph of Guardian Seal Showing the Post Elements Designed to Protect Adjacent
Labyrinth Knives (Courtesy of Turbo Parts LLC)

Turbo Parts LLC is the commercial supplier of Guardian seals. They describe the main
advantage of this design as its effectiveness at maintaining the desired clearance of all labyrinth
knives for an extended period of time between major section overhauls. The economic analyses
presented in Section 2 demonstrates the significant payback associated with extending seal
performance. For example, if in the example presented in Table 2-1, the initial 0.5% performance
improvement were to be maintained constant over the five-year period rather than decreasing as
shown, the additional payback would be $400,000.

The Guardian strip is the main feature of this labyrinth design and is fabricated from Nitronic 60
stainless steel alloy that is designed to prevent galling and exhibits low friction when it comes
into contact with the turbine shaft in the event of a rub. The vendor reports that both laboratory
tests and field experience with extended rub events have demonstrated that the Nitronic strip
does not cause shaft damage as evidenced by local areas of high hardness. The body of the seal is
made from a 12-chrome stainless steel alloy in order to provide a longer service life in the steam
environment.

The turbine manufacturer, Hitachi Ltd., has an exclusive agreement with Turbo Parts LLC to
offer the Guardian packing and Vortex Shedder spill strips (described in Section 3.6.3) in their
turbine upgrades. Hitachi performed a detailed evaluation of all third-party seals available in the
market before selecting these designs, and concluded that the Guardian could meet the
requirements of a 10–12 year major turbine overhaul interval. Hitachi provides the seals in new
turbines and upgrade packages and for normal maintenance outages.

3.6 Seals Used on Rotating Blade Tips

3.6.1 TurboCare Brush Spill Strip

The brush spill strip basically applies the brush pack design from the retractable brush packing
into the existing knife grooves in the casing near the blade tips of the turbine (see Figure 3-12).
Similar to the shaft seals, the brush spill strip is designed to run at a reduced nominal clearance
as compared to conventional spill strips, thus reducing leakage flow. TurboCare states that the
brush spill strip has been installed in over 30 steam turbines.

3-9
 EPRI Proprietary Licensed Material

Commercially Available Steam Turbine Seals

Figure 3-12
Photograph of Brush Spill Strip (Courtesy of TurboCare)

3.6.2 HELP-R Spill Strip

TurboCare offers a special spill strip element that is used in HP and IP stages that are at risk of
damage due to solid particle erosion in the tenon and coverband area. In the initial stages of the
HP and reheat turbines, oxide particles shed by the boiler tubes that enter the turbine can be
trapped in the volume bounded by adjacent spill strips, the coverband, and the casing. These hard
particles can erode the coverband tenons, eventually resulting in the need for expensive repair.
TurboCare’s high-efficiency lasting particulate removal (HELP-R 3 ) spill strip segment (see
Figure 3-13) is designed to allow these oxide particles to escape from the region above the
rotating blades, thus reducing the rate of damage. Spill strips are normally installed as several arc
segments in the casing, typically 30° arc length. In a HELP-R installation, a single-arc segment
replaces a standard spill strip. The HELP-R segment is coated to reduce the rate of erosion and
features a relief cut and vent hole through which the oxide particulates are passed downstream.

3
HELP-R is a trademark of TurboCare.

3-10
 EPRI Proprietary Licensed Material

Commercially Available Steam Turbine Seals

Figure 3-13
TurboCare HELP-R Spill Strip

3.6.3 Vortex Shedder Spill Strip

Another spill strip design used in retrofit applications is the Vortex Shedder developed and
provided by Turbo Parts LLC. The Vortex Shedder, shown in Figure 3-14, features a unique
profile to the spill strip knife. In this spill strip, the knife profile is not constant in the
circumferential direction, but instead includes of a series of dimples that protrude in the axial
direction. These dimples disturb the flow (that has a very strong circumferential component) and
produce vortices in the leakage jet. In turn, these vortices increase the average pressure within
the seal area and reduce the pressure drop across the seal. This reduced pressure drop lowers the
leakage flow proportionally. Figure 3-15 shows a diagram of this effect. Turbo Parts LLC claims
a reduction of over 5% in leakage flow due to the use of a Vortex Shedder, based on a
computational fluid dynamic simulation. Clearances used when installing the Vortex Shedder are
nominally the same as for conventional spill strips. The spill strip is manufactured from 12-
chrome stainless steel.

3-11
 EPRI Proprietary Licensed Material

Commercially Available Steam Turbine Seals

Figure 3-14
Photograph of Vortex Shedder Spill Strip for Steam Turbine Application
(Courtesy of Turbo Parts LLC)

Figure 3-15
Diagram of Vortices Created In Leakage Jet Above Rotating Blades
(Courtesy of Turbo Parts LLC)

3.7 Summary of Commercially Available Replacement Seals

Table 3-1 summarizes the options and suppliers of various replacement seal and spill strips.

3-12
 EPRI Proprietary Licensed Material

Commercially Available Steam Turbine Seals

Table 3-1
Information on Commercially Available Seals for Steam Turbine Applications

Company Types of Turbine Web Site Address Comments

Seals/Packing
TurboCare Retractable packing www.turbocare.com TurboCare is a subsidiary of Seimens-
Westinghouse but supplies turbine seals
Retractable brush seal for all types of turbine designs.
Brush spill strip
Standard labyrinth packing
Turbo Parts LLC Guardian seals www.mdaturbines.com Turbo Parts LLC is a subsidiary of
Mechanical Dynamics & Analysis, LLC
Vortex Shedder tip seal (MD&A). MD&A is owned by Hitachi, Ltd.
Brandon Engineering Sensitized packing www.brandonengineering.com Brandon Engineering contracts with
manufacturing representatives to supply
Retractable packing their packing. These companies are listed
Scooped tip seals in this table.

Steam Turbine Sensitized packing www.starturbine.com STAR is the U.S. manufacturing

Alternative representative for Brandon Engineering.
Resources (STAR) Retractable packing
Scooped tip seals
Standard labyrinth packing
Turbine Service and Sensitized packing www.serviceindustries.com TSI is the U.S. manufacturing
Supply, Inc (TSI) representative for Brandon Engineering.
Retractable packing
Scooped tip seals
Standard labyrinth packing
Energy Service Sensitized packing www.esgch.com Energy Service group is the international
Group manufacturing representative for Brandon
Retractable packing Engineering.
Scooped tip seals
Standard labyrinth packing

3-13
 EPRI Proprietary Licensed Material

Commercially Available Steam Turbine Seals

Table 3-1 (cont.)

Information on Commercially Available Seals for Steam Turbine Applications

Company Types of Turbine Web Site Address Comments

Seals/Packing
Turbo Components Standard labyrinth packing www.tce1.com
and Engineering
Brush Seals
Power Plant Services Standard labyrinth packing www.ppsvcs.com
PerkinElmer Centurion brush seals www.fluidsciences.perkinelmer.com
Siemens Siemens is the parent company of TurboCare and offers TurboCare seal designs in their steam path upgrade
packages. For older Siemens units (Westinghouse), the majority of the packing is caulked-in packing; therefore,
TurboCare gland ring packing cannot be applied.
Toshiba Toshiba applies advanced packings to newly supplied units and retrofit units where nozzle diaphragms and gland
packings will be replaced. Advanced packings are reportedly more effective when introduced together with
advanced steam path technology. Modification of existing nozzle diaphragms to introduce advanced packings may
be possible, and in that case, it is recommended that Brandon Engineering be contacted directly.
Retractable and sensitized packings are supplied to Toshiba by Brandon Engineering. Brandon Engineering
supplies these packing springs, and Toshiba supplies the packing ring itself. In the case of supplying conventional
packings, Toshiba has been and will be supplying all the packing components to their fleets.
Hitachi, Ltd. The turbine manufacturer Hitachi, Ltd. has an exclusive agreement with Turbo Parts LLC to offer the Guardian
packing and Vortex Shedder spill strips
Dresser-Rand Retractable packing www.dresser-rand.com
Retractable brush seal
Brush seals

3-14
 EPRI Proprietary Licensed Material

4
DISCUSSION

Information on the perspective of turbine owners regarding seal replacements was obtained as
part of this project. This did not involve a formal survey, but rather informal discussions with
turbine-generator engineers who have the responsibility of making decisions on component
replacement and selection. EPRI did not seek endorsements by the respondees, nor is it EPRI’s
intention to endorse any particular seal design or supplier. The resulting comments received were
considered of value to turbine owners making decisions on turbine maintenance and efficiency
improvements.

4.1 Reliability Versus Performance

Turbine operators have traditionally placed importance on maintaining seal or packing reliability
and avoiding unplanned maintenance related to seal malfunction. In some cases, this resulted in
favoring traditional (fixed-clearance) labyrinth packing designs compared to variable-clearance
or the other non-traditional designs described in the previous chapter. Apparently, the perceived
risk of unplanned maintenance related to possible packing damage outweighs the benefits for
some power producers. However, as fuel prices increase and the cost of increased steam leakage
over a sustained time period also increases, there is expected to be an increased interest in the
advanced seals that are designed to maintain low leakage rates over the increased periods
between major section overhauls. The analyses presented in Section 2 demonstrate the financial
advantages of reduced leakage over long time periods as both fuel prices and outage intervals
change.

Some companies are risk-adverse with regard to implementing new seal designs that have not
demonstrated many years of field experience. The suppliers of advanced packing designs are
highlighting the successes with field implementation of their designs, and turbine owners are
strongly encouraged to request and follow through on any referrals to actual installations of any
seal designs of interest. It is recommended that when using field experience to influence the
selection of new technology, an unbiased explanation of any reported component failures be
sought and the influencing factors be understood.

4.2 Installation Issues

Several turbine operators emphasized the importance of proper installation in relation to the use
of advanced seal designs. Most advised using the seal vendors for the installation process
because of their familiarity with their specific designs. Proper alignment is the primary issue
during installation as it reduces the risk of rubs, subsequent thermal distortion of the shaft, and
labyrinth damage. Proper clearances must also be maintained on all upper-half pins and retaining

4-1
 EPRI Proprietary Licensed Material

Discussion

keys in order to prevent any restriction of movement. In addition, the variable-clearance packing
designs require proper butt clearances between each arc section of packing. Excessive butt
clearance creates leakage through the butt joint. Insufficient clearance prevents full closure of the
packing as intended during normal operation. In some cases, leakage through the butt clearance
has resulted in a buildup of deposits on the adjoining surfaces and subsequent failure to close as
designed. This is another reason to pay particular attention to ensure that the butt joints are sized
properly.

Another area in which proper clearance is critical for reliable variable-clearance packing
operation is the neck clearance, as shown in Figure 4-1. If the neck clearance is too tight, binding
may occur that prevents closure or retraction of the packing element.

Figure 4-1
Diagram Showing Neck Clearance on a Variable-Clearance Packing

In addition to the issues of alignment and butt clearance, other installation issues associated with
older turbines include accounting for diaphragm thermal distortion, out-of-roundness, and lack of
concentricity. Again, assessing the vendor’s experience with installations in which these specific
conditions were present is recommended. The installation process must account for these
conditions.

It should be noted that some variable-clearance packing relies only on the pressure and local
steam flow in order to close the clearance once normal operation is achieved. Other designs
require that higher pressure steam be supplied by a small steam supply pipe to the packing in
order to force the seals closed under operation. The additional cost and effort to install the steam
lines should be considered.

4-2
 EPRI Proprietary Licensed Material

Discussion

4.3 Springs on Variable-Clearance Packing

Packing assemblies need to be reliable for sustained operation at very high operating
temperatures in the HP and IP turbine sections. Material that can become embrittled at these high
temperatures should not be used for springs. Inconel coil springs appear to be most reliable and
have been used successfully by one manufacturer in their variable-clearance packing design.

Turbine owners using variable-clearance packing recommended that all springs be replaced at
each major overhaul. This reduces the risk that spring characteristics such as stiffness and
strength will degrade significantly and affect the packing reliability.

4.4 Operational Issues

Many fossil plants experience problems with oxides that exfoliate from superheat/reheat tubes
and their introduction into the steam turbine. The hard particles entrained in the high-speed
steam are a potential source of erosion for the flow path blade and seal elements. Also, there is a
potential for particles to become trapped in the brush elements of retractable brush seals. There
were no specific occurrences of exfoliate trapped in brush seals reported in the discussions with
the plants. Figure 3-9 shows a brush seal in a fossil plant after seven years of operation with no
apparent buildup of particles. However, there is a significant variation from plant to plant in the
level of oxide exfoliates, so this factor should be considered when selecting replacement seals.
To reduce the risk of damage to seal elements from solid particles, operations staff should
identify and control the factors that produce significant carryover.

Another source of debris that has affected the performance of variable-clearance seals in isolated
situations is the steam extraction lines.

The most common source of seal damage is shaft rubs during unit startup and shutdown, which
experience high lateral vibration at the critical speeds. Once a rub starts, local heating of the shaft
due to friction can cause thermal distortion and worsen vibration and the associated rub. Plant
operators need to be aware of the shaft critical speeds and avoid any extended operation during
startup/shutdown at these speeds. Also, the degree of balance of the shaft system will affect the
lateral vibration amplitude at critical speed, so rotating unbalance should be maintained as low as
possible in order to protect the seal elements.

4.5 Industry Experience

Plants have reported good experiences with both retractable brush seals and the Guardian seals
[10]. The following comments provide some insight, but are not meant to be a rigorous sampling
of industry experience.

Both plants that commented on the Guardian seal have reported five years of successful
operation, with an observable reduction in the rate of performance degradation compared to
previous standard seals. In one of these two plants, there was an initial installation error
involving alignment that produced a substantial rub. In correcting this problem, no damage to the
shaft was noted. There are now over 20 plants operating with the Guardian design, with the
majority installed since 2003.

4-3
 EPRI Proprietary Licensed Material

Discussion

The TurboCare retractable packing has been installed in nearly 500 steam turbines starting in the
late 1980s through 2001. Since 1998, over 90 units have had the retractable brush packing
installed, with the majority being installed since 2001. One testimonial for the retractable brush
seal was reported [11] in which the design was installed in a 400-MW fossil unit with a noted
increase in unit output.

4-4
 EPRI Proprietary Licensed Material

5
REFERENCES

1. Advanced Steam Seal Labyrinth Design. EPRI, Palo Alto, CA: 2005. 1011932.
2. A. Vakili, A. Meganathan, M. Michaud, and S. Radhakrishnan, “An Experimental and
Numerical Study of Labyrinth Seal Flow,” Paper No. 2005-68224, Proceedings of American
Society of Mechanical Engineers Turbo Expo 2005, Reno, NV, June 6–9, 2005.
3. K. C. Cotton, Evaluating and Improving Steam Turbine Performance, Second Edition, 1998,
Cotton Fact Inc., Rexford, NY.
4. Life Cycle Management Planning Sourcebooks: Volume 8: Main Turbine. EPRI, Palo Alto,
CA: 2004. 1009071.
5. W. P. Sanders, Turbine Steam Path, Volume IIIa, Mechanical Design and Manufacture,
2004, Pennwell, Tulsa, OK.
6. Demonstration of EPRI STEEM Optical Probe for LP Turbine Steam Quality Measurement.
EPRI, Palo Alto, CA: 2001. 1004051.
7. Martin, H. M., “Steam Turbines,” The Engineer, London, 1913, p. 1610.
8. A. Egli, “The Leakage of Steam Through Labyrinth Seals,” Transactions American Society
of Mechanical Engineers, Paper FSP-57-5.
9. D. Brandon, “Retractable and Sensitized Packing Reduces Friction and Increases Turbine
Efficiency,” Power Engineering, August 1, 2004.
10. W. P. Sanders, L. H. Shuster, and G. A Clark, “Experience with Performance Upgrading of
Existing Turbine Generator Units on the TVA System,” PowerGen 2000, Orlando, FL,
November 12–16.
11. D. Robb, “Steam Turbine Repair: Turbine Fountain of Youth,” Power Magazine, November
2004, Vol. 148, No. 9, p. 7.

5-1
Export Control Restrictions The Electric Power Research Institute (EPRI)
Access to and use of EPRI Intellectual Property is granted with The Electric Power Research Institute (EPRI), with major locations in
the specific understanding and requirement that responsibility Palo Alto, California, and Charlotte, North Carolina, was established
for ensuring full compliance with all applicable U.S. and in 1973 as an independent, nonprofit center for public interest
foreign export laws and regulations is being undertaken by energy and environmental research. EPRI brings together members,
you and your company. This includes an obligation to ensure participants, the Institute’s scientists and engineers, and other leading
that any individual receiving access hereunder who is not a experts to work collaboratively on solutions to the challenges of electric
U.S. citizen or permanent U.S. resident is permitted access power. These solutions span nearly every area of electricity generation,
under applicable U.S. and foreign export laws and delivery, and use, including health, safety, and environment. EPRI’s
regulations. In the event you are uncertain whether you or members represent over 90% of the electricity generated in the
your company may lawfully obtain access to this EPRI United States. International participation represents nearly 15% of
Intellectual Property, you acknowledge that it is your EPRI’s total research, development, and demonstration program.
obligation to consult with your company’s legal counsel to
determine whether this access is lawful. Although EPRI may Together...Shaping the Future of Electricity
make available on a case-by-case basis an informal
assessment of the applicable U.S. export classification for
specific EPRI Intellectual Property, you and your company
acknowledge that this assessment is solely for informational
purposes and not for reliance purposes. You and your
company acknowledge that it is still the obligation of you and
your company to make your own assessment of the applicable
U.S. export classification and ensure compliance accordingly.
You and your company understand and acknowledge your
obligations to make a prompt report to EPRI and the
appropriate authorities regarding any access to or use of EPRI
Intellectual Property hereunder that may be in violation of
applicable U.S. or foreign export laws or regulations.

Program:
© 2005 Electric Power Research Institute (EPRI), Inc. All rights reserved.
Nuclear Steam Turbine Generator Initiative
Electric Power Research Institute and EPRI are registered service marks of Steam Turbines, Generators, and Balance-of-Plant
the Electric Power Research Institute, Inc.

Printed on recycled paper in the United States of America 1010214

ELECTRIC POWER RESEARCH INSTITUTE

3420 Hillview Avenue, Palo Alto, California 94304-1395 • PO Box 10412, Palo Alto, California 94303-0813 USA
800.313.3774 • 650.855.2121 • [email protected] • www.epri.com