Engineering Encyclopedia: Exxon Chemical and Mechanical Cleaning Manual 20) Planning and Control of A Cleaning Job
Engineering Encyclopedia: Exxon Chemical and Mechanical Cleaning Manual 20) Planning and Control of A Cleaning Job
Note: The source of the technical material in this volume is the Exxon Research
and Engineering Company Chemical and Mechanical Cleaning Manual.
Warning: This Exxon Research and Engineering Company Chemical and
Mechanical Cleaning Manual is made available exclusively for the use of
Saudi Aramco for its own use and for no other purpose. The disclosure of its
content are subject to the provisions of the Contract No. 30934/00 dated
November 1, 1986. The material in this manual may not be copied,
reproduced, sold, given, or disclosed to third parties or otherwise used in
whole or part without the written permission of the Vice President,
Engineering Services, Saudi Aramco.
Chapter : Planning And Control Of A Cleaning Job For additional information on this subject, contact
File Reference: Section 20 R.S. Sarathy on 875-3520
Engineering Encyclopedia Exxon Chemical And Mechanical Cleaning Manual
Section 20
Planning And Control Of A Cleaning Job
TABLE OF CONTENTS
Sampling................................................................................................................... 7
Testing...................................................................................................................... 7
Observation ................................................................................................... 7
Temperature .................................................................................................. 7
Total iron....................................................................................................... 8
Ferric iron...................................................................................................... 8
Inhibitor ........................................................................................................ 8
Experience will show that planning is the key to a successful chemical cleaning operation and,
when properly carried out, will minimize subsequent unnecessary work and lost time. Over 90%
of delays and incidents can be directly traced to factors which should have been examined and
settled in the planning stage. Examples are finding that there is an inadequate steam or water
supply, at the time needed, that there is no spare hose onsite, that the contractors electric pump is
not flameproof, or that the equipment being cleaned cannot be isolated from other large volume
equipment. All these and many others are incidents that have actually happened. They result in
long delays which will turn savings into losses and do not impress plant management with the
efficiency of chemical cleaning as a technique.
The check list of points to be considered may appear excessively long; however, with experience
and practice, all these points can be dealt with in two to four hours. In a refinery where many
items of equipment can be chemically cleaned each year, this is essential. This planning stage
should involve the process, maintenance and contract personnel.
There should be at least one field examination of the equipment with the contracting chemical
cleaning engineer who will be carrying out the work. This is vital to ensure that all parties
concerned are aware of the work to be done and their own responsibilities.
In planning a chemical cleaning job, the following points must be considered and decisions
reached before the work is initiated.
- What do we want to clean? - Tower alone? Tower and pipework? Is there any other
equipment that could profitably be included?
- Why do we want to clean it? - To improve plant efficiency? Preparation for repair? To
stop carryover of fouling into refined product? To inspect the equipment?
- How well do we want it cleaned? - As clean as possible? To remove production
restriction for a short period? Ready for hot work?
- What is the fouling? - What type of fouling? How much fouling? Where is it in the
equipment? Can we get samples? Have we got records?
- What is the design of the equipment? - What are the materials of construction? Can it be
isolated? Where are the undrainable areas? What is the volume of the equipment? What
size of pipework? What fittings are needed? What vents are available?
- How long before work starts? - How long will it take to make fittings?
- How long before the equipment can shut down?
- How long will it take? - Will other work be carried out at the same time? How much will
lost throughput cost?
- How much can be spent? - Will increasing chemical cost to do a better job be justified?
If the planning stage has been completed well, the stages of assembly onsite, cleaning,
neutralization, etc., should proceed smoothly, but it is most important to control and observe the
operation.
It is necessary to ensure that process supervision and operators, maintenance forces and
contractor have copies of the work plan, that necessary permits have been obtained and that the
fittings have been prepared and delivered to site. The chemicals delivered to site should be
checked for quantity, type and strength. Whether these checks should include sampling and
routine laboratory testing depends on one's confidence in the supplier and contractor. A final
check of the cleaning circuit is also advisable to ensure that vent and drain valves are operable
and that the equipment is isolated.
While the cleaning technique used will vary with each individual cleaning operation, the general
aspects of control are similar. The contractor should always check with process forces
immediately prior to injecting chemicals and receive their agreement. When cleaning has
commenced, the circuit should be rechecked for leaks, etc. Tests should be made at the agreed
time intervals by the contractor, in-house specialist or both and results tabulated in graph form
(See page 10). The critical stages of the operation, e.g., addition of acid or neutralizing
chemicals, should be carried out under the observation of the chemical cleaning specialist and
not left solely to the contractor.
The contractor should not be permitted to change the specified treatment in any way without
obtaining the agreement of the in-house chemical cleaning specialist. This is essential.
Observation of the operation is necessary in order to make decisions concerning the addition of
further chemicals, control of corrosion and deciding when cleaning has been completed. It is
only by this type of observation that future operations can be improved by reducing treating
times, optimizing circulation rates, etc. Final results can be assessed better with an exact
knowledge of how the operation was carried out and often average results can be improved by,
say, flushing at an increased rate.
A written report should be made on each chemical cleaning operation, and this should be filed
with the original work schedule. These records are invaluable in assessing the operation and in
planning and modifying the technique when the next cleaning operation is required.
These reports are also a means of transferring knowledge to other refineries and provide
comprehensive handover information when another person takes over chemical cleaning at any
given location. All the facts should be presented in a readily accessible form..
Consider the necessary facts. Many reports consist of a phrase such as "At X refinery pipestill
preheat exchangers were successfully cleaned by circulating hot hydrochloric acid for 8 hours -
results were excellent and showed 40% savings over conventional cleaning".
This appears to tell the facts until an attempt to plan on this basis shows that the temperature, the
concentration, the circulation rate, and the inhibitor are unknown. Which exchangers in the
preheat train were cleaned? Was it shell or tube side? One exchanger or two? etc., etc., etc.
It is therefore suggested that the following format is used when reporting cleaning operations.
This contains the basic facts and does not preclude adding more detailed data and comment
which may be useful, particularly practical problems with pumps, etc., etc., etc.
General:
- Reason for cleaning, with brief description of the problem and why the cleaning method
was selected
- Nature of the fouling, including any analytical and solubility test results (see Section 4)
- Whether the operation was carried out with contract or refinery personnel
- Approximate cost of the operation, and time required
Description:
Chemicals:
Sampling
When carrying out cleaning by circulation, samples are usually taken from the return line from
the equipment at the circulating tank. When several return lines are used, samples should be
taken from each of these, having allowed each to run for some minutes before sampling. Every
effort should be made to ensure that the samples are representative of the cleaning fluid in the
equipment. During onstream cleaning operations, such as acid slugging of coolers, samples may
be taken downstream of the equipment being treated.
Testing
Many samples contain undissolved solids which continue to react with the cleaning chemicals in
the sample bottle, thereby changing their strength. For this reason, field testing should be carried
out as soon as possible after sampling, even though the accuracy is probably inferior to that
obtainable by laboratory testing. It is preferable to know the acid strength +0.1% in 5 minutes
than +0.01% in 4 hours time when it is too late to make further additions of acid to bring the
strength up to a working level. Both laboratory and field tests have their function in chemical
cleaning, but it is the latter on which the operation control is based. Test results are best
expressed in graph form, as this indicates rate of change, and it is this rate of change which is
most informative.
Observation
Visual examination of the sample will often show the presence of undissolved or insoluble
fouling, and color changes in the cleaning solution will give an indication of the quantity of
fouling being removed.
Temperature
This is normally checked by an immersion type thermometer in the circulating tank close to the
discharge pipe. It is also advisable to use a surface pyrometer to check for hot and cold spots on
the equipment being cleaned, as this indicates uneven circulation flow. An experienced operator
can often tell the temperature within 5°F (3°C) by touch alone if no pyrometer is available, but
this is not advisable as standard procedure.
Strength of chemicals
For solvents, etc., testing is usually too complicated for field tests, and decisions must be made
on observation such as the separation of oil from a cleaning emulsion indicating that it is over
contaminated. In the case of acids the strength and other factors can be measured by simple
titration. Equipment consists of a 25 mls pipette, a 50 mls burette and two conical flasks.
Acid strength
Take a 25 ml sample and add 5 drops of methyl orange indicator. Titrate with a one normal
sodium hydroxide solution until the solution just turns yellow. Number of ml of sodium
hydroxide used x 0.146 = % hydrochloric acid. If the sample contains iron, the end point may
not be clear and potassium citrate should be added before carrying out the titration. This will
prevent iron precipitation from masking the end point. (Refer to ASTM D-2790 for complete
procedure)
Total iron
This is the sum total of ferrous iron (Fe++) and ferric iron (Fe+++). It may be estimated as
follows. To a 5 ml sample add 30 ml of water. Then add 0.5% potassium permanganate
(KMnO4) drop by drop until the solution turns permanently pink. When the solution is pink, add
1 ml of hydrazine sulfate crystals together with approximately 1/2 a teaspoon of potassium
iodide crystals. Add 5 drops of starch indicator and titrate with 0.0895N sodium thiosulfate until
just colorless. % Fe = 0.10 x ml of thiosulfate used.
Ferric iron
Take a 5 ml sample and add 30 ml water and 1/2 teaspoon of potassium iodide. Add 5 drops of
starch indicator and titrate with 0.0895N sodium thiosulfate to a colorless end point. % Fe+++ =
0.10 x ml thiosulfate used.
Ferrous iron
The difference between total iron and the ferric iron figure is the ferrous iron. This may also be
measured independently by titration with dichromate solution.
Inhibitor
Chemical analysis to determine the inhibitor concentration is usually complex, and results cannot
be obtained rapidly enough to prevent corrosion.
A simpler method is to measure the effectiveness of the inhibitor. If a small piece of degreased
steel wool is dropped into the acid sample, it will sink and remain at the bottom as long as the
inhibitor is effective. If it is not effective, corrosion will occur with the generation of hydrogen,
and the accumulation of these bubbles will cause the steel wool to float to the surface within one
or two minutes. This is a crude but very effective test.
Corrosion test pieces should be placed in the equipment, preferably at a point where they can be
withdrawn for visual examination during the cleaning operation. These will provide a measure
of the total metal loss during the cleaning operation. Corrosion rates on carbon steel should be
less than 0.02 mils/hour (0.5 microns/hr).
Interpretation of Tests
The graph shown in Section 20, Figure 1, illustrates the acid strength and iron content figures
which might be expected during the acid cleaning of a scaled compressor cooling jacket. Section
AC shows the rises in strength as acid is added. Point B is the theoretical maximum
concentration which is not reached due to reaction and neutralization occurring as soon as the
acid is introduced. Section CD shows the neutralization of the acid as the scale is dissolved,
rapid at first, gradually slowing as the strength drops to 2%. Section DE shows an increase in
strength as more acid is added to increase the rate of reaction and speed the cleaning. Section EF
shows further weakening of the acid as more scale is dissolved. Section FG shows little or no
drop in acid strength and indicates that no scale is being dissolved in spite of the fact that there is
adequate strength (over 4%) to react. The assumption is therefore that all the scale has been
removed. At point G the acid would be drained.
The iron graph shows a relatively rapid increase in the first three or four hours due to rust present
in the scale (A'-B'). This increase continues at a reducing rate as long as scale is being dissolved.
This too will gradually level out (C'-D'). Taking this particular example, the variations in
strength with time are largely due to the low volume of cleaning liquid relative to the amount of
scale to be dissolved.
Compare these results with the curves shown in Section 20, Figure 2, which illustrate the results
obtained during two boiler cleaning operations with virtually identical fouling. In the case of
boiler 1, good temperature control was achieved and the inhibitor remained effective. In the case
of boiler 2, temperature was excessive and the inhibitor did not remain effective.
The acid strength for boiler 1 rises steadily as the acid is introduced and then falls as scale is
dissolved; this is matched by a rise in the iron content. Both strengths then stabilize with little or
no reaction occurring. The changes in strength are not great as the total volume of acid is large
relative to the quantity of scale being dissolved.
In the case of boiler 2, a similar increase in acid strength occurs during the introduction stage
followed by a decrease as scale is dissolved and a rather rapid increase in iron content. These
rates do not level out, however, and acid strength continues to drop and iron content to rise much
higher than one would expect for this type of cleaning operation. In actual fact, the rise in iron
content and drop in acid strength was due to acid attack on the metal and severe corrosion of the
boiler occurred. This should have been suspected by observing the graph and checked by testing
the inhibitor effectiveness and examining the corrosion test piece. Tests for ferric iron would
also have shown very high values.
Figure 1
Figure 2