No 171

171/1
Shut-downs and Upsets
171/2
Low Pressure Into High Pressure Won’t Go
171/3
Another Attempt Doesn’t Work Either
171/4
Disconnection Is Sometimes Not Enough
171/5
Surplus Equipment Needs Emptying
171/6
Tournament Day
171/7
Puzzle Corner

In this Issue
An Engineer’s Casebook No. 71:-
Shut downs and
Materials for Springs in Relief Valves and
Upsets
Safety Valves
171/1 Methods for doing unfamiliar and unusual tasks need to be
Shut-downs and carefully thought through. Those who are to do the work
need to be clear about the objectives and about the
Upsets methods to be used. Co-ordination of the work of many
different groups is most important. They need to perform
The preparations for a
like one great orchestra and not like rival groups of
shut-down (or a turn
‘buskers’!
around), the work to be
done during a shut-down
and the subsequent start-
up all need careful
planning.

171/2 On a plant handling liquefied petroleum gases it was

Low Pressure into intended to remove a flowmeter from a pipeline. The man
concerned decided not to use the usual method for
High Pressure preparing the job, which was to attach a flexible connection
Won’t Go to a branch on the pipework and release the contents to a
vent stack. Instead he closed isolation valves on either side
During a shut-down It is to the section and blew the contents off to atmosphere. He
usually necessary to then closed the blow-down valve and connected a flexible
wash out process hosepipe from a nitrogen service point to a suitable
materials or to drain connection on the pipework. He opened the nitrogen supply
sections of a plant and valve and brought the pressure in the system up to the
purge them with steam or nitrogen supply pressure of 90 psig. He intended to close
an inert gas such as the nitrogen valve and then to blow-off the pressure through
nitrogen the blow-down valve. However he went away to do some
other work, meaning to return and complete the job later. In
a nearby laboratory the same nitrogen supply was used on
some of the chromatographs and they began to function
incorrectly. Other chromatographs, using nitrogen from
cylinders, were behaving normally.

The nitrogen supply to the hosepipe on the plant was

isolated and the hosepipe disconnected. Several nitrogen
service valves were opened to allow nitrogen to blow off to
the atmosphere for a time and shortly afterwards the
chromatographs in the laboratory all began to function
properly.

As pressure in the pipework being purged was found to be

120 psig, above the nitrogen supply pressure, the nitrogen
being blown off from the service points on the main was
found to contain hydrocarbons.

One of the meter isolation valves was found to be passing

and had allowed hydrocarbons at 120 psig through into the
pipework being purged. The non-return valve on the
nitrogen supply was found to have a stone in it. The valve
had been held open and hydrocarbons had leaked into the
 nitrogen system.

I am sure you can work out the proper procedures for

avoiding incidents like this.

170/3 A distillation column had been emptied and was to be

Another Attempt washed and boiled out with water to remove deposits from
the trays. The drain valve had been left open. The steam to
Doesn’t Work the re-boiler had been isolated by closing the control valve
Either only, and not by the isolation valves.

Attempts to push nitrogen An operator connected a water hosepipe to a riser which

into plant at a higher was connected at its upper end to the vapour main near the
pressure are not all that top of the column. It was intended that water should be fed
unusual. However, here is into the vapour main and thence down the column taking
another attempt with the residues with it.
some unusual features.
After some time the operator noticed that temperatures at
various points in the column were not falling as they
normally did when cold water was added. The water valve
and the inlet valve to the column were then closed and the
hosepipe disconnected so that the water supply could be
checked. Liquor sprayed out of the line onto both the man’s
arms.

In the subsequent investigation three faults were

discovered. The water pressure was lower than usual and
would not have been sufficient to push water up the riser
and into the column. Secondly, several tubes in the reboiler
were leaking. Finally, the steam control valve was passing.

This combination of faults had allowed steam to enter the

body of the column and vaporise residues in the base. The
vapour then condensed at the top of the column in the cold
vapour main and ran down into the water supply line.

If you were responsible for this operation what modifications

would you make to the equipment and the procedure?

171/4 A horizontal cylindrical dryer was used to remove the last

Disconnection is traces of a flammable solvent from a powder. The dryer was
fed from a filter through a down-comer pipe. One day, in
Sometimes Not readiness for maintenance work this pipe was disconnected
Enough and the open end pulled aside.

Sometimes, instead of Three days later most of the work had been completed and
valves and a slip plate the dryer was being re-assembled. The end cover was
being used for isolation of placed in position and bolted down. It only remained to cut
a vessel, a pipeline is off the guide pins on the cover.
disconnected and the
open end pulled aside. In the meantime the interior of the dryer was being flushed
This may not be sufficient with water and the feed pipe was being re-assembled. A
to stop material entering permit was then issued for cutting the guide pins. The area
the vessel from the pipe. round the work was tested to make sure that no flammable
gas was present. Because the dryer had been open to the
atmosphere for three days the atmosphere inside it was not
tested.

As a cutting disc was being used to cut off the guide pin on
the cover at the side nearest to the inlet rotary valve there
was a flash followed by a loud bang. Even an hour and a
half later an explosimeter gave a reading of 40% LEL inside
the drier.

There had probably been a leak into the feeder pipe.

Because the open end had not been blanked off (blinded)
some solvent had entered the drier in spite of the pipe
having been disconnected and moved aside a little.

171/5 A laboratory building was to be closed and the equipment

Surplus Equipment disposed of. Benches and cupboards were emptied and
surplus equipment neatly arranged for inspection by people
Needs Emptying interested in transferring it to other buildings. All the services
were shut off and disconnected at the mains outside the
Before services are
building.
disconnected from
redundant equipment, a Sometime later a group of people searching for useful
thorough check should be equipment decided to look in a refrigerator in the laboratory.
made to see that they are It was still switched on at the local switch though of course it
no longer needed. had warmed up to the temperature of the laboratory! When
it was opened, the refrigerator was found to contain
samples of organic peroxides which should be kept at low
temperatures to avoid risk of explosions.

171/6 During a recent shut-down there was an emergency alarm

Tournament Day requiring a plant to be evacuated. Scaffolders at work on the
plant evacuated by the shortest route and went to the
Ancillary workers on a assembly point nearby for a roll call.
plant during a shut down
are usually given strict About 15 minutes later when a supervisor was making a
instructions about the search of the plant he heard a loud bang on from the roof of
need to leave the plant a building. It was later found that a 16 ft long scaffolding
and where to go when an pole was embedded in the roof. During the inquiry a
emergency alarm sounds. scaffolder explained that he had clipped a pole to a handrail
However they do need to on a platform on a nearby high structure. This pole was then
leave their equipment used as a ‘stop’ for a number of others which he intended to
secure. use and which were not secured in any way. During the
emergency, when the poles were left unattended, vibration
in the pipes of an adjacent pneumatic transfer system had
 dislodged one of the poles. That pole had then flown
through the air like a spear in a medieval tournament and
had struck the edge of the roof of the building below it. After
all that it was discovered that the original ‘emergency’ was a
false alarm.

171/7 Do you know, or can you guess, what the photograph

Puzzle Corner illustrates?

Occasional ancillary
workers cannot be
expected to know the
functions of all the items
of equipment on a plant.
They must therefore be
carefully instructed and
supervised to ensure that
their special equipment
does not Interfere with the
ability of the plant
equipment to function
correctly
It shows part of a scaffold erected on a plant. One of the
horizontal poles has been secured immediately above a
dead weight relief valve. Had the relief valve been required
to lift the pole would have prevented it. This would not
matter in the middle of a shut-down but ancillary work often
starts when the plant is still online and continues after the
start up. Those are periods when perhaps the relief valve is
more likely to be needed than when steady operation has
been established

An Engineer’s Case Book Relief and safety valves are commonplace and yet vital
No 71 MATERIALS FOR pieces of equipment in any chemical plant operating at high
SPRINGS IN RELIEF pressure. The heart of a relief valve is the spring, normally
VALVES AND SAFETY of the helical compression type. The choice of spring
VALVES material has been the cause of much work and concern
over the recent years.

For many years the stress analysis team in Materials Group

of Engineering Department has uprated, downrated and
also designed some of its own relief valves. The methods
used for the spring design are standard. However,
considerable expertise has been developed in their
application and material choice. The trend towards buying
proprietary equipment, however, has increased the reliance
on the relief valve manufacturer for ensuring spring integrity.
 Economic relief valve design requires that each valve body
can operate on a range of duties and that body sizes are
small to keep costs down. The body size is closely related
to the size of the spring it needs to accommodate. There is
therefore a tendency to use high performance spring
materials close to their limits of torsional stress. High
strength carbon and tungsten steels are often used but it
has been found by bitter experience that these suffer failure
by hydrogen embrittlement in corrosive chemical plant
environments.

Many years ago a search for a strong material resistant to

corrosion and hydrogen embrittlement, led to the use of
precipitation hardening stainless steels of which FV520B is
our standard selection. This steel meets all the
requirements provided it is not harder than Hv 420
(equivalent to a strength of 1360 N/mm2). Otherwise it can
be susceptible to hydrogen embrittlement. This material has
been used successfully in many applications. The rare
failures have always been due to the spring being too hard.
As the material does not strain harden significantly in
service the failed springs must have been too hard on
installation.

The solution to the immediate problem is to check the

hardness of FV520B springs on critical relief valve duties
prior to service, preferably on receipt from the manufacturer.
The solution to the long term problem is to use less highly
stressed springs. Limits on hardness, and hence stress, are
at present being considered for inclusion in British
Standards specifications.

Although nearly always under static compression, the

service conditions of some relief valve springs can be
arduous. FV520B in this duty has developed a notoriety, yet
if the material is kept below Hv 420 this is totally unjustified.
Other materials were previously tried for critical spring
duties, but their performance was most definitely inferior to
that of FV520B. Another steel of this type, which is
commonly used by some oil companies is 17-4 PH

A D Tattersall

October 1983