7 replies to this topic

[recip]

Hi All and Art,

I am working on an issue of reciprocating compressor pressure fluctuation. Below are the details.Would like some advice on this issue.

Setup:

This is regarding an Ariel 6 throw reciprocating compressor driven by variable speed engine. It is a 3 stage(throws 1 and 2=3rd stage,throws 4 and 6=2nd stage,throws 3 and 5=1st stage) compressor with suction pressure of 55 psi and final 3rd stage discharge pressure of 1480 psi. There are 4 identical units running in parallel, exporting natural gas and partally also used as gas lift. There are 2 recycle control valves, one between 1st stage suction and 2nd stage discharge. The other recycle valve is between 3rd stage suction and its discharge. All the compressors run between 950- 1000 rpm in manual speed control mode. Recycle valve is always in closed position during loaded condition.

Problem Description:

Notice there is high amount of pressure fluctuation during compressor loading after a shutdown or stoppage. This pressure fluctuations is so high that sometimes the compressor trips on 1st or 2nd stage discharge pressure high. Engine speed fluctuations are not so much(maybe 5-10 rpm) Compressor valves related to the trip is inspected/replaced, but the problem is not with the valves .Loading is always done in Auto mode done through Allen Bradley PLC. Normally it is done from the local panel near the compressor instead of doing it from the Control room(CCR) as the unit is operated in Local mode.

1) My first concern is if the controller PID is set-up correctly. This is infact my major concern. Pressure transmitters is not an issue.

2) My second concern is if the original recycle control design or the set-up(example. 2 recycle controlling,recycle valve between 1st stage suction to 2nd stage discharge etc.) is correct?

3) My third concern is if the recycle line is sized small making the controller work harder and creating the pressure fluctuations.(3" line in the first stage recycle, 2" in the 3rd stage recycle). I think should not be an issue.

To confirm that my first concern is the reason for the problem and before i bring a PLC/controls guy to troubleshoot, i want to close the recycle valves manually to check how the system(pressure fluctuations) behave. If there are no problems then i can confirm that the issue is with the recycle valve/controller or the setting(PID).In my past experience at a different installation i have closed the recycle valve manually for loading in a single recycle control system but not on a twin recycle control. I thought about 3 ways to manually close the recycle valve, they are..

1)Close both the recycle valves gradually, say in 5 or 10% increment carefully watching the pressures.

2)Close 1st stage recycle valve fully and then the 3rd stage recycle valve.

3)Close 3rd stage recycle valve fully and then the 1st stage recycle valve.

I am not sure how the pressures will behave in these circumstances. What are the reasons for this pressure shifts, condition of the compressor is good. Please let me know your views on this and anymore details required.

Thank you all.

[Nirav]

Could you please provide a simple sketch of your system with location of pressure controllers, spill back valves as well as main equipments in the circuit so that it's easier to understand

[recip]

Hi,

Find enclosed Excel file with detailed schematic, recycle control logic and Compressor Setup datas(Ariel run file data).

I am adding the following additional information.

Problem Description

Recently we encountered trips due to 1st or 2nd stage dischrage high pressure. This happenned ONLY during loading of the compressor.After repeated attempts unit finally loads and operates normally.There are no pressure fluctuations once the unit is loaded.

Detailed Basic Data:

Wellhead/field gas from the production separators is compressed from 55 psig to 1480 psig by the 4 Ariel compressors with a design flow rate of 20 MMSCFD/each compressor at 1000 rpm. From here gas is is fed to the Glycol Dehydration unit and finally goes to Gas export and gas lift.

Engine speed is manually maintained/controlled between 950-1000 rpm. We dont want to change the engine speed often, due to concern of engine stability. Most of the time engines are run at maximum speed of 1000 rpm due to gas demand. Sometimes based on the recycle controller position we try to increase or reduce engine speed manually. Most of the time recycle valves are maintained to be fully closed.

We try to maintain the pressure at Production separators to be 55 psig. Discharge pressure requirement is 1480 psig for export and gas lift. I dont really know what to explain in detail about this. We need to maintain the pressures to have the correct rod load, reversals.

I really don't understand why they have designed 2 recycle controllers. This is the question i have and suspect this is also a reason for the interstage pressure swings during initial loading of the compressor. I have always seen only one recycle valve between the 1st stage suction to the final discharge.

Capacity Contro method: There are speed control(but only used in MANUAL), Manually operated VVCP(Varialble volume clearance pocket) which is always fully closed and then the 2 Recycle control valves. There are no Suction valve unloaders or spacers.

Please let me know if more details are needed. Thankyou.

Attached Files

•  Compressor Schematic Recycle logic Setup details.xls   1.46MB   189 downloads

[Art Montemayor]

At the initial outset, I quickly looked at the submitted flow diagram and because of the weight of the flow lines, I didn’t spot the intermediate natural gas feed stream coming from V-802 (which, presumably, is a field separator / surge drum at approximately 450 psig) going into the 3rd stage suction after combining with the 2nd stage discharge stream. I have highlighted the main gas stream lines on a revision of recip’s workbook. I show the intermediate feed line as a dashed one in order to draw attention to it. I also have generated a much simpler version of the flow diagram in order to show how the system basically works. This intermediate feed stream changes my initial understanding of the process. I first had the impression that the compressors were handling only one suction gas stream at the 1st stage and compressing it through the discharge of the 3rd stage. This would make the use of one recycle valve control very simple. However, an intermediate feed stream changes the basic process and mechanical design (as well as process control) and the startup and shutdown procedure of the four compressor trains. I am assuming all the trains are mechanically identical in size and application.

The main and important features and characteristics of this installation are:

The compressed gas is fed to a dehydration unit (TEG?) and this fact makes the steady and constant discharge pressure very important. The dew point of the dehydrated product depends on the pressure being maintained at a constant level. Therefore, the discharge pressure cannot degrade below the specified 1,480 psig.

The only means to automatically handle a turned-down flow rate of source gas is to recycle compressed gas back to the suction port(s) of the compressor. Since the compressor is designed to handle an intermediate stream of gas going into the 3rd stage suction, two recycle streams are necessary in order to maintain the compression ratios on all 3 stages. There are variable clearance pocket(s) reported for reducing the capacity, but these are not identified as to number, location, and capacity. Since these are manually operated, the only practical use they could offer would be for initial startup with little or no load and also for shutdowns.

Recip reports that the compressors trip out due to 1st or 2nd stage discharge high pressure. This happened ONLY during the “loading of the compressor” and there are no pressure fluctuations once the unit is loaded. By this, I presume it is meant that once the compressors are started up, they run OK. It is only during the startups that the compressors 1st and 2nd stage discharge pressures become excessive. If this interpretation is correct, then I draw attention to the compressor packager’s recommended startup procedures as written down in the unit’s operating manual. Because of the specific and special design of these compressors, the startup and shutdown instructions must be religiously followed. Ariel compressors do not write the startup, shutdown, or operating instructions for their compressors. Ariel sells compressors only to “packagers” – companies who engineer and design for the application of the compressor. It is the packager who controls how the machines will be piped up, operated, and controlled. Therefore, the packager writes the operating instructions. We, of course, have no knowledge of the mechanical design of these machines and it would be foolish - and perhaps dangerous - to presume and recommend a startup or shutdown procedure.

Reciprocating compressors are started up under “no-load” and slowly loaded with gas as they approach running speed. During this time, a blow-down and purge occur and the compressor comes on stream with 100% process gas going into the product line. With an intermediate stream coming into the 3rd stage suction, one can immediately draw the conclusion that the ability to inject the intermediate stream into the 3rd stage depends on the actual design and construction of the machine. Some compressors are capable of starting up with no intermediate gas being injected until after the machine reaches normal speed. But this should be a procedure recommended by the packager (always in conjunction with Ariel’s approval). That is why it is vital that the prescribed written startup procedures be followed religiously.

The purpose of the recycle control valve is to function as the basic capacity control for the compressor – and this can be done in a simple manner, just as shown in the attached revision of recip’s workbook. You should be controlling the discharge pressure by using a simple back-pressure control valve on the TEG absorber (if your process downstream isn’t or can’t do this).

However, to have the recycle control valve control the capacity of the compressor, you must control (or hold steady) the suction conditions into the compressor cylinder – and this will “fix” the discharge temperature if you are holding the discharge pressure constant. This is a cumbersome but simple problem that is a trade-off for using a simple and cheap capacity control method. You want the ability to recycle 100% of the compressor’s capacity in order to avoid shutting it down during periods when the gas supply is cut off or you suffer downstream shutdown(s). If this is an important part of your scope of work, the recycle valve can accomplish it – but at a price (as it is expected).

You should be able to control the compressor’s capacity 100%, using a recycle control valve designed for that purpose. However, you must design for the appropriate discharge conditions that allow for a constant, maximum suction temperature that will, in turn, allow for the desired discharge temperature.

You should also capitalize on the above ability and design the startup and shutdown of the compressor under NO-LOAD conditions by using the unloading capability. This should ensure 100% natural gas containment, without any need for flaring or venting when starting up and shutting down.

You will be consuming 100% of the required power to compress 100% of the design gas capacity – 100% of the time, whether you are on-stream or just idling off-stream. That is one of the big trade-offs in applying a recycle capacity control valve to a compressor.

The basis of the recycle capacity control valve, as I’ve drawn it on my PFD, is to maintain a positive, constant suction pressure to the 1st stage of the compressor. By doing this, what you essentially are doing is fixing the conditions (pressure & temperature) entering the 1st stage and this “fixes” the capacity of the 1st stage to a constant value and this maintains a constant flow rate through the compressor – but only up to the 3rd stage suction. All the recycle valve is doing is making up for any deficiency of gas coming from your source. The NET compression gas product going to your target is equal to what your sources are feeding your compressors. That is how this version of capacity control works. It compensates for the lack of sufficient source gas getting to the suction. It can't compensate for an excess of source gas.

This control scheme works well with a reciprocating compressor; the problems that you face are not with the compressor, but with your gas sources and your gas target.

I hope I have responded to your specific questions and that you have a good grasp on what you must do to fulfill your assignment.

Attached Files

•  CompressorSchematic Recycle logic Setup detailsRev1.xls   1.49MB   155 downloads

[recip]

Hi Art,

Thank you for the detailed reply. I have checked with the process engineer and found out that the side-stream from V-802 was never used(Vessel is isolated and all the valves closed,Not in use).So from the initial start date only gas from V-202 and V-202 B are the source for the compressors. During design stage(10 years back) they have considered one of the CASE to be using the gas from V-802, but this was never used and it will never be used also. My apologies for missing this information.

As you mentioned, there is a pressure controller after the Dehydration scrubber to maintain the pressure to 1480 psig.

There are also control valves on V-202 and V-202 B to maintain both the vessel pressures to 55 psig. Excess gas is relieved by these PCV to flare header. By this way pressure to the suction of compressors are maintained so we wont be seeing a excess pressure situation.

Note that the START and STOP logics are automatic from the PLC. We just need to give a command to LOAD and it loads itself automatically and same during UNLOADING.I am not able to find the packager's start/stop methodology.

Is the unstable pressures during loading due to this 2 recycle valve set-up?.

Please help if i have missed anything here. Hope i have provided the right information needed to analyse. Thanks.

[Babu Prasad]

Recip,

 

Still you are facing the problem or ok now?

[niharvikasmane]

  We have same set up of Compressor but only Operating Pressure is different .

we operate at first stage 19.5bara suction Pressure and discharge 52bara. second stage suction  and discharge is 128Bara.

3rd stage discharge is 220Bara.

we have done massive overhaul two compressor , but facing problem with one which is not developing Pressure out of first stage discharge .

the pressure should get at idle load is about 52 bar but getting only 33bar.even temperature not getting up more than 75DegC.

when we load ,the max Discharge pressure goes to 39 and Temperature 100DegC.

we suspect clearance in cylinder so replaced but condition unchange , any guess?? suggestion what else to be check, we have checked PSV, check valves, Replaced loading valve,VVCP is fully closed.

Appreciate if any one put light  and share experience.

 

[Bobby Strain]

You need lots of data to complete an analysis. The data will tell. At first glance, it seems that the first stage cylinder has insufficient pumping capacity compared to the other stages; hence the low discharge pressure. If you are satisfied that there is no leakage, the valves are good, and there is no recycle, then check that the rod for the valve unloder is the proper length.

 

Bobby

Edited by Bobby Strain, 14 December 2013 - 12:39 PM.