6 replies to this topic

[trinhduchanh]

Dear All Member,

 

I am a mechanical commissioning engineer for one FPSO project in Viet Nam field. We are doing commission for reciprocating compressor package . During initial start at recycle mode with  PCV - 012A at discharge cooler fully Open. I attached P&ID drawing as below for reference. During running we found that suction pressure at PIT-0041 always higher than the High High pressure of PIT-004A. Sette out pressure is 120 kpa while High High set at 80 Kpa. That why compressor always trip by High High Pressure. Any one can look on this and advice to us can increase the setting for PIT-004A above the value of settle out pressure while PCV-012A fully Open.

 

Thank's.

Attached Files

•  0909-GS-R-XB-12001_C2_PIPING AND INSTRUMENTATION DIAGRAM.pdf   870.57KB   498 downloads

[Art Montemayor]

Trinh:

 

Because you haven’t furnished a scope of your compression design, a cause and effect and a logic description of what you are trying to control (and why), it is difficult to tell you why your alarms are tripping and why your pressures are not what you expect.

 

Studying your P&ID I find a lot of issues that I am not happy with.  Whoever drew this diagram hasn’t been doing it a long time.  There are more questions that come up when looking at the drawings:

• There is no such thing as a “settle out” pressure related to a reciprocating compressor.  This term is related to a CENTIFUGAL compressor, not a reciprocating one.  The reason for this is that a reciprocating compressor is a positive displacement machine and unlike a centrifugal machine, discharge gas cannot freely flow from the discharge nozzle back to the suction nozzle by retro-flowing through the compressor.  This is because both the discharge and suction valves on any reciprocating compressor cylinder are truly nothing more than check valves - they only allow gas to flow downstream, never upstream.  Consequently, I don’t know what the designer of this system is trying to do.
• Are you merely trying to control the capacity with the recycle valve?
• Or are you trying to use the recycle valve to shut down the compressor under a minimum of load by recirculating gas around the 1^st stage while blocking any feed gas and downstream gas.  If so, where is the SDV on the discharge?  How do you isolate the 1^st stage?
• Are you using any other capacity control device on this Dresser-Rand machine?
• Did Dresser-Rand design the capacity controls, the start up and shut down procedures?  If not, who did?

Please refer to the attached simplified flow diagram that shows how I believe one would basically design a capacity control system for a one-stage compressor.  I don’t show start up or shut down hardware - such as blow down valves because I don’t know the scope of the process.  The intent is to show that basically, the first goal of a recycle valve is to control the suction pressure to a cylinder and keep it positive and steady.  Therefore, I don’t see the reason for linking the required signal with any source other than the compressor suction.  Admittedly, you may have a very good reason for having the system you show, but you haven’t explained the instrumentation logic and the reasons behind it.

 

 Reciprocating Compressor Recycle Control Valve.xlsx   45.29KB   504 downloads

[trinhduchanh]

Dear Art,

 

Thank's with your comment. I will explain to you understand what i doing here with our compressor. ROS is the vendor for this compressor package. They bought reciprocating compressor from Dresser Rand. After that they design P&ID, logic for whole package and fabrication and then handover for us doing commission prior put it online service. I attached all the information of this package as per your question. 

 

The method for capacity control performed by the recycle valve from the discharge cooler to 1st suction scrubber. During initial start compressor we have trouble with feed gas from LP seperator. It not sufficient feed for compessor run continuously.Therefore we unable to put compressor ON LOAD mode and then we run it at NO LOAD mode with 1st suction SDV Open and Block the Manual Valve at discharge of 2nd Stage. In this case compressor running wih PCV - 012A fully Open for recycle gas from discharge to suction due to 1st suction pressure low. This logic has mentioned on Control Start Up & Shutdown Philosophy document. While PCV Open then 1st suction pressure bring up above high high level. It caused trip the compressor.

Edited by Art Montemayor, 04 August 2014 - 08:39 AM.
Removed Attachments

[xavio]

Trinh,

 

What you provide is a complete set of engineering document which I would consider "sensitive materials".

I understand that your problem is real and urgent, but please be sure that you're authorized to make them public.

 

Back to your problem, I believe the settle-out pressure used for this application is just a carelessly invented term.

The settle-out condition with centrifugal comp cannot happen to recip comp.

Would you please define your settle-out condition for us?

By the way where is PT-0041 located? I cannot find it in the PID

 

In any case, I think the D-R team know best. What are they suggesting now?

 

A few years ago I had similar problem where feed gas flow was insufficient during compressor start-up.

Our compressor, unlike yours, is equipped with 50% unloader.

So, we switched to the 50% mode and proceeded the start-up normally under the supervision of vendor.

I myself unfortunately did not join the whole start-up process, so I cannot explain in details, sorry..

But, I remember that we experienced no significant problem; only some controllers were re-tuned to get more desirable responses.

 

Good luck!

 

xavio

[trinhduchanh]

Xavio,

 

As early correction from Art then my term "settle out pressure on reciprocating compressor " state that incorrect. Sorry for my mis-understood. Currently, we overrode PT - 004A for continuous running and testing. We also thinking about installation unloading valve solution for reduce 50% capacity in this case insufficient gas flow. We has raise the issuse to Dresser Rand and wait them get back advice. 

[Art Montemayor]

Trinh:

 

After looking closely and in detail at the complete P&ID and the Compressor data sheet you submitted, I am afraid I have more questions than you.  I can see now that you did not tell us ALL of the information that directly impacts on the operability of this natural gas compressor(s).

• For example, your data sheet seems to show that the compressor(s) in question is a machine that has ONE main crankcase, four throws, and four cylinders – of which, two are 1^st stage compression and the other two are 2^nd stage compression.  The fact that there is one common crank case means that there is ONE driver for each crankcase.  That means that the 1^st and 2^nd stage are both operated by the same driver and they are in tandem.  You didn’t tell us this.
• There is a SIDE STREAM from the MP (“medium” pressure?) separator that enters the 2^nd stage and by-passes the 1^st stage.  This means that the amount of gas entering the 1^st stage is LESS than that entering the 2^nd stage.  You never mentioned this.
• There are TWO RECYCLE VALVES on each 2-stage compressor: one around the 1^st stage and another around the 2^nd stage.  You also never mentioned this.

In the information you furnished there is no mention of the logic, reasons, and procedures for using the controls that are installed.  I don’t know if there are any other capacity controls and whether the recycle valves are meant for startup and shutdown as well as for capacity control.  But I now know:

• Your natural gas is very heavy (high in propane+ content.  This means that it may form liquid product upon free expansion through a pressure control valve.
• What is entering the 1^st stage is different from what is entering the 2^nd stage.
• This compressor is very susceptible to being unbalanced as far as compression requirements should the two incoming flows deviate from the design valves.
• I question whether the manufacturer (dresser-rand) is aware of the type of application and controls being used on the machine and whether they approve of this method.  Therefore, I urge you to inform Dresser-Rand of exactly what you have as controls and how the compressor(s) are being proposed to operate (startup and shutdowns included).  There is a possibility that you might have a rod overloading situation and you could generate a scenario where mechanical damage might occur.  In my experience, I would expect much more sophisticated capacity controls as well as startup and shutdown controls – particularly because this is an Off Shore application.

Please let me know if you have any doubts or have further questions.  I refer you to some of the many threads I have written about reciprocating compressor recycle valve controls in order to make sure I have explained in detail the logic and reasons for using recycle valve control and its limitations.  Note that I always make it a point to remind all of the purpose and the reasons for using recycle valves on reciprocating compressors.  I ask you to read these threads and note the reasons cited.  I am afraid that your concern is much more serious than just a "settle out" pressure.

 

http://www.cheresour...a-bypass-valve/

 

http://www.cheresour...-recycle-valve/

 

http://www.cheresour...ssor-discharge/

 

http://www.cheresour...ressor-recycle/

[trinhduchanh]

Dear Art,

 

Thank's for your well explaination. After worked with Dresser-Rand factory then they recommendation us using one PCV are sufficient for recycle gas from 2nd stage discharge to 1st suction scrubber instead for using 02 PCV for each stages. I also over look my document then i founding in the another project they use multistage reciprocating compressor of Ariel with design only one PCV for recycle gas from last stage to 1st suction. 

Attached Files

•  Using one PCV for bypass method of reciprocating compressor capacity control.pdf   65.46KB   331 downloads

Edited by trinhduchanh, 08 August 2014 - 09:32 PM.