Reciprocating Compressor Piping - trouble shooting 1

[clooney]

This is really a troubleshooting question:
A personal friend who works at this plant brought this up in a casual conversation and made me curious as to what can be done about it.
A piping system connected to a reciprocating compressor which runs at a constant speed of 427 RPM with 2 pistons. The capacity is controlled by 4 valves. Opening of each valve adds 25% to the capacity. The 10” dia. discharge lines leaving the compressor dampers are connected to a single header which is connected to an exchanger. The outgoing pressure from the compressor is about 30 Bars when it runs at 100% capacity. The inlet pressure is at 19 bars.

The commissioning team is in the process of testing the system with nitrogen. The actual service fluid is supposed to be Ethylene (similar properties as far as density, gamma, ....). Since all the related piping has not been installed, they have closed off the piping system at a certain point and tested the suction / discharge piping in a closed loop. As I said earlier, compressor discharge piping is connected to an Exchanger and a rather “strong pulse” is felt on the exchanger wall. Apparently, one can feel the pulse by simply putting his hand on the exchanger wall. The strongest pulse is felt when the compressor is operating at 75%.

Now considering that we’re dealing with a reciprocating compressor, I don’t see it as odd feeling a certain pulse on the piping and the exchanger. Of course, the pressure pulsation has to be within an allowable range which is usually verified by a pulsation study during the design phase in which the piping system and the equipment are modeled and the flow dynamics is simulated using computer softwares. Based on this study certain “restriction orifices” have also been recommended within the piping. I should note that these simulations were done by a third party. Some company in the Netherlands called TNO which is somewhat of a research institute too. They use a software called Pulsim.
The client claims all the recommended orifices have been installed, yet they find the existing pulsation worrisome.

I have never been to this plant and the owner of the plant and my friend are both not very experienced in dealing with such scenarios and I don’t know what their notion of “strong pulse” is.

It is also noteworthy, that the pipe supports have been stiffened at certain points and the flow induced vibration in the piping has been reduced to an acceptable level. Before they make the necessary adjustments in the Pipe Support Restraints, the level of vibration was so high that the compressor would actually trip and shut off when they left it running at 75% capacity.

I know this is not a whole lot of information to go on and that quite a number of parameters need to be considered when dealing with such situations. I need some input from the guys who have had dealt with such problems up close in the field and have a feel for unacceptable ranges of pulsation in the piping and connected vessels. When is it a concern?

All I can personally think of is that it must have something to do with the Test Crew closing the piping system at a juncture and running the compressor in a closed loop. I haven’t seen the report by TNO but I am suspecting that they did the simulation of the system as a whole (i.e.; all the piping included). I think running the system partially most likely changes the dynamics of the flow and levels of pulsation.

Anyhow I am sorry. I talked too much.

What do you guys think?

Thanks

 

[zdas04]

The "science" of pulsation dampening on recips is similar to the permanant bend radius when fabricating dies for vehicle body work - a lot more art than science.

Pulsation dampners are never perfect and they always take credit for the entire piping system. Short-cycling the compressor discharge back into the suction is a very common activity that tells you a lot about the set-up of the compressor and almost nothing about the pulsation and vibration of the entire system because the bottles and the orifices are tuned for a system, not a compressor. Between sympathetic vibrations, vibration amplification, and syncronization you can end up with a system that looks like it is going to shake itself off the foundation during the test and works just fine with the normal piping (unfortunately, the inverse is also sometimes true).

I can't tell you that this machine will be just fine in normal service, but it may be. Sorry to be so vague, art does not allow precision.

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The harder I work, the luckier I seem

 

[pmover]

the pulsation study that was done during design should have included all operating scanerios (i.e. 75%, 100%, or otherwise). based on results, the compressor inlet/outlet pulsation dampening devices are installed.
as with recip compressors, pulsations in piping systems are generated due to the generated pulsating flow. this pulsation wave travels through the piping system until attenuated. obviously, the pulsation wave is noticeable at the exchanger, hence i suspect there is a piping bend at or near the exchanger and the pulsation wave is not attenuated by the dampening devices for the compressor operating condition. certainly, installing more or stronger piping system anchors may help, but it does not provide a long-term solution.
suggestions, revisit the pulsation study done during design and compare as-built conditions with the pulsation study and analyze input data used in the pulsation study. determine compressor operating scenarios and were these scenarios included in study. if satisfactory, perhaps some pressure & vibration data can be collected at various locations at various operating scenarios and then an dampening device can be installed to reduce the levels of vibration. btw, there are limitations provided in api-618 for various type of pulsation studies, perhaps this reference may provide you more help.
good luck and this matter can be a challenge, but remain focused on the source of pulsation.
good luck!
-pmover

 

[LSThill]

I have done this type of piping analysis for API 618 compressor using BOS FLUIDS API 618 CHECK. FOR MORE TECHNICAL information go to

http://www.paulin.com/

 

[Montemayor]

What I think is the following:

1. A reciprocating compressor has no capability of increasing its capacity by the “opening” of any valves. By your description of the capacity controls I have to assume you don’t have any experience or knowledge of how a recip works and how it is regulated with respect to capacity. This is not a put-down; what I’m pointing out is that your description of how the compressor is controlled is defective. I’m going to have to assume that the compressor has clearance pockets and that the “4 valves” that you describe are, in reality, pneumatic or solenoid valves that admit instrument air to a piston actuator that “loads” the related cylinder by closing its clearance pocket. If you have clearance control on the first cylinder, you must also have clearance control on the second if the cylinders are in tandem (in series). You haven’t stated if this is a 2-stage machine with 2 cylinders in series or a single-stage machine with the 2 cylinders in parallel. I’m going to assume the latter, since a compression ratio of 1.57 is a very low and conservative one. (note that you haven’t said if your pressures are gauge or absolute; this makes a difference in the compression ratio calculation).
2. Without correct, detailed, and accurate information it is impossible to trouble-shoot a reciprocating compressor operation. However, there are some indications this may not be needed.
3. I note that no information is given about the action (or reaction) of the discharge pressure gauge(s). Since this is not mentioned, I presume that no extraordinary action is being exhibited by the gauge(s). These gauges are the first and most important indication of any excessive gas pressure pulsations in the discharge system. What you report is a “feeling”, which is a fantasy people have about gas pulsations. You can’t physically “feel” gas pulsating. What your hand feels is pipe vibrations – which are a result of internal discharge gas pulsation. If you have proper piping anchors or tie-downs, the piping will not vibrate, but will still be sustaining internal pulsations and the pressure indicating devices are the first and most important instruments that demonstrate this effect – and ultimately pay the price by self-destruction if they are not dampened.
4. Pulsations are a result of poor capacitance capacity within your discharge piping system. You’ve given the 10” discharge pipe size, but you fail to indicate what the total compressor discharge capacity is. Experience is often the best ingredient in estimating the proper discharge piping size and configuration in order to bring down pulsations to a tolerable level. You will always inherit pulsations with a reciprocating compressor. The RPMs don’t give an indication of a potential pulsation problem; the system’s capacitance does.
5. In my experience, I’ve never seen any success of pulsation dampening by the use of restriction orifices. To me, this is all a waste of precious HP energy. If your discharge piping is of a sufficient size and it is installed with a minimum of turns and convolutions, valves, fittings, and has a discharge snubber vessel with a liberal capacitance volume, you should have no problems. I would not tolerate any piping vibration caused by the machine. Here it is important to make sure that the vibration is not cause by a faulty mechanical operation on the part of the compressor due to bad bearing, loose parts, bad balancing, or even mis-alignment. This can all cause pipe to vibrate terribly and give the impression that gas pulsation is the culprit behind all this.
6. The best people to get ideas and recommendations on bad pulsation problems is from the compressor manufacturer (which you haven’t mentioned). This is where you should be establishing if you have the correct and recommended piping installed for this machine. This is the start. The manufacturer will also give you their acceptable method of measuring any suspected gas pulsations problem and you should apply that method to establish that, in fact, you do have a pulsation problem.

As I stated, the discharge pressure gauges are the tell-tale signs of a bad pulsation problem – usually caused by incorrect discharge piping size and design, coupled with a lack of capacitance within the system or the lack of a snubber vessel (or discharge separator). I always mount my pressure indicators (or the pressure signal) on the discharge snubbers or separators of each stage of my reciprocating compressors. In a properly designed system I have never had to throttle down on the gauges or used a snubber device to calm down the pressure gauge needle. This is what should be occurring in your installation as well.

I hope this experience helps you (or your friend) out.

 

[clooney]

Dear All,

Sorry but that was really all the information that I had regarding the piping system at hand. And I was simply quoting the friend that works at this plant. Anyhow, thanks for the replies. Unfortunately, I have no access to the relevant documents and won’t be able to get a hold of that “pulsation study” report.

Montemayor
1)Touche…. You’re right. I don’t have an in depth knowledge of compressors in general. My work limit really stops at the F.O.F. of equipment packages and I leave the rest to the guys who deal with rotary equipments. I wished I did though.
2)Since you were somewhat blunt in your reply, may I say that my initial reaction was to snap back at your comment saying “…you don’t have any experience or knowledge of how a recip works …” but by the time I got to the end of your elaborate reply I just wanted to thank you for sharing your experience.

3)On the subject of “feeling a pulse” on the exchanger wall, I suppose this friend was simply “fantasizing”. May I then infer from your reply that you lean more towards the possibility of the piping being inadequately restrained?

Thanks again and I apologize for the lack of sufficient data on this problem. But that was all I had and I was just expecting some general insight on systems connected to reciprocating equipments not a definitive answer.
C

 

[Montemayor]

Clooney:

There is nothing to apologize for – at least not to me. If I initially came across as overbearing and pompous, I can assure you that that is the farthest from my principal concern or intent. If I dedicate my time to answer you, I make it a point not to waste your time and mine building myself up or tearing others down. I don’t need that kind of input or output. My main concern is aiding or helping you out in what you consider to be a problem or lack of experience. I just happen to have the experience – but not because I’m smarter or cleverer than you. I simply have been around much longer (& believe me, I can feel it in my bones!) and have been through the wringer on most of what I write about. I consider it an accomplishment if I can help others avoid the missteps and shortcomings I experienced early in my career due to my errors or lack of guidance. Most, or all, that I’ve accumulated as helpful experience is due to some timely and great engineering mentors as well as lucky opportunities to be in the right place at the right time – things that will also happen to you and others as time goes by.

Experience makes me repeat that a “pulsation” problem should be investigated first for proper mechanical alignment and compressor parts’ operation. Mechanical vibration is often mistaken for gas pulsation due to the resultant pipe vibrations that develop. Mechanical alignment and proper operation is much more important than gas pulsation – although I’m not tolerant of either. A mechanical problem is a menace to the machine proper; a pulsation is normally a pain in the backside that ruins a lot of good instruments. Both also contribute to undesirable piping vibration. The mechanical problems are a constant maintenance concern. Once resolved, pulsation problems rarely are seen again.

I hope you get an opportunity to work intimately with reciprocating compressors. They are unique in the roles they play in process needs and they are a forgiving breed if treated and operated in good engineering fashion.

 

[clooney]

Montemayor:

I also hope I get a chance to work closer with these equipments.
Montemayor, you're the man - Really. I always look forward to reading the threads that you are involved with.
regards,
C