14 replies to this topic

[procengsas]

Good day,

I am designing a compressor system. The compressor will be used to supply consumers with 300bar Hydrogen. The existing compressors that we have are reciprocating compressors which take in hydrogen at 200mbar = 20 kPa in the first stage. The compressor is a 5 stage piston compressor. The system will comprise of a knock out pot, filters with dp cell, accumulator and a kickback to the knockout pot. The pressure control will be based on the kickback line i.e. if there is no consumption of hydrogen then the kickback will recirculate the hydrogen in the system with a solenoid valve on the main hydrogen supply line closing. My question to those who have this sort of design experience:
1. Do i need an expansion vessel? Will the knock out pot be sufficient as an expansion vessel in addition to a the solenoid valve which closes the main hydrogen supply to the system when it sees a high pressure in the KOpot. I also have a concern with the relief valve on this vessel which will be set to 25kPa. I am afraid the relief valve will constantly pop. Do i have anything to fear. Should I put in an expansion vessel in which the kickback at 300 bar feeds into.
2. I need to tie the relief valve line on the knockout pot into the flare system but due to the low pressure release I will definately have problems with the backpressure of our flare system. The flare system was designed for a backpressure of 50kPa. I could vent but that will not be safe especially during lightening. What do i do? I could size the relief valve for a back pressure of 50kPa but I am not sure whether that would be the correct way to go about it.
The flowrate of hydrogen to the compressor will be 20 kg/hr.

It would be greatly appreciated if anyone could assist. Thanks

[Art Montemayor]

Procengsas:

Your query is not clear because you furnish only a written description – which is lacking in accurate details. The best and most accurate way to describe a compressor system is undoubtedly with an engineering drawing – preferably with a P&ID, very much like the one I have originated and submitted with my comments on the thread titled “Dresser Rand Reciprocating Compressor”. You will find an Excel workbook showing a 4-stage Hydrogen compressor flow diagram on post #9. This thread can be found in the Refining, Hydrocarbon, Oil & Gas Forum. You would do well by reading this thread and downloading the workbook. With my doing most of the work by now, you should be able to easily expand the system to 5-stages and show all relevant equipment and instrumentation.

Your ideal compression ratio between stages is 3.019097, and this is a little higher than I would normally design the machine. Please show how you arrived at a 4-stage machine to do the job and what your existing compressors have as interstage compression ratios. Also show the discharge temperatures. I presume your existing machines are oil-lubricated in the cylinders and that the new one will be also. Is that correct? Always state your pressure reference. We don’t know if you mean absolute or gauge pressure.

What “pressure control” are you referring to? Do you actually mean compressor capacity control? I believe that is what you are trying to state. That is why the P&ID is so important and makes for clear communications. What I believe you are trying to describe is the universal problem with all reciprocating compressors: How to regulate the compressor’s capacity. There are a variety of ways to do this and what you vaguely describe seems to be the conventional, low-capital cost (high energy consumption) method of recycling a portion of the compressor’s last stage discharge back to the first stage suction drum based on first stage suction pressure demand. This should be confirmed when you furnish a complete and detailed sketch as I have requested.

There is no solenoid valve involved in controlling the capacity of the compressor as I’ve described above. If you are intending to furnish a solenoid valve to shut the suction gas, you are making a mistake in my opinion. That’s another reason to generate an accurate and detailed sketch of your complete system.

On which suction drum are you going to set the PSV at 25 kPa(g)? Please indicate the PSVs used on each of the suction drums on the compressor (a total of 5 suction drums) and their respective set pressures.

What is the purpose of “filters with dp cell” being used on this system? I have never seen a need for a “filter” on a Hydrogen reciprocating compressor. If you have gas that dirty, it should not be subjected to a reciprocating compressor. Please explain.

Await your reply.

[procengsas]

Hi Art

Thank you for your quick response. I have attached the PID. Please review and comment with my initial post in mind. I hope this provides clarity. I stand corrected regarding the pressures, its 14 kPag and 220 barg.

The compressor is a Sulzer type C5C with interstage aircooling. Each stage ties into a catch pot where the condensate and oil is drained. This occurs at 20 min intervals.

Thanks again

[procengsas]

Hi Art

I forgot to mention that we have water in the hydrogen supply gas and, to prevent the entry of water into the piston chamber, we try to remove as much water as possible. This prevents damage to the pistons and their associated parts.

thanks

[Art Montemayor]

procengsas:

Thank you for your timely response.

However, your sketches are not clear. They don't show all the details of each of the interstages. Please respond to each of the questions that I raised. What is Compressor 1? Your symbol for a reciprocating compressor is unorthodox and does not allow one to know where the gas enters and where it exits. Your process lines are mixed up with equipment lines. This makes your drawing very difficult to interpret. Use my spreadsheet graphic method and show all details as I requested. If you insist on taking short cuts you are depriving us of details and information.

Are you saying that the suction hydrogen gas is saturated with water vapor? Or are you saying that the suction hydrogen gas is entering as a 2-phase mixture with liquid water? And is this occuring in all 5 stages? I believe it is, because it should. You should not be trying to remove as much of the liquid water as possible; you should be removing ALL of the liquid water before each compression stage.

Await your reply.

[JoeWong]

Just chimed in to add my 2-cents worth information...

Your query is not clear because you furnish only a written description – which is lacking in accurate details. The best and most accurate way to describe a compressor system is undoubtedly with an engineering drawing – preferably with a P&ID, very much like the one I have originated and submitted with my comments on the thread titled “Dresser Rand Reciprocating Compressor”. You will find an Excel workbook showing a 4-stage Hydrogen compressor flow diagram on post #9. This thread can be found in the Refining, Hydrocarbon, Oil & Gas Forum. You would do well by reading this thread and downloading the workbook. With my doing most of the work by now, you should be able to easily expand the system to 5-stages and show all relevant equipment and instrumentation.

Click here to reach the said post.

May consider dynamic simulation to check if expansion is required. This is commonly done to verify the provided inlet and outlet volume is sufficient or overly provided.

[procengsas]

Hi Joe,

How would i go about doing the dynamic simulation? Can i do this simulation using Aspen? Is there any guides that i can follow? Your advice will be greatly appreciated.

Art, there is 0.2 mole percent water in the hydrogen gas and we try to remove as much of the water as possible. For your other questions, i will get back to you asap.

Thanks

[JoeWong]

Conducting needs certain level of expertise. Generally you may engage third party i.e. GENESIS, TRIDENT, etc to conduct such studies if your company do not have the inhouse expertise. Nevertheless, dynamic studies can be conducted using FLOWMASTER, HYSYS, etc.

[Art Montemayor]

procengsas:

Thank you for your work effort and response. I have looked at your submitted process sketch and have made my comments on it with "call outs". Please study the attached Rev1 I made to your submitted sketch.

Briefly, I would seriously recommend you NOT proceed with what you are proposing. You are incorporating what I recognize as serious, potential safety hazards - and for no given reason. What you depict is not the normal, safe, conventional way that a reciprocating compressor - particularly a hydrogen compressor - is controlled or piped up to operate.

I believe my comments are self-explanatory, based on the information you have supplied.
 Reciprocating_Compressor_SchematicRev1.zip   347.25KB   119 downloads

[Art Montemayor]

You have not made an error yet - at least not if you are at the design stage. By consulting with this Forum, you are making positive gains, not errors.

Your previous workbook should be Rev2. I am enclosing Rev3 with comments. Please note the engineering methodology of document control. We employ Revision numbers to keep a control on what is being done during a project. Otherwise confusion reigns.

You are adding bits of scope of work as you make comments. This is not wise. By not telling all of the story at the onset, the comments and remarks will continue and will be very confusing since we are not working on the same plane of information. You now reveal that you have customers for the compressed product and that the demand may be reduced. If additional new information to the scope of work continues, a lot of work will either be modifieded or wasted. One of the worst things to incorporate in a proposed engineering design is CHANGES. Changes cost a lot of engineering money. That's why a proposal should be well thought-out and designed from the very onset. Changes have been known to multiply the final cost of engineering projects.
 Reciprocating_Compressor_SchematicRev3.zip   355.26KB   103 downloads

[Art Montemayor]

Procengsas:

I am attaching Rev 5 of the workbook with calculations and responses to your questions and comments.

I am going on vacation Friday and will check in the Forum from time-to-time. However, I won't be in position to respond positively during this time. Although you don't say what position you find yourself in the project, you have your hands full with studying all in the workbook, obtaining information from Sulzer and making sure your process is operable, controllable, and SAFE.

Check out my calculations for the Adiabatic discharge temperatures.

Good Luck.
 Reciprocating_Compressor_SchematicRev5.zip   365.93KB   94 downloads

[procengsas]

Hi Art,

I have adjusted the system with your comments in mind. Could you please review. I have added more questions and call-outs.
Thanks

Attached Files

•  Reciprocating_Compressor_SchematicRev6.zip   466.65KB   71 downloads

[Art Montemayor]

Procengsas:

I am sorry that I can't reply to your comments and questions added onto the latest worksheet version. I am on vacation and won't be back until the 19th. I have tried to work with the workbook but I am using my son-in-law's Apple computer which is not my cup of tea with respect to the way it handles Excel. I have tried twice and it is taking just too much time to work with Apple's Excel version.

You continue to show errors and non-workable propositions in your P&ID. I will address each one when I return to my own machines at home or at work. Sorry.

[procengsas]

HI Art

I found the a reciprocating compressor scheme on the net. Is this the way i should design my compressor system. Can you advise.

The website is http://www.simtronic...spm/spm1300.htm

Thanks

[Art Montemayor]

Procengsas:

Attached find my comments on the Rev7 workbook. Read all my remarks and comments on the Proposed flow sheet and in Comments flowsheet.

I am also attaching a copy of the article you are suggesting as a "go-by" for controlling your machines. Please read my remarks in RED carefully. Note that my remarks are consistent in philosophy and content - while the author of the article wavers in what is written and stated in several sections. This is typical of technicians or engineers that don't have a total grasp of the process know-how. Do not interpret my comments here or on the article as being a criticism of the article or as my expressing my superior knowledge. This is totally not my intent. I am applying what I have accumulated while working with compressors over the last 48 years and I don't have the time nor patience for ego trips. The article is basically well put together, but lacks hands-on and experienced know-how. This will be obvious to all who read my red remarks. However, in all fairness, the author(s) is probably an instrumentation or controls individual and obviously is not specialized in positive displacement machinery - especially gas compressors. The best thing for you to learn is the good schematic drawing that is furnished - which is what I told you to do from the very outset of this thread. I hope you have learned that it is essential to communicate and document your engineerring design this way in order to study and logically explain what it is that you propose to do. All this should be formally documented in a detailed and accurate P&ID drawing.

I hope this helps you (& others) out.
 SPM_Reciprocating_Compressor.doc   136KB   145 downloads
 Recip_Compressor_Rev7.zip   510.81KB   129 downloads