4 replies to this topic

[obsidian_dust]

In normal condition, air instrument in our refinery is supplied by five air compressors, with one additional compressor is available for back up. From those six compressors, three of them are identical recip compressor (design capacity 23 Nm3/min@) installed in parallel. Normally, two of those three recip compressors run in parallel. There is no fixed flow indicator installed in the system, either at the discharge of each compressor or at the pipe header.

Several weeks ago we tried to use portable flow indicator to measure flow come out from each compressor. During the measurement (we do it on one compressor at a time), we observed that each compressor could achieve its design capacity, but it's only stable for several minutes. During the measurement, the flow tends to be fluctuated and in many occasions even the display indicated negative value (the flow indicator technision informed us that it caused by backpressure). Just like the flow came out from each compressor was affected by some kind of pulsation.

Once I read that such a phenomenon could happen when we run recip compressor and positive screw compressor in parallel without receiver vessel between their interconnected discharge.Should we expect to see similar phenomenon by running two identical recip compressor in parallel?

Any idea or sharing of experience will be highly appreciated.

[Art Montemayor]

First of all, please furnish COMPLETE and detailed basic data. You tell us you have 6 air compressors installed in parallel, 3 of which are reciprocating types. But you don’t tell us what the other compressor types are.

You can spare yourself (and your readers) a lot of confusion and errors by simply generating a detailed flow diagram (preferably in an Excel spreadsheet) and identifying where the flow measurements have been taken and how the compressors are connected and tested.

In my opinion, it is a waste of time and effort to try to accurately measure the discharge flow rate of a reciprocating compressor - especially with a portable flow meter. The machine is basically a positive displacement device – something that is a flow measurement device itself when working in reverse flow. This type of machine generates inherent pulsations in flow (and pressure) that require pulsation drums to reduce the inherent effects. Even with the best of pulsation drums installed, you will still obtain some pulsation.

If you want to know if your compressor is working according to design, the best way to detect this is to monitor your compressor valves’ conditions. Their discharge temperature is one quick way. The other is to always inspect and maintain your compressor valves and piston rings in top running condition. This is done with timely and well-done maintenance. There are established and standard ways to measure the delivery capacity of a reciprocating compressor. Manufacturers of such machines do this all the time. The method is outlined in the literature or by the compressed air industry. My copy of “Compressed Air Data” by F. W. O’Neil; 5^th Edition, sixth printing; Compressed Air Magazine Company, New York City gives all the details for rating such machines. I bought my copy in 1960 for $3.00 and it is probably not in existance anymore. But other sources have to exist and you should research this out. The basic test is based on the ASME test Code for Displacement Compressors.

[obsidian_dust]

Dear Mr. Montemayor,

Thank you for your enlightenment and sorry for this inconvenience. You could see the attached excel sheet, the recip compreeors I mentioned previously is K1,K2, and K3 (red circled). You were made a very good point in this following statement: "In my opinion, it is a waste of time and effort to try to accurately measure the discharge flow rate of a reciprocating compressor - especially with a portable flow meter." Actually we tried to do it because when we run two of those three compressors in parallel (K1+K2, K1+K3, or K2+K3), we see that the total flow detected by the fixed flow indicator located at the downstream of those three compressors never show any value close to 2 x design capacity of each compressor (usually we only get 30 Nm3/min by running two compressors in parallel instead of 46 Nm3/min). Based on the measurement using portable flow meter, we saw 23 Nm3/min can be achieved by every single compressor (K1,K2, and K3) even within pulsation. FYI, during measurement we installed the portable flowmeter at the upstream of the receiver. Yes, you absolutely right, it explain that the pulsation cannot be avoided during measurement.
Please let me know if you think that installing a fixed flow indicator in the downstream of a reciprocating compressor is meaningless. Any other information will be highly appreciated.

Thank you very much and have a beautiful day

Attached Files

•  Overall simplified diagram - IA system (2).xlsx   97.12KB   77 downloads

[Art Montemayor]

Obsidian:

Thank you for the nice schematic flow diagram that details out what you have and how you are operting it. It certainly saves us all a lot of time and effort in understanding what is being discussed.

I stated it before, but I will reiterate that I consider a fixed flow indicator in the downstream of a reciprocating compressor as meaningless and unrequired for proper and efficent operation.

Additionally, please refer to my comments included in your submitted workbook. I hope this experience helps you out.

Attached Files

•  Overall simplified diagram - IA system Rev1.xls   180KB   112 downloads

[obsidian_dust]

Mr. Montemayor,

Thank you so much for your sharing. We have a plan to discuss this issue with the system designer and your opinion will be a great point to be discussed.

Again, thanks so much and have a nice day.

Rgds