5 replies to this topic

[vne]

All,

I'm facing an issue when calculating centrifugal compressor power: discharge temperature as calculated by the method stated in GPSA (Eq 13-37, 2004 edition) is around 30% higher than that calculated by HYSYS. Below is the input:

Gas inlet properties:
Molecular weight: 35.68
Mass flow rate: 816,480 kg/hr [1,800,030 lh/hr]
Temperature: 46.1 degC [114.98 degF]
Pressure: 13.79 bara [200.01 psia]
Z factor: 0.8665
Cp/Cv: 1.241
Polytropic efficiency: 0.781

Discharge pressure: 44.82 bara [650.06 psia]

The discharge temperature as per GPSA is 154.8 degC while HYSYS gives 118.7 degC.

Attached please find the .zip file containing an excel file conforming with GPSA and an .hsc file for HYSYS calculation.

Anyone please help explain the big gap. Thank you.
 Comp_.zip   34.22KB   275 downloads

Edited by vne, 23 December 2010 - 08:46 PM.

[Zauberberg]

Check the calculation path in Excel as it seems that both Isentropic and Polytropic compression result in almost identical discharge temperatures, which cannot be true.

[ankur2061]

vne,

Strictly going by your parameters the answer of 154.8 deg C is absolutely correct but there are a few things that you need to note:

HYSYS compressor calculations are based on the following:

1. They consider the average compressibility factor based on the inlet and outlet compressibility factors which in turn is a function of the suction and discharge pressures and temperatures. The whole process is an iterative procedure.

2. Check the K or Cp/Cv value in HYSYS. It might not be the same as 1.241 which you have used with the GPSA equation. When Cp/Cv is calculated manually it is calculated as follows:

K or C_p/C_v = C_p / (C_p - (8.314/MW))

HYSYS uses a rigorous method to establish the K value based on the gas composition and the property package selected. If you select a different property package such as SRK in HYSYS you will get a different outlet temperature. You can check this out.

3. If you do not specify the efficiency HYSYS considers a default adiabatic efficiency value. You need to provide your own polytropic efficiency value in the efficiency tab and then complete the simulation. You need to check this out.

In my previous experience also HYSYS has always provided lower temperatures compared to manual calculations because of all the reasons mentioned above.

Hope this helps.

Regards,
Ankur.

[vne]

Ankur,

Thanks for your reply. I think the result given by HYSYS is more reliable since it uses Schultz method (as stated in the HYSYS Operations Guide) which is recommended by ASME PTC-10: Performance test code on compressors and exhauters.

2. For the k value (Cp/Cv) as input to GPSA method, I take it from HYSYS (Cp/Cv of the inlet flow). I'm not sure if this is the reason making the big gap in results. Please advise.

3. The polytropic efficiency is already an input to HYSYS.

[ankur2061]

2. For the k value (Cp/Cv) as input to GPSA method, I take it from HYSYS (Cp/Cv of the inlet flow). I'm not sure if this is the reason making the big gap in results. Please advise.

vne,

The Cp/CV value does not change dramatically from inlet to outlet and the inlet value can be considered the average Cp/Cv value. So your considering the inlet k value is absolutely right.

Did you check out the difference in the calculated gas power and the outlet temperature for the same inlet conditions and discharge pressure if you use a different property package such as "SRK" for your case in HYSYS. You will definitely find a difference.

I have developed my own calculation spreadsheet for centrifugal compressor calculations and I have compared it with HYSYS and always found that the discharge temperature predicted by HYSYS has been lower than what I have calculated. This spreadsheet is available as a spreadsheet for sale on the "online store" of "Cheresources" and has been specifically developed for engineers who don't have access to simulation software such as HYSYS. An additional feature of this spreadsheet is that it provides the number of stages and the rpm for the compressor. The spreadsheet can be previewed at:

http://www.cheresour...|&cat=33&shop=1

Hope this helps.

Regards,
Ankur.

[vne]

Did you check out the difference in the calculated gas power and the outlet temperature for the same inlet conditions and discharge pressure if you use a different property package such as "SRK" for your case in HYSYS. You will definitely find a difference.

I have just checked by switching from PR to SRK, the result difference in outlet temperature is 0.08% increase, in compressor power is 2.2% increase.

I have developed my own calculation spreadsheet for centrifugal compressor calculations and I have compared it with HYSYS and always found that the discharge temperature predicted by HYSYS has been lower than what I have calculated. This spreadsheet is available as a spreadsheet for sale on the "online store" of "Cheresources" and has been specifically developed for engineers who don't have access to simulation software such as HYSYS. An additional feature of this spreadsheet is that it provides the number of stages and the rpm for the compressor. The spreadsheet can be previewed at:

http://www.cheresour...|&cat=33&shop=1

I also compared the discharge temperatures among your sheet, my sheet, and HYSYS result for the set of input in your sheet preview:
Your sheet: 99.9 degC
My sheet: 99.9 degC
HYSYS (SRK EOS): 93.78 degC (which is 6.1% less than the sheets result).

For the set of input in your sheet preview, the results difference can be neglected but for the set of input in my example (see my attachment in the above post), it is more than 30% difference. I believe if HYSYS is available, result from it should be used over the one from GPSA.