Compressor Performance Curves Simulation 1

[AliZirgach]

I am simulating a centri. compressor via Aspen One. I know how to do this in general, but want to clear up a confusion.

I have two types of performance curves from manufacturer: Pressure Ratio vs Std Flow and Pressure Ratio vs Actual Flow.

Aspen One gives two options to you for specifying curves: Mass Flow and Actual Vol Flow.



Assuming an ideal gas and using the given inlet MW, I can just convert the Std Flows from curves to Mass flows and then use them in Aspen One, right? So I am a bit confused as to why both curves are given to me by the manufacturer? Cuz if I use the mass flow curves in Aspen the discharge pressure would become independent of inlet condition, but if I use the actual flows, the discharge pressure would change with a change in inlet condition (assuming all else the same; aka same flow and speed for all cases).



Thanks a lot for addressing the point only.

 

[georgeverghese]

"Cuz if I use the mass flow curves in Aspen the discharge pressure would become independent of inlet condition"

In this case, the discharge pressure at any mass flow will still be influenced by suction pressure.

There is no difference whichever way you input the data into the simulator, but using actual volume flow is safer, it avoids potential errors in calculations which can occur when you convert Sm3/hr to kg/hr.

 

[AliZirgach]

Thanks for your reply. Yes, it would be still dependent on the inlet pressure but independent of inlet temperature. In case of an actual flow though, with the actual flow being dependent on both pressure and temperature, the discharge pressure would become dependent on both inlet pressure and temperature. So I still don't understand exactly why I have both curves.
Good point on the actual flow being the safer way as well. The issue I have is that it seems when I use the mass flow curves my results more accurately match the data I have from operations.

 

[georgeverghese]

It really doesnt matter which curve you use from the compressor manufacturer. Frankly, I dont see why they need to give you two sets or curves, when they both say essentially the same thing. On the other hand, the use of pressure ratio is prone to misuse - it would be better to have polytropic head on the y axis - that way, it is more or less independent of feed gas mol wt.

 

[rotw]

- Discharge conditions are not dependent upon which flow unit you are using, they are dependant upon [inlet pressure, inlet temperature and gas composition] (and to be more rigorous, the equation of state used as this influences compressiblity and isentropic exponent calculations).
- It happens that a vendor provides curves in both units of actual volume flow and massflow. It is not systematic but it is not odd neither.
Sometimes it is on request of users to have both set of curves. If you look on API compressor data sheet massflow and volume flows are both specified as value at inlet. Which one you would pick as far as machinery design is concerned is obviously massflow (typically kg/h or Nm3/h in S.I.) ; the reasom is simple: actual volume flow is impacted by which equation of state is used. Both the user and vendor can be on same page and choose same equation of state, but even when the same equation of state is picked up by both vendor and user / specifier), it can still lead to different outcome (for example because binary interaction coefficients are not alined). I know you refered to ideal gas.

But why should all this be a problem? just ignore or discard the set of curves that are not relevant and proceed with your work.

 

[PaoloPemi]

I agree with georgeverghese and rotw,
one can convert (via actual density or volume) from mass flow to actual flow, see also my answer to your post here

https://www.cheresources.com/invision/topic/28817-compressor-simulation-performance-curves/

basically you should obtain the same results

 

[AliZirgach]

Thank you everyone.