Exit corrected massflow

Raku · August 23, 2016, 11:03

Hello,
I am currently working on my master thesis about centrifugal compressors and I need to calculate speed lines of course.

Therefore I want to use the Exit corrected massflow as the outlet bc which is calculated as following:

where

It is a function in CFX itself by now.

I obtain speedlines which go through the choke limits which was not possible to calculate with Pressure and constant massflow conditions before.

Can you tell me, why is that? How come that I get speedlines, which fall down after the choke-limit (which I found out before by using Pressure-Massflow BCs).

I somehow don't really find out, what the exit corrected massflow work and why it works better than the other options.

Furthermore I wanted to ask if someone can give me a good reference or a link to a paper where the exit corrected massflow is used successfully to convince my supervisor of the advantages.

Thanks in advance!

Opaque · August 23, 2016, 12:35

If you obtained a converged solution using the exit corrected mass flow, you should be able to use the "actual mass flow" and run using the Mass Flow boundary condition. If you use the previous results file, the solution should converge to the same flow field.

The above is proof the solutions are the same. I do not think the implementation details would be explained anywhere (intellectual property ? proprietary? ). Why does it work ? You should try to understand how the actual mass flow is being updated as the solution proceeds. Hint: The value is not the same between the iterations until it has converged, correct ?

Jiricbeng · August 24, 2016, 04:11

It is interesting, my experience with centrifugal compressor was that when I used total pressure inlet and mass flow outlet, it diverged! I was unable to figure out why this happened, as I initialized it from results file. The result file was from mass flow inlet and static pressure outlet, which always works very fine, but when switching the BC to total pressure inlet and mass flow outlet, it diverged even if the previous result file was pretty close to the new boundary conditions.

turbo · August 25, 2016, 11:44

Exit corrected mass BC is only recommended in turbomachinery CFD (but I do not enjoy it at all). Basic concept is that if the physical mass flow BC at exit is considered, along choke lines there are multiple identical mass flows telling we cannot use it there. If static pressure exit BC is considered, near surge there are multiple mass flows with the pressure specified telling we cannot use it either. Exit corrected mass BC can get the same physical mass flow along choke lines because exit corrected flow is changing due to different p and T at exit despite the same physical mass flow. The same principle is true near the surge, i.e., exit corrected flow is changing due to different physical mass flow despite the same pressure.

My preference is to use the exit static pressure along choke lines up to points before the map starts to be flat, and to switch to the physical mass flow BC near surge. However, you can often find the solution near surge/stall from physical mass flow show a certain amount of discontinuity from the solutions with exit static pressure. That is why you cannot trust fully the solution using mass flow BC at exit. Compressible CFD is based on static pressure specified at the exit boundary (while total p and total T at the inlet BC), which is truely consistent with physics behind the subsonic hypobolic type equations.

Saima · October 18, 2016, 17:42

Chk thesis on multidisciplinary design optimization published in Ecole Poly-technique.

Quote:

Originally Posted by Raku

Hello,
I am currently working on my master thesis about centrifugal compressors and I need to calculate speed lines of course.

Therefore I want to use the Exit corrected massflow as the outlet bc which is calculated as following:

where

It is a function in CFX itself by now.

I obtain speedlines which go through the choke limits which was not possible to calculate with Pressure and constant massflow conditions before.

Can you tell me, why is that? How come that I get speedlines, which fall down after the choke-limit (which I found out before by using Pressure-Massflow BCs).

I somehow don't really find out, what the exit corrected massflow work and why it works better than the other options.

Furthermore I wanted to ask if someone can give me a good reference or a link to a paper where the exit corrected massflow is used successfully to convince my supervisor of the advantages.

Thanks in advance!

rafr894 · November 7, 2022, 07:28

"Basic concept is that if the physical mass flow BC at exit is considered, along choke lines there are multiple identical mass flows telling we cannot use it there."

Could you please explain the above bit in a little more detail?

August 23, 2016, 11:03	Exit corrected massflow	#1
Raku New Member Raphael Kuckuck Join Date: Feb 2015 Location: Dortmund, Germany Posts: 2 Rep Power: 0	Hello, I am currently working on my master thesis about centrifugal compressors and I need to calculate speed lines of course. Therefore I want to use the Exit corrected massflow as the outlet bc which is calculated as following: where It is a function in CFX itself by now. I obtain speedlines which go through the choke limits which was not possible to calculate with Pressure and constant massflow conditions before. Can you tell me, why is that? How come that I get speedlines, which fall down after the choke-limit (which I found out before by using Pressure-Massflow BCs). I somehow don't really find out, what the exit corrected massflow work and why it works better than the other options. Furthermore I wanted to ask if someone can give me a good reference or a link to a paper where the exit corrected massflow is used successfully to convince my supervisor of the advantages. Thanks in advance!

August 25, 2016, 11:44		#4
turbo Senior Member Join Date: Jun 2009 Posts: 174 Rep Power: 17	Exit corrected mass BC is only recommended in turbomachinery CFD (but I do not enjoy it at all). Basic concept is that if the physical mass flow BC at exit is considered, along choke lines there are multiple identical mass flows telling we cannot use it there. If static pressure exit BC is considered, near surge there are multiple mass flows with the pressure specified telling we cannot use it either. Exit corrected mass BC can get the same physical mass flow along choke lines because exit corrected flow is changing due to different p and T at exit despite the same physical mass flow. The same principle is true near the surge, i.e., exit corrected flow is changing due to different physical mass flow despite the same pressure. My preference is to use the exit static pressure along choke lines up to points before the map starts to be flat, and to switch to the physical mass flow BC near surge. However, you can often find the solution near surge/stall from physical mass flow show a certain amount of discontinuity from the solutions with exit static pressure. That is why you cannot trust fully the solution using mass flow BC at exit. Compressible CFD is based on static pressure specified at the exit boundary (while total p and total T at the inlet BC), which is truely consistent with physics behind the subsonic hypobolic type equations. xxfholic likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Exit Corrected Mass Flow Rate Mesh Sensitivity Study	s__s__s	CFX	4	July 20, 2016 12:46
request for surfaceScalarField corrected from objectRegistry region0 failed	immortality	OpenFOAM Running, Solving & CFD	7	May 7, 2015 22:00
Why RNGkepsilon model gives floating error	shipman	OpenFOAM Running, Solving & CFD	3	September 7, 2013 09:00
MassFlow Function? Absolute MassFlow? Failure in Caclulation	eRzBeNgEl	CFX	0	May 5, 2011 10:46
Errors running allwmake in OpenFOAM141dev with WM_COMPILE_OPTION%3ddebug	unoder	OpenFOAM Installation	11	January 30, 2008 21:30

August 23, 2016, 12:35		#2
Opaque Senior Member Join Date: Jun 2009 Posts: 1,880 Rep Power: 33	If you obtained a converged solution using the exit corrected mass flow, you should be able to use the "actual mass flow" and run using the Mass Flow boundary condition. If you use the previous results file, the solution should converge to the same flow field. The above is proof the solutions are the same. I do not think the implementation details would be explained anywhere (intellectual property ? proprietary? ). Why does it work ? You should try to understand how the actual mass flow is being updated as the solution proceeds. Hint: The value is not the same between the iterations until it has converged, correct ?

August 24, 2016, 04:11		#3
Jiricbeng Senior Member Jiri Join Date: Mar 2014 Posts: 221 Rep Power: 13	It is interesting, my experience with centrifugal compressor was that when I used total pressure inlet and mass flow outlet, it diverged! I was unable to figure out why this happened, as I initialized it from results file. The result file was from mass flow inlet and static pressure outlet, which always works very fine, but when switching the BC to total pressure inlet and mass flow outlet, it diverged even if the previous result file was pretty close to the new boundary conditions.

November 7, 2022, 07:28		#6
rafr894 New Member KristianRanta Join Date: Oct 2016 Location: Sundsvall Posts: 29 Rep Power: 10	"Basic concept is that if the physical mass flow BC at exit is considered, along choke lines there are multiple identical mass flows telling we cannot use it there." Could you please explain the above bit in a little more detail?