rhoSimpleFoam MRF cyclicACMI centrifugal compressor

dcook · June 16, 2017, 11:56

Hi,

This is my first post...I have been experimenting with OpenFOAM for a few years and have been able to find a wealth of information via Google, these forums, research papers, tutorials, etc to solve just about any problem I have run into...until now.

Using OpenFOAM-4.1

Geometry Summary
Centrifugal Fan/Compressor with impeller and spiral regions
Regions are meshed separately and coupled through cyclicACMI

Setup
Impeller region is MRF at -309 rad/s, spiral region is stationary
Impeller inlet is totalPressure at 101325
Spiral outlet; have tried several with same results (fixedValue, inletFlowRateVelocity, coded BC to act as restriction, etc).
Turbulence: k-omegaSST
All walls employ wallFunctions
Interface patches are cyclicACMI
Thermophysical Properties hePsiThermo, pureMixture, perfectGas, const transport, hConst, sensibleEnthalpy, standard air (nMoles 1 | molWeight 28.9 | Cp 1007 | Hf 0 | mu 1.8e-05 | Pr 0.7

Symptoms
I have been able to run simpleFoam and rhoSimpleFoam successfully for various boundary conditions but the results are non-physical.
For a typical system (no work), p/rho^gamma is constant for multiple states. I have checked this on the simulation and this seems to be true and this is confirmed by using inlet and outlet patch run-time post processing averages of p,U,T,rho,h and finding the the enthalpy is approximately equal at the inlet and outlet. Problem is that when work is done on the fluid (as would happen in a compressor), the total enthalpy at the outlet should be higher than the total enthalpy at the inlet. In all cases that I have tried, the outlet temperature is slightly lower than the inlet temperature but this is likely due to a slightly higher velocity on the outlet.

I have a higher pressure at the outlet than I do at the inlet (~97 kPa at inlet, ~110 kPa at outlet)...using the relation that T1/T0 = (P1/P0)^((gamma - 1)/gamma), we should expect at least isentropic compression temperature of the gas to about 320K but this is not observed.

I have tested this with OpenFOAM-Extend-4 version using the steadyCompressibleMRFFoam solver and the enthalpy/temperature behave as expected. I have tried porting the rhoCompressibleMRFFoam into an OpenFOAM-4.1 compatible version (changing as little of the extend code as possible) but the issue returns. This makes me think that possibly something in the extend version of MRF or psiThermo is different that is causing this non-physical behavior but I am just not as adept at scouring the code to determine the differences.

Can anyone provide some insight here?

SchuKingR · August 25, 2017, 07:21

Hi dcook!

Did you found a solution for your problem? Because I have the same problem on a turbine. The temperature at the outlet is nearly the same as at the inlet.
In my case the temperature need to be lower at the outlet, because of the energy transfer from the fluid to the turbine, to let the turbine spin.

Sorry for my bad english.

Thanks in advance,
Toni

dcook · August 26, 2017, 17:32

Short answer is no. The extend version uses a modified solver that seems to utilize PIMPLE as a main outer loop for its *steady* solver. The closest I can come to an explanation for this is that they also use a dpdt term in EEQN.H that you don't find in rhoSimpleFoam (since derivatives of time are non-sensical in SIMPLE. Perhaps this has something to do with the change in enthalpy.

Though, I still don't completely understand the energy imbalance... wish I did.

Dan

anthony761 · February 5, 2021, 11:13

Hello dcook, hello foamers,

Did you succeed to perform any compressible flow simulation for turbomachinery application? What is you feedback? I have also encoutered some problems you were talking about. This is my case:

OpenFOAM-v2006
Geometry: Centrifugal compressor, only one blade passage with cyclic/cyclicAMI (with diffuser vanedless and without volute)
BCs: flowRateInletVelocity with fixedValue temperature inlet and fixedValue pressure outlet
MRF model
Turbulence: kOmegaSST
Thermophysical: hePsiThermo, pureMixture, transport const, hConst, perfectGas, sensibleInternalEnergy

Tried Solvers: rhoSimpleFoam, rhoCentralFoam, sonicFoam

Results are strange, the pressure increased as expected at the exit but not the temperature which seems remain almost constant. Do you have any idea of the problem?

June 16, 2017, 11:56	rhoSimpleFoam MRF cyclicACMI centrifugal compressor	#1
dcook New Member D. Cook Join Date: Apr 2013 Posts: 4 Rep Power: 13	Hi, This is my first post...I have been experimenting with OpenFOAM for a few years and have been able to find a wealth of information via Google, these forums, research papers, tutorials, etc to solve just about any problem I have run into...until now. Using OpenFOAM-4.1 *Geometry Summary* Centrifugal Fan/Compressor with impeller and spiral regions Regions are meshed separately and coupled through cyclicACMI *Setup* Impeller region is MRF at -309 rad/s, spiral region is stationary Impeller inlet is totalPressure at 101325 Spiral outlet; have tried several with same results (fixedValue, inletFlowRateVelocity, coded BC to act as restriction, etc). Turbulence: k-omegaSST All walls employ wallFunctions Interface patches are cyclicACMI Thermophysical Properties hePsiThermo, pureMixture, perfectGas, const transport, hConst, sensibleEnthalpy, standard air (nMoles 1 \| molWeight 28.9 \| Cp 1007 \| Hf 0 \| mu 1.8e-05 \| Pr 0.7 *Symptoms* I have been able to run simpleFoam and rhoSimpleFoam successfully for various boundary conditions but the results are non-physical. For a typical system (no work), p/rho^gamma is constant for multiple states. I have checked this on the simulation and this seems to be true and this is confirmed by using inlet and outlet patch run-time post processing averages of p,U,T,rho,h and finding the the enthalpy is approximately equal at the inlet and outlet. Problem is that when work is done on the fluid (as would happen in a compressor), the total enthalpy at the outlet should be higher than the total enthalpy at the inlet. In all cases that I have tried, the outlet temperature is slightly lower than the inlet temperature but this is likely due to a slightly higher velocity on the outlet. I have a higher pressure at the outlet than I do at the inlet (~97 kPa at inlet, ~110 kPa at outlet)...using the relation that T1/T0 = (P1/P0)^((gamma - 1)/gamma), we should expect at least isentropic compression temperature of the gas to about 320K but this is not observed. I have tested this with OpenFOAM-Extend-4 version using the steadyCompressibleMRFFoam solver and the enthalpy/temperature behave as expected. I have tried porting the rhoCompressibleMRFFoam into an OpenFOAM-4.1 compatible version (changing as little of the extend code as possible) but the issue returns. This makes me think that possibly something in the extend version of MRF or psiThermo is different that is causing this non-physical behavior but I am just not as adept at scouring the code to determine the differences. Can anyone provide some insight here? artymk4 likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Centrifugal Compressor Efficiency	albertotato	CFX	3	May 2, 2020 03:54
Temperature for Centrifugal compressor	nnkd	FLUENT	0	February 11, 2016 02:55
Centrifugal compressor mass flow error	Attesz	CFX	18	May 27, 2012 11:17
how to solve the diverage of high speed centrifugal compressor, CFD code is STAR CCM	layth	STAR-CCM+	3	May 21, 2012 06:48
[ICEM] Meshing a centrifugal compressor impeller	Mitpostdoc	ANSYS Meshing & Geometry	8	February 25, 2011 11:51

August 25, 2017, 07:21		#2
SchuKingR New Member Join Date: Aug 2017 Posts: 1 Rep Power: 0	Hi dcook! Did you found a solution for your problem? Because I have the same problem on a turbine. The temperature at the outlet is nearly the same as at the inlet. In my case the temperature need to be lower at the outlet, because of the energy transfer from the fluid to the turbine, to let the turbine spin. Sorry for my bad english. Thanks in advance, Toni

August 26, 2017, 17:32		#3
dcook New Member D. Cook Join Date: Apr 2013 Posts: 4 Rep Power: 13	Short answer is no. The extend version uses a modified solver that seems to utilize PIMPLE as a main outer loop for its steady solver. The closest I can come to an explanation for this is that they also use a dpdt term in EEQN.H that you don't find in rhoSimpleFoam (since derivatives of time are non-sensical in SIMPLE. Perhaps this has something to do with the change in enthalpy. Though, I still don't completely understand the energy imbalance... wish I did. Dan

February 5, 2021, 11:13		#4
anthony761 New Member Join Date: Oct 2019 Posts: 13 Rep Power: 7	Hello dcook, hello foamers, Did you succeed to perform any compressible flow simulation for turbomachinery application? What is you feedback? I have also encoutered some problems you were talking about. This is my case: OpenFOAM-v2006 Geometry: Centrifugal compressor, only one blade passage with cyclic/cyclicAMI (with diffuser vanedless and without volute) BCs: flowRateInletVelocity with fixedValue temperature inlet and fixedValue pressure outlet MRF model Turbulence: kOmegaSST Thermophysical: hePsiThermo, pureMixture, transport const, hConst, perfectGas, sensibleInternalEnergy Tried Solvers: rhoSimpleFoam, rhoCentralFoam, sonicFoam Results are strange, the pressure increased as expected at the exit but not the temperature which seems remain almost constant. Do you have any idea of the problem?