RMS P-mass no-convergence for low pressure ration,low rotor speed turbine

Ton · November 5, 2010, 03:21

Hi,
i am simulating a flow about low pressure ration,low rotor speed turbine with CFX. The settings are as follows:
1)Boundary condition: pressure condition
2)RS interface: Mixing Plane Method/Frozen Rotor Method
3)Material:Ideal Gas(Comprssible)
4)Turbulence Model:SST

i have done many tests about the mesh density and RS setting, but the results are the same: the RMS p-mass is divergent, but the other monitor paramters are convergent. i find that the RMS p-mass non-convergence occurs in the rotor domain.

The reason that RMS p-mass is not convergent should be caused by the unstable flow in rotor domain. But now i donn't want to do the transient analysis. i want to find a method to stabilize the flow. Who can help me?Thanks!

Ton

ghorrocks · November 5, 2010, 05:11

Use small physical timescales to start it off. Once you have a few iterations done ramp the timescale up. When the timescale is large enough the theory is that it can "average out" the fluctuations to a steady state result. It does not always work.

There are lots more hints for convergence in the documentation and here: http://www.cfd-online.com/Wiki/Ansys...gence_criteria

Ton · November 5, 2010, 18:40

[ Thanks ghorrocks,

i have done the tests about changing timescales.The results are the same. if no results is used as initial condition, the RMS p-mass smoothly drops to below 1.0E-4 at before 150 steps, but the RMS p-mass gradually diverges as the calculation is going on. if previous results is used as initial condition,the divergence occurs earlier.

There are serious flow separation in rotor . The boundary conditions is also far away design point. The calcuation is indeed difficult.

Ever i used the incompressible gas, the convergent process was good. But the mach number of most domain zoon exceeds 0.3. So i must choose the compressible gas.

whether are there some expert parameters to stabilize the unstable flow.

ghorrocks · November 6, 2010, 06:21

Forget the expert parameters. If the p-mass always diverges no matter what you do then you have a problem in your simulation setup.

Does it converge using a mass flow inlet and a pressure outlet? Just guess a representative mass flow. Pressure boundaries for inlet and outlets can be numerically unstable.

November 5, 2010, 03:21	RMS P-mass no-convergence for low pressure ration,low rotor speed turbine	#1
Ton New Member Join Date: Apr 2009 Posts: 7 Rep Power: 16	Hi, i am simulating a flow about low pressure ration,low rotor speed turbine with CFX. The settings are as follows: 1)Boundary condition: pressure condition 2)RS interface: Mixing Plane Method/Frozen Rotor Method 3)Material:Ideal Gas(Comprssible) 4)Turbulence Model:SST i have done many tests about the mesh density and RS setting, but the results are the same: the RMS p-mass is divergent, but the other monitor paramters are convergent. i find that the RMS p-mass non-convergence occurs in the rotor domain. The reason that RMS p-mass is not convergent should be caused by the unstable flow in rotor domain. But now i donn't want to do the transient analysis. i want to find a method to stabilize the flow. Who can help me?Thanks! Ton

November 6, 2010, 06:21		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,697 Rep Power: 143	Forget the expert parameters. If the p-mass always diverges no matter what you do then you have a problem in your simulation setup. Does it converge using a mass flow inlet and a pressure outlet? Just guess a representative mass flow. Pressure boundaries for inlet and outlets can be numerically unstable. Ankit.Dave likes this.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
static pressure from mass flow rate	mukhtarjussa	FLUENT	0	April 12, 2010 10:39
pressure input, mass flow output	Venugopal Bachu	Main CFD Forum	0	January 11, 2007 10:01
pressure boundary VS. mass flow boundary	dalton	Main CFD Forum	12	April 3, 2005 23:52
static pressure from mass flow rate	SAM	Main CFD Forum	4	October 27, 2004 13:15
what the result is negatif pressure at inlet	chong chee nan	FLUENT	0	December 29, 2001 05:13

November 5, 2010, 05:11		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,697 Rep Power: 143	Use small physical timescales to start it off. Once you have a few iterations done ramp the timescale up. When the timescale is large enough the theory is that it can "average out" the fluctuations to a steady state result. It does not always work. There are lots more hints for convergence in the documentation and here: http://www.cfd-online.com/Wiki/Ansys...gence_criteria

November 5, 2010, 18:40		#3
Ton New Member Join Date: Apr 2009 Posts: 7 Rep Power: 16	[ Thanks ghorrocks, i have done the tests about changing timescales.The results are the same. if no results is used as initial condition, the RMS p-mass smoothly drops to below 1.0E-4 at before 150 steps, but the RMS p-mass gradually diverges as the calculation is going on. if previous results is used as initial condition,the divergence occurs earlier. There are serious flow separation in rotor . The boundary conditions is also far away design point. The calcuation is indeed difficult. Ever i used the incompressible gas, the convergent process was good. But the mach number of most domain zoon exceeds 0.3. So i must choose the compressible gas. whether are there some expert parameters to stabilize the unstable flow.