Contra rotating rotor

csmistry · November 13, 2010, 08:43

Sir, i would like to do analysis of contra rotating rotors in Turbo mode... out of two options (Stage and Frozen rotor ) which option i use and why?
while opting option of stage i found difficulty in convergence like for so many iterations(400) it will not converge to 10E-5.
same case for frozen rotor also... reason behind it?

ghorrocks · November 13, 2010, 23:42

The difference between stage and frozen rotor is described in the documentation. Have a look here for tips on convergence issues: http://www.cfd-online.com/Wiki/Ansys...gence_criteria

Attesz · November 14, 2010, 03:50

Probably you have to use stage, because if there are big changes in the flow structure on the interfaces, this is the recommended interface type. Rotating frames are always hard to converge, mostly if you use pressure boundary conditions. Probably it is not related to the mesh. If you post more details (geometry, boundary conditions, convergence history, out file), we can help you more. Post mesh pic from the interface also.

csmistry · November 14, 2010, 05:41

i herewith sending you convergence plot

ghorrocks · November 14, 2010, 05:50

I will say it again - this FAQ covers the general issues here: http://www.cfd-online.com/Wiki/Ansys...gence_criteria

If you post more specific details we might be able to narrow it down a bit.

csmistry · November 14, 2010, 07:55

i herewith attached out file for your consideration

ghorrocks · November 14, 2010, 17:55

Why are you using the alternate rotation model?

I note you have adiabatic walls all round. Obviously you will need to update that when the flow is working.

But other than that it seems like a classic case, as discussed in this FAQ: http://www.cfd-online.com/Wiki/Ansys...gence_criteria

This is the third time I have quoted this FAQ on this thread, I hope you eventually read it.

Attesz · November 15, 2010, 05:09

Have you checked if you use periodic, you have to divide the outlet mass flow?
Use 1 bar reference pressure, and 0 to relative and so on.
Instead of alternate model put the rotor domains into rotating frame of reference, and take care of the rotation direction.
What is R1 to R1 internal interface??
Between the rotors use stage.
Set time scale to aggressive to converge faster. I it doesn't diverge, ramp up the factor (in the solver, during the run) to 5, 10 or maybe higher.
Are you sure that 1.2million elements will be enough to capture a difficult physics like this? (it can be, but it depends on your problem)

November 13, 2010, 08:43	Contra rotating rotor	#1
csmistry Member mistry Join Date: Apr 2009 Location: India Posts: 69 Rep Power: 17	Sir, i would like to do analysis of contra rotating rotors in Turbo mode... out of two options (Stage and Frozen rotor ) which option i use and why? while opting option of stage i found difficulty in convergence like for so many iterations(400) it will not converge to 10E-5. same case for frozen rotor also... reason behind it?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
CFD analysis on wind turbine rotor	Ken (Wind Turbine CFD Super Rookie)	Main CFD Forum	45	February 9, 2016 15:07
Rotating Impeller	Naith	FloEFD, FloWorks & FloTHERM	22	November 5, 2012 09:53
question about governing equation in CFX using rotating/non rotating reference frame	rystokes	CFX	0	January 12, 2010 07:14
Transient Rotor Stator Model	Farid	CFX	2	May 23, 2008 22:37
Adiabatic and Rotating wall (Convection problem)	ParodDav	CFX	5	April 29, 2007 20:13

November 13, 2010, 23:42		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	The difference between stage and frozen rotor is described in the documentation. Have a look here for tips on convergence issues: http://www.cfd-online.com/Wiki/Ansys...gence_criteria

November 14, 2010, 03:50		#3
Attesz Senior Member Attesz Join Date: Mar 2009 Location: Munich Posts: 368 Rep Power: 17	Probably you have to use stage, because if there are big changes in the flow structure on the interfaces, this is the recommended interface type. Rotating frames are always hard to converge, mostly if you use pressure boundary conditions. Probably it is not related to the mesh. If you post more details (geometry, boundary conditions, convergence history, out file), we can help you more. Post mesh pic from the interface also.

November 14, 2010, 05:50		#5
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	I will say it again - this FAQ covers the general issues here: http://www.cfd-online.com/Wiki/Ansys...gence_criteria If you post more specific details we might be able to narrow it down a bit.

November 14, 2010, 17:55		#7
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	Why are you using the alternate rotation model? I note you have adiabatic walls all round. Obviously you will need to update that when the flow is working. But other than that it seems like a classic case, as discussed in this FAQ: http://www.cfd-online.com/Wiki/Ansys...gence_criteria This is the third time I have quoted this FAQ on this thread, I hope you eventually read it.

November 15, 2010, 05:09		#8
Attesz Senior Member Attesz Join Date: Mar 2009 Location: Munich Posts: 368 Rep Power: 17	Have you checked if you use periodic, you have to divide the outlet mass flow? Use 1 bar reference pressure, and 0 to relative and so on. Instead of alternate model put the rotor domains into rotating frame of reference, and take care of the rotation direction. What is R1 to R1 internal interface?? Between the rotors use stage. Set time scale to aggressive to converge faster. I it doesn't diverge, ramp up the factor (in the solver, during the run) to 5, 10 or maybe higher. Are you sure that 1.2million elements will be enough to capture a difficult physics like this? (it can be, but it depends on your problem)