set temperaturelimits in CFX

ThomasN · October 14, 2013, 07:24

Hello,

sorry nothing found related to this question in the Forum.

I'm simulating an oil/air multiphase flow with high dynamic. The air is an isothermal fluid, the Oil is set to total energy. If the Oil surrounds the air then everything is ok. But later in the simulation I get areas with air and any
oil drops. On these points I've some trouble with temperature hotspots. The Temperature becomes higher and higher until the simulation crashes. Is there a way to set temperature limits for the calculation in CFX? It would be great if the solver would cut the temperature value on these Points to my limits.

best regards

Thomas

ghorrocks · October 14, 2013, 17:16

Rather than artificially putting on temperature limits wouldn't it be better to stop whatever is causing the temperature to go wrong in the first place? Hard limits of variables are convergence nightmares and lead to unphysical results themselves.

Have a look at your assumptions - is isothermal appropriate for air? If it could absorb some heat then would it stop the hotspots? Do you have appropriate inter-phase heat transfer models? Are you properly converged?

ThomasN · October 15, 2013, 02:53

Yes you are right it would be better and i will try it, but i only wanted to ask whether it is possible to do in CFX or not.

The mass and momentum converges quite well, the h-energy is sometimes bouncing but finally it reaches the convergence criteria. But as I said the calculated oil temperature on air dominated areas falls down to 0 K and then it stops. On these points the temperature has no influence to my results but only causes crash.

And yes I set a heat transfer model but I'm not sure whether it is active if the air is set to isothermal?

ghorrocks · October 15, 2013, 05:55

I do not know of an nice way to do it - but you can probably do it by using a source term with a step function - so it does nothing if the temperature is OK, but if it exceeds a threshold it turns on a source term to define a maximum temperature (or add a local cooling effect might be numerically nicer).

Temperature already has a built-in minimum of 0K, so it already is clipped for the minimum.

So it you are hitting both upper and lower temperature limits it sounds like your simulation has fundamental problems which need to be resolved by correcting the interphase heat transfer, convergence, or whatever is causing the heat spots.

ThomasN · October 15, 2013, 06:48

Ok thanks for the help.

October 14, 2013, 07:24	set temperaturelimits in CFX	#1
ThomasN New Member Join Date: Oct 2013 Posts: 21 Rep Power: 12	Hello, sorry nothing found related to this question in the Forum. I'm simulating an oil/air multiphase flow with high dynamic. The air is an isothermal fluid, the Oil is set to total energy. If the Oil surrounds the air then everything is ok. But later in the simulation I get areas with air and any oil drops. On these points I've some trouble with temperature hotspots. The Temperature becomes higher and higher until the simulation crashes. Is there a way to set temperature limits for the calculation in CFX? It would be great if the solver would cut the temperature value on these Points to my limits. best regards Thomas

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
How to set AOA for an Airfoil in CFX?	fender26	CFX	0	September 18, 2013 02:19
[ICEM] Question about the use of ICEM tcl scripting	lnk	ANSYS Meshing & Geometry	9	January 16, 2013 09:24
How to show the transient case?	H.P.LIU	Phoenics	7	July 13, 2010 04:31
OF 1.6 \| Ubuntu 9.10 (64bit) \| GLIBCXX_3.4.11 not found	piprus	OpenFOAM Installation	22	February 25, 2010 13:43
CFX 4.4 installation problem	Pandu Sattvika	CFX	1	December 1, 2001 04:07

October 14, 2013, 17:16		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Rather than artificially putting on temperature limits wouldn't it be better to stop whatever is causing the temperature to go wrong in the first place? Hard limits of variables are convergence nightmares and lead to unphysical results themselves. Have a look at your assumptions - is isothermal appropriate for air? If it could absorb some heat then would it stop the hotspots? Do you have appropriate inter-phase heat transfer models? Are you properly converged?

October 15, 2013, 02:53		#3
ThomasN New Member Join Date: Oct 2013 Posts: 21 Rep Power: 12	Yes you are right it would be better and i will try it, but i only wanted to ask whether it is possible to do in CFX or not. The mass and momentum converges quite well, the h-energy is sometimes bouncing but finally it reaches the convergence criteria. But as I said the calculated oil temperature on air dominated areas falls down to 0 K and then it stops. On these points the temperature has no influence to my results but only causes crash. And yes I set a heat transfer model but I'm not sure whether it is active if the air is set to isothermal?

October 15, 2013, 05:55		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	I do not know of an nice way to do it - but you can probably do it by using a source term with a step function - so it does nothing if the temperature is OK, but if it exceeds a threshold it turns on a source term to define a maximum temperature (or add a local cooling effect might be numerically nicer). Temperature already has a built-in minimum of 0K, so it already is clipped for the minimum. So it you are hitting both upper and lower temperature limits it sounds like your simulation has fundamental problems which need to be resolved by correcting the interphase heat transfer, convergence, or whatever is causing the heat spots.

October 15, 2013, 06:48		#5
ThomasN New Member Join Date: Oct 2013 Posts: 21 Rep Power: 12	Ok thanks for the help.