Impact temperature on an aircraft

Santos-Dumont · August 17, 2011, 13:01

Hi,

I would like to simulate the temperature of the surface of a flying aircraft. The main point of interest is the impact temperature on the leading edges to determine icing datas.
I didn't succeed in modeling it with Fluent. I used different models (Laminar, Spalart Allmart, k-Espilon) with a coupled solver but I get unsatisfying results.
The thing is quite disturbing because after like 20 iterations I get a result that looks pretty good with a higher temperature on the leading edges. But when I iterate more the aircraft surface temperature become constant except for some small hot and randomly located zones.
I activated the viscous heating, and I set the wall and the external temperature at the same value (240 K). The flow speed is 180 m/s and the air density 0.5 kg/m3.

What would you do to solve such a case in terms of model, boundary settings....? There is no tutorial that is about the impact temperature.

Santos-Dumont · August 17, 2011, 19:44

I made some extra tests on a simple cube but I get divergent results after 200 iterations, no matter which solver I chose.
Do somebody have some pieces of advice to share about this viscous heating on a wall ?

Santos-Dumont · August 18, 2011, 13:01

My test on the cube finally succeed with the Coupled k-epsilon solver.
Still, when I try it on the real geometry, I get the message "Time step reduced for xx cells" after like 200 iterations, and the residuals are diverging a few iterations later...

I read that it's due of sudden changes of temperature between iterations and that it could be caused by wrong solver setup...

Can someone help me out ?

August 17, 2011, 13:01	Impact temperature on an aircraft	#1
Santos-Dumont Member Charles Join Date: May 2011 Location: France Posts: 77 Rep Power: 15	Hi, I would like to simulate the temperature of the surface of a flying aircraft. The main point of interest is the impact temperature on the leading edges to determine icing datas. I didn't succeed in modeling it with Fluent. I used different models (Laminar, Spalart Allmart, k-Espilon) with a coupled solver but I get unsatisfying results. The thing is quite disturbing because after like 20 iterations I get a result that looks pretty good with a higher temperature on the leading edges. But when I iterate more the aircraft surface temperature become constant except for some small hot and randomly located zones. I activated the viscous heating, and I set the wall and the external temperature at the same value (240 K). The flow speed is 180 m/s and the air density 0.5 kg/m3. What would you do to solve such a case in terms of model, boundary settings....? There is no tutorial that is about the impact temperature.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Problem with zeroGradient wall BC for temperature - Total temperature loss	cboss	OpenFOAM	12	October 1, 2018 07:36
Calculation of the Governing Equations	Mihail	CFX	7	September 7, 2014 07:27
Water subcooled boiling	Attesz	CFX	7	January 5, 2013 04:32
Bulk temperature Tf is obtained from total or static temperature?	NPU_conanxie	FLUENT	0	March 30, 2011 06:56
chemical reaction - decompostition	La S. Hyuck	CFX	1	May 23, 2001 01:07

August 17, 2011, 19:44		#2
Santos-Dumont Member Charles Join Date: May 2011 Location: France Posts: 77 Rep Power: 15	I made some extra tests on a simple cube but I get divergent results after 200 iterations, no matter which solver I chose. Do somebody have some pieces of advice to share about this viscous heating on a wall ?

August 18, 2011, 13:01		#3
Santos-Dumont Member Charles Join Date: May 2011 Location: France Posts: 77 Rep Power: 15	My test on the cube finally succeed with the Coupled k-epsilon solver. Still, when I try it on the real geometry, I get the message "Time step reduced for xx cells" after like 200 iterations, and the residuals are diverging a few iterations later... I read that it's due of sudden changes of temperature between iterations and that it could be caused by wrong solver setup... Can someone help me out ?