Not being able to get results

Jorge Valdivia · May 15, 2020, 23:28

Hi,
I am an undergraduate student, I'm trying to simulate the flow field on a model I made of an aeronatuical combustor. The thing is that I am not considering combustion, just the hot flow and the cold flow around the material (image). My model is compressible and turbulent k-omega (according o a tutorial I saw). I have 3 inlets 2 of cold flow (580 K) of 0.79 and 0.82kg/s, and a hot one (2300K) of 0.414kg/s. The fluid is air. I don't know why fluent is not responding when i turn on the ideal gas feature in material properties. It worked well when the density was fixed. I have refined my mesh to 6 million cells and now my system isn't even able to initialize with hybrid initalization. What could be the problem? Is ANSYS not reading parameters with such high. I will post the images.

[IMG] https://imgur.com/a/N0UDQ4H [/IMG]

Edit: the pressure on the inlets is 9.5 atm

vinerm · May 16, 2020, 07:14

What do you mean by Fluent is not responding?

Jorge Valdivia · May 16, 2020, 10:19

Hi Vinerm,

I mean that the iterations are taking to much time to calculate the parameters. Much of the time this problems have to be with pressure limitations and turbulent viscosity ratio limitations, and then eventually fluent detects a floating point exception when the calculations are advanced (in coarsed meshes). I have tried different sizes of elements.

Thank you for replying.
Jorge

vinerm · May 16, 2020, 11:04

With ideal gas, you should set up operating pressure as 0 and use absolute values of pressure at inlet and outlet. Try with a coarser mesh. If you are using cooling holes in the liner, then ensure that there are at least 18 nodes along the circular edge. Furthermore, $k-\omega$ is not meant for such applications, but for wall bounded flows. Though you are not modeling combustion still the mixing of cold gas with hot gas is better predicted with $k-\varepsilon$ . Or use SST $k-\omega$ .

Jorge Valdivia · May 16, 2020, 11:16

Thank you, I will try with those turbulence models also I am reading the faq about fluent, very helpful. I am a new member so maybe my thread was not aligned with the community standards. I will post a more specific new thread when I encounter a more specific problem. I feel really optimistic about my simulation now that you gave me that advice.

Regards,
Jorge

May 15, 2020, 23:28	Not being able to get results	#1
Jorge Valdivia New Member Jorge Valdivia Join Date: May 2020 Posts: 3 Rep Power: 6	Hi, I am an undergraduate student, I'm trying to simulate the flow field on a model I made of an aeronatuical combustor. The thing is that I am not considering combustion, just the hot flow and the cold flow around the material (image). My model is compressible and turbulent k-omega (according o a tutorial I saw). I have 3 inlets 2 of cold flow (580 K) of 0.79 and 0.82kg/s, and a hot one (2300K) of 0.414kg/s. The fluid is air. I don't know why fluent is not responding when i turn on the ideal gas feature in material properties. It worked well when the density was fixed. I have refined my mesh to 6 million cells and now my system isn't even able to initialize with hybrid initalization. What could be the problem? Is ANSYS not reading parameters with such high. I will post the images. [IMG] https://imgur.com/a/N0UDQ4H [/IMG] Edit: the pressure on the inlets is 9.5 atm Last edited by Jorge Valdivia; May 16, 2020 at 01:22.

May 16, 2020, 07:14	Fluent's Response	#2
vinerm Senior Member Vinerm Join Date: Jun 2009 Location: Nederland Posts: 2,946 Blog Entries: 1 Rep Power: 36	What do you mean by Fluent is not responding? __________________ Regards, Vinerm PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.

May 16, 2020, 11:04	Ideal Gas	#4
vinerm Senior Member Vinerm Join Date: Jun 2009 Location: Nederland Posts: 2,946 Blog Entries: 1 Rep Power: 36	With ideal gas, you should set up operating pressure as 0 and use absolute values of pressure at inlet and outlet. Try with a coarser mesh. If you are using cooling holes in the liner, then ensure that there are at least 18 nodes along the circular edge. Furthermore, $k-\omega$ is not meant for such applications, but for wall bounded flows. Though you are not modeling combustion still the mixing of cold gas with hot gas is better predicted with $k-\varepsilon$ . Or use SST $k-\omega$ . __________________ Regards, Vinerm PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
OpenFOAM - Validation of Results	Ahmed	OpenFOAM Running, Solving & CFD	10	May 13, 2018 19:28
lid driven cavity varying results	yasmil	OpenFOAM Running, Solving & CFD	2	October 6, 2016 22:42
interFoam simulation yields inconsistent results for alpha1 surface	Ralinus	OpenFOAM Running, Solving & CFD	8	January 13, 2014 09:54
CFD results not close to experimental results	cider	STAR-CCM+	0	July 8, 2013 08:53
Different Results from Fluent 5.5 and Fluent 6.0	Rajeev Kumar Singh	FLUENT	6	December 19, 2010 12:33

May 16, 2020, 10:19		#3
Jorge Valdivia New Member Jorge Valdivia Join Date: May 2020 Posts: 3 Rep Power: 6	Hi Vinerm, I mean that the iterations are taking to much time to calculate the parameters. Much of the time this problems have to be with pressure limitations and turbulent viscosity ratio limitations, and then eventually fluent detects a floating point exception when the calculations are advanced (in coarsed meshes). I have tried different sizes of elements. Thank you for replying. Jorge

May 16, 2020, 11:16		#5
Jorge Valdivia New Member Jorge Valdivia Join Date: May 2020 Posts: 3 Rep Power: 6	Thank you, I will try with those turbulence models also I am reading the faq about fluent, very helpful. I am a new member so maybe my thread was not aligned with the community standards. I will post a more specific new thread when I encounter a more specific problem. I feel really optimistic about my simulation now that you gave me that advice. Regards, Jorge