[Other] Diverging after good initialization when conducting CFD of Diffuser using ideal gas

waseeqsiddiqui · October 24, 2018, 09:06

ANSYS Fluent.
I have Total Temperature, Total Pressure, Static Pressure, and Mach at inlet of duct. I need everything mentioned above for outlet, high priority of need is total pressure at outlet. So I used Pressure inlet (Total Temperature, Pressure and Static Pressure and direction normal to surface) and mass flow rate outlet (I have it same as inlet). Also Operating Condition set to Zero. When density is set constant, the solution works good but since it's Mach 0.6, ideal gas is required. So I used Ideal gas and it gives floating point error at first iteration after an incredible initialization to e-08!
Please guide.

TomD · October 24, 2018, 09:28

First of all, this post should probably be moved to the Fluent sub-forum
can you post pictures of your mesh, geometry and BC specifications?

Have you tried using your constant density solution as the initialization for the ideal gas solution? meaning, running few iterations until you converge with constant density, than switching to ideal gas model.
Also, try incompressible ideal gas, that means that the density changes only due to temperature changes and not pressure changes. this is incorrect of-course but it will show you if your problem originates from the temperatures within your domain. do you have heat transfer other than friction heating?
Also, try to look at your inlet and outlet conditions and make sure you dont over-constraint the solution

waseeqsiddiqui · October 24, 2018, 09:31

Sorry for not posting in Fluent sub forum.
Very well noted. Thank you for the suggestion.
I'll try both methods.
If doesn't work, will post specs.

waseeqsiddiqui · October 24, 2018, 09:32

No I dont have any heat transfer.
Actually only mass flow rate is common for both inlet and outlet.

waseeqsiddiqui · October 24, 2018, 09:52

Mesh and Geometry are of a S shaped Diffuser
At Pressure Inlet, I placed
Total Pressure at the respective field.
Static Pressure at the initial Guage Pressure
Total Temperature at the respective field.
Direction Vector as normal to boundary.

At mass flow rate outlet, only the respective value of mass flow rate.
Operating was set to Zero.

October 24, 2018, 09:06	Diverging after good initialization when conducting CFD of Diffuser using ideal gas	#1
waseeqsiddiqui Member MWRS Join Date: Apr 2018 Location: Islamabad Posts: 99 Rep Power: 8	ANSYS Fluent. I have Total Temperature, Total Pressure, Static Pressure, and Mach at inlet of duct. I need everything mentioned above for outlet, high priority of need is total pressure at outlet. So I used Pressure inlet (Total Temperature, Pressure and Static Pressure and direction normal to surface) and mass flow rate outlet (I have it same as inlet). Also Operating Condition set to Zero. When density is set constant, the solution works good but since it's Mach 0.6, ideal gas is required. So I used Ideal gas and it gives floating point error at first iteration after an incredible initialization to e-08! Please guide.

October 24, 2018, 09:28		#2
TomD New Member Tom David Join Date: May 2017 Posts: 17 Rep Power: 9	First of all, this post should probably be moved to the Fluent sub-forum can you post pictures of your mesh, geometry and BC specifications? Have you tried using your constant density solution as the initialization for the ideal gas solution? meaning, running few iterations until you converge with constant density, than switching to ideal gas model. Also, try incompressible ideal gas, that means that the density changes only due to temperature changes and not pressure changes. this is incorrect of-course but it will show you if your problem originates from the temperatures within your domain. do you have heat transfer other than friction heating? Also, try to look at your inlet and outlet conditions and make sure you dont over-constraint the solution waseeqsiddiqui likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
CFD Salary	CFD	Main CFD Forum	17	January 3, 2017 18:09
Ideal vs constant gas model	chris12345	STAR-CCM+	1	January 2, 2013 07:49
Is it worth it?	Jason Bardis	Main CFD Forum	47	July 27, 2011 05:52
Which is better to develop in-house CFD code or to buy a available CFD package.	Tareq Al-shaalan	Main CFD Forum	10	June 13, 1999 00:27
Good CFD Books, FAQ	Jonas Larsson	Main CFD Forum	11	April 16, 1999 17:43

October 24, 2018, 09:31		#3
waseeqsiddiqui Member MWRS Join Date: Apr 2018 Location: Islamabad Posts: 99 Rep Power: 8	Sorry for not posting in Fluent sub forum. Very well noted. Thank you for the suggestion. I'll try both methods. If doesn't work, will post specs.

October 24, 2018, 09:32		#4
waseeqsiddiqui Member MWRS Join Date: Apr 2018 Location: Islamabad Posts: 99 Rep Power: 8	No I dont have any heat transfer. Actually only mass flow rate is common for both inlet and outlet.

October 24, 2018, 09:52		#5
waseeqsiddiqui Member MWRS Join Date: Apr 2018 Location: Islamabad Posts: 99 Rep Power: 8	Mesh and Geometry are of a S shaped Diffuser At Pressure Inlet, I placed Total Pressure at the respective field. Static Pressure at the initial Guage Pressure Total Temperature at the respective field. Direction Vector as normal to boundary. At mass flow rate outlet, only the respective value of mass flow rate. Operating was set to Zero.