[ayothicfd]

Dear all,
I am doing CFX simulation of fan performance curve prediction. In the simulation, Inlet Guide Vane (IGV) has been supposed to tilt in order to get different curve by changing the mass flow rate. Actually, I have generated IGV full open condition. While doing 30 degree inclination of IGV, I have taken three converged results on corresponding mass flow 37.8,42.6 and 48.8 kg/s. By setting mass flow as 53.5 kg/s, the result is diverging. I have gone through Ansys_FAQ. I used the same mesh for the above mentioned mass flow. I don't know exactly what would be the reason of error either quality of mesh or any other? Actually, I have given the result file (mass flow of 48.8) as initial file. I have attached the out file and snapshot of solver while run stopped.
Snapshot:
http://www.cfd-online.com/Forums/att...1&d=1339923307
Outfile:
http://www.cfd-online.com/Forums/att...1&d=1339923733

Thanks in advance.

[ghorrocks]

This FAQ seems to discuss your exact issue: http://www.cfd-online.com/Wiki/Ansys...do_about_it.3F

Is your flow compressible? Have you checked that your flow has choked?

[ayothicfd]

Hello Glenn,
Thank you for your post and sorry for the reply. My simulation physics is working under Isothermal and Incompressible (Pressure rise won't be much large)flow. I have taken backup file for the same run(before diverging) in which I have verified. Still not getting a reason why the run got diverged.

[ghorrocks]

sorry for the reply

No need to apologise, the forum is here to ask questions and all thoughful questions are welcome.

In that case I would be looking for things like separations, stalling or other flow features which is making convergence harder at high flow rates.

[lffabiani]

Can you send a cross-section of the domain (I cannot open the first attachment)?

I see that you have multiple domains and interfaces. Are the mesh near the interfaces similar?

In CFD-POST, check some variables: Velocity, Pressure, Temperature, Mach Number and check if they are not out of range (for instance T=0 causes overflow). Maybe this indicates a mesh or timestep issue

One other thing to watch is the size of the mesh. When you import the mesh into PRE, check if the settings (mm, m, etc) are correct. Afterall, 50 kg/s = 180.000 kg/h (which is a lot, and it will not pass on a small geometry).

Best Regards

[ayothicfd]

Hello Fabiani,
Thanks for your reply. Actually my simulation consists of four various parts such as suction duct, IGV(Inlet Guide Vane-regulating unit), Impeller and Casing. Cross section of Impeller is 2540 mm dia and IGV is 1750 mm dia. Hope that below link may help you to view the geometry of my domain.
http://www.cfd-online.com/Forums/att...1&d=1340723262