Convergence problem with structural and thermal analysis

Altair Forum User

Hi, I have to do a structural and thermal analysis on a blade and a disk (in particular of the rotor) of a gas turbine.I created a cylindrical system and then I imposed the cyclic symmetry with the constraints. To simulate the centrifugal force I used RFORCE Card while to simulate the pressure I used PLOAD4. Between the disk and the blade there is a contact surface (blade and disk are able to slide). If I run this non-linear analysis I obtain good results. However when I add also the thermal analysis something goes wrong:
-I created a convection surface for the blade
-I imposed a known temperature for some points and for the environment
-I used MAT4 and MATT4 in order to have some material properties dependent from temperature using TABLEM1 (linear dependency)
-I used card PCONTHT for the contact surface to assure conduction between the disk and the blade 
The solver starts but at the third iterations (no matter which NLPARM's parameters I use) it stops because the error goes over the machine's precision.
I have some doubt about it:
-How can I link the collector of thermal and structural analysis in order to apply the results of thermal analysis to the structural one?
-Which Loadstep do I have to use for thermal analysis using PCONTHT?
-Could it be that I have to change something in the creation of the contact (ADJUST, MORIENT etc..)?
 

Rahul Rajan_21763

Please refer relevant files for PCONTHT and contact based thermal analysis.

Unable to find an attachment - read this blog

Altair Forum User

Hi, thank you for your help. I've tried to follow the instructions that you sent me, but some other problems come out.
I run the thermal analysis (using loadstep 'heat transfer (steady rate)') and the gap are opend, and then I link it with the structural one, but it gives me an error: ERROR # 5024 TEMP (INITIAL) refer to temperature result from heat transfer subcase. I've tried also to change the loadstep of thermal analysis, using 'nonlinear heat transfer', but it doesn't change.
So I haven't solved the problem about linking the two loadsteps and also I don't know what I have to do in order to pass the results of the structural analysis to the thermal one. In particular I'm not sure about the part when I have to specified the subcase and 'Initial' or 'Material' etc..

Rahul Rajan_21763

If possible please share .fem file to check same.

Altair Forum User

I've sent you the .fem file by dropbox. If you discover something please update me here. Thank you in advance.

Rahul Rajan_21763

Thanks for sharing files. I will get back to you.