This is the link to my project on aerodynamic analysis of non-pneumatic tire.
So we were’nt getting accurate results and our graphs weren’t converging in the end for drag and lift .
S o we wanted to know whether our geometry is too complex to be computed using 16 cores or whether we need an academic account /32 cores to get accurate results
S o we are providing high mesh finnesss using hex-dominant parametric algorithm , so we wanted to know whether to uses this we need higher core power because we tried in standard mesh as well but there again we weren’t getting accurate answers
Hi there, thank you for using the forum!
There is a similar project that had the same issue with the Hex dominant parametric mesher, so I am sending you the corresponding solution: Aerodynamic analysis of NON pneumatic tire - "Machine run out of memory" - #4 by tsite
The bounding box resolution of your project is also unnecessarily large:
Which is the Run that used the Standard mesher? It is the most efficient way to resolve any meshing issues
Best regards,
Fillia
the first link uses standard mesher with sizing manual
the second link uses automatic mesher
You can try to toggle on the ‘Potential flow initialization’ for the project with the standard mesher. Maybe you want to set the rotation axis of the wheel as (0,0,1) instead of (0,0,-1) too?
Other than that, I suggest you keep using the standard mesher
Check this post too: Aerodynamic Analysis of Non-pneumatic tires - Pressure divergence
Cheers!
Thank you for your help
Could you mention where exactly is this Potential flow initialization’ and how do I vary these values?
Also i had a doubt whether my centre of rotation of the cad model is off or not?
This helps stabilizing the flow at the beginning of the simulation, and you do not have to add input for this, just activate the option.
The coordinates of the center of rotation can be found in your CAD software
Cheers!