I am doing a project on air flow around high speed trains (250km/h) and the simulation has completed successfully. However, the results do not seem correct. For a vehicle at such high speeds the flow should be quite turbulent in the wake region but this seems to be non -existent. The flow patterns from my simulations do not show a stagnation point, the pressure remains constant over the body and the air seems to adhere to the surface (i.e. no flow separation). At such speeds, it would be expected that the air creates a high pressure region at the nose where the velocity comes to rest and large flow separation is expected (despite the streamlined shape) Can someone please help me to identify where I may have made an error or explain why this is the case? Thank you!
Baseline Model | SimScale Workbench
I will try it with the compressible solver thank you but the speed I am working at should under the Mach 0.3.
250km/h = 69.44 m/s (which lies below the recommended 100m/s).
Hi @mn21nt,
Yes, actually you’re right, I’ve made a confusion with the units - and I apologize for that. The Incompressible solver in this case should work fine, I’ll delete that first answer
That said, I’m running some tests on your simulation and get back here in a few with my findings 
Best,
Igor
Please let me know when you find anything of note
Thank you!
Hi @mn21nt, sorry for the delay,
I was running some tests but I’ve noticed something about the screenshot you posted:
The Iterations tab is set to 0, so that you’re seeing the very first step of the simulation. When you toggle it to the last step, things get better:
However, I’d still advise you to make a change to your model in order to make things converge better which is to increase the distance from the train to the inlet like this:
If I have any further inputs in the next few hours, I’ll post them here
Best,
Igor
I have changed the iterations to 1000 now and the results do look much better, thank you! I expected to observe a larger wake region considering the size and speed of the model but the results display the right characteristics. I am currently running the simulation with the recommended larger upstream region to see what difference that would make to the results.