Propeller disk in static condition simulation failed

I got some time to work on the actuator disk so I started, based on help from jousefm: Simulation of an Actuator Disk

Everything worked almost ok until trying to simulate a propeller in static condition. I changed the velocity inlet to a pressure inlet, and now the simulation is not converging. Can anybody help?

link to the project:

Dusan

Hi Dusan,

can you try using a velocity inlet and pressure outlet (with that combo) and not two times the same boundary condition and let me know how things work out?

Cheers,

Jousef

Hi Jousef,

Thanks for replying so swiftly,

I used specifically a pressure inlet, as the velocity is zero (static condition means that there is no velocity except the one created by the propeller) If I do velocity inlet having velocity 0, that means there is no intake of air from that wall and that would be wrong, the air should come from there as it is “sucked” by the propeller. Not sure how to proceed.

Thanks,
Dusan

Hi Dusan,

ah! Got the idea now, thanks! I already have another idea in mind but let me test and think through it to see if that can work.

Edit: Ok so Fan Pressure would require a table upload with two columns, one being the flow rate and the other one the pressure value which I assume is what you do not have. Another “trick” might be to set a very small inlet velocity at the beginning just to keep it a bit moving and the momentum source will accelerate the surroundings (at least that’s my hope --> testing). Last but not least the pressure reference game, would have to check which pressures to set optimally to create a suction effect only from the momentum source without additional influence from outside.

Cheers,

Jousef

I tried this, does not work. I tried setting the inlet velocity 1m/s and the disk velocity at 35m/s. It fails to converge. Anyway I think this is wrong because at the moment you set the inlet velocity, that is fixed and cannot be changed by “suction” so even if it converges, the result will be skewed and not what is happening in real life.

Regards,
Dusan

Hi Dusan,

in general the Pressure Inlet/Outlet should work. I actually could make it work with a velocity of 0.001 and a pressure outlet condition. Don’t know what happened in your case but I will try one run today using pressure inlet/outlet and get back to you asap.

Cheers,

Jousef

Thanks Jousef! Really appreciated!

Dusan

Ok so total pressure inlet and static pressure outlet (both 0 Pa) work, not sure what you have changed in your case (maybe numerics, did not check). However it works for me!

One additional hint. Start the next run(s) with a coarser mesh because my sim took 180 minutes and is probably an overkill for such an application. Reduce the fineness and perform a convergence study by looking at some output results like forces or anything else that’s interesting to you - please note that residuals and convergence of result output go hand in hand. So if the forces converge but the residuals look odd there might be an issue with the numerics but it could also be a flow phenomenon that needs to be taken care of.

Best,

Jousef

Great!

Thanks Jousef!

I’ve copied it, you can delete from your Dashboard

Dusan

This is what I was afraid of: In the project you’ve set the velocity inlet to a very small value, so there is no air coming in, practically the disk is functioning into a closed “box”. Check the saved 2D plot of particle traces.

Anyway, I’m trying now to simulate a pressure inlet, let’s see how it goes.

Dusan

Hi @dusan!

You can just substitute this with the total pressure inlet and static pressure (fixed value 0 Pascal) outlet as previously mentioned to create the suction effect. Worked perfectly fine for me. The small velocity was just for testing.

Best,

Jousef

It does not work. For some reason the simulation run cancels itself. I tried 4 times. What the heck? What I’m doing wrong?

Hi @dusan,

Your BC assignments need to be adjusted. Refer to this project which will help you in what to assign.

Also looking at your geometry. Its just a box with nothing in it (refer to figure below). You might want to check on that.

Regards,
Barry

Hi @dusan!

Maximum runtime is set to 200 seconds which will cancel the run before even starting. Increase the time and run it again.

Best,

Jousef

Hi Jousef,

I increased the runtime, but the simulation is failing to converge. I’ve changed the intake velocity to pressure from your example. This is driving me nuts!

Dusan

Hi @dusan!

Shared a project with you where it worked, absolutely nothing in the numerics has been changed. Maybe you missed a point here, please counter-check.

Cheers!

Jousef