Error executing AcuSolve
Altair EmployeeHi everybody,
I’m new to AcuSolve. I have to do a CFD analisys (steady state) of a cooling fan of a coaxial pumping unit (the fan of a piston compressor), but i don’t understand why the analisys doesn’t reach the convergence and it finisches after several step (462) with an error (“error occured executing AcuSolve”). I have considered more inflow sections and more outflow sections, and the fluid inside the control volume is divided into two parts, a part around the fan (to assign the reference frame) and the part inside the carter (the control volume is a portion of the carter of the compressor).
I exported the model from hypermesh to acuconsole, and in acuconsole i set the following parameters:
• Analysis type: Steady State;
• Flow equation: Navier Stokes;
• Turbolence equation: Spalart Almaras;
• Fluid: Air (I left the default parameters for the air)
• At the fan i assigned a wall condition
• Reference frame (2850 rpm) assigned to the fan wall and to the fluid modelled around the fan
• Max time step= 1000;
• Convergence tollerance =0.01 (looking at the residual ratio i’ve seen that it doesn’t fall under 0.1)
I want atmospheric pressure in the inflow and outflow section, so in the “Simple boundary condition” i set pressure=0.0 N/m^2 for inflow and outflow(i suppose that 0.0 is the atmospheric pressure) and i’m interested to calculating the mass flux at the outflow.
This is the first time that i do a cfd analysis and it’s problably completely wrong, so i would be really happy if you helped me. I suppose that my mistake was to consider two different fluid in the control volume (two different geometry), and give to one the reference frame, but i’m not sure.
I post the .hm and .acs files, thanks in advance
Best regards
Answers
-
Hi arezzo95,
due to my experiences your case is strong turbulent, especially at the fan outlet, maybe at your inlet also if the back flows occur around the fan.
In this case I would build extensions for both, system inlet and outlet in order to get a fully developed flow on both sides. You can also create two interfaces, you dont have to seperate their names/parts, I mean: one fan inlet and outlet, to examine, for example fan inlet pressure or mass flow, ....
Best regards
0 -
Hi raskolnikov,
Thanks for your answer, therefore i have to extend inflow and outflow section, but in this case i should set a 'slip' condition in the entrance and exit area and not a wall condition, so there isn't friction force and the air is free to slide on this 'artificial' Surface, isnt't it? Anyway i try to do the analysis in this condition and i let you know as soon as possible.
Best regards
0 -
Rather than having several inlet and outlet surfaces, you'll have better luck with a single inlet surface and a single outlet surface. The fluid will then be able to distribute appropriately to the various openings for the passages. It would be something like putting the fan inside a pipe or duct for the simulation.
You may also wish to contact your local Altair support office in Italy.0 -
I forgot to mention the mesh you've included is extremely coarse - only 46 thousand mesh nodes. A model like this will probably require in the range of 1 or 2 million mesh nodes for good accuracy.
0