Issue with self-weight during optimization process

Hi,

Can you share the entire .out file?

I do not have my pc with the licence so I can't check if it is the good file. But I think it is.

These are two models in 2D and 3D of a street light floor lamp.

Thank you very much.

Unable to find an attachment - read this blog

Hi,

I see feasible design (solution converged) in both out files. 

Yes I know but the solution is not clean with a lot of checkerboard in the material and isolated areas. Is there anything to do to avoid self-weight issues ? 

It looks like this while there is a force applied at the right edge...

<?xml version="1.0" encoding="UTF-8"?>

Hi,

Do you have any member size controls (minimum membersize/maximum membersize)?

If you have member size constraint, try to add CHECKER card on DOPTPRM 

Hello,

I don't have any member size control. I tried to add one in DESVAR and modify some parameters in DOPTPRM, including CHECKER=1 but there isn't any improvement. Moreover, the result doesn't respect the maximum volume constraint. (I wonder if Optistruct is able to take the self-weight into account for topology optimization). Do I have to put a non-design  area on the right edge to get density where the force is applied ? Here is the file I'm using to modelize the problem.

Thank you very much for your advice.