Problem with HFSS when it is making initial mash on local machine

[haydar]

I do not know the problem but it stucks when it is "making initial mash on local machine" surface recovery!!!!!!!
does anybody has an idea?
one more question, I creat a part, then move several times. how can I see final position of that part?

 

[haydar]

Re: HFSS stucks

hi
I decreased the hight of the air boundary and number of samples. but after 2 hours it gived en error:


[error] Initial mesh, process mesh3d : Unable to recover all the surface triangles.

I will further decrease parameters.

 

[itaifrenkel]

Re: HFSS stucks

Mesh problems depend on the geometry.

Do you have true curved surfaces or cylinders ?

 

[haydar]

Re: HFSS stucks

the problem is that: is my drawing correct?
anyways, i will try different designs and other parameters. If I will be able to learn HFSS, I will be very happy!however, it is so hard to manage.
thanks.

 

[asdfaaa]

Re: HFSS stucks

I remembered that I have helped you modified a similar structure, only difference is the location of the 4 waveports, In you model, no ground plane for coupler and no groud reference plane for your coaxial feed. The most reasonable cause of meshing error is the coupler part, you can check, no material assigned to them. I have tried to assign them to PEC, but I think you may need to consider the metal loss and etc,. the best way is to redraw this part, assign them material prtoperty and boonlean them correctly.
Best Regards,

 

[haydar]

Re: HFSS stucks

I remembered that I have helped you modified a similar structure, only difference is the location of the 4 waveports, In you model, no ground plane for coupler and no groud reference plane for your coaxial feed. The most reasonable cause of meshing error is the coupler part, you can check, no material assigned to them. I have tried to assign them to PEC, but I think you may need to consider the metal loss and etc,. the best way is to redraw this part, assign them material prtoperty and boonlean them correctly.
Best Regards,

hi asdfaa
let me explain why it is "not assigned". I drew coupler as 3D pec, then because of the problem, I detached the face, and assigned it perfect E as a boundary. this is done by suggestion of wlcsp (my friend in this forum).
In addition, wlcsp said that, If one does not assign something to the outer boundary of 3D object, boundary is perfect E by default.
wlcsp made changes and sent me. it worked. but I did the same (at least I think so), but it did not worked properly.

 

[asdfaaa]

Re: HFSS stucks

yes, the model I modified for you, I have correctly set up four waveports, and everything is fine, we can do simulation get the S parameter, am I correct? The model you attached in this post, you didnt correctly set the waveports, no ground plane.
Yes, to assign that planar shape to PEC is the only choice I can choose to do simulation, and some consideration can not be included(metal loss), and also I can not unite this coupler with pin of coaxial feed. I believe port process should be OK, maybe you face the meshing error, it comes from the model, I suggest you redraw that part, carefully set every propety and boonlean, other benifit is that we can do parametric studying about this coupler.
Best Regards,