FastFieldSolvers Forum
FastFieldSolvers Forum
Home | Profile | Register | Active Topics | Members | Search | FAQ
Username:
Password:
Save Password
 All Forums
 FastFieldSolvers
 FastHenry2
 Corner Holes
 New Topic  Reply to Topic
 Printer Friendly
Author Previous Topic Topic Next Topic  

todotson

USA
4 Posts

Posted - Nov 08 2021 :  17:50:18  Show Profile  Reply with Quote
Hi,I'm running a simple FH2 simulation:

**************************************************
**************************************************
**************************************************

*This is an example Preamble. Please replace me with a seeded ID.
.units mm
.default sigma = 3.5e7
*
*
*

**************************************************
gNet3 x1 =3 y1 = 0      z1 = -6.940000e-02 
+  x2 = 9   y2 = 0      z2 = -6.940000e-02 
+  x3 = 9   y3 = 7   z3 = -6.940000e-02 
+  thick = 0.03470
+  seg1 = 21 seg2=25
+  ng_Component2Terminal2_GP (3.500000e+00,5,-6.940000e-02)
+  ng_Component3Terminal1_GP (7,5.000000e-01,-6.940000e-02)
+  ng_Via1 (3.500000e+00,5,-6.940000e-02)
+  ng_Via2 (7,5.000000e-01,-6.940000e-02)
*+ hole circle (5.50000,7.00000,-0.03470,0.90000)
*+ hole circle (7.00000,3.50000,-0.03470,0.90000)
+ hole circle (8.50000,7.00000,-0.03470,0.90000)



**************************************************

*Vias for lower layer nets
**************************************************
nVia1node1 x=3.50000 y=5.00000 z=0.01735

nVia1node2 x=3.50000 y=5.00000 z=-0.06940

.equiv nVia1node2 ng_Via1
EVia1 nVia1node1 nVia1node2 w=1.00000 h=1.00000


**************************************************
nVia2node1 x=7.00000 y=0.50000 z=0.01735

nVia2node2 x=7.00000 y=0.50000 z=-0.06940

.equiv nVia2node2 ng_Via2
EVia2 nVia2node1 nVia2node2 w=1.00000 h=1.00000


**************************************************

*Vias for Excitations


**************************************************

nExc_Via3Node1 x=3.50000 y=5.00000 z=0.01736

nExc_Via3Component2Terminal2_GP x=3.50000 y=5.00000 z=0.01735

.equiv nExc_Via3Component2Terminal2_GP ng_Component2Terminal2_GP

EExc_Via3 nExc_Via3Node1 nExc_Via3Component2Terminal2_GP w=1.00000 h=4.00000


**************************************************

nExc_Via4Node1 x=7.00000 y=0.50000 z=0.01736

nExc_Via4Component3Terminal1_GP x=7.00000 y=0.50000 z=0.01735

.equiv nExc_Via4Component3Terminal1_GP ng_Component3Terminal1_GP

EExc_Via4 nExc_Via4Node1 nExc_Via4Component3Terminal1_GP w=4.00000 h=1.00000



**************************************************

*Place External Excitations
**************************************************
.extern   nExc_Via3Node1 nExc_Via4Node1

*Simulation Frequency
**************************************************
.freq fmin=6e8 fmax=6e8 ndec=1
.end


**************************************************
**************************************************
**************************************************

This is a simple planar conductor with 2 vias, and a hole in one corner. Excitations are assigned from one via to another. FastHenry2 crashes when I attempt a simulation, and doesn't crash when I comment the + hole circle (8.50000,7.00000,-0.03470,0.90000) line.

Can I get some advice on how to proceed with this simulation?








PhD student at North Carolina State University

Professor: Dr. Srdjan Lukic

Topic: Design Space Exploration in Power Electronics

Enrico

479 Posts

Posted - Nov 12 2021 :  12:44:51  Show Profile  Reply with Quote
Hi,

most probable cause is the break of some mesh or partial isolation of some segments. If you are left with dangling segments / filaments, the current of course cannot flow anywhere. I agree FastHenry should not crash, in principle; and should warn you with a message. I'm afraid that at the time the code was written not all cases where handled.

I had no time to run your input file but I can suggest you to move the hole position (even if this is not reflecting your exact geometry) to understand if this is actually the issue. If so, either you can (slightly) move the hole, or you can increase the meshing of the plane, so that you have more segments possibly not isolated (of course this latter solution would increase the simulation time).

Please let me know if after these tests and modifications you still cannot simulate your structure.

Best Regards,
Enrico


Go to Top of Page

todotson

USA
4 Posts

Posted - Nov 21 2021 :  19:01:25  Show Profile  Reply with Quote
Thanks for the reply. Sorry I haven't been monitoring this. I ended up just using a square hole instead of a circular one, with edges declared at the corner of the planar conductor. I don't have the raw output on hand, but it is a workable solution for me.

It would be interesting to experiment with the mesh density to get this hole to work.

PhD student at North Carolina State University

Professor: Dr. Srdjan Lukic

Topic: Design Space Exploration in Power Electronics
Go to Top of Page
  Previous Topic Topic Next Topic  
 New Topic  Reply to Topic
 Printer Friendly
Jump To:
FastFieldSolvers Forum © 2020 FastFieldSolvers S.R.L. Go To Top Of Page
Powered By: Snitz Forums 2000 Version 3.4.06