Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Jul 24, 2014, 9:59 a.m. EDT
Dear Pu,
I had the same problem using v43b and v44 using the RF module.
After asking the support service and a long thread of messages going back and forth, it seems they have done some changes in the solver that affects the way the field is "normalized"
I did solved my problem after playing with the phase of the solution and re-normalizing the field. So for one of the versions I had to set "Solution at angle (phase)" for a value theta (0<theta<90[deg]). My suggestion, try first with theta=90[deg] and see if that fix it.
Hope this helps,
Felipe BM
Dear Pu,
I had the same problem using v43b and v44 using the RF module.
After asking the support service and a long thread of messages going back and forth, it seems they have done some changes in the solver that affects the way the field is "normalized"
I did solved my problem after playing with the phase of the solution and re-normalizing the field. So for one of the versions I had to set "Solution at angle (phase)" for a value theta (0
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Jul 31, 2014, 3:56 p.m. EDT
I had a similar issue using Structural Mechanics. Support just suggested using a finier mesh (even though the older version had a fairly coarse mesh as well) and it fixed this issue. I am not sure why, but it helped.
I had a similar issue using Structural Mechanics. Support just suggested using a finier mesh (even though the older version had a fairly coarse mesh as well) and it fixed this issue. I am not sure why, but it helped.
Please login with a confirmed email address before reporting spam
Posted:
10 years ago
Mar 12, 2015, 8:20 p.m. EDT
i have the sme problem in heat transfer simulation very good with 4.4 but completely different in 5.0 !!!
i have the sme problem in heat transfer simulation very good with 4.4 but completely different in 5.0 !!!
Please login with a confirmed email address before reporting spam
Posted:
10 years ago
Apr 2, 2015, 10:05 p.m. EDT
I was never able to get v4.4 RF to give reasonable results for any of my complex RF/Microwave problems that were solved splendidly in v4.3b. I gave up. Stayed with v4.3b until the second release of 5.0 came out (Dec, 2014) - wasn't interested in wasting time on the first release. I ran a bunch of complex test cases in it before concluding that it really was an "upgrade" - always seemed to give correct answers, that agreed precisely with v4.3b. So we went with it. Since then, our problems have become much more complex and we're having far more problems getting 5.0 to finalize the geometries than we ever had with v4.3b - but to be fair, the problems we are trying to solve now are far more complex than anything we tried in v4.3b. But after we finally get 5.0 to finalize and mesh (and we spend 70% of our time in trial-and-error attempts to get to that point, but these are 400MB problems without the solution, ~800 geometry nodes), we get results that always make sense. That was never the case with v4.4. Go back to 4.3b if you can't go forward to the latest 5.0. It was a solid release.
David
I was never able to get v4.4 RF to give reasonable results for any of my complex RF/Microwave problems that were solved splendidly in v4.3b. I gave up. Stayed with v4.3b until the second release of 5.0 came out (Dec, 2014) - wasn't interested in wasting time on the first release. I ran a bunch of complex test cases in it before concluding that it really was an "upgrade" - always seemed to give correct answers, that agreed precisely with v4.3b. So we went with it. Since then, our problems have become much more complex and we're having far more problems getting 5.0 to finalize the geometries than we ever had with v4.3b - but to be fair, the problems we are trying to solve now are far more complex than anything we tried in v4.3b. But after we finally get 5.0 to finalize and mesh (and we spend 70% of our time in trial-and-error attempts to get to that point, but these are 400MB problems without the solution, ~800 geometry nodes), we get results that always make sense. That was never the case with v4.4. Go back to 4.3b if you can't go forward to the latest 5.0. It was a solid release.
David
