Electronics

Electronics

AnsysElectronics Desktop (HFSS) solvers- HF3d error code 149

Tagged: ,

    • parhamz
      Subscriber

      Hello all,

      I have Ansys electronics desktop 2022 R2 working on ubuntu 20.04 . I made my mesh pretty coarse and I have a lot of RAM and CPU power on my local machine and I still get this error. No matter I use direct solver or iterative solver I get the same issue. What is the solution for this problem ?

      The solution happens after the mesh is created I can see the mesh and the analysus convergence. the moment solving for sweeping frequency starts it fails.

      Here is the detail of the error that I get:  Transition_wr75 (Modal Network)
          [error] Solving frequencies ... (Nondomain solver), process hf3d exited with code 149. (1:55:20 PM  Jan 17, 2023)
          [info] A discrete frequency sweep with 17 point(s) has been started using HFSS - Solving Distributed. (1:55:42 PM  Jan 17, 2023)
          [info] Discrete frequency sweep complete. (1:55:42 PM  Jan 17, 2023)
          [error] Sweep first_analysis_sweep_coarse failed (1:55:42 PM  Jan 17, 2023)
          [error] Simulation completed with execution error on server: Local Machine. (1:55:42 PM  Jan 17, 2023)

       

    • parhamz
      Subscriber

      update:

      The issue is still not resolved but when I remove the frequency sweep to a from analysis,  HFSS works fine.

      The moment any frequency sweep analysis is created the error will come back!

       

       

    • Praneeth
      Ansys Employee

      Hi,

      Please go through the following suggestions and observations on your issue -

      1. Try and see if the interpolating sweep works for you.
      2. Did you do any modifications to HPC?
      3. If so, were you able to do any frequency sweep analysis before?

      Kindly let me know if any of the above is the case for you.

      Best regards,
      Praneeth.

    • parhamz
      Subscriber

      Hello Praneeth,

      Thank you for ypur reply. I created two different design and I solved them for frequency sweep with different HPC options, solvers (direct or iterative), discrete or interpolating frequency, and saving all the fields option scenarios (there is check mark that says if you want to save all the fields for all the frequencies). The result can be found in the table below:

      Some things to note: my license give me the ability to solve in parallel using only 4 cores.

      Summary: when HPC is manual and I only ask for 1 task everything works fine. When HPC is manual and I ask for 2-4 tasks nothing works and the error code is 99. When HPC is in auto and I don't ask for savind the fields it works fine and solves 4 frequency in parallel. When I asked to save the fields it gave me an error 149. Table below can show the combination better.

      Using direct or iterative solver does not change anything. Thank you in advance for your time.

      Regards,
      Parham

       

    • parhamz
      Subscriber

      Update:

      The first two design that I created and test what has been suggested to me were two simple rectangle and circular waveguide solved with solution type: network analysis, modal.

      I created the design that i want to work on which is a custom rectangular waveguide. because of the sharp corners i needed to make a really fine mesh. This model also needs to be solved on network analysis, modal mode.

      Having said that, in the case of HPC being in auto mode/manual and hf3d solver have access to only 1 core/ 1 task everything works fine. In other cases ( the combination that i tried:1- including HPC auto with 2-4 cores and HPC manual 2-4 cores | 2- save fields on or off | 3- discrete or interpolating  ) the solver does not work and give me the code of 149.

      Regards,

      Parham Zarei

    • parhamz
      Subscriber

      Update2:

      The local system that I am running ansys electronics desktop on, have 128 GB of RAM capacity (4*32 GB). I thought hf3d cannot work in parallel might be to the RAM issues. So I went back to the two simple design (which I provided the table) that I created and increased the meshing density in multiple steps. In everythin I run I didn't ask to save the fields

      1- In the highest meshing density: With HPC set to 1 core (auto or manual) : using direct solver it used 95 GB of Ram to solved the problem and frequency sweep just fine. Also for iterative solver it used 30 ish GB of RAM and solved the the probem and frequency sweep just fine.

      2- In the medium density meshing: with HPC set to 1 core (auto or manual): everything works fine. With HPC auto up to 4 cores and 95% of RAM direct solvers started solving the adaptive meshing at 60 GB RAM capacity and transitioned to 1 frequency at a time for solving sweep frequency. The iterative solver, solves the adaptive frequency with at 20 GB and went to for solving frequency sweep with 4 cores in parallel. I was following the pid of 4 hf3d created on my machine. They started to utilize RAM up to 9 GB each and then the error code 149 appeared.

      3- Low density is provided by the table above.

      Having said that, Now I am confused. On one hand each core can access to almost whole RAM capacity (case 1, direct solver used 95 GB). On the other hand when 4 cores supposed to access around 80 GB (4*20 GB) it failed and gave me that error. 

      looking forward for your reply.

      Regards,

      Parham

       

    • parhamz
      Subscriber

      I am just following up on any possible solution/update on this matter ? I would appreciate any solutions that I can use the full capacity of my system in HFSS.

Viewing 6 reply threads
  • You must be logged in to reply to this topic.