Problems with FEA and Shape Optimization Simulation "HELIUM_ERR"

[s20007167]

Good day,

does anyone of you know the following problem? can no longer start a simulation. The error "HELIUM_ERR" always comes up.

Thanks in advance.

[ahosburgh]

I am also having this issue. My simulation precheck says everything is good, and it seams to upload to the cloud successfully, but then it fails. 
Here is my log file. There may be other simulations in it, but the last one is at least relevant. I didn't want to edit it because I don't know what is needed or not. 

Motor Mount GD v6: Simulation Model 1 - Study 1 - Shape Optimization
_____ 2022-01-19 09:02:52 _____

Motor Mount GD v6: Simulation Model 1 - Study 1 - Shape Optimization
_____ 2022-01-19 10:01:31 _____
Error: Please inspect mesh failures.

Motor Mount GD v6: Simulation Model 1 - Study 1 - Shape Optimization
_____ 2022-01-19 10:01:32 _____
29798 Tetrahedra : 100.0% of elements (100.0% of volume)
Face Angle min: 13.1, max : 136
Dihedral Angle min: 13.3, max : 163
Worst shape ratio : 7.46 on element 21112
Worst aspect ratio : 5.06 on element 13484, shortest edge: 0.000611, longest : 0.00727
Lowest collapse ratio : 0.186 on element 21112
Worst Jacobian ratio : 1 on element 1
Base mesh: 7257 nodes, 29798 elements
Job has finished successfully. Receiving results.

Motor Mount GD v6: Simulation Model 1 - Study 1 - Shape Optimization
_____ 2022-01-19 10:02:46 _____

Motor Mount GD v6: Simulation Model 1 - Study 1 - Shape Optimization
_____ 2022-01-19 10:02:46 _____
30229 Tetrahedra : 100.0% of elements (100.0% of volume)
Face Angle min: 18.5, max : 141
Dihedral Angle min: 14.8, max : 159
Worst shape ratio : 8.9 on element 15882
Worst aspect ratio : 3.22 on element 30150, shortest edge: 0.000566, longest : 0.007
Lowest collapse ratio : 0.169 on element 15882
Worst Jacobian ratio : 1 on element 1
Base mesh: 7343 nodes, 30229 elements
Job has finished successfully. Receiving results.

Motor Mount GD v6: Simulation Model 1 - Study 1 - Shape Optimization
_____ 2022-01-19 10:05:21 _____

Motor Mount GD v6: Simulation Model 1 - Study 1 - Shape Optimization
_____ 2022-01-19 10:05:22 _____
Document is being uploaded to cloud storage. Cloud job is pending.
Starting solve
Error: Unable to start the solve - Error("HELIUM_ERR"): Internal Server Error

I've also attaches a photo of the project I am trying to simulate. I have already run several different studies using similar models so I'm at least 10% sure my constraints and loads and everything else are correctly configured. Maybe 9%. 

[s20007167]

When did you get this error? Everything was fine with my program before yesterday.

[ahosburgh]

I was having some other funky issues that I haven't posted about here yet, like one model not uploading at all, but this specific issue started yesterday. 1/18/2021 for me as well. My log is from this morning but only because I was trying again on a different computer. 

[s20007167]

I hope that this is a problem from Autodesk, because I can't find any problems inside of my prototype...

[mcmillr]

Sorry you are having problems. It looks like this is affecting some users but not others. We investigated this and made some adjustments to the cloud software which might clear up the issues. Can you try again?

If you still cannot solve, please reply with your Fusion version (Help, About) should be 2.0.12157. Also your Active Plan, which is right under the version, (Student, Subscription, Personal, etc.)

We also will need your Fusion log files, you can get them under Help, Support and Diagnostics, Diagnostic log files.

It would also be good to know if you bought the Simulation extension, or activated the Sim Extension 7-day trial.

Thanks!

[s20007167]

Unfortunately the problem persists.

2.0.12157
Active Plan: Fusion 360, Student
Windows 10 Pro 21H2 x86_64 (19044.1469)

I have attached my log file.

I didn't activate anything consciously, but only with my student license

Additionally, I tried to run a simulation with a simple cube and an acting force on one side of the corner. This is still not wirking too. I added the error screenshot. The problem only occures, when I use the cloud. I can solve local, but it is not possible to do a shape optimization local.

Kind regards

Marcel

[Olly_GI]

Similar issue here - no cloud simulation available - upload on server stops rapidly after initiation - error message: Unable to start the solve - Error ("HELIUM_ERR"): Internal Server Error
Thanks

[ARonda81]

Same problem here

[2392537m]

Also having the same issues as of yesterday (19/01/22)

[mcmillr]

Thanks for the detailed reply. I am wondering if this issue may be specific to student accounts. If anyone else having this problem is NOT a student, please let me know. I obtained some good information from the log and I am working with another engineer to diagnose the problem. Hopefully we should come up with a fix soon.

[ahosburgh]

I tried running it again this morning and the problem still persists. 

Here is my information as well. I hope this helps. Thank you for your support. 

Fusion Version:
2.0.12157
Active Plan: Fusion 360, Student
Windows 10 Pro 21H2 x86_64 (19044.1466)

Diagnostic Log Files also attached

[mcmillr]

We found a problem that was affecting student users and possibly personal users as well, and this only affected studies that cost 9 cloud credits oddly enough.

Please try again, it should be working now.

My sincere apologies for the trouble this caused.

[s20007167]

Thank you for your help.

Unfortunately, the FEA simulation is still so slow and it doesn't cost any credits. Can you please check that too?

[ahosburgh]

I tried it again and it seems to be working now! 

Thank you so much! 

[mcmillr]

Different Study types have different costs, but students effectively have 'unlimited' cloud credits so the cost is not displayed.

As far as a solve job taking a long time, could you supply a bit more information? How long is 'long'? What kind of study, how complex is the model, are you comparing local solve versus cloud solve. Any details you provide would assist us in understanding your concern about performance.

[s20007167]

I did the same FEA simulations as before the error occurred and it takes longer I think. The simulation used to take roughly 20 minutes, now it was almost an hour under the same conditions. But I'll check it again tomorrow and report back.

[mcmillr]

Yes please let me know if you are conviced the solve time is slower. Make sure you check that you haven't increased the mesh density or changed other parameters that may have caused the issue.

One way to know for certain is to open an older document that has been solved, and right click on the Results entry in the browser and choose 'solver data'. Then change the dropdown at the top of the dialog to Solver Output. This will show you the actual log file written out by the solver, and if you scroll to the end, it will tell you the 'wall clock time'. You can then look at this for your later solves and see if the time has increased. Near the top of the file there is a section about "Model Database Size" and this will tell you how many gridpoints and elements there were in the model. If those numbers are similar, but the solve time has gotten a lot longer, I absolutely want to look into it. Please share a link to your older version and your newer version and I will download them and run time performance comparisons using our old and new solve pipeline implementations.

[s20007167]

Good morning,

I've tried running the analysis several times with no success. I can perform a FEA locally, but it doesn't work via the cloud. It worked in between, but now it no longer solves via the cloud, regardless of whether it is FEA or Shape Optimization.

I've attached my log file.

Unfortunately cancel doesn't work either.

[s20007167]

push