some doubles about no output data when running my input.deck

[hwckyu]

Hi devs,

the step, submitting my input.deck(the attached txt file), seems ok. But, when the calculation is finished, I find that the output fields' and particle's files are 2 KB, which seem no data stored. And indeed, when ploting the .sdf file, I do not find any varibales inside.(the example file is attached, .sdf)

inputDeck.txt fields0001_sdf.txt

Are there anything wrong in my deck? Looking forward,

Fisher.

[Status-Mirror]

Hey @hwckyu,

I've run your input deck with the version of EPOCH currently on the main github branch, and I can recommend a few tweaks.

• You include a use_collisional_ionisation key in the collisions block, but this should be collisional_ionisation instead.
• I can't actually run this input deck to the first dump. EPOCH is loading a timestep of 0.112 fs for me, and the runtime is 3 ns - it's expecting a 16 day runtime for me. I wonder if your sdf files are left-over from an earlier run attempt, before you settled on this final output block?

You might think this time-step is ridiculously small - after all, you have a cell size of 500 microns, which implies a time-step nearer $c \Delta x \approx 1.7$ ps. However, EPOCH also chooses a time-step to resolve the plasma frequency, and I think it's trying to resolve the plasma frequency of your I1 species.

You can override this time-step behaviour by tweaking the source-code, but that's a dangerous road to go down. There's probably a reason why you want to resolve the plasma frequency in a PIC code.

Let me know if you want any further support - I'd find it useful to know what you're trying to simulate.

Cheers, Stuart