Discussion thread: "Analysis of small scale fluctuations in Hall effect thrusters using virtual Thomson scattering on PIC simulations"

[tarek-ben-slimane]

Hello everyone,
I am a PhD student at Laboratoire de Physique des Plasmas at Ecole polytechnique, and I am glad to attend this ExB workshop.
My contribution to the mini-posters session is titled "Analysis of small scale fluctuations in Hall effect thrusters using virtual Thomson scattering on PIC simulations". The work consists in adapting the experimental Thomson scattering diagnostic to PIC simulation in order to analyse the millimeter scale structures propagating in a HET discharge.

• The first 2 slides give a quick overview of the methodology and of the simulation case used to test this analysis tool.
• Slides from 4 to 6 show the characterisctics of the dominant modes in k-space for fixed t and show the dispersion relations obtained from the analysis.
• The last few slides investigate the impact of the laser beam waist on the estimation of the electron density fluctuations using Thomson scattering.
  Please feel free to check it out and ask me any question.

PS : The results presented in these slides have been recently published. Here is link if ever more details or precision are needed : https://aip.scitation.org/doi/pdf/10.1063/5.0073215

[mariomerinomartinez]

Thanks Tarek for this interesting contribution.

I was wondering: the way you compute the k-spectra is by doing a wavelet analysis of density ?

Best regards,
Mario

[tarek-ben-slimane]

Hey Mario,
Thanks a lot for your question.

The k-spectra is obtained at a fixed axial position by performing a 2D gaussian windowed FFT at different azimuthal positions. The 2D FFT yields a (kx,ky) map, for every azimuthal position then the magnitude of these maps is averaged to produce the one in the slides.
Mathemathically speaking, this might look like a 2D-wavelet transform however the size of the gaussian kernel is fixed by the laser waist whereas in a wavelet transform it should stretch with k.

I hope I answered your question,

Cheers,
Tarek

[mariomerinomartinez]

That was fast :) Thank you!