Investigate effects of non-causal v_loop_efit in CMOD & D3D

[yumouwei]

In CMOD, we found that:

1. VLOOPT is not an output of the magnetic equilibrium reconstruction, but a non-causally smoothed loop voltage signal.
2. VLOOPT is heavily affected by the IAVEV parameter in EFIT which smoothes the raw input v_loop signal. For analysis & EFIT18 IAVEV=20ms; for EFIT21 IAEAV=0. These seems to agree with what we observe from the comparison.
3. Changing the IAEAV parameter does not seem to affect EFIT outputs such as Wmhd, li, and q95, so these signals are still safe to use.
   • For both analysis, efit18, and efit21, the magnetic smoothing (IAVEM) was set to 0, therefore the equilibrium reconstructions are causal.

We should consider & investigate:

1. Remove v_loop_efit from the list of outputs as it's either unreliable or a duplicate of mflux:v0
2. p_oh computation: remove the fallback of using v_loop_efit in case mflux:v0 isn't available
   • If mflux:v0 doesn't exist then where did v_loop_efit come from?
3. Investigate if this affects D3D EFIT & physics methods as well.

[github-actions]

This issue has been open for 3 months with no activity.