Electronics engineer here!

Very impressive! I’m ranked 14 kyu on OGS. DM me if you want to add each other. I’ve got some learning to do to get to SDK rank.

To my knowledge the best way right now is to bring the layout from KiCad into FreeCAD and then use the OpenEMS plugin in FreeCAD to set up the sim.

There are some awesome videos coming out from this channel: youtu.be/VcJqhsbzR3c?...

Thank you for confirming! This is what I suspected, but it’s not clear from marketing material.

I also don't really understand what's happening @ 20 kHz... My first suspect is my fixture, since that wasn't included in the calibration. TBD.

-83 dB = 1.8 mΩ
1.8 mΩ is still ~ 20% error, but
* I don't have a torque wrench yet,
* I haven't done anything to de-embed the fixture from the measurement,
* the CM choke only has -30 dB of CM insertion loss.
I think if I improve each of these points, I can get a lot closer to 1.5 mΩ.

When I bring my hand close to the core, I see a change of roughly ±5 dB - pretty unacceptable right off the get-go. I wouldn't have noticed this as quickly without the fast sweep-time of the VNA.

I ran a SOLT cal with the choke inline to see how it would look, and I immediately noticed, even limiting the span to 10 kHz ~ 100 MHz, how sensitive the choke was to cable positioning and de-tuning. It's just some 30 AWG twisted-pair wrapped around a toroid, loose on the bench.

I was going to ask: is Octave insufficient? I haven’t spent time with MATLAB in forever and don’t know Octave well, so I’m genuinely curious.