I've been working through reproducing Steve Sandler's 2-port shunt-through measurements (2019 SIJ article, Picotest slide-deck). My target is to measure capacitors with ESR down to 1 mΩ with an accuracy of at least 10%. I've had good success with a common-mode transformer with > 30 dB CM insertion loss when measuring a 1.5 mΩ resistor.
I'm now trying to figure out how to de-embed fixture parasitics (inductance and remaining ground-loop error). The 2024 issue of SIJ, page 42 shows why de-embedding is important, and shows how to configure the fixture with a short (with an ideal short in EM simulation, or with a low-resistance piece of copper of similar size to the DUT on the physical fixture), but I'm unclear on how to actually implement the de-embedding algorithm for a 2x THRU shorted fixture.
Can anyone please point me to resources on how to de-embed the shorted fixture from the measurement? I think I could subtract ESR and ESL from the DUT measurement, but I'd like to generate s2p files that I can use in an EM solver.
