If KCL is dead in the loop under test, it is also reasonable to assume it is dead in the measurement loop.
No, that's not reasonable, it's absurd. KCL doesn't 'die' here, it just gets a little behind the curve.
If it makes you feel better, you can call it that way.
My neighbor's cat was hit by a car a couple of years ago and it 'just got a little behind the curve' since then.
Sounds a lot better, I have to concede that.
It's a transitory thing
Well, yes. That's what
we some people are trying to analyze: what happens in the first few nanoseconds before the perturbation that travels along the wires reaches the load.
and that is why I've repeatedly asked you about the timeframe. What is the permittivity of the conductors in this timeframe? How large is the measurement loop compared to the test loop? I would have been more careful than AlphaPhoenix with the test setup, but for the timeframe he was displaying it was fine. Or at least OK-ish. Maybe. KCL doesn't 'work' in the main loop for a microsecond or so due to the propagation speed and self-capacitance and probably other things. The KCL issues in the measurement loop are going to be in the double-digit picoseconds at most.
Look at Ben Watson's simulation data for that 9cm x 2.5 cm strip (IIRC), you can see that even after 620 ps after the 'switch has closed' the currents in three different points of the loop are very different. To me that means that KCL
dies gets behind the curve, and it's still de--- way behind the curve after 1.8 ns, for that tiny strip.
Maybe when you say "gets a little behind the curve" you are thinking about ordinary transmission lines with periodic excitation. Then I can understand what you mean. KCL still dies, but can be seen alive in the model with a lot of distributed components - so we can say it's alive and well and just behind the curve.
But the transient between 200ps and 820 ps in Ben's simulation is a one-off due to the pulse coming from the switch closing (even if he used something different, and we could argue about that). In the case of Derek's simulation it would happen in the time comprised between t=0 (or t=d/c, if you wish, where d is the distance switch-load) and t=2L/c (where L is the length of each 'arm' of the line - I am neglecting the velocity factor for simplicity.
Also note that in the two transmission line model, the 60ps delay between the closing of the switch and first current in the load would not be visible.