[G0HZU_JMR]

At work I do a lot of broadband design covering HF through many GHz and I have used similar techniques to investigate inductor behaviour. I use E and H field probes and a VNA for free space measurements. I also use this technique to measure the Q of LC resonators.

However, I find that the most accurate way to model a typical inductor through all its resonance modes is to measure it on a full 2 port VNA and then play with the data on a simulator. In my case this is for inductors that will go on a PCB. The inductors can be anything from 0402 size to a big air spaced solenoid or a handwound inductor on a toroid.

The alternative method to model all this is to create a complicated transmission line model for the solenoid and I have a method to model/predict where the resonance modes will be based on the inductor dimensions. I can usually get the first two resonances very close but the third one is harder to model. I don't think David's model goes beyond the first free space resonance. He's demonstrated higher resonance modes but I get the impression these are expected to be at multiples of the first resonance. I don't think this is the case for typical solenoid inductors.

I think David Knight has been looking at this stuff for many years. His methods are really interesting and I'd love to be there at some of his demos. However, I think it's possible to go beyond what he is doing by using a modern VNA and an RF simulator. The resonance modes usually aren't harmonically related although sometimes it appears like this. Often the resonance modes are grouped closer together and my models predict this correctly.

I have to be able to predict/model this stuff because even a tiny resonance in a (power amplifier) bias tee up at UHF can cause a hotspot in it if the reflection coefficient of the load is in just the right place at the right frequency. This can expose the bias tee to a very high RF voltage and if this happens at a resonant frequency then the bias tee inductor can get very hot very quickly. I usually use a thermal camera and a special load I designed to let me explore various VSWR circles at the load whilst looking for hotspots in any bias tee components. Sometimes it's capacitors that can get hot if they have a resonance with a certain load at a certain frequency.