Re: IIRC these were the tests that showed what EMP could do to electronics
"not so much with valves"
That's a huge generalisation. If you made a valve the size of a MOSFET, it would have a similar inherent vulnerability to EMI, and more than a bipolar transistor. The reason that it's better "with valves" is simply the size of the things, it's harder for the induced currents to damage big metal electrodes. Mostly, it's not the technology, it's the ruggedness of the components*. Of course, if you know what you're doing, you can take advantage of the small size of transistors. The induced currents depend on the loop area of the conductors, so, with a small part, it's easier, to shrink the loop area. And put it in a little metal box, and be careful with the signals going in and out.
As for satellites, it's gonna be much easier to launch a machine made from transistors, than from valves like these:-
Anyway, I did a bit of research. It seems most of the damage to the satellites was to their solar arrays, stuck out flying through the soup of MeV electrons, rather than to their internal gubbins.
* Of course, it's a bit more complicated than that. Valve circuits are generally high impedance circuits. Not much current flowing, high voltages. The electric field dominates. Transistor circuits are usually the opposite, it's all about the H-field. A lot of digital electronics engineers were taught by people who grew up with valve circuits, and hence don't get that current flows in loops. This helps to keep me in a job.
p.s. For clarification, in that America, a thermionic valve is a vacuum tube.