I know that inhomogenous magnetization of the iron wire will change the delay and alter the compressed pulse produced by the bifilar coils.
So, you take a magnet and get rid of that little nonuniform magetization issue.
Can we move on now? I'm tired of hashing over the same stuff year after year.
Forget the NMR of iron "ruse", and all of the BS about transmutation. Nothing need be transmuted, that is just a means of explaining something that doesn't appear possible with accepted laws.
===========================================================================================
We can be reasonably sure, by now, that SM used several of these bifilar coils, around the ring. That means iron wires wrapped with these coil as delay elements, one for each coil. We also know that we do not need iron, as we can use delay circuits (like Peter showed).
We can be reasonably sure that the collector wires are horizontal. You could stick two pieces of wire in the side of the TPU and measure power. After all, SM did say the collectors are layed horizontal, then wrapped with control wires and then more control wires.
Based on the two assumptions above, the pulse created by the bifilar (don't forget it is biased by about 500v) somehow induces curent in the collector.
So, we need some circuits for a controllable test:
1. tunable delay of each segment
(passive elements are simple enough and we are not using the iron delay method)
2. sequential adjustment of each segment is via #1
(assuming he used one main square wave or sine wave signal for all of them and
then adjusted delay of each segment to make them sequential)
(I want to stay away from sequential trigger at this time to closer match the TPU rather than AVEC)
3. supply for the 500v bias (we switch over it) or we cheat and use a static coil
(I vote for cheating for sake of time and easy of supplying power)
4. HV supply (1500v or more)
Damn. That just too easy. (I never bothered to lay this out on paper before)
Hmm. Lest we put the cart before the horse, we can test one segment coil with several loops of wire through the middle, still need the DC coil or 500v offset for bias (or a magnet -

)
I have everything but have to check on the delay element values, as I probably need some of the chip type, trimers caps, inductors, etc. Peter found the pulse compression effect with a ~220ns delay.
Anyone else game?EDIT:
Hmm. I'll need a 1500v square wave source. I don't want to pull one off another project. I have an old one, but it's not all that great. I can just build antoher one though.
Man, this is going to be fun!
Anyone else?