Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated

[T-1000]

Hi all,
please note the number of wires coming from Pierre's induced coil.
The attached image is an enlarged detail from his 1st video at ca. 6:30

Due square wave-like with spikes going on induction (because the field rotates in 6 steps resolution) the collecting coils must have more interwinding capacity in order to consume spikes. In low resolution of magnetic field movement simulation the flat collection coils made from foil might be even better due spike rise/falll times against wire length and inductance.

[Thaelin]

Has consideration to the direction of injected signal from the recovery diodes been given? Could this be driving a H/V spike on top of the rail?  TDR reflections play havoc in certain situations but may well be "USED" in this one creating extra voltage on the main rails.

thay

[Jeg]

If a half bridge with MOSFET's were used with a rating of say 20A or greater, the body diodes would have the same rating and this would not be an issue.

In this case in which we use mosfets or IGBTs do you think that an extra diode is needed? (To prevent an opposite flow of current  through the mosfet/IGBT body diode).

ps. I have the suspicion that thyristors are best for this work.

[Jeg]

To achieve that scheme electronically you have to use a half bridge configuration.

I plan to use a  much more robust half bridge with 45A 45V rating and built in opto isolation, over current protection etc.

Can you provide a type of such an isolated bridge? 

[Slider2732]

This may be obvious and I may have missed it in the technical discussions.
But if we view the supercaps of Pierre's as the condensers in the Benitez Patent, doesn't that make sense of why there are 72 relays and not just 36 ?
Every time there is a collection from a coil sequence, the correct supercap is grounded.
All supercaps can be initially charged and then at the switch on of the device their common ground to the charging source is removed, replaced by the switching action in and out of those relays.
Control timings need not be in any Arduino code and, we don't see it there in Pierre's published code, instead tied to the relevant ordering of the coil relays.
Each coil winding is a transformer of the Benitez Patent ?