So now comes the question for Dog,
We have a rise and fall time of 100V/ns... with f.i. 500ns delay beteeen both.
We can trigger one kacher cycle.
Crucial is to synchronise both mosfets ( kacher and impuls) as narrow as possible.
I think I purchase the Raspberry today
Personally I bet synchronisation first
I have an order put together as well and can fund it here shortly--
being on a fixed income now is a bugger.
I think synchronization needs to be solved, then we can at least
have a controlled platform to conduct some meaningful tests and
hopefully find the magic sauce. Once I get the parts on my bench
I can code something up to generate a structured bitstream that
will fire off events with the best precision I can squeeze out of
these little microcontrollers. I did look at what SL presented and
it does indeed do PWM real well, but that's not quite what we are
after here, plus there will be a longer learning curve to get results.
Rise/fall times are a signature of impedance and I want to draw
your attention to something I think is important, at least it was
for Akula. Below is a snippet of Akula's 30 Watt Lantern schema.
I circled the areas where it's clear he was carefully tuning this
device. The most critical area is in red. This is a RC time constant
with a diode so that it only operates in one direction. If you notice
when VT1 is off, both L1 & L2 have a positive charge and the screen
inside which is acting as a capacitor plate is grounded. Again,
I stress we are looking at something more than a transformer,
transmission line. We are looking at a reactor of some sorts.
Same thing with the grenade coil, only it has no magnetic core,
making it simpler in some respects and more complicated in
others. For the lantern, flyback is being redistributed creating
some sort of electron void, which gets filled from the ground,
in this case from the air or wherever free electrons can be
gathered. That diode along with R2 and C4 make it possible.
You'll also notice there is a one to three turns ratio among the
reactor which clearly alters the impedance unlike a typical
boost/buck converter.
Anyway, there is a lot going on in that circuit and knowing
what to look for can go a long way towards getting the higher
power units running. So first we get the events to happen
when they should, then we see how those events interact.
With a little intuition, maybe we can get the components to
respond how we like.