[Open_electroporator] pulser progress, transformers are good enough if fast pulser used

[John Griessen]

This photo just taken today with a good "old school" DSO, a Tek 2230 with peak detect so it doesn't miss events in between samples 
and assign low values to them.  The upper trace is the microcontroller low volt output signifying where program
and timer events happen.  The lower is the HV output at 500 volts per division.  This was with a DC in of 11 Volts,
so it shows that with close pulse spacing, the doubler will work fine to hold up higher volts as in that last pulse
that goes to 1000V.  I didn't even program that one -- it was spurious to the program -- an artifact of interrupts
-- the collision of too many interrupts to be handled in time.  So that program needs rework.  I need to learn the
low level details of the timers even more and not use python to program them on the fly in interrupt service routines,
but only to set them up to need no interrupts at all, (which is possible by chaining them to each other and using count 
prescalers, etc.), soft-wiring their clocks to each other to do the burst-of pulses logic without program polling or interrupts.

So this photo shows the first crude version of what I had envisioned doing -- a programmable volt profile in high voltage.  It 
still needs a recording method built into the microcontroller circuit, but that's easy.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: culture_shock_pulse_train-1.jpg
Type: image/jpeg
Size: 420355 bytes
Desc: not available
URL: <http://cibolo.us/pipermail/open_electroporator/attachments/20160821/18371b4f/attachment.jpg>