So I finally
got around to some testing today.
I ran the unit above which has the SMD film cap supplying the FET driver, and the new power pcb having 8 MOSFETs.
I have to say I'm pretty pleased with the results.
It went better than expected - last time I tested I had home built flyback diodes and a single TVS. Also I was using a leisure battery with my 45ah Bosch car battery. This time I had a decent schottky, 3 TVS diodes and two good quality 45ah car batteries connected in parallel.
I set P1 = 0.2ms, D = 1ms, P2 = 4ms, this was enough to weld a 0.1mm piece of nickel.
Yellow trace = gate drive
Blue (green?) trace = Drains
Zero volts is 2nd division from bottom for both traces.
zoom out shot of both pulses
Zooming in to the pulse2 turn off
a little more
and a little more
The difference between these shots and those on page one are clear. Last year I had managed to get the avalanche current (flat top of drains waveform at ~27V) down to 22us (from 70us without any diode in circuit). This is now down to barely over 1us. To quote my 9 year old, Sarah, "Booooomm!"
A big Thank You to all for your input and kind words of encouragement on this - I would have given up before now without your interest along the way
Also to those whose ideas I built on for the power side - mainly crossbreak, okashira, riba and everyone on the other development thread.
I shot some video of a weld at P1=0 D=1 P2=15 for a 0.15mm weld. It'll take a while for me to edit on my slow machine.