Kapanadze Cousin - DALLY FREE ENERGY

[verpies]

is it?

It is not surprising that the Power MOSFET is getting overloaded with long pulse widths because when it is loaded only with an inductor, then the current through it will increase with time up to the V/R_DS-ON current limit, eventually blowing it up because this limit is very high.

I found on digikey and newark can sort by reverse recovery time, can go from 200us to 2ps... some say the speed of the swich is what's improtant?  should it be the pulse width?  200ns?  100ns?  50ns?

Unfortunately the reverse recovery time does not tell us much about the DSR effect in a diode, because it is measured after the diode has been conducting forward for a very long time.
The DSR effect requires forward conduction times below 500ns before the reverse conduction is attempted.

I find experiments using diodes, but nothing describing what characteristic really determines the effect...

Unfortunately there is not much in the diode datasheets that will allow you to predict if the given diode exhibits good DSR effect.
You can exclude any diodes that do not have PN or PIN semiconductor junctions (e.g. Shottky diodes).
I guess, theoretically, shoddily manufactured PIN diodes would have the best chance of working as good DSR Diodes.

But practically, for the DSR effect, you can trust only specific part numbers with specific date codes, that were experimentally verified by you or someone else.

[verpies]

... that pulse drains all my produced 150+.... 

...because your energy transfer in T2 is still very low.
Last time you posted your T2 primary current waveform it looked like the energy is being transferred only 1% of the time !!!

Proceeding without fixing that first, makes no sense.

[d3x0r]

I found a paper on designing SRDs and considers DSRD designs....



New Approaches For Designing
High Voltage, High Current
Silicon Step Recovery Diodes
for Pulse Sharpening Applications



page 12 has ideal requirements... Ts should be greater than Tr and Tr should be short....


power rectifier has a large Tr vs Ts


Page 3  has a table of tested SRD/WFSRD/DSRD ... Tr is 5ns or less
Page 111 is chapter 8 which is the throery of DSRD

Page 130 has a table of switching times for various DRDs ...


It goes from 500V (2x what gets fed to this one) Tf=5.5ns, Ts=2.6ns Tr=1.0ns up to 5000V and Tf=1183ns, Ts=127ns, Tr=9.9ns


Although I'm beginning to think that SRD for improving steepness of rising signals isn't the same effect?  DSRD is an excessive reverse pulse, right?

[verpies]

I will try to stay withing the 37 - 500ns range and hunt for saturation by increasing the drain voltage

It's better to increase the number of primary turns.

[d3x0r]

It's better to increase the number of primary turns.

On the 7-4-2 toroid, that's not really an option, only 18 turns of 24 gauge wire even fit through it....