Kapanadze Cousin - DALLY FREE ENERGY

[verpies]

My transformer primary is about 3.5 uH and has 10turns. ...
The toroidal I'm using has an AL of 35 nH/turns squared and dimensions are external diameter 12 mm internal diameter 4.5 mm and height  4 mm.

Using the formula for the relative permeability of a toridal inductor with rectangular crossection: μ_r = L / (0.0002 * TH * N² * ln(OD/ID) )
and the formula for the AL Value as A_L = L / N² ,we get:

                    N     OD [mm]    ID [mm]    TH [mm]          L [μH]           Î¼_r        A_L [μH/N²]     Saturation Current [A]
------------------------------------------------------------------------------------------------------------------------------------------
Hoppy         12           11.0            5.0             4.0       312.000       3435          2.167                      ? ? ?
Hoppy           6           11.0            5.0             4.0         88.000       3115          2.444                      ? ? ?
Black_Bird   10            2.0             4.5             4.0           3.500           45          0.035                      ? ? ?
Itsu             12                                                                 ? ? ?
Itsu               6                                                                 ? ? ?

Feel free to correct or add to the table above.

[verpies]

Increasing the pulse width from 11ns till 500ns shows nice increasing ramp up current across the CSR as in the blue line on saturation diagram in post #325, so no saturation.

Unless that nice ramp is already on the red line, like in the snapshot below, not on the blue line like in Reply #458.
The slope of that ramp will tell us if it is on the blue or red line

But i think i am pretty sure i have seen no saturation of the little toroid during these exercises.

I am not so sure.
If your calculated permeability will be close to 1 (like in air or vacuum) then it will mean that the core is saturating so early that we cannot see it, as on the attached snapshot at 2sec. index of this video (note that the yellow line was added artificially):

[verpies]

Itsu,
Try a transorb between gate and source and drain to source. The gate transorb (TVS diode) should be rated at around 300W or more at a breakdown voltage under the max vgs which is 30V abs max for the IRF 840.

The Transorb or Transil can have several nF of internal capacitance that needs to charged and discharged, even if the Transorb does not avalanche and this capacitance will degrade the gate signal. I read somewhere about Transorbs with pF internal capacitance, but I never used them.

I don't think Itsu has a problem with overvoltage at the gate.  His MOSFET driver cannot exceed the maximum voltage rating of his MOSFET's gate and any eventual Miller effect, caused by the drain-gate capacitance, is effectively attenuated by the 10Ω between the gate and ground.

You could also increase the gate resistance up to say 22R (which I'm using) to reduce any gate oscillations due to your very fast waveform edges.

That 22Ω resistor would form a voltage divider with Itsu's 10Ω gate-source resistor and that would degrade the gate voltage significantly during the on-time.
Itsu's gate driver is strong enough to ovecome the 10Ω resistor and charge the gate quickly during the on-time, but without this 10Ω resistor Itsu's driver is not strong enough to discharge the gate quickly and turn-off the MOSFET quickly.
Some of the newer MOSFET drivers address this problem by providing 2 times more gate discharge current than gate charge current (e.g. the UCC27511 )

Ferrite beads can be used on the gate lead but not too much as they will slow the edges.

Ferrite beads might help, if Itsu has problems with gate breakdown due to high voltage spikes on the gate, ...but does he?

You may also need a more powerful mosfet.

Maybe. He already is pulling 25A for 40ns with his MOSFET - that's pretty good IMHO.
Do you think he needs more current?

[verpies]

Also, in my circuit I am using a snubber with a 3.3nF in series with a 47 ohms resitor from the drain to ground, besides the diode ( a MRU820) in parallel with the transformer primary.

@Itsu
It might be a good idea to experiment with lower resistances of that snubber resistor, down to Black_Bird's 47Ω.

A fast flyback diode connected in antiparallel across the T1's winding, is also a good idea to kill the flyback HV spike eventually appearing between the drain and source of your MOSFET.

[itsu]

If I knew it had a limit of 5MHz I would have suggested using a 4.7nF and 10nF capacitors in parallel with T1.

Ok,  here the results with 10 and 4.7nF:

12t   10nF 1723Khz
12t  4.7nF 2463Khz
  6t   10nF 2593Khz
  6t  4.7nF 3663Khz

All had nice peaks on this frequency only.

Thanks all for your additional suggestions, i have some ucc27511 drivers which i will try to hook up to my MOSFET

Regards Itsu