How to turn ON a Mosfet about half way or less?

[ChrisP58]

Last I saw, he was using N-channel mosfets. How do the diodes tied to the upper rail (20 VDC) help turn them off?

The usual design has them draining toward the driver, which hastens turn-off.
John

In Gary's original, non-pwm design, the mosfets were turned on by pullup resistors to +12V, and turned off by the schottkey diodes from the gate of one mosfet, to the other drain of the other mosfet. Basically, as one mosfet turns on, it turns the other mosfet off. The LC resonant timing controls the switching frequency.

In a design using PWM, the mosfet gates must be exclusively driven by the PWM controller. The pullup resistors, and the diodes to the opposite drain, need to go away. In PWM systems, diodes can by used to provide independent control of turn on and turn off transition times.

 

[jjw]

Last I saw, he was using N-channel mosfets. How do the diodes tied to the upper rail (20 VDC) help turn them off?

The usual design has them draining toward the driver, which hastens turn-off.
John

He asked about the original circuit where the diodes where connected from the gate to drain of the other fet and the fets where connected to the ends of a center tapped coil.

 

[jpanhalt]

He asked about the original circuit where the diodes where connected from the gate to drain of the other fet and the fets where connected to the ends of a center tapped coil.

???

I was referring to the circuit in post #18 by Gary350 and your subsequent post #19 that describes the diodes' function, presumably in that schematic too.

John

 

[gary350]

They are not zeners, but schottky diodes and they turn the fets off.
I think the pwm frequency should be synchronized to 60kHz resonance frequency with a pll.

I read some place online Schottky diodes are the same as zener diodes. I should have known you can not believe every thing you read on the internet.

I'm not sure I like the idea of PWM not running at resonance frequency.

Maybe a small preamp in the gate circuit will be better than PWM.

When I first started experementing with this circuit over a month ago I did not have 240 ohm gate resistors so I used 200 ohm, then I put resistors in series to get 220 ohms, then 240 ohms with 3 resistors. All I can remember is 240 ohms worked bests but at my age my short term memory has gone down the toilet, I need to start keeping good notes.

I checked the gate voltage with my digital metal and it reads 20.5 vac. I should be getting 12 vdc from the 7812 voltage regulator? Meter much be reading power supply voltage through the Schottky diodes.

 

[ChrisP58]

I checked the gate voltage with my digital metal and it reads 20.5 vac. I should be getting 12 vdc from the 7812 voltage regulator? Meter much be reading power supply voltage through the Schottky diodes.

The only way to get an accurate read of the voltage of a switching waveform is with an oscilloscope. The most that a voltmeter will give you is some quasi average reading. Even a true RMS meter will only give you one number. To understand what is going on you really need to know what the waveform looks like. Then you can see the peaks, the rise and fall time, how much dead time there is, the ringing, etc.