Using a MOSFET to activate an induction heater

[Alberto]

Hello. Prepare youself for a lot of dumb questions and a very bad english.

Some time ago I built this induction heater and with your help I made some modifications ( in red)



The MOSFETS are IRFP250N. The question is you said I should connect first the 15volts power supply to the gates and then the main power for the rest of the circuit. I know one day I´ll forget about that, so to make it fail safe, mi idea is to put a mosfet or relay that only allow the current in the main circuit once the 15V are connected. To run the main circuit I´m using 30V and 14amps (for short periods of time, maybe 20 or 30 seconds.

I have done some research but I´m not sure how to do it. I suppose a need the mosfet in saturarion and put a 10k resistor between source and gate. But this is all I know. I don´t know what MOSFET should I use (I have some here, like IRFP260N or IRL540N).

Should I connect the source to ground in the power supply and the drain in the negative of the indcution heater? And the source and the gate to the 15V power supply but I suppose I´ll need a resistor between the gate ant he +15VCC rigt? What value?

Thank you

best regards

[AstRii]

I have a whole article on my website about trying to "improve" the ZVS driver and one of the modifications is turn on/off the power supply for the LC tank with the help of P and N channel FETs.

https://www.uhvlab.org/zvs-driver

This modification turned out to also work as sort of "soft-start" since the leakage current of the P FET starts the oscillation at low power.

[davekni]

I know one day I´ll forget about that, so to make it fail safe, mi idea is to put a mosfet or relay that only allow the current in the main circuit once the 15V are connected.

Might be easiest to have 15V power a relay (mechanical or solid-state) on AC line input to your 30V power supply.

[Alberto]

Thanks for your answers.

astRii my electronic knowledge are insufficient to understand how to run that. So I'll go for the relay.

[petespaco]

RE:

Might be easiest to have 15V power a relay (mechanical or solid-state) on AC line input to your 30V power supply.

Dave-  Could there be a problem  getting oscillations started if the 30 volt DC supply comes up slowly?  If so, maybe put that relay in the DC ckt?

[davekni]

Dave-  Could there be a problem  getting oscillations started if the 30 volt DC supply comes up slowly?  If so, maybe put that relay in the DC ckt?

There are lots of post on the induction heater thread about needing to apply DC power abruptly.  However, the reason abrupt application of DC power is needed is because gate power shares the same DC supply.  With gate power applied first, slow ramp up of main 30V supply should be better.  Slow start minimizes overshoot in oscillation amplitude so reduces risk of frying FETs.
I have personal experience with slow ramp on a couple of my own projects.  However, my circuitry is slightly different.  There is a chance that I'm missing some issue.  I believe a couple others have tried gate power first and slow (at least not abrupt) application of DC with the conventional ZVS circuit.  You'd need to search the Chinese 48V induction thread to be certain.

[AstRii]

I'm afraid that Mazilli's ZVS driver usually has a very high inrush current and your relay contacts might fuse together rather quickly.

How about using two relays? One that would connect the power through a power resistor, ensuring some small oscillations and then one relay to connect the low impedance input power after some milliseconds.

[petespaco]

Maybe the easiest answer for Alberto is to NOT modify the original circuit at all.  (No separate gate power supply at all)
It has worked quite well for me for several years. as long as the main DC power supply can  deliver at least 20 amperes or more at his 30 volts.
In my own experience,  the gating problems seem to be related to situations where the DC power supply's voltage drops below 12 upon startup.
I usually switch the DC power on with a DC rated circuit breaker AFTER I am certain that the power supply is up to full output voltage.

[davekni]

I'm afraid that Mazilli's ZVS driver usually has a very high inrush current and your relay contacts might fuse together rather quickly.

That's true only when gate power is not applied first.  Gate power actually starts low-amplitude oscillation powered through gate resistors.  Since oscillation is already started, main power can ramp up smoothly without any inrush.

[Alberto]

Thanks for your answers. The relay won't be responsible of initiate the current in the main circuit. This will be done with a switch.