MOSFET problem / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

MOSFET problem

previous 1 2 3 next

Move Thread

LAN_403

HV Boy

Sun Oct 06 2013, 05:48PM

Registered Member #33572 Joined: Wed Aug 28 2013, 09:02PM
Location:
Posts: 9

Dr. Dark Current wrote ...

HV Boy wrote ...

Also as I said, a normal NPN transistor doesn't let through the square wave too.

This is expected.

But I was able to pass it through KT872A and feed it to my FBT. And everything was fine until now.

Patrick

Mon Oct 07 2013, 03:32AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

to be of further assistance id need to see pics and schematics and see a whole lot more than just words. i cant fully see whats preventing your circuit from working.

HV Boy

Tue Oct 08 2013, 10:59AM

Registered Member #33572 Joined: Wed Aug 28 2013, 09:02PM
Location:
Posts: 9

Sorry that I took some time to post but here we go. This is the picture of the schematic. And here's some pictures of the built circuit on breadboard. Heres the MOSFET's and the transistor that I used. Here's the normal healthy square wave and here's the square wave when I introduce it to a MOSFET. Here's the datasheet of the MOSFET's that I used IRF630B, IRF634B. I hope that's enough detail. If needed I can tyr and provide more. Thanks for the help.

Patrick

Tue Oct 08 2013, 07:45PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

the fact your square wave changes when unconnected to connected is troubling. the gate of the Mosfet is effectivley a very tiny capacitor plate, so the result should be a nearly unchanged square wave.

i think two things, first the 555 timer may not be working right. or second, as Dr Dark Current says, youve punched through the gate and killed the mosfet. (Yet the first thought is problematic as your 555 works when unconnected.)

--you do have a conductive path between the 555 timer pin 1 and the Soruce pin on the MOSFET, right?

--what freq is your square wave? ( i dont know what those Cyrillic letter mean

)

IamSmooth

Tue Oct 08 2013, 09:47PM

Registered Member #190 Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1567

Shouldn't you have a small resistor, like 10R, between your 555 output and the gate?

Patrick

Wed Oct 09 2013, 03:16AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

IamSmooth wrote ...

Shouldn't you have a small resistor, like 10R, between your 555 output and the gate?

5 to 15 -ish ohms is common, but not the cause or reason for it not working at all.

2Spoons

Wed Oct 09 2013, 04:16AM

Registered Member #2939 Joined: Fri Jun 25 2010, 04:25AM
Location:
Posts: 615

With that circuit the MOSFET will die on the first cycle. The back EMF from the primary of the transformer will kill it the moment it tries to switch off. Thats if the primary current hasn't already ramped up to the point where the MOSFET melts.

You have 1] no voltage clamping on the MOSFET drain; 2] no current control.

Patrick

Wed Oct 09 2013, 04:20AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Perhaps he should use a power resistor for the initial setup to take the place of the primary.

EDIT: 200 and 250 Vds is pretty low to drive a HV transformer primary. TVSs or diodes would be needed for a healthy transistor.

HV Boy

Wed Oct 09 2013, 11:38AM

Registered Member #33572 Joined: Wed Aug 28 2013, 09:02PM
Location:
Posts: 9

Patrick wrote ...

the fact your square wave changes when unconnected to connected is troubling. the gate of the Mosfet is effectivley a very tiny capacitor plate, so the result should be a nearly unchanged square wave.

i think two things, first the 555 timer may not be working right. or second, as Dr Dark Current says, youve punched through the gate and killed the mosfet. (Yet the first thought is problematic as your 555 works when unconnected.)

--you do have a conductive path between the 555 timer pin 1 and the Soruce pin on the MOSFET, right?

--what freq is your square wave? ( i dont know what those Cyrillic letter mean

)

I just connected source directly to ground. And the frequency is either 2khz or 20khz. But anyways thanks for the help guys I'm just gonna reside to either make a PWM driver of a ZVS driver.

Patrick

Wed Oct 09 2013, 04:20PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

HV Boy wrote ...

I just connected source directly to ground. And the frequency is either 2khz or 20khz. But anyways thanks for the help guys I'm just gonna reside to either make a PWM driver of a ZVS driver.

im not sure you've chosen a wise path.

If you cant get this to work, Id really caution you to learn its lesson before abandoning it. I have the very same (though more sophisticated) 555+flyback circuit and its the best, most well behaved circuit I've ever built. (And I've built a lot of HV stuff over 15+ years.)

EDIT: look at my thread here, and try to understand what ive done before you move on:

[youll notice on the above link, my schematics include all the tactics we EE's are supposed to use on HV switches and what not, As 2Spoons, IamSmooth pointed out earlier...]

this quote shows what I was able to get (eventually):

Patrick wrote ...

Proud Mary wrote ...

Thinking about that 850 Vpp auto tap on the primary, you can expect quite a bit of back EMF looking for weaknesses in your driver circuit

Well ill up the gate protection then, my mosfet can handle 1000V, but Its TVSed down to 600V, STmicro always puts double zeners across the gate, but ill use a TVS there too.

The flyback (and the bottom pins) itself is pretty well insulated with the addition of epoxy and oil, so ill eat your best top hat if a spark or arc develops from the bottom pins... the semi's however are always a questin mark though.

EDIT: ive got the transistor/ heat sink up too 156 degree F... core to 110 F.

EDIT: the fall time (100% to 0%) seems to be between 800 nS and 1.2 uS i could probaly do better with a lower gate resistance.

EDIT: Fall time is 1.1uS (100% to 10%), so that gate drive IC is really working, as the 555 was only capable of 3.8 uS when directly connected to the gate.

New pic...

97degrees F, at the top surface of the acrylic tank. The small bubbles on the center and left are from the heat of expansion from oil filling the now hot ferrite gap, the larger bubble on right is for thermal expansion of the oil, the air bubble shrunk about 0.2 inches in overall diameter while at 97 degree F (ferrite is at 118 peak temp F). Time to let it cool off. I do worry that the pressure will blow out the gasket sealer.

EDIT: Its 9 PM, im too tired to continue, ive captured "a" high voltage scope screen shot, not sure how credible it is though. Shorted out the main capbank 4000uF @ 82 Volts across the drain-source, whole apartment smells like burned metal, oscilloscope was undamaged, im unharmed but pisst off. Im done for the night, time to drink new castle.

Whoops it just gave up the purple smoke! MOSFET is definitely KIA, had another 20 seconds of run time before it finnally died. Apartment stinks even more now. I had 12, 11 left now.

EDIT: 11:43 PM totally drunk right noW, no more hV for me. I have new pics to post.
HV scope pics...

Current pulse, zoomed in. 6.6V means 6.6V / 0.5Ohms = 13.2 amps.

Average at 10.8 kV (1000:1 Probe, 30.6M ohms) I have no idea if this is credible or bogus.

Peak at 24.4kV (1000:1 probe, 30.6 Mohms) I have no idea if this is credible or bogus.

_________________________________________ ___________________________________________

EDIT: 1:03AM .... neEd sLeEp....ZzZZzzzzzZzzZZZzZzzz....

Does any one know what a flyback output looks like on scope?
maybe on the last page of this PDF?

From the above PDF and my Screen caps...

_________________________________________________ ___________________________________

>>>

previous 1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob