I have this transistor: http://www.fairchildsemi.com/ds/FG%2FFGL60N100BNTD.pdf

I'm trying to use it to discharge a 350 V capacitor (capacitance isn't labeled, about the same as one you'd find in camera flash circuit). I have the positive terminal connected to the collector and right now, I just have the negative terminal connected to a resister and then to the emitter. The idea is to discharge the capacitor with a small voltage applied to the gate. However, while charging up the capacitor, the circuit shorts at a certain voltage (close to Vmax for the cap).

I've tested the transistor by putting the emitter to ground then putting the collector to a circuit that activates when applied to ground. I apply 4.5 V to the gate and the low-activated circuit turns on. This tells me that I either don't know how to read the product sheet or there's something I'm missing for when I need to discharge a capacitor. I've tried googling, but can't find anything on using high-voltage transistors as a switch for cap discharges. So could someone point me in the right direction?


That's essentially the simplified version. In parallel with the capacitor is a charging circuit. I also simplified the triggering circuit to the switch. And keep in mind it is an IGBT, so I'm assuming I properly accounted for the bipolarity of the transistor when positioning the polar capacitor.
The desired result is just to have the current drain either through a thin wire filament, heating it up. Another possible result would be to rapidly discharge it through a wire loop to create a magnetic field. The point of the transistor is to have the action triggered by another circuit (e.g. serial port output from a computer, microcontroller output, certain logic conditions met, etc.) and thus avoid direct human interaction. This IGBT is just one that I happened to have access to. I don't have any experience with transistors, so if this one isn't suited for what I want to do, then I'm fine with buying something. However, if what I do have will work, I'd like to learn what I'm doing wrong.

 

That's a very curious problem, since your transistor is rated for 1000V. And the fact that you've tested it at circuit voltage makes it even more curious. I'm not an expert with IGBT's, but I think there must be something else in your circuit that isn't being shown in your simplified diagrams that is causing this strange behavior.

__________________
You don't need a quadraphonic Blaupunkt -- you need a curve
ball.

 

Hi dliqdzovuk, I'm doing something similar (home built EDM). What I found was that you need two transistors(I'm using mosfets)

The first one goes between your power supply and the capacitor, call it CHARGE.

Second is between capacitor and ground, call it DISCHARGE.


CHARGE is on -- DISCHARGE is off. capacitor charges.

CHARGE is off -- DISCHARGE is on. Capacitor discharges.

Without the CHARGE transistor the capacitor charges up when the power is first turned on, but when YOUR transistor turns on power just flows from the power supply instead of the capacitor.

Kind of hard to put into words. Hope it makes sense to you.

Cary

 

There's a charging circuit in parallel to the capacitor, that's about it.

If I'm understanding you correctly, I think I have the same or similar situation with the charging part of the circuit I'm not showing. Due to the behavior I was receiving though with the IGBT I'm using, I didn't think it was this circuit so I didn't really deem it necessary to include it in the sketch.

Somebody somewhere else had an idea. They proposed that I increase the gate voltage to 10 or 15 volts and that I move the ground closer to the emitter (to the left of the resister). Though I still don't fully understand the physical characteristics of transistors, it seems like this might be the solution.

 

Hi dliqdzovuk, OH, I didn't even notice the gate value in your circuit! Yes IGBT and mosfet's need at least 10VDC for a complete turn on (unless they are logic level)

If you look at the data sheet for your transistor you'll see that most of the ratings are for Vgs (voltage gate to source) of 10VDC. If you look in the graphs shown in the data sheet,(called "output characteristics" or something similar) most time they show no advantage to go over 15VDC on the gate.

At 5V you are just barely turning on the transistor. It is acting more like a very high ohm resistor.

I still stand by my charge/discharge idea though.

Cary

 

Attached is a modified circuit from a flash camera. I'm not too sure how this is working and am still trying to figure it out. But maybe you can get some ideas from this since your trying to accomplish almost the same thing. There's a transformer in this design and not sure why it's needed.

Attached Thumbnails

 

 

Hi dliqdzovuk, looking at your circuit again I see something else.

You need a gate driver. As it's drawn it is a high side switch. Your load is between the IGBT and ground so the gate voltage has to be 10-15V higher than the emitter voltage.

If you don't mind me asking, what are you building?

Cary

 

The transformer is there to make high voltage for the flash. It takes the
1.5V battery and makes, IIRR, around 400V.

Cary

 

Oh so the flash requires the high voltage. 1.5V wouldn't be enough to activiate the gas inside the tube?

With the capacitor storing that much energy I wonder if dliqdzovuk idea can be used to discharge the cap.