Linear power supply mosfets and drivers.

[fastline]

I am still working on a board that appears to have an unregulated linear PS. Basically AC comes in, is filtered a small amount, goes through a bridge, then to a pair of mosfets and their output is taken to a transformer for further distribution. The PS has a VERY hard short in it.

One of the mosfets was obviously shorted so I removed both and tested the circuit for shorts. I later resinstalled just the 1 mosfet to see if the board would at least power up. It shorted out hard and killed the fuse again. This time, it took out the gate resistor.

I am now wondering if I have a driver problem that is killing the mosfets and that is causing all the problem. So, my question is, if there is a gate resistor failure, is that resistor usually there specifically to die upon driver issues or shorts or would the problem usually go back even further?

I did some board testing before reinstalling the single mosfet and when testing against either the drain or source (cannot remember), I had 160VDC. Obviously if the gate saw that kind of voltage, that would be a problem but I was not sure if this would drop out once any load was applied.

There is a driver daughter board for the mosfets that has several resistors and zeners on it. I am suspecting a zener failure but not too familiar with testing them. So, when a mosfet fails, what does it usually take with it?

 

[fastline]

I removed all 5 zeners from the driver board circuit for testing. I realize their breakdown should be tested but with a diode test with a DMM, I have .5V on one direction and OL in the other. I am suspecting they are all good at this point. ??thoughts

 

[ben7]

I am still working on a board that appears to have an unregulated linear PS. Basically AC comes in, is filtered a small amount, goes through a bridge, then to a pair of mosfets and their output is taken to a transformer for further distribution. The PS has a VERY hard short in it.

One of the mosfets was obviously shorted so I removed both and tested the circuit for shorts. I later resinstalled just the 1 mosfet to see if the board would at least power up. It shorted out hard and killed the fuse again. This time, it took out the gate resistor.

I am now wondering if I have a driver problem that is killing the mosfets and that is causing all the problem. So, my question is, if there is a gate resistor failure, is that resistor usually there specifically to die upon driver issues or shorts or would the problem usually go back even further?

I did some board testing before reinstalling the single mosfet and when testing against either the drain or source (cannot remember), I had 160VDC. Obviously if the gate saw that kind of voltage, that would be a problem but I was not sure if this would drop out once any load was applied.

There is a driver daughter board for the mosfets that has several resistors and zeners on it. I am suspecting a zener failure but not too familiar with testing them. So, when a mosfet fails, what does it usually take with it?

The driver seems pretty guilty right now. Did you check the rectifier on the secondary side of the transformer-that is another suspect. I think you are working with a Switch Mode Power Supply(SMPS), it uses Pulse Width Modulation(PWM) to control the mosfets, which send pulses of current through the primary of the transformer. Google Pulse Width Modulation and SMPS to see how they work. When a mosfet fails, it is usually in a high current circuit, so it could take resistors or transistors out easily.

 

[ben7]

I removed all 5 zeners from the driver board circuit for testing. I realize their breakdown should be tested but with a diode test with a DMM, I have .5V on one direction and OL in the other. I am suspecting they are all good at this point. ??thoughts

I think that is normal for zeners. Your DMM doesn't have enough voltage on the probes to make the zener regulate, so the zener may read OL on your DMM. Are you sure those are zeners!? They could be regular diodes too!

 

[fastline]

Yes, I have been removing lots of things from circuit for testing. the bridge works flawlessly. There is a ICL thermistor in series on the AC side just before the bridge and it has sustained damaged. Has 2 pin holes in it but still readin 10 ohms at room temp and drops to near zero with heat. I think it is only their to limit start current for closer fuse matching.

I am familiar with PWM systems for sure but not much in power supplies. My experience is mostly in motor drives.

 

[fastline]

dup. deleted

 

[fastline]

As I was trying to say, I cannot find any dead shorts on the driver daughter board. Mostly resistors, 1 transistor and these diodes. However, I cannot find any IC or similar to control a PWM input to the mosfets so I am real confused as to how that would be working. Another guy I have talked to indicated the PS probably was linear. I am really not sure. The mosfets are obviously there for a reason and I would think you would not need them in a linear supply???

 

[Nigel Goodwin]

As I mentioned ages back, in your previous thread, draw the circuit out and post it - you can't get technical advice on a completely unknown circuit.

 

[fastline]

That simply just is not feasible on this circuit. half the stuff you just cannot even draw out without removing components. I understand what you are saying and even a partial would probably help but I would undoubtedly write something out wrong..

 

[dknguyen]

What on earth did you build it from then? Your mind's eye or winging it as you're going along? There has got to be a schematic you used from somewhere.

EDIT: Oh, I think I see. You're trying to repair the board.

 

[ben7]

As I was trying to say, I cannot find any dead shorts on the driver daughter board. Mostly resistors, 1 transistor and these diodes. However, I cannot find any IC or similar to control a PWM input to the mosfets so I am real confused as to how that would be working. Another guy I have talked to indicated the PS probably was linear. I am really not sure. The mosfets are obviously there for a reason and I would think you would not need them in a linear supply???

Power supply seems to be culprit, I think it is an SMPS too. Can't you just draw the circuit for the power supply part...if not, at least a few good pictures might help!

 

[ben7]

Did you check the tranny's secondary rectifier diode?

Note: it is (should be) a schottky diode too, it's different than a 1N4001