Hello, I'm having trouble with this mofet http://www.irf.com/product-info/datasheets/data/irfp9240.pdf always being on.

I've attached the schematics down below and to me it seems right. It's a passive pull-up resistor between Gate and Source to keep the mosfet off. It is 32VDC at the gate, Vin is 32VDC which means VGS=0V so it should be off, but still it lets voltage through the mosfet... Can it be burnt/shorted even if my multimeter says that there is no connection between Source and Drain(measuring when it's desoldered)? Am I wiring it wrong? Am I missing something?

PS: There is an opto-coupler between R2-R3 and Gate, forgot to put it in the schematics. The resistors in the voltage divider is 1k each, not 1k together.

 

are you sure you did not swap S and D?
http://www.vishay.com/docs/91239/sihfp924.pdf

what is your load? did you consider leakage current?

 

Mosfet is always on without gate connected

It is entirely possible that without the gate connected a mosfet can be fully on.

But your schematics have the gate connected so no one knows what exactly you are talking about.

 

It's like @dannyf said; you cannot simply disconnect a MOSFET gate from its surrounding circuitry to turn it off. With the gate floating the MOSFET may be off, or on, or partially on.

PS: There is an opto-coupler between R2-R3 and Gate, forgot to put it in the schematics.

If that's the case, I expect the MOSFET is being held on by leakage through the opto's output transistor.

If your question hasn't been answered so far, I would suggest posting a complete schematic including all parts, so we can figure out what's going on.

 

Or, as panic mode suggests, he could have the drain and source are swapped, that would mean that the parasitic diode is forward biased when the gate is biased off. The datasheet to whcih he linked does not show which pin is which. I found a VISHAY document that shows that the TO-47 package has the traditional pinout for the more familiar to me, at least) TO-220.



So Aivar, please check your connections against the drawing above (top view, by the way).

 

As panic said, Drain and Source was swapped. >.< I had just checked the datasheet that I linked, which didn't show what pin was Source and Drain, so I used my multimeter to see what way the diode was connected, but I guess I did something wrong. It works now though, but when the mosfet is off it lets 6V through, but that drops to 0V as soon as I connect something, in this case a motor. Could the opto cause this 6V leakage, OBW0549?

 

No current through the opto, removed it completely and the voltage was still there.

 

It works now though, but when the mosfet is off it lets 6V through, but that drops to 0V as soon as I connect something, in this case a motor. Could the opto cause this 6V leakage, OBW0549?

When you say it "lets 6V through," do you mean with nothing connected to the output but your multimeter? If so, MOSFET drain-to-source leakage alone (a few microamps, or even less) could cause your meter to register that much voltage, even with zero gate-to-source voltage. In that case, loading the output even slightly, like with a 10 kΩ resistor, should make that voltage almost disappear.

But do mind the comments above about not leaving the MOSFET's gate floating, ever; there should always be a DC path of some sort between gate and source (even if it's only a 100 kΩ resistor) to keep the MOSFET off when it is not being driven on.

 

Exactly, nothing but my multimeter on the output. I still have the pull-up resistor so it's not floating, correct?

 

Pull-up resistor, where? Between gate and source? If so, that's correct.

 

Regarding what I said in #8 above, looking on the data sheet for the IRFP9240 I see its Idss is specified at 100 μA max, so this could easily be why your meter is registering an output voltage even when the MOSFET is not being driven.

 

Thanks for all the answers everyone, especially you OBW0549! Helped alot.

 

You're welcome, anytime...

 

Next time, post a FULL schematic so we can try to ascertain what you are trying to accomplish.

 

Your second photo in your first post might require something like a 100 ohm resistor in series with the gate.

 

Your second photo in your first post might require something like a 100 ohm resistor in series with the gate.

I've written "1k resistor" under the gate line because I forgot to put a resistor on the schematic, but it's there on the actual mosfet.