Page 1 of 1

Have I fried my N-drive MOSFET?

Posted: Mon Jul 08, 2013 1:27 am
by jars121

I bought an N-drive MOSFET to turn on/off a 12V strike latch from my Raspberry Pi. The Pi outputs a 3.3V logic signal, which goes to the Gate of the MOSFET. I've since read the documentation on the N-drive from Freetronics, and it says that the maximum Gate voltage is 2.5V...

I've used a voltage-reducing resistor to bring the 3.3V signal down to 1.8V, but the MOSFET still doesn't seem to be working. Would an input of 3.3V have killed it?

Also, just to clarify, this is my setup:

Gate: 0V/1.8V signal coming from pin 4 of Raspberry Pi
Drain: Negative terminal wire coming from 12V strike latch
Source: Circuit ground

The positive wire from the 12V power supply goes to the positive terminal on the strike latch itself, but where does the negative wire from the supply go? I'm assuming it also goes to the circuit ground?


Re: Have I fried my N-drive MOSFET?

Posted: Mon Jul 08, 2013 1:56 am
by cef
Firstly, you shouldn't have killed it.

The MOSFET itself should turn on when the voltage at it's gate is somewhere between the minimum and maximum gate threshold voltage. That means you need between 1.5V and 2.5V at the gate on the MOSFET to switch it from the off to on state (and less when you want to switch it off). You can exceed the max gate threshold voltage, as it's just the threshold where it switches from off to on, and on to off.

It's good practice to make sure you exceed the threshold voltage by a small margin, so that your on/off's are crisp. There are (not many) reasons you may want to drive a MOSFET in the gate threshold area, but for common switching like this, you will find that it's not a good idea, as it's inefficient and the MOSFET will heat up a lot more than it otherwise would.

You shouldn't need an in-line resistor between the input and the gate, as there is one on the N-Drive board (R1, 1k). Adding a resistor may actually reduce the amount of voltage the MOSFET gets and could be causing your issue. R1 and R2 (10k) form a voltage divider, so the input voltage at the gate will be 10/11th's the voltage on the input. At 3.3V that should be 3.0v, which "should" be more than enough to switch the MOSFET on.

If you still can't get it working connected direct to the Pi, and you have a variable supply on hand (0-5V is enough), you can try driving the gate manually using your variable supply. Disconnect the N-Drive from the Pi and try putting a higher voltage on the N-Drive gate input and see if it energizes the latch. It should work with 3.3V, but it will definitely work with a 5V input also. This will at least allow you to pin down what is at fault.

And yes, the negative from the 12V supply to drive the latch needs to go to circuit ground (same as Source in this case).

Re: Have I fried my N-drive MOSFET?

Posted: Mon Jul 08, 2013 2:40 am
by jars121
Legend, thanks for your detailed response.

I posted this same question elsewhere, and it was suggested that I use a pull-up resistor to bring the signal voltage from 3.3V to 5V. I might try this and see if it works. I tried connecting a 5V supply to the Gate, but the latch activates as soon as I connect the negative 12V supply wire to the circuit ground, which is the issue I've been having the entire time...

Does this mean that the current is bypassing the MOSFET entirely and completing the circuit?

I'm so confused haha!

Re: Have I fried my N-drive MOSFET?

Posted: Wed Jul 10, 2013 4:49 am
by cef
Ok, if it's stuck always on, then that's a different sort of issue.

First off, confirm that there is a resistor between the Gate and Source on the board. It should be 10K (if you're going to measure it, disconnect the N-Drive from your circuit first) and is labelled R2.

If this is missing, you may find that the MOSFET will stay switched on all the time.

If you've got a volt-meter, check the voltage on the gate pin of the MOSFET itself, and confirm it varies with your signal from the RPi. Also worth checking that the RPi signal is going up and down as you command it.

Also check that there aren't any bridged connections anywhere on the N-Drive. This could be anything from a small off-cut from a piece of wire to a tiny blob of solder.

You can also do what I suggested previously, but try with Ground on the Gate input to the N-Drive module. In this case it should be "off" and the latch should not energize.

Basically, you have 2 circuits:
Control: RPi goes to the Gate on the N-Drive module, which through the MOSFET goes to the Source, which is connected to Ground (completing the circuit). A voltage less than Vgs(min) should mean the MOSFET is off. A voltage more than Vgs(max) should mean the MOSFET is on.
Output: +12V from the supply goes to one side of the latch (it's an electromagnet). The other side of the latch goes to the Drain on the N-Drive module. The Source of the N-Drive module goes to Ground (which is also the Ground of the 12V supply). The MOSFET should only complete the circuit if the Control side is in the on state, and should not complete it if it's in the off state (*).

* - Note that there are exceptions to this, such as using a large voltage (eg: >60V) across the Drain/Source, etc. Basically exceeding the specifications of the device.