5V Power supply problems.

Thread Starter

electronice123

Joined Oct 10, 2008
339
I have a mastech power supply which has (2) 30V 0-5A connections and (1) 5V connection.

I am using a function generator to pulse a NTE2987 FET which drives a small coil.

I have been having problems with the FET turning on even when the function generator is not connected. I finally traced it down to the power supply 5V connections.

So, how do I solve this probelm? Do I need to add a cap to across the 5V power supply terminals? Should I just try a seperate power supply altogether?

Thanks in advance for any help!
 

Thread Starter

electronice123

Joined Oct 10, 2008
339
After more testing the FET will turn on even when not connected but when I touch the gate. When I connect the function generator to the gate the fet turns on even when the function generator is not on.

It also turns on when the FET gate is connected to either terminal of the 5V power supply?

Don't know what I'm doing wrong here, but I am very confused?

Also, after I drew the circuit up I did realize I forgot to add a diode to the coil....Perhaps I just destroyed my FET?

On the drawing I only show the outputs from the function generator as it has it's own power supply.
 

Attachments

bertus

Joined Apr 5, 2008
22,270
Hello,

What voltage does the function generator have as output?
I do not see a connection from the generator to the source connection of the fet.
The voltage should be applied between the source and the gate of the fet.
This voltage should be 5 Volts min. to have the fet fully on.
Also a back EMF diode should be used across the relays coil.
The cathode should go to the + side of the coil.

Bertus
 

Thread Starter

electronice123

Joined Oct 10, 2008
339
The function generator outputs a square wave that is about 4.9V high and -.1V low.

I don't have any connection from the function generator to the source of the FET...I didn't know that I needed one since the coil uses a higher voltge and current than the function generator can provide.

I will try that and add a diode and see what happens. Thanks Bertus!
 

crutschow

Joined Mar 14, 2008
34,280
You can not leave the gate of a MOSFET floating as the MOSFET will be in an indeterminate state. Place a resistor (say 10k ohm) from the gate to source to insure that the gate never floats.
 

JMac3108

Joined Aug 16, 2010
348
Do what Crutschow said!

You've discovered something important ... ALWAYS (or almost always) put a pull-down resistor on the gate of your N-channel MOSFETS, or a pull-up on the gate of your P-channel MOSFETs. If not, the gate can float high and turn on the MOSFET. 10K or even 100K works fine.

I've been designing with MOSFETs for many years and always include the resistor. In my current design I left them off in a couple places where I was sure that there was never a condition where the gate was not being driven. I did this because the space constrains on my board were so severe that even a single 0402 SMD resistor took up more room than I could spare. WELL, it worked fine in most cases, but in one case I had not considered a certain condition that could occur - and of course it did occur and I smoked a MOSFET. Should have squeezed in that resistor!
 
Top