0
\$\begingroup\$

As a now-professional EE, I'm slightly embarrassed to be asking what I feel like is a simple question. But I'm stumped and google isn't giving me good answers.

I'm trying to PWM a string of LEDs via a low-side NMOS switch (a PMV40UN2). On the surface this seems like a simple thing to accomplish. However what I'm seeing is that the LED string is always on, regardless of the state of the MOSFET's gate.

Here's the circuit:

enter image description here

My first test was to pull out the LEDs + resistor and throw in a 10K resistor from 12V to the FET's drain, to see if it was a bad solder joint or something else on the PCB causing the issue. Here's the plot for the gate (yellow) and the drain (purple) in that setup:

enter image description here

So the MOSFET is properly pulling its drain to ground when +12V is applied to the gate. Perfect. Now we look at the same test points with the 10k resistor still in palce, but with the string of LEDs and the current limiting resistor added back in:

enter image description here

I can see the drain trying to go high in the off-state, but for whatever reason it's not able to do it and instead stays low. The result is what I originally saw, the LEDs are always on.

I'm an RF engineer by trade, so I don't deal with DC all that often. However it's pretty obvious something simple(?) is going on here that's totally going over my head.

\$\endgroup\$
10
  • \$\begingroup\$ Do you have an NMOS available to try out? Normally an NMOS is connected to the negative rail, or a PMOS is connected to the positive rail when used as a switch. Using a PMOS on the negative rail or (more commonly) an NMOS on the positive rail requires a special gate drive circuit for the MOSFET. If you for some reason want to use a PMOS, you'd normally put it on the high side of the circuit and either attach the control voltage to the positive rail or switch the PMOS with an NMOS drive circuit or MOSFET driver. \$\endgroup\$
    – K H
    Mar 10, 2021 at 2:48
  • \$\begingroup\$ Sorry - that was a typo on my part. PMV40UN2 is an N-channel FET. \$\endgroup\$
    – Patrick
    Mar 10, 2021 at 2:49
  • 1
    \$\begingroup\$ Do you have the oscilloscope ground lead (s) connected anywhere other than a common ground between the 12V supply ground and whatever it is that is producing the PWM? \$\endgroup\$ Mar 10, 2021 at 2:49
  • 1
    \$\begingroup\$ I might've found the culprit. I've actually got 3 identical strings of LEDs + resistors running through the single FET. I've isolated the issue to a single string, which seems to be grounded somewhere else for whatever reason. It's covered in vinyl tape so the fault isn't obvious, but I'll do some digging. \$\endgroup\$
    – Patrick
    Mar 10, 2021 at 3:18
  • 1
    \$\begingroup\$ Close question with your update as your accepted answer. Pls. \$\endgroup\$ Mar 10, 2021 at 5:42

1 Answer 1

0
\$\begingroup\$

Solution was simple and dumb. String of LEDs was being grounded external to the circuit, and I was trusting they were all assembled properly from the get-go. Trust but verify!

\$\endgroup\$
1
  • \$\begingroup\$ Patrick - Hi, Thanks for coming back with the answer to your question. In order to mark the topic as solved, please "accept" your answer (click the "tick mark" next to your answer, to turn it green). Then it is shown as having an accepted answer in various lists, and we don't get nagged for it being a question with an unaccepted answer. Note that you must wait 48 hours after you asked the question before you can accept your own answer. So please come back in a couple of days to accept it. Thanks! \$\endgroup\$
    – SamGibson
    Mar 10, 2021 at 17:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.