I'm using an FQP30N06L to dim a 1W LED being driven by a constant current supply. This MOSFET has a \$V_{GS}\$ threshold of 2.5V, which is why I picked it since I'm driving it from a 3.3V PWM pin on an Arduino. It works fine when the drain-source voltage is under 6V or so. When it gets to 9V, it stops turning on. I always understood N-channel MOSFETS to be the way to go in this scenario precisely because they can be driven from a low voltage relative to drain (as opposed to having to drive down from a high one for a P-channel). What am I missing? What parameter should I be looking at to understand?

\$V_{GS}\$ chart for reference:

enter image description here

Schematic of how I'm using it for reference:


simulate this circuit – Schematic created using CircuitLab

In the schematic above, M1 is getting stuck off, but I can turn it on by bringing Gate to Drain


You need resistors in series with the LEDs.

That is not a 3.3V logic-level MOSFET and cannot reliably be driven fully on with 3.3V. It is specified at 5 and 10V so it's best to drive it with at least 5V. If you don't have 5V available, you might want to buy a different MOSFET.

All Vgs(th) guarantees you is that with 2.5V you'll have at least 250uA flowing with gate tied to drain.

Rds(on) is guaranteed to be < 45m\$\Omega\$ with 5V drive at 25'C (a fair bit higher when hot).

enter image description here

Your graph above is a typical characteristic and not something you can depend on except perhaps the general shape.

  • \$\begingroup\$ sigh. thanks. also - LEDs are 1W, driven off a constant current supply, hence no resistors. Really irritating since SparkFun advertises these as logic level. Trust but verify, I guess. \$\endgroup\$ – kolosy Dec 23 '17 at 0:13
  • 1
    \$\begingroup\$ @kolosy "LEDs are 1W, driven off a constant current supply" um that can make a huge difference to what the mosfet is going to do to... \$\endgroup\$ – Trevor_G Dec 23 '17 at 0:19
  • \$\begingroup\$ @Trevor care to elaborate? I did mention the 1W part in the post. Will add the CC part. \$\endgroup\$ – kolosy Dec 23 '17 at 0:30
  • \$\begingroup\$ @kolosy if you have a constant current driver trying to deliver NmA down that chain you may find discharging the gate drain capacitor rather difficult. \$\endgroup\$ – Trevor_G Dec 23 '17 at 0:32
  • 1
    \$\begingroup\$ Well, 5V is logic level, just not yours. The schematic symbol you used is for a 9V constant voltage supply. There is a different symbol for a constant current supply. CC supply may be why you are seeing strange things happen. \$\endgroup\$ – Spehro Pefhany Dec 23 '17 at 2:04

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.