0
\$\begingroup\$

I am using Mosfets to power an RGB LED light strip with a Raspberry Pi:

https://www.sparkfun.com/datasheets/Components/General/RFP30N06LE.pdf

The I though the threshold voltage for the above Mosfet was low enough for the 3.3V supply from the Raspberry Pi to the gate pin but the LEDs are dim (the blue is the dimmest maybe because of different internal resistance in the strip?). I measured the voltage across the drain and source for each of the 3 Mosfets (red, blue, and green) at 7, 6, and 5 volts respectively. I attached the gate pin to the 5V pin on the Raspberry Pi and the lights lit up full brightness. Is it possible to wire up a circuit that uses the Mosfet's I currently have in place at 7, 6, and 5 volts to control the gate pins of three other Mosfets that then power the lights? i.e. I want to open a Mosfet's channel from Drain to Source by providing voltage to it's gain pin from another Mosfet's output. If not possible, is there something I could be doing wrong with the wiring or do I ultimately need to find another Mosfet with even lower threshold voltage? Below is the fritzing diagram I have been using with a 100k pull-down resistor. Power Source is a 12VDC 1A supply

enter image description here

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
  • 1
    \$\begingroup\$ Welcome to EE.SE! No such thing as a gain pin. Do you mean gate? Also, please draw a schematic. Click on edit and the schematic symbol and a fairly full-fledged schematic editor will open. \$\endgroup\$ – winny May 7 '20 at 19:19
  • \$\begingroup\$ Yes, gate pin, trying to apply a gain to the gate pin didn't translate from brain to finger. \$\endgroup\$ – jwd0015 May 7 '20 at 20:21
  • \$\begingroup\$ There is transconductence gain in the MOSFET but it’s a far stretch to apply gain to the gate. Again, please draw a schematic and we’ll help you from there. \$\endgroup\$ – winny May 7 '20 at 20:35
1
\$\begingroup\$

Generally speaking there are many reasons why a microcontroller/single board computer GPIO pin might not be suitable to drive the gate of a MOSFET. You may be able to use a single lower power 'logic level' MOSFET as a gate driver, but I don't recommend this. For a low side driver such as this you will have the best results using a MOSFET gate driver IC, which is a device purpose built for exactly this kind of application.

\$\endgroup\$
  • \$\begingroup\$ Any particular driver for this type Mosfet? \$\endgroup\$ – jwd0015 May 7 '20 at 20:20
  • \$\begingroup\$ This is a shopping question and technically off topic. It's hard to pick one that won't work though, so don't sweat your decision too much. Also, I wouldn't completely rule out the possibility that your circuit isn't working because of some other mistake you've made. You might want to try using another microcontroller (such as an arduino if you have one lying around), and verify that you're really toggling your Raspberry Pi GPIO pins correctly \$\endgroup\$ – Ocanath May 7 '20 at 20:31
  • 1
    \$\begingroup\$ @jwd0015 The TC4427 comes in inverting and non-inverting versions, comes in DIP8 too. microchip.com/wwwproducts/en/TC4427 ... you also don't need the pull-down on the gate to discharge cap since that's built in. nice clean edges, especially on big power n-fets like IR630s. \$\endgroup\$ – PeterT May 7 '20 at 21:22
  • \$\begingroup\$ @alanwaring's recommendation seems solid to me, I second it \$\endgroup\$ – Ocanath May 7 '20 at 23:55
  • \$\begingroup\$ @alanwaring I notice that the TC4427 has an operating range from 4.5 to 18V, does that mean it needs at least 4.5V from the microcontroller or is that just what I need to provide power to the TC4427? \$\endgroup\$ – jwd0015 May 8 '20 at 14:00
0
\$\begingroup\$

Based on the datasheet, your 3.3V gate drive should be enough. Are you absolutely sure that you are using the MOSFET that goes with that datasheet? Have you verified that you are actually getting 3.3V out of the RPi GPIO?

If you want to use the 12V to pull up the gate drive to the MOSFET, you can add another transistor and some resistors to do that. Here is a circuit that does that with a BJT. You could do it with a MOSFET, but I already had this circuit drawn.

schematic

simulate this circuit – Schematic created using CircuitLab

R1 is there so that if the RPi is not connected, the LED is off. Otherwise, it is not necessary. R4's value is chosen high enough to limit the base current of Q1, but low enough so that when the RPi GPIO voltage is low, the base voltage is low enough to turn off Q1.

Edit: Adding a MOSFET version

schematic

simulate this circuit

\$\endgroup\$
  • \$\begingroup\$ The reason I recommended a gate driver is that the circuit you posted will have really fast fall times and really slow rise times. The RGB strip is controlled via. PWM, so for anything besides solid white you will have high switching losses, and potentially a lot of heat build up \$\endgroup\$ – Ocanath May 7 '20 at 21:43
  • \$\begingroup\$ I am certain that is the Mosfet I am using and based on research I too thought it would be capable of handling this task. I measured the voltage from gate to ground and it was 3.25 at max output so unless that 0.05 V is having that big of an effect on the system, I don't know. \$\endgroup\$ – jwd0015 May 7 '20 at 23:08
  • \$\begingroup\$ @Ocanath Good comment. I didn’t know the RGB was PWM driven. This circuit I posted is meant only for very low-frequency or simple on/off switching. \$\endgroup\$ – rpm2718 May 7 '20 at 23:57

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.