ESP32 Forum

Hey all, hope you are well

I've made a basic 12V LED strips dimmer using a couple of MOSFET that worked really well. Having 2 power sources for the ESP and LED seems unnecessary, so I added a cheap buck converter to step down the LED power to 5V.

As far as I know the wiring is fine, I measure all the right voltages at the right places. However, the ESP gets really hot after a few seconds of use, then gets buggy due to too much heat. The buck converter is cool, mosfets too, and the Vin measured on the ESP and out the buck is 5.06V. All GNDs are connected, any idea what I do wrong?

In case it matters:

- MOSFET: RFP30N06LE 30A 60V N-Channel Mosfet
Gate: 10k res, GPIO pin
Drain: LED GND wire
Source: common GND

- Buck converter: MP1584EN
Input voltage: 4.5 V to 28 V; Output voltage: 0.8 V to 20 V.
Output current: 3 A (maximum); Conversion efficiency: 92% (maximum).
Output ripple: less than 30 mV; Switching frequency: 1.5 MHz (highest), typically 1 MHz.


Thanks for the help, just discovered ESP32 and I'm having a ton of fun with it!

Hello, first I want to make sure power to ESP is 3.3 V, right? Second, GPIO of ESP will not directly connect to 5 V interface, right? Third, could you show me schematic how you drive LED strips? Finally, could you measure the current provided by ESP?

1- Power to ESP comes into the 5V VIN pin, and should be regulated down to 3.3 if I understand correctly.

2- GPIOs 14 & 27 (white wires) are connected to the MOSFET gate pins

3- Wiring is basically like this with an esp instead of the arduino, I can make a custom schematic after work http://adam-meyer.com/arduino/images/20 ... htbulb.png

4- I measure 3.3 between the 3v3 and GND pins of the esp, and 5v between VIN and GND (red and yellow in pic) also measuring 5v out the buck converter, 12v in, and 12v in the LEDs

The 1-1.5MHz (plus harmonics) ripple from the buck converter can't be good for the ESP. I suggest to add a few capacitors between the ESP's 5V and GND inputs, preferably ceramics because electrolytics won't do much @ 1MHz, to try and reduce the interference on the supply line. Additionally/alternatively an inductor in the 5V line may also help.

Ha, spot on, each capacitor I added made it noticeably cooler. Thanks @microcontroller, it did the job!