ESP32 Pico custom PCB flashing error

[jpfaguirre]

I have developed a custom PCB board for a personal project which I had assembled in China via JLCPCB. I have problems programming the board via UART using a CP2102 board.

I have tested strapping pins:
* GPIO0: LOW (PCB pull-down)
* GPIO2: LOW (PCB pull-down)
* GPIO5: HIGH (internal pull-up)
* GPIO12: LOW (internal pull-down)
* GPIO15: HIGH (PCB pull-up)

When I try to program the ESP32 Pico I push the button connected to EN so it is LOW (measured with multimeter). I put a probe of my oscilloscope (Analog Discovery) in the TX pin of the ESP32 pico and it is always HIGH, except when I press the EN button. In this case I see high frequency noise (MHz) which makes me believe that the chip is not broken.

The schematics can be found in this image:


I have tried powering it with the CP2102 board and also using a 12V 2A connected to the AMS1117-3.3 in my board. In both cases the results were the same.

Any ideas where the problem could be or what else could I test?

[jwktje]

What error do you get when you try to flash it? And how are you flashing it? From ESP-IDF or Arduino IDE?

[chegewara]

GPIO12: LOW (internal pull-down)

It will do only if you burn efuse.
https://github.com/espressif/esptool/wi ... -Selection

[jpfaguirre]

I tried flashing it both with ESP-IDF and Arduino. In both cases the error was timeout. When I reset putting IO0 to GND and pressing enable nothing is sent by the ESP32 via UART (nothing seen on serial terminal nor in the TX Pin using an oscilloscope).

[mspider65]

Connect a terminal like putty and check if you are able to receive some text during boot process.
If not, it could be a problem of the GND PAD soldering. The GND pad footprint shoul not be solid but similar to the one in the following image.

Footprint.png (14.41 KiB) Viewed 5759 times

I had also a similar problem with JLPCB because JLPCB normally doesn't correct automatically footprint errors like this.