Get custom board with PICO-V3-02 SoC running

[Ing-Dom]

Hey folks,

I designed a custom board with a ESP32-PICO-V3-02 SoC in QFN48 package.
I used a onoard CH340X for programming via USB including reset & bootloader logic.

The result is, Im not able to program it.

esptool.py v4.5.1
Creating esp32 image...
Merged 2 ELF sections
Successfully created esp32 image.
Configuring upload protocol...
AVAILABLE: cmsis-dap, esp-bridge, esp-prog, espota, esptool, iot-bus-jtag, jlink, minimodule, olimex-arm-usb-ocd, olimex-arm-usb-ocd-h, olimex-arm-usb-tiny-h, olimex-jtag-tiny, tumpa
CURRENT: upload_protocol = esptool
Looking for upload port...
Auto-detected: COM7
Uploading .pio\build\esp32dev\firmware.bin
esptool.py v4.5.1
Serial port COM7
Connecting......................................

A fatal error occurred: Failed to connect to ESP32: No serial data received.
For troubleshooting steps visit: https://docs.espressif.com/projects/esp ... oting.html
*** [upload] Error 2

I checked the 3.3V - ok - board is supplied by a 5V to 3,3V LDO. USB analyzer shows 10mA @5V.
I pulled IO0 manually LOW
IO2 is floating

There may by Layout issues, there may be soldering issues, I'm simply stuck because there are to many options what could be wrong.

Lets assume the chip has sufficient power supplay and is like received from the factory.
What should happen when is is powered up and what can I use to check it?
Should I be able to connect via UART0 / esptool and get a connection and see if that chip is alive?
What should be the power consumption with an empty chip?

[aliarifat794]

Maybe the CH340 chip is faulty. Try another one.

[Ing-Dom]

fortunately, I made a Experimentalboard where I can disconnect and connect all parts with jumpers.

So, I have really only the plain ESP32. Some C at the 3,3V pins, thats it. the rest is disconnected.
The strange thing is, the chip does not draw any current (but the pins measure 3,3V).
I checked the VDD pins, all of them have 3,3V (I did not connect VDD3P3_RTC and VDD_SDIO, I think thats is correct.

Pin 9, EN, is also measured 3,3V (external Pullup).

2024-07-10 00_05_08-Window.png (84.19 KiB) Viewed 2330 times

I tried different USB2Serial which I tested with another esp before, but as the chip does not draw current, that is most likely not the cause...

I also tried another chip from different supplier (one from LCSC, one from mouser).

That is driving my crazy!

[ESP_Sprite]

You missed pin 19, that also needs to get 3.3V. Also, the datasheet shows an 100nF capacitor from VDD_SIO to ground.

[Ing-Dom]

thanks for your hint.
You are correct, I didn't connect Pin 19 because I do not need RTC - I thought then it would not be neccessary.
Now I see also flash is sourced by that power domain, so I really should connect it.
I already connected VDD3P3_RTC pin with some wire to 3v3, unfortunately, It didn't solve the problem.


I used that

2024-07-10 21_44_40-Window.png (95.3 KiB) Viewed 2269 times

as reference (https://www.espressif.com/sites/default ... eet_en.pdf).
There is no C shown at VDD_SDIO.
It seems that 100n is already integrated into the SoC (see Figure 8 in the datasheet).

Another hint I got was that MTDI/IO12 must pulled low while boot.
But as stated in Table 9, the default should be pull-down, so that should be fine.

[ESP_Sprite]

Ah, you're right, I may have looked at the wrong datasheet or picture wrt the vdd_io pin.

Some things to test:
- Can you double-check the 'EN' pin of the chip to see if it's actually high?
- If you have a 'scope, can you see if U0TxD has serial output when you power on or reset the chip?
- Are you sure you connected the Pico's TxD to the USB-serial RxD and vice versa?

[Ing-Dom]

EN Pin was indeed the first thing I checked. It is high.

Yes, I can connect a scope to U0Tx, but I'm sure there will be no output because I think the chip cannot output anything when it consumes no power.

- Are you sure you connected the Pico's TxD to the USB-serial RxD and vice versa?

I also double checked that, and as always, tried to swap the pins
But as above, I believe the probleme is more fundamental, it is not the communication but its the power-up of the chip.

I'm quite out of any ideas. I would say the chip is dead, but because I also exchanged that, I would not believe that.

This evening I will try again with additional pull-down on MTDI.

The only thing I could think of is that the GND-Pad has no connection to the chips GND-Pad. But this is very unlikely, because it is soldered with proffessional equipement (solder plate) from an experienced guy.. and also re-soldered it.
Any idea how one could check this?

[ESP_Sprite]

The only thing I could think of is that the GND-Pad has no connection to the chips GND-Pad. But this is very unlikely, because it is soldered with proffessional equipement (solder plate) from an experienced guy.. and also re-soldered it.
Any idea how one could check this?

You can, pretty trivially. Because of a bunch of reasons, every GPIO acts like there is a diode connected to it with the cathode connected to the GPIO and the anode to GND. In other words, with your multimeter in diode mode, you should measure a voltage drop of 0.5-0.7V in one direction. (Note this trick works for 99% of all ICs, it's not ESP32-specific.)

[Ing-Dom]

I finally got it running.
I turned out that the missing connection on Pin 19, VDD3P3_RTC was the reason.

I had some other issues with the "one click download" but I could fix that also.

Thank you all for your help! No I can start getting Ethernet working...