Custom ESP32-C3 board bring-up problem

[FeherAron6]

Hello everyone!

I've used ESPs since the venerable ESP8266, but only as pre-built modules.
I wanted to design an ESP32-C3 WiFi solid-state relay (I've attached the schematic and the PCB file).
So far, everything except the power electronics is populated.
I've measured the pins of the ESP32 using a multimeter based on the internal clamping diodes, so everything is soldered correctly.
The power LED is on, so 3.3V is present, yet the USB does not appear on the PC.

I know that I forgot to put pull-up resistors in GPIO2 and GPIO8, but I added them during assembly with wiring. I've tried to connect a USB/UART module to the RX/TX pins, but I see no activity.

My oscilloscope only has passive probes and goes up until 50MHz, so probing the 40MHz crystal is out of the question.

How could I try to debug this annoying problem?
Any help would be appreciated.

[ESP_Sprite]

Can you probe U0TXD (IO7) to see if there's serial chatter there on bootup? Also, the 20pF caps on your crystal seem a bit high; how did you calculate those?

(Also, interesting choice trying to design your own RF circuitry if you only have a 50MHz scope at your disposal... if your antenna system somehow doesn't match, you won't have anything to diagnose why; you normally want a VNA for that.)

[FeherAron6]

Thank you for your reply.

I will edit my post as soon as I check the U0TXD signals.
Regarding the capacitor, yes, that value seemed a bit high to me, but I was using a 40MHz Crystal from LCSC that I had at hand (C5917310), which stated that the maximum load capacitance is 15pF. According to C = 2(CL - Cs) with CL = 15pF, I get 24pF for a stray capacitance of 3pF and 20pF for a stray capacitance of 5pF.

Regarding the antenna, you are right. I don't have the testing equipment at hand, but at the university research lab, we have VNA. I will have access to it in a month or so when I return to work.

I started the assembly and bring-up at home since, with STM, PIC, and NRF MCUs, I have not had problems so far, at least starting the units and being able to program them.

EDIT: I've tested the U0TX with a proper Saleae logic probe, and at reset, it jumps to a high level and stays there.

[ESP_Sprite]

If you have a VNA, you can probably also use it to see if the 40MHz osc is doing its thing. Alternatively, if you have something else that can generate 40MHz, you can insert it on one of the clock pins to see if that makes it come alive. Generally, no response on U0TXD on an otherwise new chip means either a power issue, something wrong with CHIP_PU (ground instead of high level) or an issue with the 40MHz crystal.

[FeherAron6]

So far no luck.

I've tried to change the crystal capacitors 15pF, 10pF - nothing. On other MCUs, I can sense the clock just with a loop above the crystal; here, it's flat.
Since my signal generator goes up to 10MHz only, I configured an STM32 to output 40MHz on its MCO (the frequency and voltage level are good). I tried injecting the signal into the P and N oscillator pins (not both at once), but no luck.
At this point, I am thinking that the IC is broken.

I will have access to proper test equipment in a month, then, I'll try resoldering the ESP, and I'll order a minimal board from AliExpress, just so that I can at least work on the firmware part.