computer not detecting ESP32C3(com and USB)

[LeeHooo]

Hello everyone, I get a tough problem.I am a newbie of esp32 so this problem has been bothering me for a long time.

The situation is: I drew a board with esp32c3(the chip is purely new),and only use USB(have no uart in my board,only usb) to load my code,no uart was used.Then I hooked it up to the computer and it just wouldn't recognize the usb device and there was no serial port recognition either. here is my board

1726742092163.png (64 KiB) Viewed 1218 times

1726742092163.png (64 KiB) Viewed 1218 times

I drew a board with an esp32c3 (the chip is purely new and nothing else) and the only code loaded on the board mainly was usb, no uart was used. then I hooked it up to the computer and it just wouldn't recognize the usb device, and there was no serial port recognition either.

First of all, I think it might be a soldering problem, I tried it on a couple new boards and it didn't work. (But it worked sometime a long time ago, downloaded the program and then it didn't work again, almost the same circuit.) but subsequent consistent failures make me unclear if that's the cause.
Next may be the crystal problem (not familiar with it), I used passive crystal 40m,then tried 8uf,15uf and 22uf respectively, but none of them work at the moment. xtal_p and xtal_n pins are 0.82v and 0v (may not have been vibrating?). A few times before it was 0.82v on both pins), but I'm not sure if this is related to the USB not being recognized or not, as I see the manual provides the USB with that 3P3 power supply.

Does anyone have some ideas?Any thoughts would be appreciated.

Thanks

[ESP_Sprite]

Can you post your entire schematic? Did you check your power supply? Is CHIP_EN high?

[LeeHooo]

Thank you for your reply , ESP_Sprite

Here is my entire schematic:

607f1f72f545df9c4f7c76ed9dd13e4e.png (178.28 KiB) Viewed 1061 times

And my part of pcb (I am not sure it has some problem or not)

2ab9be0c1d054b565b00f561bddbe643.png (284.88 KiB) Viewed 1061 times

There's nothing wrong with the power supply. I tested it with a multimeter.And the CHIP_EN pin is high(3.3V)

I then re-soldered it a few times and the crystal is currently vibrating and the pins are all around 0.8V. But there is a phenomenon, that is, when I use the multimeter probe or directly by hand to cover the crystal xtal_p connected pins, the computer does not report an error (error of not recoginized device), but nothing can be recognized. I don't know why.
There is still a question, is it possible to burn a brand new chip without firmware directly with USB, or do I have to use the serial port to burn it first?

[Mia Barn]

Hey, I think the board or port is not fully installed on your PC. Have you checked the drivers?

[lbernstone]

It is most definitely possible to burn a fresh chip over the usb pins, and if you hold down the boot pin, you should at least see it show up as a usb device when the device boots.
Comments on your design:
1) You are using pins that are already in use by the onboard flash. From the datasheet:

Screenshot from 2024-09-20 21-30-39.png (51.9 KiB) Viewed 1022 times

2) Your board is a bit light on capacitors. C11 on the EN pin should be 1uF. This is a boot delay (10k/1uF is 10 msec tau).
3) You should have some bypass capacitors at the load points on the cpu and lora (might as well be 10uF since you already have them on board).
For testing, hook it up to a bench power that shows how much draw you are getting. You should see the board draw ~100mA, then drop 60-70 stable if it is actually booting. If it never draws more than a few mA, then the crystal is not ticking.

[LeeHooo]

Hey, I think the board or port is not fully installed on your PC. Have you checked the drivers?

Yes, my computer has the relevant drivers installed.

1727026403823.png (48.41 KiB) Viewed 786 times

[LeeHooo]

It is most definitely possible to burn a fresh chip over the usb pins, and if you hold down the boot pin, you should at least see it show up as a usb device when the device boots.
Comments on your design:
1) You are using pins that are already in use by the onboard flash. From the datasheet:
Screenshot from 2024-09-20 21-30-39.png
2) Your board is a bit light on capacitors. C11 on the EN pin should be 1uF. This is a boot delay (10k/1uF is 10 msec tau).
3) You should have some bypass capacitors at the load points on the cpu and lora (might as well be 10uF since you already have them on board).
For testing, hook it up to a bench power that shows how much draw you are getting. You should see the board draw ~100mA, then drop 60-70 stable if it is actually booting. If it never draws more than a few mA, then the crystal is not ticking.

Actually, after holding down the boot pin, it really doesn't recognize the USB device over and over again anymore. Rather, it only remained unrecognized.
And then 1) I actually hadn't connected the lora when I burned it, so these pins were floating
2) I tried 10uf, it didn't work
3) Meaning replace that 0.1uf capacitor with 10uf?
Then measure the on-board current can not be connected to a component to see, my board power is too small adjustable power supply can not see!

[LeeHooo]

oH!! thank god,I slove it in the end.
A couple solutions to possible causes:
On the original board 1. I replaced all the 100nf capacitors with 10uf (could be a current issue Don't really know but it works,maybe men who know can help explain)
2. still soldered falsely (USB differential circuits are not equal length but not a big problem)
By the way, I'm using an 8pf capacitor for the crystal.

[lbernstone]

In future designs, you should have both a 100nF and a 10uF capacitor. The 10uF is buffering energy and preventing voltage dips. The 100nF is acting as a line filter, preventing noise from causing stray signals (at a specific frequency).
The capacitance on the crystals is going to affect your timing, so the way to tune this is to hook it up to an oscilloscope and make sure that your 40MHz is giving you as close to that rate as possible. The rated capacitance on the part is the suggested starting point, and will increase based on how difficult the pathway is. There is a pretty decent range where the capacitors are going to be OK- if they are too far out of range you will probably notice this most as having issues with WiFi, or corrupted bits in SPI.