search

pimoroni / enviro

MIT License
101 stars 79 forks source link

Enviro Urban won't work, LED always on, It says usb powered when it is powered by batteries. #210

Open A1a01 opened 4 months ago

A1a01 commented 4 months ago

Hello my Enviro Urban works fine for a couple of hours and then it stops working, leaving the white LED turned on. Checking the logs it says it was powered by usb (which is not, it is powered by a LiPo battery) and I think that's the issue with the board. Any solution to the error of usb powered? The logs where it shows usb powered: 2024-02-10 22:48:18 [info / 115kB] - shutting down 2024-02-10 22:48:18 [debug / 113kB] - on usb power (so can't shutdown). Halt and wait for alarm or user reset instead 2024-02-10 22:54:00 [debug / 111kB] - reset 2024-02-10 22:54:03 [info / 138kB] > performing startup 2024-02-10 22:54:03 [debug / 135kB] - running Enviro 0.0.10, MicroPython 856e08b1931b88271816a2f60648f6ff332235b2, enviro v1.20.4 on 2023-08-04 2024-02-10 22:54:03 [info / 115kB] - wake reason: rtc_alarm 2024-02-10 22:54:03 [debug / 113kB] - turn on activity led 2024-02-10 22:54:03 [debug / 108kB] > 99 blocks free out of 212 2024-02-10 22:54:03 [debug / 106kB] > taking new reading 2024-02-10 22:54:03 [info / 133kB] - seconds since last reading: 360 2024-02-10 22:54:04 [debug / 130kB] - starting sensor 2024-02-10 22:54:04 [debug / 128kB] - wait 5 seconds for airflow 2024-02-10 22:54:09 [debug / 109kB] - taking pms5003i reading 2024-02-10 22:54:09 [debug / 106kB] - taking microphone reading 2024-02-10 22:54:16 [info / 119kB] > going to sleep 2024-02-10 22:54:16 [debug / 117kB] - clearing and disabling previous alarm 2024-02-10 22:54:16 [info / 115kB] - setting alarm to wake at 23:00pm 2024-02-10 22:54:16 [info / 113kB] - shutting down 2024-02-10 22:54:16 [debug / 110kB] - on usb power (so can't shutdown). Halt and wait for alarm or user reset instead 2000-01-01 00:00:04 [info / 138kB] > performing startup 2000-01-01 00:00:04 [debug / 135kB] - running Enviro 0.0.10, MicroPython 856e08b1931b88271816a2f60648f6ff332235b2, enviro v1.20.4 on 2023-08-04 2000-01-01 00:00:04 [info / 129kB] - wake reason: usb_powered 2000-01-01 00:00:04 [debug / 127kB] - turn on activity led 2000-01-01 00:00:04 [info / 125kB] > clock not set, synchronise from ntp server

sjefferson99 commented 4 months ago

The USB sense is done in enviro/_init.py It simply reads wlgpio2 (which is connected to vbus) and if there is a voltage then that means a USB cable is supplying voltage. The only other way I know that it would report USB connected is if somehow you have connected the lipo or any other positive rails to vbus (pin40).

I have also seen 100% consistent behaviour when reporting on USB power or not which supports the simplicity of the code.

Any chance you have something connected to vbus somehow?

https://www.raspberrypi.com/documentation/microcontrollers/images/picow-pinout.svg

A1a01 commented 4 months ago

I haven't connected anything to vbus and I tried implementing #144 and #199 and it seemed to work fine for a couple of weeks but today the white LED happened again. I added a photo to show how I found the board. imagen And also the log file doesn't show any indication to what seems to be the issue, 2024-02-22 23:21:43 [info / 108kB] - uploaded 2024-02-22T23_21_35Z.json 2024-02-22 23:21:43 [info / 106kB] > Disconnecting wireless after upload 2024-02-22 23:21:43 [info / 104kB] > going to sleep 2024-02-22 23:21:43 [debug / 102kB] - clearing and disabling previous alarm 2024-02-22 23:21:43 [info / 100kB] - setting alarm to wake at 23:28pm 2024-02-22 23:21:43 [info / 98kB] - shutting down 2024-02-22 23:21:43 [debug / 96kB] - on usb power (so can't shutdown). Halt and wait for alarm or user reset instead 2000-01-01 00:00:05 [info / 131kB] > performing startup 2000-01-01 00:00:05 [debug / 129kB] - running Enviro 0.0.10, MicroPython 856e08b1931b88271816a2f60648f6ff332235b2, enviro v1.20.4 on 2023-08-04 2000-01-01 00:00:05 [info / 123kB] - wake reason: usb_powered 2000-01-01 00:00:05 [info / 121kB] - setting default alarm to wake at 00:03am 2000-01-01 00:00:05 [debug / 118kB] > delayoff set on gpio2 for 120000 ms

sjefferson99 commented 4 months ago

I've had a look at the schematics for the enviro and the pico-w and you should only get a digital 1 on USB if there is significant voltage (quite near to 5v if I've understood the voltage divider on wl_gpio2 and assuming a standard threshold for the pin detecting a 1) on vbus (bottom left in your picture).

Enviro schematic Pico-W power system schematic Page 18

I have measured voltage between ground and vbus and vsys on my board under battery power and I get 4v at vsys (which is the battery output right now) as expected and 0.75v on vbus. I think this is a result of the mosfet connecting vbus and vsys and should be well below the digital 1 input hence my board not reporting on USB power.

If you can measure vbus when it's on battery power and see if you get ~0.75v or something nearer 5v, maybe somehow 3.3v is getting passed to it which could potentially register as a digital 1.

If you have a voltage on vbus on battery only then I think that might be a board fault, if not then that is very mysterious.