RU-V207/rover-pi
5
1
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

rpi-usb-gadget works better but needs an offline-switch

Browse Source
This commit is contained in:
foley
2026-09-06 20:07:41 +00:00
parent 39c0f158fe 040a6ca33f
commit b0d632f0c5
12 changed files with 486 additions and 122 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+12 -2
View File
@@ -6,13 +6,22 @@ Scripts and documentation for setting up a Raspberry Pi (Zero W 1 or 2) for use
1. [Raspberry Pi Imager procedure to setup ssh and Raspberry Pi Connect](imager.md)
1. Find out your Pi's ethernet addresss in case you need to find it on `iot-research` again. Type `ip addr show dev wlan0` and for something like `link/ether 01:23:45:67:89:ab`. Write this information (the 12 characters and colons) in your notebook. In T-411-MECH we want you to put this information in a spreadsheet links from the main page: this allows us connect Pi's to people in case of any issues.
1. [Solder the GPIO header pins](solder-gpio.md)
1. [Gadget Mode to setup a serial connection over USB](gadget.md)
1. [Marcel's script on github to show the IP address on the OLED screen](https://github.com/mkyas/cpsy-display-ip)
1. Login to the pi using ssh or serial. (Pi Connect breaks wihle upgrading the Pi Connect program). Now update the packages on your pi:
```
sudo apt-get update
sudo apt-get -y upgrade
```
1. [Gadget Mode to setup a network connection over USB](https://github.com/raspberrypi/rpi-usb-gadget)
1. [Setup the OLED screen to print the IP](oled-display/README.md)
Based upon [Marcel's script on github to show the IP address on the OLED screen](https://github.com/mkyas/cpsy-display-ip)
1. Install Blinka by following the guide [CircuitPython Libraries on Linux and Raspberry Pi](https://learn.adafruit.com/circuitpython-on-raspberrypi-linux/installing-circuitpython-on-raspberry-pi)]
## RU student-only guides at ROB-IOT
- [Starting on Raspberry Pi Zero](https://reykjavik.instructure.com/courses/6589/pages/starting-on-raspberry-pi-zero-w)
## References and Links
- [Sparkfun GPIO pinout](https://cdn.sparkfun.com/assets/learn_tutorials/6/7/6/PiZero_1.pdf)
- [Guide to creating ssh keys](https://pimylifeup.com/raspberry-pi-ssh-keys/) Of note, this guide is a little dated as it suggests using RSA keys and many people are moving toward EDCSA keys.
- [Raspberry Pi Connect](https://connect.raspberrypi.com)
- [Setting up Raspberry Pi Connect](https://www.raspberrypi.com/documentation/services/connect.html)
@@ -29,6 +38,7 @@ Scripts and documentation for setting up a Raspberry Pi (Zero W 1 or 2) for use
## TODO
- Finding the Pi using `nmap` or `arp`
- Create a interface HAT that makes it easy to attach to a digital analyzer (v207) and an Analog Discovery
# Some Tricks
Here is a page of tricks we have discovered that may involve advanced techniques.
Scripts/gadget.sh
+14
View File
@@ -9,11 +9,25 @@ if [ $(id -u) -ne 0 ]; then
exit 1
fi
if [ -f /boot/firmware/GADGET_MODE ]; then
systemd-cat -t gadget.sh -p info echo "/boot/firmware/GADGET_MODE exists: GADGET mode enabled"
else
systemd-cat -t gadget.sh -p warning echo "/boot/firmware/GADGET_MODE missing: GADGET mode disabled"
systemctl disable serial-getty@ttyGS0.service
exit 1
fi
# Is the dwc2 overlay installed? Nothing will work if it isn't.
<<<<<<< HEAD
if lsmod | grep -q "dwc2"; then
echo "DWC2 module loaded, setting up OTG gadget serial and ethernet" | logger -t $0 -p info
echo "DWC2 module loaded"
=======
if [ -d /sys/class/udc ] && [ -n "$(ls -A /sys/class/udc)" ]; then
systemd-cat -t gadget.sh -p info echo "DWC2 precompiled, setting up OTG gadget serial and ethernet"
elif lsmod | grep -q "dwc2"; then
systemd-cat -t gadget.sh -p info echo "Pre 6.18.39 kernel. DWC2 module loaded, setting up OTG gadget serial and ethernet"
>>>>>>> master
else
echo "DWC2 module missing. Aborting OTG setup." | logger -t $0 -p info
echo "DWC2 module missing"
Scripts/serial-ether-gadget.service
+2 -1
View File
@@ -1,6 +1,7 @@
[Unit]
Description=Serial and Ethernet over OTG USB
After=systemd-user-sessions.service
#After=systemd-user-sessions.service
Before=serial-getty@ttyGS0.service
[Service]
ExecStart=/root/rover-pi/Scripts/gadget.sh
[Install]
bluetooth.md
+79
View File
@@ -0,0 +1,79 @@
# Bluetooth on the Raspberry Pi Zero W (1 or 2)
The Zero W series has Bluetooth capability because of the Wifi chip on the board.
**Warning: This is a work in progress!!!**
# Reference
- "Exploring Raspberry Pi" by Derek Molloy pg 537-544
- [Connecting to a Headless Raspberry Pi using Bluetooth](https://medium.com/@tomw3115/connect-to-a-headless-raspberry-pi-using-bluetooth-0e61c05e1b68)
- https://github.com/sorah/bluetooth-getty
- [Use dbus insteadd of sdptool](https://linuxvox.com/blog/bluez-adding-services-attributes-and-profiles-without-sdptool-command/)
# Getting Bluetooth working
1. Login to the pi via a USB-serial port or monitor-keyboard-mouse. Configuring bluetooth can sometimes kick you off the wifi.
1. Tell `rfkill` to unblock bluetooth. Perhaps this is a security thing?
```
sudo rfkill unblock bluetooth
```
1. Restart the bluetooth service and check it's status
```
sudo systemctl restart bluetooth
sudo systemctl status bluetooth
```
1. Figure out the address using `hcitool`
```
hcitool dev
```
1. If you have a laptop set to be discoverable, you can check if bluetooth is working
```
hcitool scan
```
1. You can then query that device for what services it offers
```
sdptool browse ADDRESS
```
1. Now we want to make the Pi discoverable to our phone, tablet, or laptop
```
sudo hciconfig hci0 piscan
sudo hciconfig hci name RaspberryPi
```
1. A Serial Port Profile(SPP) is needed for our serial port
```
sudo sdptool add SP
```
1. For some reason, we have to put the bluetooth daemon in compatibility mode. (True as of 2016-2026 --foley). Find the line with `ExecStart` and put `--compat` at the end
```
sudo nano /lib/systemd/system/bluetooth.service
```
1. Test the SPP is setup
```
sudo sdptool browse local
```
1. Time to pair it with a device
```
sudo bluetoothctl
discoverable on
agent on
pairable on
scan on
```
and once you have paired it
```
discoverable off
exit
```
1. If this works, you can install the `bluetooth.services` and `rfcomm.service` files to automate some of this setup.
```
sudo cp bluetooth.service /etc/systemd/system/bluetooth.target.wants/.
sudo cp rfcomm.service /etc/systemd/system/.
sudo systemctl daemon-reload
sudo systemctl enable rfcomm
sudo reboot
```
# Discussion
Apparently Bluez's `sdptool` is now depricated and they are telling people to use this new GATT via `busctl` or `dbus-send`. [linuxvox article on depricated sdptool](https://linuxvox.com/blog/bluez-adding-services-attributes-and-profiles-without-sdptool-command/)
gadget.md
+146 -119
View File
@@ -1,119 +1,146 @@
# Gadget Mode to setup a serial connection over USB
Prepare Raspberry Pi OS connection over a microUSB serial cable
Based upon Adafruit guide: https://learn.adafruit.com/turning-your-raspberry-pi-zero-into-a-usb-gadget/serial-gadget:
The author discovered the guide was out of date, consulted the [2025 Raspberry Pi whitepaper on OTG](https://pip-assets.raspberrypi.com/categories/685-app-notes-guides-whitepapers/documents/RP-009276-WP/Using-OTG-mode-on-Raspberry-Pi-SBCs), found additional errors, and was finally able to correct them thanks to this blog https://www.isticktoit.net/?p=1383
## Important :
- **Using the Gadget Mode will make the usb port unable to connect to a mouse or keyboard!!!** You will need to disable it make them work again.
- You will want to disable the serial console on RX/TX in `raspi-config` if you are trying to use that serial port for communicating with other devices.
## Procedure
Make sure you have your USB cable plugged into the left port **USB** and not the right port marked **PWR**.
1. Open a connection to the device via ssh or Raspberry Pi Connect
1. Become root (the superuser)
```
sudo su
```
1. Enable the DWC version 2 USB peripheral device tree overlay by appending it to the end after the `[all]` category (which is at the end)
```
echo "dtoverlay=dwc2" >> /boot/firmware/config.txt
```
1. We add the required module `dwc2` for loading at startup
```
echo "dwc2" >> /etc/modules
```
1. Check that we can load the libcomposite module
```
modprobe libcomposite
```
1. Put a copy of this repository in root's home directory using git
```
apt install -y git
git clone https://gitea.cs.ru.is/RU-V207/rover-pi.git
```
1. Go to Scripts and test if the `gadget.sh` runs without errors and hopefully a Serial port appers in Device Manager or equivalent
```
cd rover-pi/Scripts/
./gadget.sh
```
1. Install the systemd
```
cp serial-ether-gadget.service /lib/systemd/system/.
```
1. Enable the service
```
systemctl enable serial-ether-gadget.service
```
1. Enable the getty (which connects a login terminal to the serial port)
```
systemctl enable serial-getty@ttyGS0.service
```
1. Reboot and see if the serial port appears
# Connecting to the serial terminal
You will need a serial terminal program (which used to be very very common) in order to communicate over your serial port
## Windows
- Putty: https://www.chiark.greenend.org.uk/~sgtatham/putty/latest.html
- MobaXterm: https://mobaxterm.mobatek.net/
## Linux
If using a Linux VM, you will need to pass the Pi USB connection there.
This is similar to the procedure for accessing an Arduino.
1. Start the Linux VM if you haven't already
1. Menu: Devices > USB
1. Look for "Silicon Labs CP210x USB to UART Bridge" and click on it.
1. check if port /dev/ttyUSB0 is found. The Pi will most often be accessible over this port.
```
ls -l /dev/ttyU*
```
1. The user must be in the dialout group to access it. Use the Linux command: `uid` -to check if you are in the group. If not in the dialout group, use:
```
sudo usermod -a -G dialout <username>
```
1. You will use the GNU `screen` application in Linux to access the Pi. You can install it with:
```
sudo apt install screen
```
1. You should now gain access to the Pi with:
```
` sudo screen /dev/ttyUSB0 115200
```
1. Press enter and you will be prompted to log in to the Pi with default credentials. User: pi - PW: raspberry
# Disabling the OTG mode to use a keyboard and mouse again
**WARNING: Untested**
1. Comment out the dwg2 line in `/boot/firmware/config.txt` using a `#` in front of the line
```
[all]
#dtoverlay=dwg2
```
1. Comment out the dwg2 line in `/etc/modules` using a `#` in front of the line
```
#dwg2
```
1. Disable the gadget service
```
systemctl disable serial-ether-gadget.service
```
1. Disable the getty
```
systemctl disable serial-getty@ttyGS0.service
```
# Connecting over the USB ethernet to ssh into your Pi
I have not tested this procedure because the serial console and normal network connection is sufficient for me --foley
https://www.raspberrypi.com/news/usb-gadget-mode-in-raspberry-pi-os-ssh-over-usb/
# Untested
Here are items that bear investigating but haven't been tested yet
- [Modify Imager manifest to install otg](https://github.com/raspberrypi/rpi-imager/tree/main/doc/local_json)
# Gadget Mode to setup a serial connection over USB
Prepare Raspberry Pi OS connection over a microUSB serial cable
Based upon Adafruit guide: https://learn.adafruit.com/turning-your-raspberry-pi-zero-into-a-usb-gadget/serial-gadget:
The author discovered the guide was out of date, consulted the [2025 Raspberry Pi whitepaper on OTG](https://pip-assets.raspberrypi.com/categories/685-app-notes-guides-whitepapers/documents/RP-009276-WP/Using-OTG-mode-on-Raspberry-Pi-SBCs), found additional errors, and was finally able to correct them thanks to this blog https://www.isticktoit.net/?p=1383
## Important :
- **We are still working out the kinks with the gadget mode switching on the Pi Zero W 2**
- **Using the Gadget Mode will make the usb port unable to connect to a mouse or keyboard!!!** You will need to disable it make them work again.
- The blue USB cables included in the kit in 2026 are flakey. We recommend replacing them ASAP.
- You will want to disable the serial console on RX/TX in `raspi-config` if you are trying to use that serial port for communicating with other devices.
## Procedure
Make sure you have your USB cable plugged into the left port **USB** and not the right port marked **PWR**.
1. Open a connection to the device via ssh or Raspberry Pi Connect
1. Become root (the superuser)
```
sudo su
```
1. Check that we can load the libcomposite module
```
modprobe libcomposite
```
1. Reboot
```
restart now
```
1. Become root again
```
sudo su
```
1. Put a copy of this repository in root's home directory using git
```
cd /root
apt install -y git
git clone https://gitea.cs.ru.is/RU-V207/rover-pi.git
```
1. Go to Scripts and test if the `gadget.sh` runs without errors and hopefully a Serial port appers in Device Manager or equivalent
```
cd rover-pi/Scripts/
./gadget.sh
```
1. Install the systemd
```
cp serial-ether-gadget.service /lib/systemd/system/.
```
1. Enable the service
```
systemctl enable serial-ether-gadget.service
```
1. Enable the getty (which connects a login terminal to the serial port)
```
systemctl enable serial-getty@ttyGS0.service
```
1. Reboot and see if the serial port appears
# Connecting to the serial terminal
You will need a serial terminal program (which used to be very very common) in order to communicate over your serial port
## Windows
- Putty: https://www.chiark.greenend.org.uk/~sgtatham/putty/latest.html
- MobaXterm: https://mobaxterm.mobatek.net/
## Linux
If using a Linux VM, you will need to pass the Pi USB connection there.
This is similar to the procedure for accessing an Arduino.
1. Start the Linux VM if you haven't already
1. Menu: Devices > USB
1. Look for "Silicon Labs CP210x USB to UART Bridge" and click on it.
1. check if port /dev/ttyUSB0 is found. The Pi will most often be accessible over this port.
```
ls -l /dev/ttyU*
```
1. The user must be in the dialout group to access it. Use the Linux command: `uid` -to check if you are in the group. If not in the dialout group, use:
```
sudo usermod -a -G dialout <username>
```
1. You will use the GNU `screen` application in Linux to access the Pi. You can install it with:
```
sudo apt install screen
```
1. You should now gain access to the Pi with:
```
` sudo screen /dev/ttyUSB0 115200
```
1. Press enter and you will be prompted to log in to the Pi with default credentials. User: pi - PW: raspberry
# Enabling/Disabling the OTG mode to use a keyboard and mouse again
## Disable
1. Remove the SD card and access it on your computer
1. Delete the file called `GADGET_MODE` in the boot partition/drive usually called `bootfs`.
1. If the Pi is booted, you can `sudo rm /boot/firmware/GADGET_MODE` and reboot.
## Enable
1. Remove the SD card and access it on your computer
1. Create a file called `GADGET_MODE` in the boot partition/drive called `bootfs`.
2. If the Pi is booted, you can `sudo touch /boot/firmware/GADGET_MODE` and reboot
# Connecting over the USB ethernet to ssh into your Pi
I have not tested this procedure because the serial console and normal network connection is sufficient for me --foley
https://www.raspberrypi.com/news/usb-gadget-mode-in-raspberry-pi-os-ssh-over-usb/
# Untested
Here are items that bear investigating but haven't been tested yet
- [Modify Imager manifest to install otg](https://github.com/raspberrypi/rpi-imager/tree/main/doc/local_json)
# Old DWC Instructions (for older versions)
## Setting up
1. Enable the DWC version 2 USB peripheral device tree overlay by appending it to the end after the `[all]` category (which is at the end)
```
echo "dtoverlay=dwc2" >> /boot/firmware/config.txt
```
1. We add the required module `dwc2` for loading at startup
```
echo "dwc2" >> /etc/modules
```
## Disabling the OTG mode to use a keyboard and mouse again
**WARNING: Untested**
1. Comment out the dwg2 line in `/boot/firmware/config.txt` using a `#` in front of the line
```
[all]
#dtoverlay=dwg2
```
1. Comment out the dwg2 line in `/etc/modules` using a `#` in front of the line
```
#dwg2
```
1. Disable the gadget service
```
systemctl disable serial-ether-gadget.service
```
1. Disable the getty
```
systemctl disable serial-getty@ttyGS0.service
```
imager.md
+9
View File
@@ -44,3 +44,12 @@ We don't have many monitors and keyboards in V207 for students to use, so we wil
![imager-writeimage-verifying.png](Graphics/imager-writeimage-verifying.png)
1. The process is complete! Now you can [connect to your pi](connect.md)
![imager-complete.png](Graphics/imager-complete.png)
1. Take the SD card and put it into your pi.
1. When you first boot it, it can take 10 minutes to finish configuring itself. Usually when the LED stops blinking for at least 30 seconds, it is done.
## Connecting using Raspberry Pi Connect
1. Login to https://connect.raspberrypi.com if you haven't already
1. Look at the list of devices, find the one you want, and click on "Connect"
![connect-devices.png](Graphics/connect-devices.png)
1. A remote shell window will pop up
![connect-shell.png](Graphics/connect-shell.png)
imu.md
+13
View File
@@ -0,0 +1,13 @@
# The Intertial Measurement Unit
As of 2025, we started using an AliExpress version of https://wolles-elektronikkiste.de/en/icm-20948-9-axis-sensor-part-i
This has an unmarked jumper on the back. Joe believes that it shorts the VDDIO and VDD pins in order to allow it to be used with 1.8V devices.
Marcel has tested that the pads connect to both of those pins.
The IMU runs on 1.8V, but our Pi has GPIO voltages of 3.3V. Marcel reports that students in 2025 had no problem using it as long as AD0 and NCS were connected to VDD.
According to the specification, the maximum pin voltage is VDDIO+0.5, which would put it at 2.3, a whole volt below 3.3. On his board, he connected the pins to 3.3V though a 10K resistor to protect it from overcurrent.
For I2C at least, it appears that we are saved by the fact that it is open-drain. Each device communicates to the master by pulling lines low, which should work with both voltages.
oled-display/LICENSE
+21
View File
@@ -0,0 +1,21 @@
MIT License
Copyright (c) 2024 Marcel
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
oled-display/README.md
+38
View File
@@ -0,0 +1,38 @@
# cpsy-display-ip
Demonstrates the SSD1306 OLED by showing the IP address on the system boot.
Originally at https://github.com/mkyas/cpsy-display-ip and customized for Mechatronics 1 students.
This project contains the file `display-ip.py` that displays an IP
address on the SSD1306 OLED screen and the file `display-ip.service`
that executes the script on boot after the Raspberry Pi is connected
to a network.
To use the script, you first must connect the screen to the Pi's
I2C bus.
## Setup
Do the following steps on your Raspberry Pi.
1. Enable I2C in `raspi-config`
1. Go to your home directory: `cd ~`
1. Clone this repository to your raspberry pi `git clone https://gitea.cs.ru.is/RU-V207/rover-pi`
1. Install [pillow](https://python-pillow.org/), numpy, and venv:
> sudo apt install python3-venv python3-pil python3-numpy
1. Create a virtual environment, here called cpsy:
> python3 -m venv cpsy
1. Edit the generated `pyvenv.cfg` to set `include-system-site-packages = true`
1. Activate the environment `source cpsy/bin/activate`
1. Install the display driver: `pip install adafruit-circuitpython-ssd1306`
1. Edit `display-ip.service`:
- Replace `foley` with your user name on the Pi.
- the `WorkingDirectory` points to the directory containing `display-ip.py`
- the path to python points to your virtual environment
1. Copy the edited `display-ip.service` to `/etc/systemd/system`
```sudo cp display-ip.service /etc/systemd/system/.```
1. Enable the service `sudo systemctl enable display-ip.service`
1. Test by starting the service `sudo systemctl start display-ip.service` which should turn on the screen.
1. If nothing happens, check on the status with `sudo systemctl status display-ip` and look for error messages.
1. If you make additional changes to the service file, you may have to reload the `systemd`:
```sudo systemctl daemon-reload ```
oled-display/display-ip.py
+78
View File
@@ -0,0 +1,78 @@
# SPDX-FileCopyrightText: 2021 ladyada for Adafruit Industries
# SPDX-License-Identifier: MIT
"""
This demo will fill the screen with white, draw a black box on top
and then print the device's IP address in the center of the display
This example is for use on (Linux) computers that are using CPython with
Adafruit Blinka to support CircuitPython libraries. CircuitPython does
not support PIL/pillow (python imaging library)!
"""
import board
import digitalio
from PIL import Image, ImageDraw, ImageFont
import adafruit_ssd1306
import socket
# Define the Reset Pin
oled_reset = None
try:
oled_reset = digitalio.DigitalInOut(board.D4)
except:
pass
# No reset pin
# Change these
# to the right size for your display!
WIDTH = 128
HEIGHT = 64
BORDER = 5
# Use for I2C.
i2c = board.I2C() # uses board.SCL and board.SDA
oled = adafruit_ssd1306.SSD1306_I2C(WIDTH, HEIGHT, i2c, addr=0x3C, reset=oled_reset)
# Clear display.
oled.fill(0)
oled.show()
# Create blank image for drawing.
# Make sure to create image with mode '1' for 1-bit color.
image = Image.new("1", (oled.width, oled.height))
# Get drawing object to draw on image.
draw = ImageDraw.Draw(image)
# Draw a white background
draw.rectangle((0, 0, oled.width, oled.height), outline=255, fill=255)
# Draw a smaller inner rectangle
draw.rectangle(
(BORDER, BORDER, oled.width - BORDER - 1, oled.height - BORDER - 1),
outline=0,
fill=0,
)
# Load default font.
font = ImageFont.load_default()
# Get the local IP address as a string.
with socket.socket(socket.AF_INET, socket.SOCK_DGRAM) as s:
s.connect(("8.8.8.8", 80))
text = s.getsockname()[0]
# Draw it into a box.
bbox = font.getbbox(text)
(font_width, font_height) = bbox[2] - bbox[0], bbox[3] - bbox[1]
draw.text(
(oled.width // 2 - font_width // 2, oled.height // 2 - font_height // 2),
text,
font=font,
fill=255,
)
# Display image
oled.image(image)
oled.show()
oled-display/display-ip.service
+13
View File
@@ -0,0 +1,13 @@
[Unit]
Description=display-ip
After=network-online.target
[Service]
Type=simple
User=foley
WorkingDirectory=/home/foley/rover-pi/oled-display
ExecStart=/home/foley/cpsy/bin/python3 display-ip.py
Restart=never
[Install]
WantedBy=multi-user.target
pwm.md
+61
View File
@@ -0,0 +1,61 @@
# Getting the PWM working
The Raspberry Pi zero (1 and 2) have 2 PWM peripherals, allowing you to pulse pins without the CPU doing any additional work. This is particuarly useful for controlling motor speeds when connected to a motor driver/H-bridge chip.
# References
- [CircuitPython Libraries on Linux and Raspberry Pi](https://learn.adafruit.com/circuitpython-on-raspberrypi-linux/installing-circuitpython-on-raspberry-pi)
- [Raspberry Pi Official Overlays information (github)](https://github.com/raspberrypi/firmware/blob/master/boot/overlays/README)
- Line 4235 starts the description of the "pwm" overlay configuration
- [Adding Basic Audio Ouput to Raspberry Pi Zero](https://learn.adafruit.com/adding-basic-audio-ouput-to-raspberry-pi-zero/pi-zero-pwm-audio)
- [CircuitPython pwmio documentation](https://docs.circuitpython.org/en/latest/shared-bindings/pwmio/index.html)
# Important Points
1. The `gpio` program is no longer available. It has been replaced with `pinctrl` which works a little differently.
1. The Blinka/CircuitPython interface uses `lgpio` to do hardware PWM on the pins. This requires the `pwm-2chan` overlay to be setup correctly, which it's documented procedure does. Details on relevant pins on line 4267 of the [overleays README](https://github.com/raspberrypi/firmware/blob/master/boot/overlays/README)
# Discussion
My raspberry pi zero W 2 wasn't working reliably with the PWM.
The `gpio` command has been depricated and is now replaced by the `pinctrl` commend. Running it did not list any PWM0 on any of the pins, which was concerning.
[This forum post](https://forums.raspberrypi.com/viewtopic.php?t=388352) indicated the need for a `pwm-2chan` device tree overlay to get it to show up. After adding `dtoverlay=pwm-2chan` to the `config.txt`:
```
$ pinctrl
0: ip -- | hi // ID_SDA/GPIO0 = input
1: ip -- | hi // ID_SCL/GPIO1 = input
2: a0 -- | hi // GPIO2 = SDA1
3: a0 -- | hi // GPIO3 = SCL1
4: op -- -- | hi // GPIO4 = output
5: ip -- | hi // GPIO5 = input
6: ip -- | hi // GPIO6 = input
7: op -- -- | hi // GPIO7 = output
8: op -- -- | hi // GPIO8 = output
9: a0 -- | lo // GPIO9 = SPI0_MISO
10: a0 -- | lo // GPIO10 = SPI0_MOSI
11: a0 -- | lo // GPIO11 = SPI0_SCLK
12: ip -- | lo // GPIO12 = input
13: ip -- | lo // GPIO13 = input
14: ip -- | lo // GPIO14 = input
15: ip -- | hi // GPIO15 = input
16: ip -- | lo // GPIO16 = input
17: ip -- | lo // GPIO17 = input
18: a5 -- | lo // GPIO18 = PWM0
19: a5 -- | lo // GPIO19 = PWM1
```
Which looks much better.
There is a pwm chip `in /sys/class/pwm`
After running the `lgpio/pwm-test.py`
the pin is back to being listed as an input, so I'm confused.
The PWM pin (under load) did the right values. Without a load on the Analog Discovery 2, I just saw noise.
I commented out `dtoverlay=pwm-2chan` and `lgpio/pwm-test.py` worked.
The pwm chip is missing from `/sys/class/pwm` so I'm guess that's not needed.