Towards a low cost harp-board for ecosystem dissemination

[bruno-f-cruz]

A low-cost harp board for ecosystem dissemination

Goals

• Create a low-cost harp board with basic functionality
• Provide a competitive alternative to run Bonsai courses with instead of Arduino
• Provide a cheap "first harp board" for people to get started with harp and try out the ecosystem
• Provide a demo board that can be used to play with custom firmware and build new devices

Features

I think a minimal viable product includes:

Hardware

• 4 digital inputs
• 4 digital outputs
• (these pins could potentially be configurable between input and output)
• 1 PWM cable output (frequencies compatible with servos would be a plus)
• 1 Analog input (250 Hz sampling rate would suffice)
• 1 harp-compatible synch input (for syncing with other harp boards)
• 1 USB port for power and communication
• 1 LED for harp status

Firmware

• Implement core harp protocol + harp synch protocol
• USB communication
• Control over digital output lines (eg. Set, Clear, Toggle, State)
• Asynchronous events on digital input line state changes (i.e. Rising and Falling)
• Configurable PWM output (frequency, duty cycle) and a register to start and stop it
• A periodic event with an analog input read

Implementation

I think the obvious choice for this board is the pico. Without going into too much detail, I think the pico is a good choice because:

• An existing harp core implementation is available
• It's cheap
• Easy to source
• It's performant
• It's easy to program (at the limit, just click a button and drag and drop a uf2)
• Lots of IOs
• Good community support
• Great documentation
• An ecosystem with available interface boards

Hardware

At the level of the hardware I think the most viable strategy is to try to use something that already exist in the market to interface with outputs. Discussing currently available options, we came up with the following options:

• Gravity
  https://www.dfrobot.com/product-2393.html

• Grove
  https://www.seeedstudio.com/Grove-Shield-for-Pi-Pico-v1-0-p-4846.html

The obvious problem is that these boards do not include a physical layer for the clock synchronization input. To solve that, @Poofjunior suggested that we could adopt a "shim board" strategy that sandwiches between the pico and the shield board. Something like this.. This would obviously co-opt a couple of pins from the pico that may be available to the shield board, but I think it's a reasonable trade-off.

Given the prices of the individual parts, I believe we could probably offer a complete solution for <20$.

Firmware

Not a lot to say here just yet. But I will just signal that this seems like a good opportunity to write a small tutorial on "How to generate your own device using the pico harp core"

[jfrazao]

I would prefer to have more analog and pwm/servo eventually reducing the number of digital io.
Raspberry pico can do:
3 ADC
3 PWM