Arduino Adapter Board

Hi all,

I’m opening up a discussion on the development of an Arduino adapter board. There’s been at least three inquiries about compatibility so far.

Honestly, we don’t think that it would be difficult to design an adapter board that sits between Arduino shields and our device. It would pretty much consist of level-shifters and analog-to-digital chips – on a correctly dimensioned PCB and pin layout. This would easily cost less than $10. If enough people are interested, we might create a reward tier for it before our campaign ends – otherwise, we’ll set up a sales site for after our KickStarter launches.

We’re going to whip up a preliminary design this weekend. Ongoing feedback, requests, or concerns would be well appreciated!


Also I would like to get some clarifications on what people want. In my mind, I see two different boards.

  1. A board that sits between our WebFPGA device and your existing Arduino shields.
  2. A board (really, a glorified socket) that allows the WebFPGA device to sit on top of an Arduino.

Let me know your thoughts,
Ryan

Hi, I obviously don’t know the details of the original inquiries, but personally, I’d be way, way more interested in Teensy or Feather compatibility than Arduino. (Obviously, the electrical and mechanical challenges are somewhat different.) I plan to use the Teensy and FPGA boards together. I could be wrong, but I suspect that many (most??) intermediate+ users will choose either that or the Feather interface/form factor.

I’d probably be game for designing at least one of the adapter boards, since I planned on doing something similar for my own work anyway.

Anyway, that’s my 2¢.

I believe a single board with a teensy/feather socket, Arduino layout both for and as a shield, and webfpga socket could unify them, similar to the Papilio DUO. The host MCU/shield and the level shifters would have to appropriately tristate shared pins, but the same PCB could accomplish both host and shield hobbyist purposes in that way, as well as attaching a full-size shield to the MCU.

Similarly for experimenter purposes an Arduino mega profile could provide a high I/O count, with the majority of the I/O provided at low speed from ADC/DAC/PWM/GPIO chips, and a limited number of full speed level-shifted pins.

Those would both push the board area too high for some uses on the other hand, so a teensy-/feather-specific board might be better as a dedicated “mini” board, perhaps with the WebFPGA and MCU on opposite sides.

I spun up a GitHub repository with an Arduino Uno shield footprint. Basically, you would take your existing shields (say a Bluetooth shield or an SD card shield, etc.) and plug it next to the FPGA board, letting them talk to each other. (The Arduino Shield socket is missing female headers on the render, but we would definitely solder those on.)

The chip on the board is a 6-channel ADC. I haven’t added the level-shifters or done any routing yet.

So what I’m hearing from you guys is that we should create something similar with the Teensy/Feather footprint.

1 Like

Sounds interesting. I haven’t given it a huge amount of thought yet, but gut feel is that a single board could be made for the three types of MCU hosts without too much trouble, at reasonable size, with a couple assumptions.

The connectors wouldn’t _need to be populated for the two hosts that _aren’t being used. In addition to saving components, it should save space, as the MCU board footprints could then be allowed to overlap.

Of course, all those 0.1" PTHs would make routing a bit more challenging, due to the Swiss cheese factor. Hopefully, there are enough shared signals in similar enough locations to make it manageable.

Oops. I should clarify two things. In the above, I meant to convey the notion that MCU hosts could be in the mix, along with the wings, shields, etc.

But, that said, the primary, self-imposed assumption/constraint made was that it wouldn’t necessarily be a requirement to host a heterogeneous mix of wings and shields, for instance. This may or may not be how others see potential solutions. Thoughts?

I created a new thread to sort-of test my assumptions about anticipated uses of the board: “FPGA as a “host” or a peripheral device?” It’ll be interesting to see what responses it gets. Hopefully, those answers will help guide the direction on the board/s being proposed here.