We’ve been working on this and designed a shield that fits the V3.6 pin rows (the two long rows only).
Until we have some more info from our partners on open source hardware licenses I can only share the schematics with you. The layout / full git repo may follow later but you can always comment or pm me for questions.
Thanks At this moment I believe our team implements some loose modules for this.
We designed a faraday cover to be placed over the odrive with the shield being mounted on top of that. In the cage, we have created a narrow gap for the 2.54mm pins to go through. This is not ideal in terms of faraday but at least something. The shield has an L shape so there is ‘air’ above the odrive’s caps (if you do not mention the Faraday cage). This leaves space for this separate module to be fitted.
Maybe @Jorijs could tell you more about this module for the SPI.
for now, I’ll not include this in the schematics but might be doing this later if I have time but because of this solution for us it is not needed to do so and I will not spend time on it for the moment.
We don’t really really use any transceivers for the encoders using SPI, as the lines are single-ended instead of differential. Though differential signals are more resistant to EMI, it does not seem as beneficial to us as the distance our wires have to cover is relatively small. So achieving proper use out of them would require properly shielded driver and transceiver modules as well, which did not seem worth the effort. (Though this could change in the future if we encounter serious EMI issues).
Just this last period we integrated a connector that allows a differential signal to be connected to our design. The board converts this to single-ended signals and works with our new odrive firmware.
