A Few Nuggets of Progress

So I've just changed job to one that's a little more time consuming, so these posts are going to become a little more infrequent than they already were for at least until the end of the year.

I'll give a small update on a few pieces of progress that I've made since last time though. Nothing incredible, but it's all you're gonna get (unless Henry can finally get round to posting something).

Cable Assemblies

I've put together a handful of cable assemblies for development purposes - I'll need to make some more eventually that are to the correct lengths as these ones are longer than required and one-sided so that I can bodge things together with some solder-spliced joints:

I did say nuggets - I really hope you weren't expecting much...

Rev-B Motor Controller & Encoder Board Release

So I made a post about this a while back, but I only got round to committing to a "formal" release of both of these boards with their litany of issues resolved a few weeks ago. Some might say I was putting it off and only pulled the trigger because I was about to lose access to the Altium license I had through work.

Anyway, after many incremental, and likely virtually useless, minor improvements it's ready for purchase. The one thing that's stopped me ordering a batch of them (and a stencil so that I can at least re-flow the top side rather than hand soldering every single component again) is that I have found a slight issue that might warrant fixing.

The crux of the issue is that, motor control work, position sensing works, but motor control and position sensing at the same time causes a large amount of errors to be reported. I'm reading at a high enough rate that ignoring the errors and ploughing on gives a good enough result (that I showed in a previous post). It is, however, indicative of an underlying problem with the signal integrity of that connection. I was hoping that the LVDS interface would simply solve anything like this on it's own, but it might also be interference picked up on the board rather than the cable that are giving me grief.

I'm hoping that an RC filter or two will sort it, or even better - a software/FPGA solution. Either way, I haven't had time to investigate the root cause of this, and with the lack of pressure (with no Pi-Wars 2026 announcement), these boards have been left in limbo somewhat.

Here they are though:

What next?

Mostly the same things as my last post probably?

I've got another post in the works, diving into some signal integrity topics (thrilling stuff I know), which is nearly complete so maybe I'll try to find time to complete an publish that. It'll be to a similar level as the capacitor post, so probably not of great interest to most people that currently read these.

The other tasks on my mind are:

• Investigate the cause of the encoder reading errors
• Implement velocity-based FOC control (maybe position too while I'm at it?)
• Implement CAN communication
• Actually get round to developing more of the CAN pi-hat with improved power supply so that it doesn't brown itself out occasionally like the old one did...