Show HN: I designed my own 3D printer motherboard (opens in new tab)

Earlier journal log entries mentioned TMC2209_SILENTSTEPSTICK, but most recent schematics removed this designation. Seems like an oversight.

I thought it was pretty clever but they admit it was tricky to make work at all, let alone get good results.

It's also just double the price, so I can get 4 layer boards for like $8 from JLC and it just makes everything much more easily.

You still do want to build up to it though, I made a macropad, then a keyboard, and then made this, so it's definitely not just an immediate jump, but I built those 2 projects within the span of a couple months!

The Mini Sparkle is a 4 layer and available on their github https://github.com/adafruit/Adafruit-Mini-Sparkle-Motion-PCB

There are many more available at OSHWA : https://certification.oshwa.org/list.html

I have a couple idea's on how I wanted to do it: - Belt printer fitted into a briefcase (the harbor freight case form factor would be good for that!) - Positron style - Maybe mess around with double four-bars

Making it self-contained with a battery is also a really cool concept I'll have to explore!

I think we're only a few years away from BLDC servo motors taking over from steppers in 3d printers.

Ideally control algorithms for them would go into the MCU so there is proper force feedback too - ie. The system will know that there is an extruder clog by the increased extrusion force, or even set print speeds to be 'the fastest you can follow this path' rather than a fixed number of mm/sec. Ie. If the bearings get a little stiff it'll go slower rather than skipping a step.

There are some patents on sensorless servo control expiring which should cut the price of this stuff almost in half since the position sensor is one of the most expensive bits.

Power supplies are one of the more expensive parts of a 3d printer, and by having BLDC motors which can do regenerative braking, that same energy can be reused in the head and bed heaters, which should allow significantly smaller power supplies too - again with significant software complexity to make sure the bed heater primarily heats when the head is decelerating and stops heating whilst accelerating to not exceed the power budget.

The problem with most AI powered tools is that they don't understand the context of your board, which is absolutely crucial for routing. Like so-and-so trace could be switching really fast and an AI powered design tool wouldn't understand that and wouldn't route it properly. Or like power for example, different things have different power requirements and I feel like we're not at the point where these tools are referencing every datasheet on the planet to figure out trace/fill sizes and whatnot.

There's also some concepts I don't think these tools can wrap around quite yet. For example, minimizing loops, consistent ground planes, proper impedance control, and just all of these factors lead to an inferior board.

Not saying that humans are perfect and understand all of these too, I just feel like if you understand what's happening with your board, you can do a much better job.

At the end of the day, you'll be spending more time fixing the mistakes, which are actually pretty hard to spot without checking every single trace sometimes, then actually just routing it yourself, but maybe one day we'll get there!

I'd love to hear some other takes though ;)