search

mcbridejc / GaussSpeedway

25 stars 3 forks source link

Using TMC Stepper Motor driver instead? #5

Open beniroquai opened 10 months ago

beniroquai commented 10 months ago

Hey @mcbridejc, this project is amazing! Thank you so much for making it open-source and providing such nice documentation! I'm a little lazy and was wondering if - instead of the electronics in the current version - I could simply use a TMC stepper driver. What would be the drawback here? I can imagine that micro stepping would allow me to have a more precise motion. i was thinking of building a small xy microscope stage based on this design. Do you have a rough estimate on the force that is available for pulling stuff over a surface? Or in other terms, what's the maximum load of the small "car" can carry in your track setup?

mcbridejc commented 10 months ago

Thanks @beniroquai!

You can definitely use stepper drivers -- in fact that's what I used on the first prototype. The only concern that pops into my head is making sure you have enough inductance for the driver. I think some of them expect/require considerable inductance. I ended up adding 68uH inductors to keep the ripple down.

I am doing microstepping (8X, 16X? I forget exactly, but some amount) in the code on the controller. I don't think it makes a huge difference, to be honest.

I don't have a very quantitative estimate, but the force is quite low. The cars are printed quite thin, and if they get perhaps 2 or 3X as heavy I think they won't move reliably (with the same magnets anyway). I tried to weigh a car, but my scale seems to have lost its calibration and I seem to have lost my calibration weights at the moment.

For precise positioning, the friction is usually too high. I can't move the car back and forth within a step period reliably at all. I have seen people use paramagnetic films to get some lifting force and either lower the friction or even levitate a bit, so that's something to consider...but I can say if you tried to move a 75mm microscope slide with my setup it's not even close to working -- although there's nothing to say you can't get bigger magnets and more current!

beniroquai commented 10 months ago

Thanks for the detailed response!

I feared the lack of force to move a sample eventually. After doing some research I also found some projects that could move a magnet in XY. It could even move smaller parts like LEGO pieces, so this sparked the idea of having an ultra low-cost and perfectly planar XY stage for scanning blood smears to detect malaria parasites eventually. Here, price matters :) The current prototype looks like this and I can imagine that the XY stage can be replaced (if you are curious: UC2): image

Regarding "more current", I found this implementation https://benwang.dev/2022/08/09/PCB-Linear-Actuator.html. He uses a 3 phase synchronous motor. Pretty cool, but I'm lacking the knowledge to get this to work eventually. I'm also more familiar with ESP's than STMs ;) But perhaps I have to go for the "bigger is better" path. I'll order some PCBs generated with your beautiful scripts to get a feeling for their performance. If you're keen to support this crazy idea, perhaps we could have a chat :)

Thanks a lot again!

EDIT: Regarding microstepping: You say, it does not affect the stepsize at all?