I am also pretty excited… the first problem is that the drivers have a dip switch for 5V and 24V signal input voltage, it is set to 24V by default so it can‘t be shortened by accident, but I think I need 5V (I don‘t have a jackpot, but still, 24V seems pretty high for those small boards we are using)? The Arduino Mega I am using has 5V, but I don‘t know whether that’s the voltage the whole OpenCNCShield provides.
/edit: For FluidNC it‘s 5V, should be the same for mine I guess. If it does not move at 24V I am going to set it to 5V.
So a closed loop stepper just knows when it skips steps so it can stop before ruining your project?
I’ve never pushed hard enough for it to be an issue but curious how this works out for you.
I wonder how far just using a little bit stronger NEMA 17 for the X axis goes. I guess I’m not sure how that compares to upping the amps on the existing motor since I don’t think we’re at the motor limits. Maybe a bigger motor can provide the same torque at lower amps and temperature?
Theoretically it knows when it skipped steps and can adjust for it because it gives feedback to the driver, so basically the stepper motor skips a step, the driver gets informed and rectifies it. You have the option to include a brake that stops the CNC if it can‘t correct, but that‘d be the next step.
I am only considering doing this because @stevempotter had so much success with his Z axis on the MPCNC (or didn‘t he do it and I am only imagining it? I can‘t find it any more…). If it does not work I am going to be sad.
I had always hoped coolstep would be the game changer for TMC’s but I don’t think they have ever worked that way. I thought you could set the current to something like 1.5A and it would only ever run at what is needed (~0.4A) to keep it from skipping and only ever increase the current if needed. To me that is ideal!
Here is the post about my closed-loop Z stepper, with a link to a video of its amazing speed and repeatability. I have yet to push it hard enough while carving to cause it to give up and throw an error. But if it does, it triggers a Door signal that halts gracefully and allows restarting. Z speed - Closed loop stepper is AMAZING
I got one of those from Sourcingmap on Amazon and it was crap. It was beveled incorrectly so it just pushes the wood rather than cutting it. Here is an end-on photo where you can see the problem. https://photos.app.goo.gl/2NCbq4cZfnRvtnTHA
The bounce back… It’s incredible. I did really not expect that. I expected not to be able to move the steppers by hand, as @stevempotter said, but that I could but it would bounce back was the most amazing thing to happen this week. It will only stall if it can’t back to the position I think, but I wasn’t able to simulate that.
The bounce should not be happening because that is like, 1000 missed microsteps? It should immediately go back to the right path as soon as it deviates a tiny bit.
Seeing that, and if the tests in wood are also great, I would love to be able to add closed loop steppers to the other axis’ as well, especially Z. Adding speed there would be so rad.
I guess you thought I was exaggerating when I said closed-loop steppers are amazing
I was only expecting the correction. But also got much more torque, speed, and cool temperature too. It allows these terrain carvings to go 4 times faster.
I’ll figure it out.
