So I am just finishing up on my build. So far so good. I used 2 sheets of Birch play 48X48" and made the table (40X40")
I was using the ramps setup but decided to use Mach4 (overkill I know)
I’m hoping someone can help me determine the “Revolutions per inch” number that I need to set up the motors in Mach. Is there a calculator somewhere I can use?
I found this
http://buildlog.net/cnc_laser/belt_calcs.htm
It shows 156.250 Steps to inch Using no microstepping ( I want to try that first because of the huge loss in torque when micro stepping).
Just an FYI
For Marlin, I found this calculation in the this forum somewhere accidentally, it was posted by Ryan,
'with a pitch of 2mm… and your stepper is 1.8 degrees per step. then….
at 16 microsteps it is 1600 steps per mm [ ((360/1.8) * 16) / 2mm = 1600 ]
at 32 microsteps it is 3200 steps per mm’
My M8 AllThread rod has 1.25mm pitch (360/1.8) * 32) / 1.25 = 5120
My T8 LeadScrew was 2m pitch, which was 3200 as above
Maybe this is helpful somehow…
Cheers
There is a calculator tab towards the top, it does belts & lead screws.
Thanks Cehbab and Jim. Unfortunately I’m using the evil “US” system right now…
The value of 156.250 is dead on with the Nema motors with 0 microstepping.
This is a very interesting topic. I’m curious if this is testing that Ryan has done yet or not, running in full steps or half steps vs microsteps? I’m eager to see how this plays out and what if any previous testing has been done on the topic. Good luck!
It has come up a few times. IIRC, the answer was basically, he did test at 1/8th steps and 1/4th steps and saw a noticeable difference in the path of a circle, so he bumped it up even higher. The other thing I’ve read (here or somewhere else) is that the torque reduction is a true thing, but only on the microsteps, so when the motor gets pushed out of the way of the 1/32nd microstep toward the location of the full step, the torque increases, so if you are approaching that torque limit, the microstepping will still finish with the same total torque, but it will try to get more precision when you are at lower resistance. I know there are other issues with microstepping, when we are doing significant speeds. That’s all theoretical on my part though, I haven’t ever tried anything but 32.
Hefe has it right from what I understand. There are conflicting results out there but from what I’ve tried I’ll stick with 1/16-1/32 stepping.
With the recommended setup, gt2, 16 tooth, 180 degree steppers. The lowest number to get even steps would be 1/4 stepping. That gives you a max resolution of 0.04mm, at full stepping your max resolution would be 0.16mm. Small circles turn into hexagons and smooth curves get less smooth. Full steps sound horrible by the way, it sounds bad and you will also get oscillations as it snaps into the next step.
It is super easy to play with just change your jumpers and flash the firmware with the appropriate steps. It is interesting to try yourself I highly recommend it. I have done a few test over the last few years to reassure myself.
