OK, got a rookie here. When setting up a motor, how do I calculate the correct travel speed/distance settings? I think I have an idea what I need to know, just not how to use it.
Hi Calculating the steps per mm settings is easy, see C-Beam™ Machine - Plate Maker The leadscrew is likely to be spot on from the calculation, but belts will need to be calibrated because they stretch a bit. As for travel speed, start with 1000mm/minute and double it until it stops working, then go to halfway between the last one that worked and the one that did not work, and try again. When you find a speed that only just works, reduce to 80% of that value. As speed gets higher, you will need to increase acceleration as well just to keep acceleration times reasonable. Once you have a reliable top speed, double acceleration until it stops working, then reduce, and so on. Note that a bare motor will be able to rotate and accelerate MUCH faster than one that is driving a mechanism.
I'm having a heck of a time getting my machine up and running consistently. I can't seem to figure out a Base Setting for my Axis'. I'm using the TR8 x 8 ( 4 Start ) Leadscrews on all 3 Axis and a G540 Driver which is a 10 Microstep Driver. I've tried the Axis Calibration in Mach3 but it gives me weird numbers to accept as Gospel. I've tried starting with 2000 Steps Per, Accel at 25 and Velocity at 30. But all I get is jerky motions of less than an inch. The only Calculator I have been able to find is the RepRap Calculator and it doesn't do 10 Microsteps. What is everyone using as their base Steps Per, Accel, and Velocity? Also, I'm trying to convert everthing into Imperial Measurement.
I remember thinking as a kid I would never use algebra again. But apparently I end up having to do more math for my hobbies than my real life job. The steps/mm can then be calculated like this: steps per mm = (steps per revolution x microsteps)/mm per rev. (formu21-122--xyla from gnea/grbl) So: 200 x 10 = 2000 2000/8 = 250 steps/mm *someone check my math. I suck at it. This assumes you are using 1.8 degree steppers and the TR8 x8 tadvance a nut EXACTLY 8 mm/revolution. You may have to calibrate further once you get it dialed in to theoretical steps/mm. Here is how in a low production value video :
Thanks for the Math Update. You'd think some site would be available with all of the theoretical math necessary available!
good point, must add 10 and other base10 microstep settings to GRBL settings calculator and calibrating is a resource right here on this site a note on calibrating GRBL based controllers. I found that changing the $ values before sending it back to the start point broke 'something' and I had to reset after that to get good travels. so do this process: jog to where you need to start set 0 for that axis (lets use X) G0 X500 measure G0 X0 ; go back to start point $100=newcalib value G0 X500 ; should get you to the right point, but you may want to do a soft reset before this
Craig: oh man your acceleration is slow ! (-: I slowed my OX down to 650 from 1300 because it was jerking the table around, but it never lost steps.