I've been running 116oz-in stepper motors with the Xylotex/TurboCNC setup.
My programs will be over 50,000 lines in length and I have been experiencing lost steps while running parts.
My thinking was that I needed to slow my feedrates down or get bigger motors. I slowed the feedrate down from 30" per minute to 20", but still lost steps.
I decided then to buy the largest stepper motors that I could find that met the Xylotex board requirements.
Here's what I bought. NEMA 34 mount, 680 oz-in holding torque, 18 oz-in detent, bipolar series, 2 Amps DC, 6.8 Volts DC, 1.8 degree step angle.
I set my Vref to 2.75 for the 2 Amps DC, hooked up the motor and applied power. 24volts. The motor locked up perfectly and I couldn't overcome the power of the motor by trying to twist the shaft.
My problem came when I tried to jog the motor. I figured that the power requirement specs for the motor were very similar to the 116oz-in motors so I did not change anything in TurboCNC and tried to jog the motor. All I got was a bunch of noise and no movement.
After much fiddling around, I found that I had to add a value of 400 to the pulsewidth (uS) in the motion setup to get the motor to turn consistently.
The motor spun smoothly and was very strong. I couldn't stop the motion of the shaft at all. The problem is that with that high of a pulsewidth, my motor is only turning at about 60RPM. For my ball screws with a pitch of .203, that equates to about 12 IPM. Not at all what I want.
If I lower the pulsewidth, say to 200. The motor will turn, but its very weak and I can stop the motion with my fingers.
By the way, the Xylotex board is step for 1/8 micostepping. Max Speed 15000, Accel 20000, Max start speed 5000. This should give me a top end of 112.50 IPM
I am stumped and any help is greatly appreciated.
I'm not really familiar with turbocnc, but: 400us pulse width would be approx 2500hz max, so your rates don't make sense to me. If it were me I'd drop the microstepping first and start at lower rates then try faster. The xylotex shouldn't need nearly that high of a pulse width unless the filter is now that slow.
You might try reducing the ACCEL to something below the MAX. I had a problem with that. I wanted instand ACCEL but which worked ok while jogging and testing but when running a program there were so many of them close together I was losing a lot of steps. I reduce the ACCEL and the problem went away. Don't know what computer you are using but mine wont work very good if I go above 12000 MAX. If you add all the motors pulses together you will get 45000 which is more than Turbocnc can handle at least on my computer, yours may be different. You might try something like 12000 for ACCEL. The START seems ok. I was using that too. You might reduce that to around 3000 and see what you get. Sometimes it takes a little fooling with to get things working. You should definitely NOT need 400 us for the pulse. Something between 2 and 5 should be ok.
What Marv said. I wouldn't count on getting 112IPM with those motors, though. 6.8V probably won't let them spin too fast.
Mach3 2010 Screenset
(Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)
Thanks everybody for the suggestions.
Here is an update.
I got this info from the motor supplier:
" If you could increase your power supply to 28V that will help also. But the easiest thing is to reconfigure the stepper motor wires for center tap to end. This cuts the inductance by a factor of 4 increasing the speed. I would kick the xylotex board up to 2.5A
connect red to A
connect red/white to A#
connect black to B
connect black/white to B#
insulate the ends of the remaining wires but do not connect them to anything"
I tried this and was able to increase my speed from 12" per minute to 32" per minute and still get good torque. I am looking for a 28v/30v power supply as that would give me 15% more voltage.
Xylotex send me a similar response:
" Your motors have a very high inductance. This is the reason they
can be so powerful, yet only require 2 Amps. The problem you are
seeing is due to the time it takes to charge the coils in your
motor. The pulse width really isn't changing anything, except it is
allowing more time for the coils to charge. It thus makes it so you
can't step as fast.
The opitmal (maximum) motor size (torque vs. inductance) for the
Xylotex drive is around 250 oz.in. at around 2.3A to 2.7A per
phase. Higher torque at similar current will require more time to
charge, or a higher voltage.
Although only 2Amps, the inductance of your motor needs a voltage
around 40 to 60VDC to be effective (Gecko territory). You will get
the full rated *holding* torque with the drive, but as step rate
increases, torque drops. This isn't a problem with my drive (other
than not handling 60VDC), it's really a matter of physics."
The interesting point here is the recommend motor size of 250 ox-in max. for good speed performance. That's important to know when developing a DIY machine, (matching components for optimal performance.)
Marvin, thanks for your practical advise. Ignoring the issues with the larger motors, I was jogging and moving my machine at over 100" per minute with the 116oz-in motors no problem, but when I ran a program with lots of little steps I would see a definite loss of steps. I'll experiment some more before I switch over to the big motors entirely.