Hello Guest it is March 28, 2024, 06:26:55 AM

Author Topic: Motors stalling, losing steps  (Read 28709 times)

0 Members and 1 Guest are viewing this topic.

Offline Zaae

*
  •  120 120
    • View Profile
Motors stalling, losing steps
« on: December 25, 2008, 06:21:55 PM »
Hello all!

I have a smaller machine that has a set of 3 Keling  KL23H276-30-8B steppers on it.

I recently started doing some different machining that's requiring more speed than I usually run, but a problem that's always existed has now come back to bite me. Since I got the machine, it's always had this little glitch. You'll be jogging an axis, and as it goes along, it will suddenly make a very short 'stalling' sound, for lack of a better term. Normally the axis will continue on its way happily, but now I've been trying to ramp up the speeds and when it hits these 'spots', the axis starts losing steps (axis stops, whining noise from stepper like it's stuck)

This can happen on any axis, and in any location on the axis, however it seems to usually happen in the middle area, as if it's determined by how long the motor has been running. Unfortunately, the controller is some generic thing that doesn't appear to have brand name on it.

Anyway, I've been scouring the forums, and tried tons of tricks I've read, but can't get this to go away. As far as I can tell, it's not mechanical, everything moves nice an free, no binding. I'm running a 3.0ghz pentium desktop with nothing installed but mach and windows xp, though it's entirely possible it's the pc causing the problem I suppose, I have no proof either way.

I'm kind of at a loss here. The steppers seem to sail along flawlessly at high speed other than these glitches. I know I haven't covered all the bases as far as testing goes, but I thought I'd ask here, hopefully you guys can save me some time.

TYIA ~
Z

Offline RICH

*
  • *
  •  7,427 7,427
    • View Profile
Re: Motors stalling, losing steps
« Reply #1 on: December 25, 2008, 07:07:26 PM »
Zaaephod,
As the steppers speed increases torque decreases. You should test your particular setup to see where it will start skipping and set the max velocity such that it's approx 30 to 40% below the found skipping max speed. You need a
to leave some torque for cutting and some safety margin. You also can't have instant acceleration so ramp appropriately from start. Don't know what kind of drives you have but you may be able to increase the stepper torque by using a higher amp setting on your drives.
RICH

Offline Zaae

*
  •  120 120
    • View Profile
Re: Motors stalling, losing steps
« Reply #2 on: December 25, 2008, 07:22:33 PM »
I've done tons of testing with different velocities and acceleration settings, and it seems as though these steppers should be able to handle the speed. I've moved on to testing software and signals, as it really doesn't seem to be a mechanical problem. I've run the X axis until it stalled (took 8-9 full passes across the table, cutting air), then turned off the controller and the screw turns easily with my fingers, it's not even the slightest bit bound.

I should also mention that throughout the course of the day, I had the machine in 1/8, 1/4, and 1/2 step mode (recalculating steps, velocities, and acceleration for each) and none of that seems to have helped. Neither has pulse width, sherline, or anything else.

I just ran the Driver Test software, and I do see some spikes here and there. I'm going to go through the xp optimization again, swap out parallel cables, check grounds, and re-route some cables... it really seems like some sort of signal noise. If none of that helps I'll try the amp settings you mentioned.

Thanks for the reply on Christmas night !

Offline RICH

*
  • *
  •  7,427 7,427
    • View Profile
Re: Motors stalling, losing steps
« Reply #3 on: December 25, 2008, 08:25:06 PM »
Look up the specs on those steppers and see if you can find motor curves. That will give you prefomance
data at maybe two different voltages and wiring. It will show you speed vrs torque curves and that puts you into the
correct ball park very quickly. You should not be on the far end of the velocity for the motors.
It does not take much to miss a step. When missed you will know it and you back down. If they can be run solid at some lower velocity then work your way up. Yes, noise and a number of other things can cause problems. Some are easy to find others can be a PITA. So find some velocity anad acceleration value that's solid so you have a place to look for other problems if they exist.

My Christmas is a day of rest, tomorrow starts the travel to Bethlehem ( Pennsylvania   ;) ). Till then there is still piece on Earth.
H ;D H ;D H ;D,
Rich

 

Offline vlmarshall

*
  •  23 23
  • G28 G91 X0 Y0
    • View Profile
Re: Motors stalling, losing steps
« Reply #4 on: December 25, 2008, 10:09:59 PM »
Take a look at this thread on CNC Zone, about stepper resonance problems and building dampeners to cure it.
http://www.cnczone.com/forums/showthread.php?t=32284

I had the same troubles you describe, on a little Sherline running a Xylotex board and 269in/oz steppers, until I built dampeners like the ones described in that thread. My maximum rapid speeds jumped from 20 IPM, and 12 on Z-axis, to 60+ ipm on all three axis.
« Last Edit: December 26, 2008, 05:00:07 PM by vlmarshall »

Offline RICH

*
  • *
  •  7,427 7,427
    • View Profile
Re: Motors stalling, losing steps
« Reply #5 on: December 26, 2008, 07:27:36 AM »
Zaaephod,
How fast do you want to run?
How much torque is required to move the axis?
What is stepper rpm range works?
Start velocity slow and gradualy increase velocity until it misses steps, then try going up and down and note where
it seems not to miss steps. There is mechanical and electrical resonance.

I have attached a motor curve for the stated stepper. Note the torque values to speed.
RICH

 
« Last Edit: December 26, 2008, 07:50:21 AM by RICH »

Offline DAlgie

*
  •  314 314
    • View Profile
    • Algie Composite Aircraft
Re: Motors stalling, losing steps
« Reply #6 on: December 26, 2008, 12:34:39 PM »
Increase the power supply's voltage by getting a new power supply, this always works, but there is a limit on how much the stepper controls can take, a Gecko has a limit of 80 V, yours might be a lot lower than that I bet. The steppers are safe until they get too hot to the touch, usually around 120F is about it.
   DaveA.

Offline Sam

*
  • *
  •  987 987
    • View Profile
    • hillbillyhilton.com
Re: Motors stalling, losing steps
« Reply #7 on: December 26, 2008, 01:35:52 PM »
Zaaephod, the problems your describing sound so similar to the one that I had. There were little "spikes" during travel. No certain time, no certain place, completely random. If your traveling fast enough, then the motor would stall due to the decrease in the torque at higher speeds. The problem drove me insane trying to figure it out. I mean, it WAS working just dandy a week ago. I KNEW it wasn't mechanical. I swapped computers. No luck. I took the break-out board off. No luck. I really didn't think it was my power supply. I bought some Gecko drives, and everything worked as smooth as butter. They outperformed those cheap combo boards by far, even when they did work correctly. So, if you have access to a good driver, you might want to try that.
"CONFIDENCE: it's the feeling you experience before you fully understand the situation."
Re: Motors stalling, losing steps
« Reply #8 on: December 26, 2008, 04:38:12 PM »
One other thing to check is that the drives themselves are not heating up too much.
No heat sinks and/or poor ventilation can cause intermittent problems too.
FWIW,
RC

Easy enough to verify, just leave the panel door open and let a large fan blow in. Sheilded from chips, dust and debris of course.
« Last Edit: December 26, 2008, 04:40:59 PM by Overloaded »

Offline Zaae

*
  •  120 120
    • View Profile
Re: Motors stalling, losing steps
« Reply #9 on: December 26, 2008, 10:51:10 PM »
Thanks a bunch for all the replies guys, all of the ideas here give me some direction. I'll be trying each of these until I either go bald from pulling hair, or until the problem is fixed.

:)