Hello Guest it is November 22, 2019, 09:22:11 PM

Author Topic: After using the “Steps Per Unit” in “Axis Calibration” my CNC wont run properly  (Read 28433 times)

0 Members and 1 Guest are viewing this topic.

Using Mach3

Am in the process of fine tuning my CNC and have hit a snag while setting my “Steps Per Unit” in “Axis Calibration”.

I have been air cutting and using a pen to draw the various Gcode files, everything was repeating and running smoothly except that everything was half size. Sounds like I need to adjust the Steps Per Unit.  My drivers are set to "Full Step"  And in "Motor Tuning" the Steps per Unit was showing 2000. I am using a 10 TPI Acme so this sounds correct.  ? ? ?

I went into the Settings Tab and selected “Steps Per Unit” in “Axis Calibration” then the X axis, deleted the 0 and input 2”. Ran that and measured and it was 1”, input that figure and saved it. Did the same I think for the Z axis.

Went back to the Motor Tuning the Steps per Unit was still showing 2000. Hmmmm Thought this should be a different number now after what I did above.

Tried a X and Y axis manual jog and the stepper makes about one revolution and stops, and buzzes if I hold the jog button. It does this in both directions. It does this every time I press X jog.

Tried Z axis manual jog and it moves a few stepper rev. but only about 1 turn in the down direction every time I hit jog.

In the Diag screen each axis moves toward home slowly as long as i hold the button.

As I said everything was OK before I tried the “Steps Per Unit” in “Axis Calibration” utility.

Any Ideas on how to fix this?

Thanks
Mr. Chips
Retired 3M'r
The buzzing is undoubtedly the stepper stalling because you're trying to step too fast.  You doubled the steps/unit, but didn't change the max speed, so the step rate during jog was doubled, and the motors can't handle it.  First, reduce the max speed on the motor tuning dialog to a low number, like 10IPM.  Then check the jog operation.  I think you'll find it works fine, just slow.  While you're there, check the move distance is correct.  Once both of those are OK, then play with max speed and acceleration, and set them low enough that the motors don't stall.
But, sounds to me like you're really in half-step mode, not full step.  Either that, or you have a 2:1 mechanical reduction you've left out of the calculation.  For 10-pitch screws, full-step 200PPR motors, steps/unit *will* be 2000 in inch mode.

Regards,
Ray L.
Regards,
Ray L.
The buzzing is undoubtedly the stepper stalling because you're trying to step too fast. You doubled the steps/unit, but didn't change the max speed, so the step rate during jog was doubled, and the motors can't handle it. First, reduce the max speed on the motor tuning dialog to a low number, like 10IPM. Then check the jog operation. I think you'll find it works fine, just slow. While you're there, check the move distance is correct. Once both of those are OK, then play with max speed and acceleration, and set them low enough that the motors don't stall.
But, sounds to me like you're really in half-step mode, not full step. Either that, or you have a 2:1 mechanical reduction you've left out of the calculation. For 10-pitch screws, full-step 200PPR motors, steps/unit *will* be 2000 in inch mode.

Regards,
Ray L.


Thanks Ray,

I slowed all the axis IPM from 40 (I had ran at 40 for days without any stalling) to 10.  And they all moved normal but naturally slowly.
I am in full step.  And steps set to 2000 per in.   I input 2" and it moved 2"

I have a fixed gantry machine.

With the steppers turned off the amount of force required to turn both the X and Y acme threads by hand is about the same.  However on the Y axis it will now JOG at 100 IPM no problem, but the X axis stalls at over 40 IPM.  If I run the X axis at 30 and jog it is impossible to stall it by pushing on the axis as it moves, but set it to 40 and I can push on it an dstall it easily.

That is strange all I did was slow the axis down to 10 ran all the axis, now i can run the Y at 100+ without any hesitation????

What should the acelleration and G force be?  Or is this a factor?
Mr. Chips
Retired 3M'r
Acceleration is a function of the motors, the machine, the drivers, etc., etc., and will be different for every axis, and every machine.  X will generally be quite different from Y, because it's fighting a higher mass and hence inertia.  Start low, and work your way up, checking at each step that it will reliably rapid to the correct position.  If you go too far, you'll find you lose position randomly.  There's really little to be gained by going to max acceleration except for bragging rights, so I wouldn't push it.
You may also be hitting mid-band resonance.  Are you using Geckos?  If so, you should adjust the resonance compensation.  It will be different for the two axes.

Regards,
Ray L.
Regards,
Ray L.
Acceleration is a function of the motors, the machine, the drivers, etc., etc., and will be different for every axis, and every machine. X will generally be quite different from Y, because it's fighting a higher mass and hence inertia. Start low, and work your way up, checking at each step that it will reliably rapid to the correct position. If you go too far, you'll find you lose position randomly. There's really little to be gained by going to max acceleration except for bragging rights, so I wouldn't push it.
You may also be hitting mid-band resonance. Are you using Geckos? If so, you should adjust the resonance compensation. It will be different for the two axes.

Regards,
Ray L.

How will I know when I'm hitting mid-band resonance?  I am not using Geckos but would a damper work, I have been reading about this on the CNCZone

I would post a photo but an new and don't know how.
Background:  200 in/oz steppers, Probotix isolated BOB and drivers 24V PS.
My homemade CNC is a fixed gantry, with a 24" X 24" work area, using 1/2" 10 TPI, 1 start acme screws, the X axis is running on 1" dia fully supported lin brgs, It is moving the Z axis and a wood router.  My Y axis is running on gas pipe with roller bearings on the top and side of the pipe, the pipe is supported under the X axis travel area.  I am a wood hobbyist and this was the most rigid machine I could make and have the lightest loads on the steppers.   I have linear bearings and ball screws for all axis for my next machine and it will be a movable gantry with each side screw driven.  That is if I ever get the bugs out of this one and enough confidence to start cutting parts.

Mr. Chips
Retired 3M'r
Nice looking machine!  Mid-band resonance occurs at moderate speeds, and when you hit it, motor torque falls drastically.  One way to deal with it (sometimes) is to just power your way through it - accelerate fast enough that you get through it, to higher RPM, before the motor has a chance to stall.  But, that isn't always practical.  Mechanical dampers are another way to deal with it, and can be quite effective.  Micro-stepping can also help, as it reduced vibration.  From what you've described, I do think this is likely your problem.

Regards,
Ray L.
Regards,
Ray L.

Offline RICH

*
  • *
  •  7,371 7,371
    • View Profile
MrChips,
The power supply min amperage should be at least equal to the sum of all the amps required for x,y,& z
along with proper voltage output. That said:
 
Would set the X and Y at the lower speed of the two. Think I am correct in that Mach will default to the lower speed of the two axis's when doing circular moves. Reliable movements are more important than speed. If there are minor differences in the acclerations between two axis set the x & y  at the lower of the two.
As the stepper speed increases the torque goes down. It is a trade off, faster speed with less torque or slower speed with more torque. So if you are cutting some wood  and say you hit a knot you'll need  some power to get thru it. Power is 
a function of speed and torque. So you run at a feed rate which leaves some reserve while cutting.

As far as acceleration goes, start low, increasing by 2x until your motors skip, then set at say 60% of that acceleration value, then tweek up or down to refine.
Just some thoughts ,
RICH
Nice looking machine! Mid-band resonance occurs at moderate speeds, and when you hit it, motor torque falls drastically. One way to deal with it (sometimes) is to just power your way through it - accelerate fast enough that you get through it, to higher RPM, before the motor has a chance to stall. But, that isn't always practical. Mechanical dampers are another way to deal with it, and can be quite effective. Micro-stepping can also help, as it reduced vibration. From what you've described, I do think this is likely your problem.

Regards,
Ray L.

I have changed the step from full to 1/8 step and changed my step units to 32,000.  (With my Probotix drivers I can go down to 1/16 step would this be considered Micro-stepping)  On a command of 2" it moves about 2 1/4"  I am afraid to go into the “Steps Per Unit” in “Axis Calibration” because this is where I ran into trouble last time.  I can just test and measure until I get in shape.

When I moved from full to 1/8 step I could no longer run the Y axis at 100 had to drop to I think 60.  Why would that be?  I also lowered the acceleration to 20.   Also in the Motor Tuning page is Step and Dir Pulse, it says 1-5, currently it is on 0, as it was the default, what should this be set to?   There are so many options in Mach 3 and I tried watching the videos, but seems some are missing, I printed the manual but what I would like to see is a typical setup for a machine of a given machine type.

I know high speed is just bragging rights, but I was kind of using speed as a proof of good alignment and i would like to have a not too slow jog speed.
Mr. Chips
Retired 3M'r
MrChips,
The power supply min amperage should be at least equal to the sum of all the amps required for x,y,& z
along with proper voltage output. That said:
 
Would set the X and Y at the lower speed of the two. Think I am correct in that Mach will default to the lower speed of the two axis's when doing circular moves. Reliable movements are more important than speed. If there are minor differences in the acclerations between two axis set the x & y  at the lower of the two.
As the stepper speed increases the torque goes down. It is a trade off, faster speed with less torque or slower speed with more torque. So if you are cutting some wood  and say you hit a knot you'll need  some power to get thru it. Power is 
a function of speed and torque. So you run at a feed rate which leaves some reserve while cutting.

As far as acceleration goes, start low, increasing by 2x until your motors skip, then set at say 60% of that acceleration value, then tweek up or down to refine.
Just some thoughts ,
RICH

When you say "until your motors skip", you mean loose steps don't you?  What would be a good test for this?  Cutting MDF (because of the consistant composition) at say 1/4" deep with a 1/4" 2 flute bit, and I'm guessing 20 IMP with my setup.

Thanks Rich
Mr. Chips
Retired 3M'r

Offline ger21

*
  • *
  •  6,289 6,289
    • View Profile
    • The CNC Woodworker


When I moved from full to 1/8 step I could no longer run the Y axis at 100 had to drop to I think 60.  Why would that be?

Because Mach3 can't output steps fast enough, unless you change to a faster kernel speed.
Gerry

2010 Screenset
http://www.thecncwoodworker.com/2010.html

JointCAM Dovetail and Box Joint software
http://www.g-forcecnc.com/jointcam.html