585,717 active members*
4,139 visitors online*
Register for free
Login
IndustryArena Forum > Machine Controllers Software and Solutions > CamSoft Products > Help with backlash compensation - position display showing correction
Results 1 to 6 of 6
  1. #1
    Join Date
    Oct 2003
    Posts
    128

    Help with backlash compensation - position display showing correction

    I have a somewhat critical job coming up so I am using backlash compensation for the first time. All good on the X & Y, but Z seems to be showing the additional movement in the position display.

    With a A G0 Z.010 the display reads Z .0112 for example.

    We are using stepper motors that also have encoders on them if that matters. I cannot see what is differentrant on the Z axis except that the gearing is set to -2 and x & y are 1

    This is an old Bridgeport Series 2 so there is no homing routine. Am i correct that the first movement in each axis after startup will be somewhat random as the control is applying compensation but does not really realy know the position of the motor?

    And does it apply compensation to the hand-wheel moves?
    Thanks
    Marc

  2. #2
    Join Date
    Apr 2003
    Posts
    332

    Re: Help with backlash compensation - position display showing correction

    mbam,

    This is complicated, so no one else may be able to explain this in detail. We'll give it a shot.

    Backlash does apply to Hand Wheel's in position / increment mode. Not in free wheeling Tach mode.

    It also does know "how much" and "if to" add/subtract steps in stepper mode on the first move. Click on the button to the left of the white BACKLASH box in CNC SETUP under DESTINATION CONTROL. This will explain how it does this using positive or negative values entered into these boxes to tell it which way the first move would be travel after homing.

    What is most likely happening in your case is complicated because you have encoders that report Z axis positions in the readout displays using a different number of encoder counts per inch (per rev) than the number of micro steps the same axis motor uses to move 1 inch. The display reading you see Z.0112 is directly related to the values from the encoders, which may not match up with the step count. The G code program Z.01 dictates the number of steps to move. There maybe another variable applying itself too. You're Z dimension is close a(0.0012) nd maybe showing Z differently than X,Y because of gravity. The physical head maybe in a little different position than the stepper motor positioned it at.

    See (Closed loop position correcting Steppers) by clicking on the button to the left of SERVO-STEP box under GENERAL SETTINGS in CNC SETUP.

    You may have to adjust , test and re-adjust the Z values for both step count under RATIOZ and encoders per inch under AUX ENCODER SETTINGS, plus figure in that the BACKLASH value, may or may not get added to this. If this were a Closed Servo it would self correct to the Z readout display automatically. Reading the encoder position of Z , 12,000,000 times a second. Steppers can position, but they go to fixed positions units based on (steps per inch). Unless the encoders report they are out of tolerance, they will not remove the difference of 0.0012, where servos would strive to do position error correction to +/- 1 encoder count accuracy despite the tolerance you've entered.


    Tech Support
    CamSoft Corp.
    [email protected]
    PH 951-674-8100
    Fax 951-674-3110
    PC Based CNC Control For The Machine Tool CNC Retrofit And CNC Controller OEM Market
    (Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)

  3. #3
    Join Date
    Oct 2003
    Posts
    128

    Re: Help with backlash compensation - position display showing correction

    Thanks for the response.

    I don't think the encoders are doing anything in my current setup.If I run out of travel the motor of course stalls but the readout continues as if it is moving. I for sure will take a look at Closed loop position correcting Steppers!

    If I have no backlash comp in the Z axis the display agrees with the commanded position. It's only when I enter a value that the discrepancy occurs.

    I currently do not have any homing routines. As everything on the machine is movable I couldn't figure out how to implement one. Actually I suppose I could pick an arbitrary point on each of the axis and use that for home. I have to take a look at my encoders and see if there is a reference position on them.

    My X axis actually has less than .0001 backlash so that is set to 0, Y however (going from memory at this time- which is shot btw) has about .0009, it worked out that 35 counts or so took care of it.
    Thanks
    Marc

  4. #4
    Join Date
    Apr 2003
    Posts
    332

    Re: Help with backlash compensation - position display showing correction

    mbam,

    It does make it much easier without the encoders. This reduces the number of reasons now.

    We can think of 3 things:

    1. If this happens on the very first Z move, then reverse the BACKLASHZ value, if positive make negative or vise versa.

    2. Also you may not have the full BACKLASH error or as much as you think you have for Z and you're over compensating.

    3. Under DESTINATION CONTROL in the box USE BACKLASH enter a -1


    We'll continue this directly with you by e-mail if need be.


    Tech Support
    CamSoft Corp.
    [email protected]
    PH 951-674-8100
    Fax 951-674-3110
    PC Based CNC Control For The Machine Tool CNC Retrofit And CNC Controller OEM Market
    (Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)

  5. #5
    Join Date
    Oct 2003
    Posts
    128

    Re: Help with backlash compensation - position display showing correction

    Thanks, but a few last questions for now - will any of the above affect the position display? I thought that without feedback the display was just doing the math for the counts? And the comp will just add the counts sent to the stepper? And why does Y work OK with same setup?
    Thanks
    Marc

  6. #6
    Join Date
    Apr 2003
    Posts
    332

    Re: Help with backlash compensation - position display showing correction

    mbam,

    Yes, they should correct the Z axis position display. The settings we asked you to change should correct the display while still
    doing the physical comp.

    You're correct that without feedback it just counts steps. In your case the steps are counted plus extra steps "backlash comp" is added or subtracted.
    The X,Y&Z work the same way, so there's another factor affecting Z.

    We're thinking that X,Y must have the right amount of comp in the right direction and the BACKLASHZ needs to be reversed. The USEBACKLASH=-1 really only allies when using encoders, but it won't hurt.

    We suggest making a move Z1 rather than Z.01. It will show up more and the position Tolerance can't be the reason.

    There are several other factors:
    (A) Home Z, Fixture & Tool offsets add steps
    (B) Check to see the Z is part of the coordinated group of axis and not an independent axis
    (C) Check if the SETUP logic command re-defines the parameters of Z. Usually in the STARTUP.FIL file.
    (D) Gearing or screw pitch. Z may really need to have a fractional amount entered. For example we've seen 1000.5 steps per inch. Over a long distance it works out if done right. If you only entered 1000 instead of 1000.5 you can accumulate error because a stepper motor and only position to a whole step. It drops
    off the rest, so over a long distance or many moves , the error is accumulated.

    Don't worry about all the possibilities. They probably don't apply. You may find the simplest advice regarding reversing Z's back lash on the first move after home we gave you works.

    Tech Support
    CamSoft Corp.
    [email protected]
    PH 951-674-8100
    Fax 951-674-3110
    PC Based CNC Control For The Machine Tool CNC Retrofit And CNC Controller OEM Market
    (Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)

Similar Threads

  1. Backlash compensation
    By lumberjack_jeff in forum LinuxCNC (formerly EMC2)
    Replies: 5
    Last Post: 01-19-2010, 05:15 AM
  2. BackLash Compensation
    By bearracecars in forum CNC (Mill / Lathe) Control Software (NC)
    Replies: 0
    Last Post: 05-21-2009, 04:36 AM
  3. EMC Backlash compensation
    By CROSSHATCH in forum LinuxCNC (formerly EMC2)
    Replies: 9
    Last Post: 01-19-2009, 07:30 PM
  4. Backlash Compensation
    By cnccustom in forum Mach Mill
    Replies: 1
    Last Post: 11-22-2007, 02:33 AM
  5. NOTES ON BACKLASH CORRECTION IN Mach3 1.90.004
    By chris59 in forum Machines running Mach Software
    Replies: 5
    Last Post: 05-09-2007, 06:00 PM

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •