Help with repetability of absolute servo motors


Results 1 to 7 of 7

Thread: Help with repetability of absolute servo motors

  1. #1
    Member
    Join Date
    Sep 2009
    Location
    Romania
    Posts
    81
    Downloads
    0
    Uploads
    0

    Default Help with repetability of absolute servo motors

    Hi all

    I am scratching my head around an issue for which I cannot detect the cause but only witness the effects.

    I have been running a cnc touter with Delta nc300 controller, asda2 drives with absolute encoder motors. The router is stiff, steel, rack and pinion, with almost no backlash (alpha gearboxes), meaning around 0.01mm. Axes A is slave of Y. Motors are absolute single turn encoder (Delta has no multiturn) and run with 3.6V batteries. Spindle very little run out.
    For the production I make I need both gantry perpendicularity and repeatability (0.05mm max positioning error). I have to make 2 drill holes (3.175) in the table in 2 still plates, so that I use those holes to guide the aluminium composite panel sheet I machine.

    Lately I observed random errors with the repeatability. If I change the steel plates and make new holes, machine works ok, production works ok. After a few hours of work I send the router to do again the same holes, and goes through the procedure very precisely, No steel is removed, and after that the 3.175 shank of the bit still goes inside the hole very precisely, no play.
    If I turn off the router, then start it again. and try to do the holes again....surprise. they are off about 0.2-0.3mm , sometimes even more. it goes off on both Y and X axes. This drives me crazy, because it i have to turn off the router in the middle of the work, i have sometimes to scrap a 100eur sheet.

    So my takeaway are:
    - indeed spindle has no runout, so it cannot be because of it
    - general mechanical repeatability is ok because the machines returns to holes
    - thermal expansion is excluded as a cause

    I run with compensation in the controller as the racks are not that exact. now when i am writing this I think that maybe if i should disable the compensation, do the holes, restart the router and check if router return to holes. Could be path.
    The batteries on the absolute encoders are ok, actually new. No error pops up as normally does when batteries are low (or too high voltage). I do not think this could be the problem.
    there are no other compensation setups that could affect the setup.
    because the repeatability problem is on all axes, I think it is not any single of the motors or drives defective. Rather in controller something.

    So if anyone has further ideas where to search for, it would be appreciated. I can post further settings of the controller and drives if that helps.

    thank you
    Marius

    Similar Threads:


  2. #2
    Member
    Join Date
    Jan 2005
    Location
    USA
    Posts
    12254
    Downloads
    0
    Uploads
    0

    Default Re: Help with repetability of absolute servo motors

    Quote Originally Posted by nicubila View Post
    Hi all

    I am scratching my head around an issue for which I cannot detect the cause but only witness the effects.

    I have been running a cnc touter with Delta nc300 controller, asda2 drives with absolute encoder motors. The router is stiff, steel, rack and pinion, with almost no backlash (alpha gearboxes), meaning around 0.01mm. Axes A is slave of Y. Motors are absolute single turn encoder (Delta has no multiturn) and run with 3.6V batteries. Spindle very little run out.
    For the production I make I need both gantry perpendicularity and repeatability (0.05mm max positioning error). I have to make 2 drill holes (3.175) in the table in 2 still plates, so that I use those holes to guide the aluminium composite panel sheet I machine.

    Lately I observed random errors with the repeatability. If I change the steel plates and make new holes, machine works ok, production works ok. After a few hours of work I send the router to do again the same holes, and goes through the procedure very precisely, No steel is removed, and after that the 3.175 shank of the bit still goes inside the hole very precisely, no play.
    If I turn off the router, then start it again. and try to do the holes again....surprise. they are off about 0.2-0.3mm , sometimes even more. it goes off on both Y and X axes. This drives me crazy, because it i have to turn off the router in the middle of the work, i have sometimes to scrap a 100eur sheet.

    So my takeaway are:
    - indeed spindle has no runout, so it cannot be because of it
    - general mechanical repeatability is ok because the machines returns to holes
    - thermal expansion is excluded as a cause

    I run with compensation in the controller as the racks are not that exact. now when i am writing this I think that maybe if i should disable the compensation, do the holes, restart the router and check if router return to holes. Could be path.
    The batteries on the absolute encoders are ok, actually new. No error pops up as normally does when batteries are low (or too high voltage). I do not think this could be the problem.
    there are no other compensation setups that could affect the setup.
    because the repeatability problem is on all axes, I think it is not any single of the motors or drives defective. Rather in controller something.

    So if anyone has further ideas where to search for, it would be appreciated. I can post further settings of the controller and drives if that helps.

    thank you
    Marius
    As you said a good place to start is with the compensation, with you having a fixture you can always check the first hole with an indicator and make that your X0 Y0 point or make another point on your fixture the X0 Y0 point you should always have one to reference after each shut down

    Mactec54


  3. #3
    Member
    Join Date
    Sep 2009
    Location
    Romania
    Posts
    81
    Downloads
    0
    Uploads
    0

    Default

    Quote Originally Posted by mactec54 View Post
    As you said a good place to start is with the compensation, with you having a fixture you can always check the first hole with an indicator and make that your X0 Y0 point or make another point on your fixture the X0 Y0 point you should always have one to reference after each shut down
    I just renember i actually have no compensation on the X axis. I will still suspend compensation and give a try.
    Also will hook Delta pc software for asda drives to check position of the abs encoder and start restart controller.

    Any recommendation on photo/laser home switch high precission?



  4. #4
    Member
    Join Date
    Jan 2005
    Location
    USA
    Posts
    12254
    Downloads
    0
    Uploads
    0

    Default Re: Help with repetability of absolute servo motors

    Quote Originally Posted by nicubila View Post
    I just renember i actually have no compensation on the X axis. I will still suspend compensation and give a try.
    Also will hook Delta pc software for asda drives to check position of the abs encoder and start restart controller.

    Any recommendation on photo/laser home switch high precission?
    Any Home switch should work as if your control or servo drives are using the Z pulse ( index pulse ) of the Encoder you can not get a better accuracy than that for homing

    Mactec54


  5. #5
    Member
    Join Date
    Sep 2009
    Location
    Romania
    Posts
    81
    Downloads
    0
    Uploads
    0

    Default

    Quote Originally Posted by mactec54 View Post
    Any Home switch should work as if your control or servo drives are using the Z pulse ( index pulse ) of the Encoder you can not get a better accuracy than that for homing
    I agree. Encoder can do 0.0002mm acureacy at 16bit. No home sensor can do that, and i do not need more than 0.03. Still, if the drive+motor loses position more than 0.3mm, it does not work for me. So home switch doing 0.01 is more than enough. I would prefer abs motors to do the job, but was thinking of alternatives. Moreover, if position is lost and I need to square the gantry, home switch would help.
    I bought also linear magnetic encoder and will try full closed loop, asda a2 allows this, so i can avoid compensation.
    As for Z pulse, asda allows homing not only Z pulse, but use the home sensor as home position. Just need to select the parameters.
    Issue could be between controller and drives, i will look into it



  6. #6
    Member
    Join Date
    Jan 2005
    Location
    USA
    Posts
    12254
    Downloads
    0
    Uploads
    0

    Default Re: Help with repetability of absolute servo motors

    Quote Originally Posted by nicubila View Post
    I agree. Encoder can do 0.0002mm acureacy at 16bit. No home sensor can do that, and i do not need more than 0.03. Still, if the drive+motor loses position more than 0.3mm, it does not work for me. So home switch doing 0.01 is more than enough. I would prefer abs motors to do the job, but was thinking of alternatives. Moreover, if position is lost and I need to square the gantry, home switch would help.
    I bought also linear magnetic encoder and will try full closed loop, asda a2 allows this, so i can avoid compensation.
    As for Z pulse, asda allows homing not only Z pulse, but use the home sensor as home position. Just need to select the parameters.
    Issue could be between controller and drives, i will look into it
    Using the Z Pulse is the best no type of home switch is going to be better, how it works is you hit the home switch and then the servo will back off until it hits the Z Pulse

    Closing the loop when using a rack and Pinion would be a problem, the system would always be fighting with any backlash which is never a good out come, so scales in your case would be a waste of money

    If the motor and Drive is setup correct that part can never lose position, it would be a mechanical problem if you lose position, the Drive would fault out if you are not within your set value

    Mactec54


  7. #7
    Member
    Join Date
    Sep 2009
    Location
    Romania
    Posts
    81
    Downloads
    0
    Uploads
    0

    Default

    Quote Originally Posted by mactec54 View Post
    Using the Z Pulse is the best no type of home switch is going to be better, how it works is you hit the home switch and then the servo will back off until it hits the Z Pulse

    Closing the loop when using a rack and Pinion would be a problem, the system would always be fighting with any backlash which is never a good out come, so scales in your case would be a waste of money

    If the motor and Drive is setup correct that part can never lose position, it would be a mechanical problem if you lose position, the Drive would fault out if you are not within your set value
    I tried out today the repetability with an optical sensor. And could not find any fault. On/off, moving by hand the axes while system is off, long/short disconnect from power supply, high speed vs low speed, on/off compensations. Repetability was very good, within 0.01mm along the entire axis. With or without compensation the system behaved like scared of not being replaced. I wish I could make it fail.
    So i am left with the option to check if i see patterns when it will fail.
    So I installed 2 optical sensors nearby home of Y and A, and will install another 2 at the end of A and X. So before production operations I will check the repretability, thisnway avoiding failed parts.i know where it should trigger the sensors.

    Any further recommendation besides observing patterns (and luaghing now and then) is more than welcome.



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
  •  


About CNCzone.com

    We are the largest and most active discussion forum for manufacturing industry. The site is 100% free to join and use, so join today!

Follow us on


Our Brands

Help with repetability of absolute servo motors

Help with repetability of absolute servo motors

Help with repetability of absolute servo motors