Problem Bad cut on angled lines and arc - Page 2

Page 2 of 2 FirstFirst 12
Results 21 to 37 of 37

Thread: Bad cut on angled lines and arc

  1. #21
    Member Arche's Avatar
    Join Date
    Jun 2022
    Posts
    30
    Downloads
    0
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    Thanks for suggestion.
    So I did test. I added photos. It is little bit challenge because both 20° and 70° were smooth, not so perfect like straight cut but smooth.
    It was very difficult to tell what is worse. So I raised feed rate to 4000mm and spinde RPM to 24000 ( I never use that ) and it seems
    how pattern shows that 20° is worse. It shows wavy pattern that sometimes occurs on straight cut on Y (but not so intense). This wave reminds me balls inside carriage how they rolls, just my guess ?

    Attached Images Attached Images


  2. #22
    Member
    Join Date
    Mar 2015
    Location
    Netherlands
    Posts
    409
    Downloads
    1
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    This wave reminds me balls inside carriage how they rolls, just my guess ?
    The balls inside the nut rotate in the nut, every 3..5 rotations of the spindel. The nut is preloaded. A wrong diameter (to large) ball could be the cause but the effect would be largely compensated by the other balls.
    The balls inside the carriage rotate in the carriage, every 4..6 rotations of the spindel. The carriage maybe preloaded. A wrong diameter (to large) ball could be the cause the effect but the carriage would move in Z direction.


    The 20° edge shows deeper indents at a shorter interval. The interval is about 1/5 of the thickness (15 mm), so every 3 mm at 20°? That is about 8.24 mm at 90°. That "fits" the 10 mm pitch ball screw!

    Have you checked the straightness of the ball screw and measured the runout at both the machined ends?
    If you remove the BF bearing and position the part close to the BF side, is the result better?
    With the carriage moved closely to the BF size can you measure the runout at the end of the ballscrew?
    With the carriage moved closely to the BK size can you measure the runout at the end of the ballscrew?



  3. #23
    Member Arche's Avatar
    Join Date
    Jun 2022
    Posts
    30
    Downloads
    0
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    I use FF and FK. I changed all ballscrews from C7 chinese ones to C5 brand ones, also FF and FK.
    Nothing changed, what remains are rails HGR20 with HGH20 carriages, they are cheap chinese unknown quality.
    So I do not think it could be ballsrews.
    Material on photo is 8mm thick. I tried push really hard on Y gantries from side (away from X) and all I can measure is 10 microns max
    and on place where is X gantry connected to dual Y, there is 0 or maybe something, but not measurable with my tools.
    Also I pushed in way of moves and servos are fighting back to proper position with 0 backlash.
    I would bet on rails, but not sure.

    I tried to slide by nail of my finger on side of rails where balls are running and I feel a tiny tiny repetitive wavyness, but I tried to measure it with no success in difference also looks smooth.
    But I feel it by sliding. I do not have comparsion with precise rails, should it be perfectly smooth ?



  4. #24
    Member
    Join Date
    Mar 2015
    Location
    Netherlands
    Posts
    409
    Downloads
    1
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    I miscalculated the pitch of the error, the pitch of the X-error must be smaller than the pitch of the edge.
    The 20° edge shows deeper indents at a shorter interval. The interval is about 1/5 of the thickness (8 mm), so every 1.6 mm at 20°? That is about 1.51 mm in X direction at 90°.
    Now I am confused about a possible cause!
    Nothing changed, what remains are rails HGR20 with HGH20 carriages, they are cheap chinese unknown quality.
    So I do not think it could be ballsrews.
    You are right, it is unlikely the new ball screws have the same error.
    what remains are rails HGR20 with HGH20 carriages
    Have you changed the FF and FK bearings?
    I tried to slide by nail of my finger on side of rails where balls are running and I feel a tiny tiny repetitive wavyness, but I tried to measure it with no success in difference also looks smooth.
    But I feel it by sliding. I do not have comparsion with precise rails, should it be perfectly smooth ?
    You will feel this on all preloaded new bearings. It takes a wile to before wear makes things smoother. The better (harder) the bearings, the longer it takes.

    If you move the carriage using the ball screw (jog 5 times the pitch) can you measure the movement of the carriage opposite the direction of movement?
    if you mill the bad edge by 0.01 mm steps, can you estimate the depth of the grooves?

    Last edited by hfjbuis; 07-03-2022 at 10:21 AM. Reason: wrong calculation


  5. #25
    Member Arche's Avatar
    Join Date
    Jun 2022
    Posts
    30
    Downloads
    0
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    FF and FK are changed completely for new brands ones.

    If you move the carriage using the ball screw (jog 5 times the pitch) can you measure the movement of the carriage opposite the direction of movement?

    ( you mean for backlash ? I measured it and there is no measurable backlash)


    if you mill the bad edge by 0.01 mm steps, can you estimate the depth of the grooves?

    (I will try to measure it, if it is possible with dials I have and let you know)



  6. #26
    Member
    Join Date
    Mar 2015
    Location
    Netherlands
    Posts
    409
    Downloads
    1
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    If you move the carriage using the ball screw (jog 5 times the pitch) can you measure the movement of the carriage opposite the direction of movement?
    ( you mean for backlash ? I measured it and there is no measurable backlash)
    No, I don not mean backlash. If you move the carriage in Y-direction using the ball screw, there should be no movement in the X-direction. If there is movement in the X-direction it could come from the spindle and bearings (probably not because they are replaced) or from the carriage and the rails.



  7. #27
    Member
    Join Date
    Mar 2015
    Location
    Netherlands
    Posts
    409
    Downloads
    1
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    Quote Originally Posted by Arche View Post
    I would bet on rails, but not sure.
    I tried to slide by nail of my finger on side of rails where balls are running and I feel a tiny tiny repetitive wavyness, but I tried to measure it with no success in difference also looks smooth.
    But I feel it by sliding. I do not have comparsion with precise rails, should it be perfectly smooth ?
    I have gaven the rails another thought. If it would be the rails or carriage, the effect should also be seen on the 0° or 90° edge.

    if you mill the bad edge by 0.01 mm steps, can you estimate the depth of the grooves?
    You could also use a poupi indicator to measure the depth of the grooves.

    Could you disconnect 1 motor of the Y axis by removing the motor, coupler or ball screw (preferably). If you then mill the 45° edge at the side where the spindel is driven, you could see the influence of the 2 closed loop motors "fighting each other".



  8. #28
    Member Arche's Avatar
    Join Date
    Jun 2022
    Posts
    30
    Downloads
    0
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    Hi,
    Could you disconnect 1 motor...

    I already did as I mentioned before, it was the same result, same pattern. I also tried to loose bolts connected to X to have as much freeness to one side - same result (but still X is little bit heavy and sits on both sides on carriages).
    I will do more testing next week or week after, it is really time consuming...



  9. #29
    Member
    Join Date
    Mar 2015
    Location
    Netherlands
    Posts
    409
    Downloads
    1
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    I have made another test. I removed one Y ballscrew and did cut on more rigid side. Same pattern like it is with two ball screws.
    Could you disconnect 1 motor...
    I already did as I mentioned before, it was the same result, same pattern.
    You are right, I missed that remark completely.

    Have you checked the alignment of the ballscrew shaft and motor shaft. The coupler you use, allows "only" axial movements (temperature) and that is OK. For radial movements, it is pretty rigid.
    I check the alignment using a bushing that is placed on the motor shaft (without coupler) and this bushing should move freely over the ballscrew shaft.

    You could also check the "run out" of the ball screws (X, Y and A) by placing a poupi indicator on the ball screw (crest outside) and the holder of the indicator on the gantry and Z-axis. This is probably easier to do and it covers all misalignment's of the ball screw, nut, bearing and motor.



  10. #30
    Member Arche's Avatar
    Join Date
    Jun 2022
    Posts
    30
    Downloads
    0
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    Ok, so I played with machine through the weekend.

    - I tried to realign ballscrews for straightness from the motor shaft to the other side as much as I can.
    - I tried realign rails (they are HGR20 and HGH20CA carriages, both chinese ones) honestly, it is impossible to make
    them perfectly parallel.
    - I measured pattern grooves on material and the indicator vibrates around 0,01mm
    - I tried to push on Y very hard by hand with indicator placed on the end of ballscrew ( where is connected to motor ), it moves just something like 0,005 maybe.

    My estimate is that these HGH20CA are floating since they are on machine horizontally and they probably miss preloading. These
    HGH20CA were really cheap, so quality is unknown. I do not know where to find
    what size of balls should be there. I tried to buy 3,969mm for test and reball one carriage, and there is no difference when I slide carriage on rail.
    Another size I could buy is 4mm balls, but question is, is not it too much ? Is it possible to see somewhere what size of balls carriages contain or what size difference is from normal state to preloaded ?



  11. #31
    Member
    Join Date
    Mar 2015
    Location
    Netherlands
    Posts
    409
    Downloads
    1
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    Last year I experimented with the preload of a matched pair of 7000AC bearings. I found that the difference between no preload and overloaded was arround 0.01 mm.
    On YouTube I saw a video of someone preloading ball screws for a living. He repacked the ball screw with bigger balls (by try and error) until he was satisfied.

    I found this thread about replacing ball screw balls. He repacked his ball screw with 2 different ball sizes. https://www.cnczone.com/forums/balls...96640-cnc.html

    I would go for the smallest size bigger than the original balls.

    - I measured pattern grooves on material and the indicator vibrates around 0,01mm
    - I tried to push on Y very hard by hand with indicator placed on the end of ballscrew ( where is connected to motor ), it moves just something like 0,005 maybe.
    A small error on the X-axis could give a large error at the work piece, especially when your gantry is tall.



  12. #32
    Member Arche's Avatar
    Join Date
    Jun 2022
    Posts
    30
    Downloads
    0
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    Well,
    I preloaded everything, carriages and ballsrews. And result is same.
    I tried test cuts and straight X and Y are supersmooth, but something in
    angle is becoming rough, the peak of roughness is 45°, when it is closer
    to straight cut of one axis, it becomes more smooth. There is no difference
    if it is closer to Y or X, it is equal, like mirrored.

    Attached Images Attached Images


  13. #33
    Member
    Join Date
    Mar 2015
    Location
    Netherlands
    Posts
    409
    Downloads
    1
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    If it is not the ball screw, bearings, rails, or resonance, and the problem is most visible in the 45° angle, than it can be caused by X and Y steppers not running synchronously.
    If you monitor the position error of the closed loop drivers, how big is this error in encoder lines and actual position?



  14. #34
    Member Arche's Avatar
    Join Date
    Jun 2022
    Posts
    30
    Downloads
    0
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    Hello,
    you're right, it probably does not run synchronously.
    How big is error I cannot answer because I cannot read the chart.
    I will ask support how to read it properly. But on last tweak I realised
    that I forget different profiles on each motor on dual Y. It was not before, it happened at last tweak. So I tried profiles from Y and cloned it to X. And result is better than it was before. I assume
    that I need new profiles for preloaded system so I am going to contact support for new synchronized profiles and I will see.



  15. #35
    Member
    Join Date
    Mar 2015
    Location
    Netherlands
    Posts
    409
    Downloads
    1
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    If I remember right, your servo has a resolution 12800 pulses/rev, 0.00078 mm using a 1610 ball screw. Assuming that the servo has an error window (tracking error limit) of at least 4 counts, the minimal synchronization error is at least 0.0031 mm. The position error between the X and Y servo's would be at least ± 0.00625 mm.
    You could check how high the actual "Tracking Error Limit" tracking error limit is set.



  16. #36
    Member Arche's Avatar
    Join Date
    Jun 2022
    Posts
    30
    Downloads
    0
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    Ok, meanwhile...

    I think native is 6400 counts in my servo resolution, I can set more, but motors are 6400.
    I had support that tuned for me all motors, they did great job. Now it
    is almost supersmooth. Tracking errors are I think in critical maximum 5 counts, so that is really tiny.

    Anyway thank you for advices, I am still learning so I am glad for every new information I can get.



  17. #37
    Member
    Join Date
    Mar 2015
    Location
    Netherlands
    Posts
    409
    Downloads
    1
    Uploads
    0

    Default Re: Bad cut on angled lines and arc

    I am still learning so I am glad for every new information I can get.
    We all keep learning from things that don't work as expected.

    I have never calculated the influence of the servo "dead band with" before because I assumed it would be neglectable at such high encoder resolutions. It is the combination of servo's and 10 mm pitch spindles that makes the difference.
    Support has done a really good job in tuning your servo's. You also deserve a compliment because you didn't gave up and tried all possible solutions to solve this problem.
    Have fun and enjoy using the new tool.

    Regards, Huub



Page 2 of 2 FirstFirst 12

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

Bad cut on angled lines and arc

Bad cut on angled lines and arc