Need Help! Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better


Results 1 to 14 of 14

Thread: Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better

  1. #1
    Member
    Join Date
    Nov 2016
    Location
    Ireland
    Posts
    4
    Downloads
    0
    Uploads
    0

    Cool Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better

    Hi All

    This is my first post here so go easy on me,

    I have a fadal 4020 and after running a ballbar test on the machine I replaced the trust bearings on the X axis and Leveled up the machine and set my backlash, It all looks prity good except the Scaling error is -219.1 PPM on the Y and -179.8 PPM on the X . Can anybody tell me how I could fix this

    Thanks in advance

    John

    Similar Threads:


  2. #2
    Member
    Join Date
    Jun 2012
    Location
    United States
    Posts
    516
    Downloads
    0
    Uploads
    0

    Default Re: Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better

    Quote Originally Posted by IrishJ View Post
    Hi All

    This is my first post here so go easy on me,

    I have a fadal 4020 and after running a ballbar test on the machine I replaced the trust bearings on the X axis and Leveled up the machine and set my backlash, It all looks prity good except the Scaling error is -219.1 PPM on the Y and -179.8 PPM on the X . Can anybody tell me how I could fix this

    Thanks in advance

    John
    you need to adjust your pitch comp tables. on a fadal this is done under the survey 'SV' menus. best way to do this is to have the machine laser calibrated using an interferometer.

    that being said .0002 inch error is pretty much zero - get over it. i doubt you'll be able to control tool runout better than this anyhow. most parts are +/- 0.005", some +/- 0.001" so you're doing way better than necessary. maybe run that ballbar at many locations around the table, see what ppm's you get. you could use that to do some SV stuff rather than laser, but laser is going to be the best, good down to 1ppm measurement.

    fadal machines are not a makino claiming 5 micron tolerances on parts. just make parts, don't get caught up on any problems like 0.0002" error. its a waste of your money

    did i count that right? 1 ppm is 1 millionth of an inch.

    0.001 thousanth
    0.0001 ten thousanth
    0.00001 hundred thousanth
    0.000001 millionth (1ppm)

    so 100 millions is two over so tenths... yea i did count it right. just get over it



  3. #3
    Member
    Join Date
    Jun 2012
    Location
    United States
    Posts
    516
    Downloads
    0
    Uploads
    0

    Default Re: Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better

    Hey I got myself a renishaw qc10 ballbar kit. hopefully it is easy to use. we can compare results on my fadals vs. yours.

    i'll likely use it more to tune the servo amps for roundness at various feedrates.



  4. #4
    Member
    Join Date
    Nov 2016
    Location
    Ireland
    Posts
    4
    Downloads
    0
    Uploads
    0

    Default Re: Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better

    Thanks for that Mflux



  5. #5
    Member
    Join Date
    Jun 2012
    Location
    United States
    Posts
    516
    Downloads
    0
    Uploads
    0

    Default Re: Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better

    Here are two ballbar tests on my 4020 machine. One at 40ipm, another at 80ipm

    I just did a huge amount of maintenance. new waylube, new spindle, new gibbs....

    also this machine has not been laser calibrated yet (because i knew it needed gibbs and lube done). I am working on laser calibrating it now. I will post new ballbar tests after laser cal. I will then attempt to tune up the servos by messing with the trim pots on the axis amp cards and matching it up on the following error readout / etc. I will then post a 3rd set of ballbar results which will be for after laser cal and after servo tune-up.


    right now my machine looks pretty sick...

    Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better-4020-40ipm-laser-cal-11-18-2016-a

    Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better-4020-80ipm-laser-cal-11-18-2016-a



  6. #6
    Member
    Join Date
    Jun 2012
    Location
    United States
    Posts
    516
    Downloads
    0
    Uploads
    0

    Default Re: Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better

    Here's how laser cal went for the X axis - made a big difference. not sure why my ball screw is so messed up in the last few inches of the right side of the table on this X axis. the comp tables can only do 0.0008" / 1.0" updates so its as good as i can make it. you can set and verify backlash during a laser cal too...

    Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better-x-axis-laser-cal-nov-16-jpg



  7. #7
    Member
    Join Date
    Jun 2012
    Location
    United States
    Posts
    516
    Downloads
    0
    Uploads
    0

    Default Re: Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better

    This morning while the laser was still setup on the X axis I decided to run the machine in to warm up the ballscrews. I ran it at 150ipm for 1 hour x and y axis' using M11, M12 reciprocating axis motions (just back and forth infinite).

    holy sh*t the screw expanded a lot. it is 0.007" / 40" under this condition. I noticed that the fadal factory SV tables usually biased the ballscrew about 0.005" / full travel (in this case its a 4020 so 40" x) towards the short side. I see why now.

    I don't always operate in a condition where my screws experience 150 ipm all day long, and there may be plenty of downtime during the day - so I will probably just pick a number in the middle between 7thou and 0 (cold cal). and I will bias my SV table accordingly.

    here's the data of Cold vs Hot screw. what a pain in the a$$.

    Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better-x-axis-hot-vs-cold-ballscrew-nov



  8. #8
    Member
    Join Date
    Jun 2012
    Location
    United States
    Posts
    516
    Downloads
    0
    Uploads
    0

    Default Re: Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better

    Quote Originally Posted by mflux_gamblej View Post
    This morning while the laser was still setup on the X axis I decided to run the machine in to warm up the ballscrews. I ran it at 150ipm for 1 hour x and y axis' using M11, M12 reciprocating axis motions (just back and forth infinite).

    holy sh*t the screw expanded a lot. it is 0.007" / 40" under this condition. I noticed that the fadal factory SV tables usually biased the ballscrew about 0.005" / full travel (in this case its a 4020 so 40" x) towards the short side. I see why now.

    I don't always operate in a condition where my screws experience 150 ipm all day long, and there may be plenty of downtime during the day - so I will probably just pick a number in the middle between 7thou and 0 (cold cal). and I will bias my SV table accordingly.

    here's the data of Cold vs Hot screw. what a pain in the a$$.

    Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better-x-axis-hot-vs-cold-ballscrew-nov
    if steel expands at 0.0000117inch/inch/deg C then this 0.007" error was caused by a 15 deg C temp rise - not all that much. shop is temp controlled air at 25C or so, so ballscrew is up at like 40C or so.



  9. #9
    Member
    Join Date
    Jun 2012
    Location
    United States
    Posts
    516
    Downloads
    0
    Uploads
    0

    Default Re: Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better

    I've decided to split the difference 60 / 40 : Cold / Hot. I don't run at 150ipm continuous. so i guess if i ever make a 40" long part it may be off by 0.003"? seems not all that bad to me. this is a real pain when you consider the dynamics. fadal may have been onto something with coolant thru the ballscrews on newer machines than mine - great theory, you always read that the implementation was sh*t though where dowfrost leaks a bunch. just to put all this into perspective, I can expect at worst a 0.0035" / 40" error at any temperature. so on a 6" long part that is at most 0.0005" in other words - big freaking deal.



  10. #10
    Member
    Join Date
    Jun 2012
    Location
    United States
    Posts
    516
    Downloads
    0
    Uploads
    0

    Default Re: Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better

    here's my final x, and y laser calibration results. I chose to leave a slope of about -0.00015" / 1.0" on the ballscrews when cold. that should leave plenty of room for heat expansion, but also isn't balls to the wall like the fadal factory slope. My shop isnt 85°F and 100% rapids. more like 65F, 50% rapids, lots of machine down time.

    images attached.

    Attached Thumbnails Attached Thumbnails Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better-x-axis-laser-cal-nov-16-jpg   Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better-y-axis-laser-cal-nov-16-jpg  


  11. #11
    Member
    Join Date
    Jun 2012
    Location
    United States
    Posts
    516
    Downloads
    0
    Uploads
    0

    Default Re: Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better

    Here's the ballbar results after laser calibration. it looks like the backlash is different dynamically than when running slow and doing the laser calibration stuff.
    what did laser calibration improve? 1.) scaling mismatch, 2.) positional tolerance, 3.) backlash 4.) circularity to some extent


    Servo mismatch is the biggest problem now. its at about 2.0 milliseconds for both 40 and 80 ipm

    Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better-4020-40ipm-laser-cal-11-18-2016-a

    Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better-4020-80ipm-laser-cal-11-18-2016-a



  12. #12
    Member
    Join Date
    Jun 2012
    Location
    United States
    Posts
    516
    Downloads
    0
    Uploads
    0

    Default Re: Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better

    servo tuning made a big difference for my machine.

    master clock program is supposed to take 72 seconds, mine was 77. i got that put right by adjusting pot on clock card.

    then x,y,z following error adjustments to set inch ballscrews to 595 at 150ipm feed. x was 545, y 585, z 580. got it all to 595. made ballbar result go from 4 thou circularity to 1 thou. I also adjusted BL settings to get the ballbar test minimized - so its a dynamic setting not for static like when using the laser cal setup.

    here's the final results - any more tweaking is right down in the weeds / you could spend a year doing this work...

    Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better-4020-40ipm-servo-tune-11-21-2016-a

    Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better-4020-80ipm-servo-tune-11-21-2016-a

    so, after all this sh*t i went thru on my machine and ended up with 0.0012" circularity.... i think if it is even true that your fadal is down at 0.0002" thats probably better than factory - would love to see your plots to prove it.



  13. #13
    Member
    Join Date
    Nov 2016
    Location
    Ireland
    Posts
    4
    Downloads
    0
    Uploads
    0

    Default Re: Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better

    Thanks a million for all this info mflux_gamblej, you are sure putting a lot of work into this, Can you tell me which Laser calibrator you are using. Maybe its time I invested in one



  14. #14
    Member
    Join Date
    Jun 2012
    Location
    United States
    Posts
    516
    Downloads
    0
    Uploads
    0

    Default Re: Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better

    Quote Originally Posted by IrishJ View Post
    Thanks a million for all this info mflux_gamblej, you are sure putting a lot of work into this, Can you tell me which Laser calibrator you are using. Maybe its time I invested in one
    i have the HP 5528, with material temp sensors, ambient air sensors, several types of optics for it, and gpib to usb from national instruments / hooks up to labview to automate the data collection. the HP is old, very reliable, and technologically superior to the renishaw systems. The HP is more similar to the NIST standard in the way they did the interferometer using a 2 frequency laser, dopler shifts, and a heterodyne system for using the laser wavelength as the length standard. It is a stabilized helium neon laser.

    that being said, the renishaw is user friendly. it is also about $15000 more than the HP (which of course the HP is no longer made / its from the 80s). If you can get your hands on an HP, i can recommend it for it's physics, but not so much for ease of use. its very heavy, and you have to write some software on your own for automation.

    you can probably just have someone come out to do it for you once every 2 years. that being said - if you want to custom tune your axis then you likely want your own one.

    --

    about the time i spent on this, it was a long time coming on this 4020 machine. i did all this waylube rebuild and so on / needed to laser cal anyhow. I also didn't have a ballbar but had considered getting one many times. was excited to see someone report ballbar test results on a fadal. usually people talk about a makino ballbar test which has 5 micron circularity - so what's a fadal going to be able to do when you try and really set it up right?

    i still would like to see your ballbar plots, for a frame of reference. I do believe that getting my 4020 machine down further than I did just now is going to be a whole lot of effort for not so much return on investment.

    I have a VMC20 as well, will ballbar that one when I find time.



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

Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better

Ballbar test on 4020 Scaling Error -219.15 on the Y and X not much better