584,812 active members*
5,228 visitors online*
Register for free
Login
Page 3 of 4 1234
Results 41 to 60 of 63
  1. #41
    Join Date
    May 2006
    Posts
    4043

    Re: Rough Finish on 45 Degree wall

    Hi Troy,

    Wonder if it could be the low integrator limit? Maybe its not quite high enough to get you into the exact position, and then when you send it another 0.001 it moves the 0.001 plus previous error? Might be long shot but could be worth trying. I didnt run the math on counts etc, maybe the error im thinking about would be way smaller then a thou and not really noticable.
    I think Murray is right. We know the friction is about 500 DAC counts and stiction could be higher. So with an error stuck at say 0.001 inch the Integrator would ramp to 500 and with a resolution of 20325 and P of 6.5 the DAC would be at:

    500 + 0.001 x 20325 x 6.5 = 622

    If that isn't enough to overcome stiction the error would persist. It would be easy to verify. With some error of say 0.001 inch look at the Analog Screen and observe the DAC. Is it stuck at some value like 600 yet the motor isn't moving?
    TK
    http://dynomotion.com

  2. #42
    Join Date
    Dec 2007
    Posts
    578

    Re: Rough Finish on 45 Degree wall

    Quote Originally Posted by TomKerekes View Post
    Hi Troy,

    I think Murray is right. We know the friction is about 500 DAC counts and stiction could be higher. So with an error stuck at say 0.001 inch the Integrator would ramp to 500 and with a resolution of 20325 and P of 6.5 the DAC would be at:

    500 + 0.001 x 20325 x 6.5 = 622

    If that isn't enough to overcome stiction the error would persist. It would be easy to verify. With some error of say 0.001 inch look at the Analog Screen and observe the DAC. Is it stuck at some value like 600 yet the motor isn't moving?

    I re tuned the drive using DMM Tech's latest software that has autotuning. After this i re tuned in Kmotion and was getting much better results, about .0005" to .0007" of error. Then i done a test feed move at 1IPM of .001" ,the DAC sets at about -40 to -50 after it has moved, its dithering a little, but KMCNC dro does not show any movement. So i tryed another test move of .0005" and measured movement with an indicator is .001"

  3. #43
    Join Date
    Dec 2007
    Posts
    578

    Re: Rough Finish on 45 Degree wall

    So i just tried same move test of .001" within Kmotion's Gcode. In Kmotion Tool Setup i configured Y axis( this is the axis i have been testing.) as 20325cpi with Velocity of 15 and Accel of 35, just like it is in KMCNC. Made a small Gcode in Kmotion of .001" and my .0005" indicator barely moved. So i did a move of .005" and indicator showed .0005". Last test was a move of .010" and indicator measured .001". Did several tests, re zeroing axis using Kmotion Step Response window , and re ran same Gcode move of .01" with exact same results , indicator measured .001" movement every time. And appeared to be smooth movement when watching indicator.
    Also, after this move of .01" the DACs stop at -60 and Axis window show -8 on Dest and -8 on position for axis 1. And when i check the DRO in KMCNC, which is set to Display Encoders, there is no movement shown.

    Do i have something configured wrong in KMCNC that could be causing all this?

    Troy

  4. #44
    Join Date
    May 2006
    Posts
    4043

    Re: Rough Finish on 45 Degree wall

    Hi Troy,

    I think you are in mm mode rather than Inches.
    0.01 x 20325 / 25.4 = 8 counts

    The way to troubleshoot these types of issues is to break it down to categories of:

    1 - resolution/scaling errors (make a big move and check distance - if correct this is not the problem)
    2 - servo tuning errors (check if commanded Destination and Encoder Position match - if matches, servo tuning is not the problem)
    3 - mechanical errors (use indicator and make small repeated moves in same direction to determine if error accumulates or varies, then reverse to check for backlash)

    Note I would use KMotion.exe's GCode Screen. It hasn't been updated in a long time. I think it will just add confusion.
    TK
    http://dynomotion.com

  5. #45
    Join Date
    Dec 2007
    Posts
    578

    Re: Rough Finish on 45 Degree wall

    Hi Tom,
    Yup it was in mm.

    Found a couple things had me chasing my tail. One, after i would adjust the drive Gear and Line number then reboot drive the Line number was changing from the 4000 i was trying to test to 2048. Second, was my Brown&Sharp .0005" test indicator was not taking the sudden movement/shock of axis and was causing it to bounce. Did a test Gcode with .0005" moves with a M01 after each move then return to 0.The indicator would read .01" to .0105" after running all moves then would return to 0 within .0005" ,tested this several times and would repeat within .0005" every test. Ugh
    Then decided to reluctantly use my brand new Interapid .0001" indicator. Reluctantly, because i just knew i would get a run away condition a crunch my indicator. Anyhow after checking with my Interapid i can move .0001" and it shows .0001" with about .0002" back lash.

    Also, the ADC values are around 1.784 Volts, should these be higher? On my lathe they are around 5volts.

    Attached is my latest tune on Y axis, any suggestions on what i should change? The DMM auto tune software has 3 different settings for Inertia i can try also.



    Troy

  6. #46
    Join Date
    May 2012
    Posts
    537

    Re: Rough Finish on 45 Degree wall

    Good to hear you have that sorted out.

    DMM told me that none of the auto tune stuff applies when using torque mode, so probably no point in trying those. I think thats torque mode tends to work better. Kflop is in complete control.

  7. #47
    Join Date
    May 2006
    Posts
    4043

    Re: Rough Finish on 45 Degree wall

    Hi Troy,

    As Murray points out in Torque mode the drive should just be applying commanded current so there shouldn't be any tuning for load.

    Also, the ADC values are around 1.784 Volts, should these be higher? On my lathe they are around 5volts.
    Not sure what you are referring to. Do you have ADCs connected to anything?

    Tuning looks good to me.
    TK
    http://dynomotion.com

  8. #48
    Join Date
    Dec 2007
    Posts
    578

    Re: Rough Finish on 45 Degree wall

    Quote Originally Posted by mmurray70 View Post
    Good to hear you have that sorted out.

    DMM told me that none of the auto tune stuff applies when using torque mode, so probably no point in trying those. I think thats torque mode tends to work better. Kflop is in complete control.
    Thats what i thought but if you look at there manual (pages attached)almost all parameters apply to Analog Torque mode, unless i am misunderstanding it. And before i ran the autotune my error count in Kmotion was around 35 to 45. After i ran autotune i got a much lower error as above plot shows. Also had to reduce D and the feed forward settings in Kmotion.

  9. #49
    Join Date
    Dec 2007
    Posts
    578

    Re: Rough Finish on 45 Degree wall

    Quote Originally Posted by TomKerekes View Post
    Hi Troy,

    As Murray points out in Torque mode the drive should just be applying commanded current so there shouldn't be any tuning for load.

    Not sure what you are referring to. Do you have ADCs connected to anything?

    Tuning looks good to me.
    Sorry, disregard that, was confusing myself.

  10. #50
    Join Date
    May 2006
    Posts
    4043

    Re: Rough Finish on 45 Degree wall

    Hi Troy,

    Regarding DMM Parameters:

    Max Acceleration - maybe changes the torque response time? I suppose it could be somehow autotuned?

    On Position Range - can't imagine how that could be used in torque mode. The drive doesn't even know the desired position.

    Line Num - yes sets the encoder resolution.
    TK
    http://dynomotion.com

  11. #51
    Join Date
    Dec 2007
    Posts
    578

    Re: Rough Finish on 45 Degree wall

    I dont think these drives are completely thought through.

    Anyhow, I have good position and jogging results now from Z and Y. The X axis makes a crackling(kind of like wood burning) sound when making small moves of .0001" to .001". And does not move as smooth as Z andY. When moving .0001" it sometimes takes 2 moves before table shows movement with an indicator and then will jump .0002" to .0003" . The DACs show about -70 to -80 during the move. The crackling sound also gets louder until axis table actually moves.If i turn down 'P' the sound gets quieter (same type of sound) but movement of table is still the same. When idle this motor makes more noise than the other 2.

    The X axis motor was replaced last year and had a different size shaft. So wondering if something was updated/changed from the older motor and or drive. Reason i changed motor was a corner of one of the magnets inside broke off for what ever reason. Thinking about testing the old motor even tho it might be weaker now. You think there would be any issue? The piece that broke of was about 3/16 square.

    Troy

  12. #52
    Join Date
    May 2006
    Posts
    4043

    Re: Rough Finish on 45 Degree wall

    Hi Troy,

    The X axis makes a crackling(kind of like wood burning) sound when making small moves of .0001" to .001"
    Strange. Maybe make a Move on the Step Response Screen for clues. Maybe try swapping the X Y Drives and Motors to isolate whether it is the Motor or Drive.


    The X axis motor was replaced last year and had a different size shaft. So wondering if something was updated/changed from the older motor and or drive. Reason i changed motor was a corner of one of the magnets inside broke off for what ever reason. Thinking about testing the old motor even tho it might be weaker now. You think there would be any issue? The piece that broke of was about 3/16 square.
    Lower torque shouldn't be an issue except potentially reducing maximum acceleration some. But if torque is uneven as a function of rotor position (torque ripple) that might be an issue.
    TK
    http://dynomotion.com

  13. #53
    Join Date
    Jan 2005
    Posts
    15362

    Re: Rough Finish on 45 Degree wall

    Quote Originally Posted by Need TECH Help! View Post
    Hi Tom,
    The 150 is slightly smoother but nothing like the finish of a single axis cut.

    Attached is a picture of closed loop at the Drives only. Even tho picture does not show it well,there is now a very visible pattern about every .100 to .125". Checking with a .0001" indicator as before, the high to low spots is easily .0008"
    With closed loop still at drives only , watching the encoder readout in KMCNC, X and Y are off by .0002-.0004" from what was commanded.
    So i would say when KFLOP is in closed loop mode that it is doing what it can with the way the drives are in Step/Dir mode.
    Next test will be in analog mode.
    Troy
    I have the same servo system on one of my machines and have never seen any cutting problems like this, here is what I get in A36 1" steel plate, with a not new cutter full 1" depth of cut x .020" cut

    What oil pressure do you have for your way lube
    Attached Thumbnails Attached Thumbnails Spindle Plate-7.jpg   Spindle Plate-6.jpg  
    Mactec54

  14. #54
    Join Date
    Dec 2007
    Posts
    578

    Re: Rough Finish on 45 Degree wall

    I have the same servo system on one of my machines and have never seen any cutting problems like this, here is what I get in A36 1" steel plate, with a not new cutter full 1" depth of cut x .020" cut
    What does a 45* side mill cut look like for you?

  15. #55
    Join Date
    Dec 2007
    Posts
    578

    Re: Rough Finish on 45 Degree wall

    Hi Tom,

    Strange. Maybe make a Move on the Step Response Screen for clues. Maybe try swapping the X Y Drives and Motors to isolate whether it is the Motor or Drive.
    Noise is from the motor. Tried swapping the drives and same issue and noise on X motor using Y axis drive. After this i decided to swap motor with Z axis. And realized/remembered that the newer motor on X axis does not only have a smaller shaft but is 3/4" shorter and about 3lbs lighter.
    The tuning is way different with this newer motor on Z axis. Before i was getting an error of about 14 now i cant get it better than 17-18. Previous PID was P10, I24, D2 . The newer motor tune so far is P22, I34, D0 with less feed forward, a LowPass Filter of 400 and lower acceleration. Still has the same type noise when used on Z axis only louder. Plus accuracy is about .0005". With older motor it was within .0002" and smoother motion. Think i might try damaged motor on Z axis. Then latter down the road try a whole different servo and drive set up.

    Also, did a quick 45* test cut with newer motor and the older motor of Z axis.With newer motor on X axis finish was better than before with drive in Step/Dir. When using the older Z axis motor on the X axis the finish was even better with only about .0002" high and low spots. Will get a pic tomorrow.

    Troy

  16. #56
    Join Date
    May 2012
    Posts
    537

    Re: Rough Finish on 45 Degree wall

    Any chance the noise is from the coupling rocking on the shaft? You mention shaft is smaller, did you sleeve the old coupler maybe? Improper fit? Something off center a little and rubber coupler making a noise because of it? Could try greasing the rubber coupler too, sometimes they make a little noise. Not much to make a noise inside those motors but if it sounds much different then the others id certainly get it checked out. Brushless motors should hardly make a sound.

    Glad to hear its cutting better.

  17. #57
    Join Date
    Dec 2007
    Posts
    578

    Re: Rough Finish on 45 Degree wall

    The new motor shaft was about 1 mm smaller, i had to replace the coupling. Couplings are 2025 T6 aluminum.Bore is very good fit around shaft with a split and bolt to clamp.Just the spring tension alone of the split bore clamps on shaft very tight.

    These motors have never been quiet like other manufactures servos. They have a slight sound of a stepper motor when running at slow feeds. Been looking at specs and user feedback of Deltas, might end up going this route when iam ready.

  18. #58
    Join Date
    Jan 2005
    Posts
    15362

    Re: Rough Finish on 45 Degree wall

    Quote Originally Posted by Need TECH Help! View Post
    What does a 45* side mill cut look like for you?
    Just the same that is why I show the corner and the inside bore holds better than .0001" on the bore X Y interpolating and very smooth
    Mactec54

  19. #59
    Join Date
    Jan 2005
    Posts
    15362

    Re: Rough Finish on 45 Degree wall

    Quote Originally Posted by Need TECH Help! View Post
    The new motor shaft was about 1 mm smaller, i had to replace the coupling. Couplings are 2025 T6 aluminum.Bore is very good fit around shaft with a split and bolt to clamp.Just the spring tension alone of the split bore clamps on shaft very tight.

    These motors have never been quiet like other manufactures servos. They have a slight sound of a stepper motor when running at slow feeds. Been looking at specs and user feedback of Deltas, might end up going this route when iam ready.
    The motors I have run quiet they make almost no noise at all, I have Yaskawa also and they run just the same, noisy servos is usually from setup programing or are over loaded and are working harder than they should be to move your machine axis, what type of couplings are you using if they are not diaphragm type then you will be getting windup in the coupling from loading and then you have the stick slip to deal with, you have so many problems on these old machines everything has to be perfect to get them working how you want

    Yes Delta make a good servo system but if your machine is not in perfect condition the same thing is going to happen
    Mactec54

  20. #60
    Join Date
    Jan 2005
    Posts
    15362

    Re: Rough Finish on 45 Degree wall

    Quote Originally Posted by Need TECH Help! View Post
    What does a 45* side mill cut look like for you?
    Here is a aluminum part with angles cut
    Attached Thumbnails Attached Thumbnails HPIM2560.jpg  
    Mactec54

Page 3 of 4 1234

Similar Threads

  1. Only MeshCam has option to rough and finish?
    By estes53 in forum Uncategorised CAM Discussion
    Replies: 5
    Last Post: 07-18-2015, 04:46 AM
  2. VMC Side wall surface finish, Newbie question
    By BlueTwo in forum Uncategorised MetalWorking Machines
    Replies: 10
    Last Post: 10-07-2011, 10:06 AM
  3. Problems with rough finish
    By 44-henry in forum Tormach Personal CNC Mill
    Replies: 11
    Last Post: 06-02-2009, 06:43 PM
  4. How to rough/finish threadmill in MCX?
    By John_B in forum Mastercam
    Replies: 4
    Last Post: 07-27-2008, 04:25 PM
  5. Very rough finish when turning 6061-T6 ... why?
    By cnczoner in forum Uncategorised MetalWorking Machines
    Replies: 18
    Last Post: 03-16-2008, 11:24 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
  •