Adjustments: Inaccuracy of Cut Length.


Results 1 to 6 of 6

Thread: Adjustments: Inaccuracy of Cut Length.

  1. #1
    Registered
    Join Date
    Apr 2016
    Location
    Canada
    Posts
    9
    Downloads
    0
    Uploads
    0

    Default Adjustments: Inaccuracy of Cut Length.

    Hi all, I'm trying to get my Momus CNC to cut more accurately. Currently to test accuracy I am cutting a 3" square, 0.5" Straight Cutter, 50 IPM, 75% stepover, Climb Cut. I am consistently getting a measurement of 2.98" for the x, and between 2.96 to 2.97 for the Y axis. Here is the things I have tried.

    I have tried changing the speed of cut, no appreciable difference between 80 IPM, 50 IPM, and 20 IPM.
    I have tried using exact stop mode, no change noted.
    I tightened both the x and y belts, and the belt clamps. I also tightened the pulley set screws, but I can not tighten them too much because the set screw is tiny and is threaded in thin aluminum. No change here.
    I tried loosening the bolts on the bearings. Initially the force needed to break the gantry/ carriage loose when disconnected from the belts was 6 to 10 pounds, measured with a fishing scale. Now with the bolts loosened the break loose force is reduced to under 5 lbs for both axis. Machine moves a little more smoothly but the dimensions did not improve.
    I switched from 4x microstepping to 2x microstepping. No change in the x, still came out 2.98, y came out 2.95. I only did one cut like this so I'm not sure if that was a fluke or not.

    So let me present what I think is wrong. Considering that changing speeds did not affect the result, I don't think the motor is loosing steps because its being over worked. The fact that loosening the bolts on the bearings and changing from 4x to 2x microstepping had no effect also leads me to believe that the motor is probably not missing steps. I'm running Kelling 425 oz in steppers, driven by Kelling KL 4030 drivers, using a UC 100 and Mach 3. Step and DIr pulse length is set to 4 micro seconds for both.

    The most likely suspect is that the pulley is moving on the shaft. I am considering loctiting the pulleys on. Does anybody know how hot a stepper motor can be before the seals/insulation are damaged? I am wondering in case I have to remove the loctite.

    Secondly, I heard that the spindle being poorly trammed can cause inaccuracy in the length of cut. The Momus is kind of a pain in the ass to tram the spindle, so mine is currently out of square by a bit. On the squares the ridge left behind is about as thick as a piece of paper. Somehow I doubt this is causing the cuts to come out inaccurately.

    Your advice is much appreciated. Feel free to bring up anything I may have missed. I'm just looking for second thoughts before I start taking things apart.

    Similar Threads:


  2. #2
    Member ger21's Avatar
    Join Date
    Mar 2003
    Location
    Shelby Township
    Posts
    35538
    Downloads
    1
    Uploads
    0

    Default Re: Adjustments: Inaccuracy of Cut Length.

    How are you programming the cut? There shouldn't be any "stepover" settings in a profile cut.

    What size do you get if you cut a 6" square? If the error is double, then adjust your steps/unit.

    Gerry

    UCCNC 2017 Screenset
    [URL]http://www.thecncwoodworker.com/2017.html[/URL]

    Mach3 2010 Screenset
    [URL]http://www.thecncwoodworker.com/2010.html[/URL]

    JointCAM - CNC Dovetails & Box Joints
    [URL]http://www.g-forcecnc.com/jointcam.html[/URL]

    (Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)


  3. #3
    Registered
    Join Date
    Apr 2016
    Location
    Canada
    Posts
    9
    Downloads
    0
    Uploads
    0

    Default Re: Adjustments: Inaccuracy of Cut Length.

    It is a pocket cut. Now that I look back at previous cuts, I may have to adjust the steps per inch on the Y as the error does increase with the side length for the Y axis, but this will still not completely eliminate the problem because just looking at the numbers I can see that if I plotted cut length as a function of model side length, the y intercept would not be zero, indicating some backlash.
    Model Length / Measured Cut length for y axis.
    1 / 0.981
    3 / 2.971
    5 / 4.965
    Plotting this, the line of best fit is y = 0.996x - 0.016.
    I'll report back after the long weekend.



  4. #4
    Registered
    Join Date
    Apr 2016
    Location
    Canada
    Posts
    9
    Downloads
    0
    Uploads
    0

    Default Re: Adjustments: Inaccuracy of Cut Length.

    I'm using Cam Bam, is it possible to reduce the acceleration during the actual milling movements (ei not rapids). As far as I know the acceleration settings in mach 3 only apply to rapids. I tried replacing my drive pulley with one that I thought was higher quality. It made a marginal difference. Reducing the error from ~0.03" to ~0.02 to 0.015". Now I am thinking maybe it is an issue with missing steps? I also tried tightening the crap out of the y belt, did not seem to make a difference, if anything the accuracy decreased slightly.

    What sort of accuracy are you guys getting with your Momus machines? I would be happy if I could get consistently ~0.005" but finding the source of this inaccuracy is proving quite illusive.



  5. #5
    Registered
    Join Date
    Apr 2016
    Location
    Canada
    Posts
    9
    Downloads
    0
    Uploads
    0

    Default Re: Adjustments: Inaccuracy of Cut Length.

    Update, I'm going to try doing a fine finishing cut and see if that improves the accuracy. I'm thinking maybe 0.03" stepover and 0.01" DOC at 40 IPM. If this improves the cut accuracy, I will know that gantry racking is likely causing inaccuracy.



  6. #6
    Member
    Join Date
    Apr 2007
    Location
    USA
    Posts
    1955
    Downloads
    0
    Uploads
    0

    Default Re: Adjustments: Inaccuracy of Cut Length.

    I wondered how you are doing with narrowing down the accuracy question?

    A few years ago I looked at building a belt driven system and the implications. A couple of things stood out:
    - The concept makes the assumption that the belt teeth and pitch are exactly the same over the entire length of the belt. This may not always be reality and for most belts it is unlikely.
    - If the pulley or shaft are even slightly eccentric, the impact is amplified. For example, if the center of the pulley or pulley circumference is just 0.001 inch off, as it rotates the impact on distance will be (0.001 x pi) so 0.00314 inches and will happen in both directions as it spins. One way to test this is with a dial indicator, but still this is really a measurement of the edge of the pulley, not the actual pulley pitch diameter.
    - The build I was looking at was for a 4 x 8 ft machine using a conventional router, and I was surprised when my calculations showed that I needed at least a 6 inch wide belt to avoid "bounce" as the directions changed and that still was not zero. One way to tell if this is happening is to give the system a chance to "settle" before it changes direction by adding in a delay.

    I am not criticizing the design - it is a low cost build with DIY rails. It is actually very clever. These are DIY rails and motion, not professional grade carriages like a real cnc router would use, so you might be at the upper end of the concept already.

    The frame design is pretty good, but at the same time, it is built from wood and assembled from pieces. It "might" be slightly more rigid if built up from fewer, larger pieces. Temperature and humidity can affect this as well.

    IIRC, Mach contains a setting that allows you to plot the "actual distance" vs "programmed distance" and add in corrections. I think it is used to map ball screws for very precise motions. Perhaps this will help a bit more.

    Your goal is nearly 5x where you are now - I am not sure if it is possible to get there with the existing rails.



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

Adjustments: Inaccuracy of Cut Length.

Adjustments: Inaccuracy of Cut Length.