603,979 active members*
1,336 visitors online*
Register for free
Login
Results 1 to 13 of 13
  1. #1
    Join Date
    Mar 2004
    Posts
    636

    Thumbs up Positional Issues

    Recently completed my CNCRP4896 and cut my first parts. Did not go good. The last line of code returned the machine to X0Y0 and it faulted out by reaching the y axis limit switch. The DRO was about .75" from reaching zero. The program was about 45 minutes long and a combination of circular pockets (100 ipm) peck drilling and profile cuts (200 ipm). Needless to say the parts were quite a ways off. I suspect some very serious lost steps.

    So tonight I went and investigated..

    I homed the machine and ran the first segment of the program - about 3 minutes. Using the verify button in Mach3 the error was +.0005 in x and -.0486 in Y

    Ran the second segment of the part - to rule out program issue. Error of -.0005 in X, -.0378 in Y.

    Looked for mechanical issues on Y (loose pulleys, pinion engagement, added some tension, checked for any binding - did not find any), reduced accel values from 40 to 20 and ran a 3rd program error was x +.001 y -.0113

    Changed the Look ahead in CV from 20 to 200 and ran a 4th program (little longer about 6 minutes)- Error x + .001, Y - .0839.

    Chnaged CV on angles from 0 to 89. Ran 1st program again Error x +.0005, Y - .0501
    Up until now all above programs were circular pockets at 40 IPM (air cuts)
    Changed to Program 5 which was profiling moves 200 IPM feed. Error X +.0015, Y - .0167
    Reduced Accell values from 20 to 10 Error x 0.0000 Y -.0162
    Reduced accell values to 5 on Y - Error X -.0005 Y -.0201
    Upped accel values to 50 on x and y and upped rapids to 500 ipm Error x -.0005 Y -.0172

    Real pleased with the X axis, not so the Y

    4 things: I feel these trials rule out the accel values being to high, I feel it is not a mechanical thing (binding etc) and the longer the program runs, the greater the error. Also, the error is always in the same direction on Y

    So what do I check next? Is my next step to swap a stepper motor out on Y and check that? Check voltage perhaps...Suggestions?

    Thanks for the Help everyone.

    Dan
    Check out what I am working on at www.routerbitz.com!

  2. #2
    Join Date
    Mar 2011
    Posts
    584
    Is your steps per inch off on the Y? Did you actually measure the steps per inch or just use the default?

  3. #3
    Join Date
    Mar 2004
    Posts
    636
    Just used the default. I have not actually measured it yet. I believe that if it was that, it would be off equal amounts both directions, but return to zero.

    Please correct me if I am wrong.
    Check out what I am working on at www.routerbitz.com!

  4. #4
    Join Date
    Feb 2004
    Posts
    1086
    Dan,

    I agree that a scaling issue should still let you return to zero. What sort of PC / OS are you using? It may be necessary to up the pulse width on the step and direction bits in Mach -- the Y driver may be on the edge of detecting these pulses if your parallel port doesn't put out a real clean signal, and could be missing one every now and then. Other than that, the main loss of motion we see is from loose set screws on the motor pulleys on the R&P drives. I know you said you checked that, but it might be worth checking one more time.

    Best regards,

    Ahren
    CNCRouterParts

  5. #5
    Join Date
    Mar 2004
    Posts
    636
    I am running an older pc 2.6 GHZ processor with 512 RAM.. I ran this same computer with another machine. I did reformat everthing, did a clean install of windows XP and Mach 3 since running it on the older router I had.

    I will look again for mechanical issues and up the pulse width today and report back.

    Dan
    Check out what I am working on at www.routerbitz.com!

  6. #6
    Join Date
    Mar 2004
    Posts
    636
    I pulled the Y axis stepper off again and rechecked the pinion on the stepper shaft etc, etc...no signs of issues. Everything seems tight.

    Ran the driver test and it gives me a "system excellent" message. Very flat line. I recorded most of the numbers on that screen if there is anything in particular that would be useful for diagnosis.

    I also thought maybe noise on that axis from the router power cord running along side of it so I unplugged that and it did not help.

    Turned the screen saver off and all the power saving items in windows that I could find...

    Ran program 1 again ....-.0486 error obvioulsy haven't found the issue yet.

    Help?

    Thanks to all in advance.
    Check out what I am working on at www.routerbitz.com!

  7. #7
    Join Date
    Feb 2004
    Posts
    1086
    Dan,

    Did you increase the pulse length in config-motor tuning? Even on a stable system, your driver may have difficulty with the parallel port signal if it is not long enough, and the driver test can't tell you anything about this. As a more general test, you could also switch the motor plugs between the Y and Z axis drivers and re-assign pins in Mach to see if the issue is related to the driver.

    Best regards,

    Ahren
    CNCRouterParts

  8. #8
    Join Date
    Mar 2004
    Posts
    636
    Sorry Ahren, I should have mentioned that. I have tried the Y axis pulse length from 3 - 7 and it did not help. It was set a 5 durign the tests yesterday. I did notice the motors not running as smooth (audibly) when I was down to 3, but 5 makes them sing.

    Tomorrow I will try and swap some parts around as you mentioned to see if I can narrow it down a bit more.

    Thanks

    Dan
    Check out what I am working on at www.routerbitz.com!

  9. #9
    Join Date
    Mar 2004
    Posts
    636

    Think I found it....

    Today I was able to find a bit more time to investigate the issue and it appears that I may have found it. While double checking the wiring I noticed that I had the wire from pin 10 on the y axis gecko drive going to the pc +5 on the PMDX. I switched this to match the other axis and plugged it into the PC Ground terminal for the Y axis.

    I have ran 4 programs the shortest of which was 2:34 and the longest 15:11 and the greatest error I have gotten was .0025 on a program with alot of .265 holes circularly interpolated with a .25 bit. Lots of small moves. I will tune my accel numbers a bit to see if I can get that any smaller, but even if I cant I am not to concerned.


    By the way, the cumulitive error of the 4 programs was .004. I can live with that on a wood router. Especially since I have not really tuned it at all.

    Thanks for everyones help.

    Dan
    Check out what I am working on at www.routerbitz.com!

  10. #10
    Join Date
    Jul 2013
    Posts
    608
    hi, where is this "verify" button in Mach 3 that you are talking about?

  11. #11
    Join Date
    Mar 2003
    Posts
    35494
    MDI page next to the Machine Coordinates button.
    Gerry

    UCCNC 2017 Screenset
    http://www.thecncwoodworker.com/2017.html

    Mach3 2010 Screenset
    http://www.thecncwoodworker.com/2010.html

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

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

  12. #12
    Join Date
    Sep 2012
    Posts
    228
    Hey Gerry... Whey you try the verify button... and it shows you some error... will it then recalibrate itself, or does it do nothing? I have run multi-tool jobs with complicated toolpaths, and seem to have started to develop some error, when I hit the verify button.... but then after the verify, the machine goes back to the point before I hit the verify button, and I don't know if it recalibrated or if it doesn't do anything about it... it seems it would be helpful, if it could automatically recalibrate...

  13. #13
    Join Date
    Mar 2003
    Posts
    35494
    I think it just tells you if there's an error, but doesn't fix the error. I've never used it. The only way to recalibrate would be to re-home the machine.
    Gerry

    UCCNC 2017 Screenset
    http://www.thecncwoodworker.com/2017.html

    Mach3 2010 Screenset
    http://www.thecncwoodworker.com/2010.html

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

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

Similar Threads

  1. Simple 5 axis positional not correct.
    By blakemachine in forum Mastercam
    Replies: 14
    Last Post: 10-15-2013, 05:03 AM
  2. positional feedback to stepper?
    By SwampDonkey in forum Linear and Rotary Motion
    Replies: 11
    Last Post: 09-18-2012, 02:04 AM
  3. Can a DC motor be used with positional feedback as a servo?
    By fle33.com in forum Servo Motors / Drives
    Replies: 7
    Last Post: 12-15-2009, 01:23 AM
  4. Unrealistic Positional Accuracy Specs?
    By LazyMan in forum DIY CNC Router Table Machines
    Replies: 3
    Last Post: 06-03-2009, 10:04 PM
  5. Positional 5-Axis Machining
    By PrecisionD in forum Surfcam
    Replies: 1
    Last Post: 10-06-2008, 10:25 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
  •