Z-axis overshooting sometimes - Page 2


Page 2 of 4 FirstFirst 1234 LastLast
Results 13 to 24 of 45

Thread: Z-axis overshooting sometimes

  1. #13
    Member
    Join Date
    May 2005
    Location
    USA
    Posts
    3850
    Downloads
    0
    Uploads
    0

    Default Re: Z-axis overshooting sometimes

    Coupling slippage is a huge issue with the lower cost clamp types. However you wouldn’t hear the slippage thus there was or could still be lubrication or binding issues. In any event considering previous posts I suspect that your whole machine could use a bit of maintenance.



  2. #14
    Member
    Join Date
    Jun 2010
    Location
    Australia
    Posts
    3885
    Downloads
    4
    Uploads
    0

    Default Re: Z-axis overshooting sometimes

    Is this a Chinese machine? I ask because you often find that the Allen head screws they use inside China are rather soft. The hex hole burrs around and the tip of the bolt squashes. Strangle, the Allen head bolts I have bought from China/eBay have been good.
    So ... try replacing the bolts with a better brand, and tighten them up.

    By the by: I have a small Chinese lathe. It goes OK now that I have replaced just about every Allen head bolt in it! Especially in the QCTP!

    Cheers
    Roger



  3. #15
    Member
    Join Date
    Jan 2008
    Location
    USA
    Posts
    87
    Downloads
    0
    Uploads
    0

    Default Re: Z-axis overshooting sometimes

    I took the z axis off tonight to check it out, there were a couple loose nuts there. I tightened it all back up and it seems to be working like a champ now. I think the threads are slightly stripped on one of the couplers, I may at one point do a through bolt drill through both the coupler and acme screw just so that I know it'll never slip on me.

    Regarding maintenance, what all maintenance needs to be done other than cleaning off and oiling the cold rolled steel and the acme threads?

    It is a fineline automation machine,



  4. #16
    Member
    Join Date
    Jan 2008
    Location
    USA
    Posts
    87
    Downloads
    0
    Uploads
    0

    Default Re: Z-axis overshooting sometimes

    I thought I had the issue sorted out, and went to try again today and the bit dove right into my piece. I took the Z-axis apart to install a new anti-backlash nut, and noticed when tightening down all the bolts that one of the acme coupler screws was broken. Not sure how I missed that before, but instead of messing with a screw, I tig welded the acme thread right to the coupler, there is no way that they will slip now. Honestly, I really doubt that it was slipping before because I tried spinning the other end of the amce rod with pliers hard enough to torque it a bit and mar up the threads and it wasnt budging.

    I tried started a gcode (the same one that gave me trouble this morning) about 10 times and the Z seems to have stuck for me. On one run though, it looks like I lost/gained about 1/16" in the X&Y axes as I never touched their offsets when repeating the code over and over. I double checked the tension on the R&P drives, its nice and tight. At this point I think I may be having a computer issue. Is that possible? I am running the newest software 1.2111 with your 2017 screenset Ger, because I was having an EMI issue with my VFD so I run 2ms debounce ignore so the spindle doesn't trip the limits.



  5. #17
    Member
    Join Date
    Jun 2010
    Location
    Australia
    Posts
    3885
    Downloads
    4
    Uploads
    0

    Default Re: Z-axis overshooting sometimes

    2 mS is awfully long. With that sort of filtering you could even miss encoder pulses.
    How about setting the speed and acceleration to painfully low values (5%) and rerunning the test programs? If the system is then OK, your filter length may be a problem.

    Cheers
    Roger



  6. #18
    Community Moderator ger21's Avatar
    Join Date
    Mar 2003
    Location
    Shelby Township
    Posts
    34534
    Downloads
    0
    Uploads
    0

    Default Re: Z-axis overshooting sometimes

    The encoders go directly to the drivers, so wouldn't be affected.

    As I mentioned before, you shouldn't be able to lose position with servos, as the drives should be faulting.

    Also, keep in mind that 1.2111 is a test version which may contain bugs. If you are having any issues, you should be trying the stable release.

    And I don't think any computer issues would cause a loss of position.

    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)


  7. #19
    Member
    Join Date
    Jan 2008
    Location
    USA
    Posts
    87
    Downloads
    0
    Uploads
    0

    Default Re: Z-axis overshooting sometimes

    The encoders go directly to the drivers, so wouldn't be affected.

    As I mentioned before, you shouldn't be able to lose position with servos, as the drives should be faulting.

    Also, keep in mind that 1.2111 is a test version which may contain bugs. If you are having any issues, you should be trying the stable release.

    And I don't think any computer issues would cause a loss of position.
    That's true, and I don't see these motors losing position at all with their strength and the speeds I have dropped them to.

    True as well, unforunately, I need a version with the debounce option otherwise my vfd is tripping the limit switches.

    2 mS is awfully long. With that sort of filtering you could even miss encoder pulses.
    How about setting the speed and acceleration to painfully low values (5%) and rerunning the test programs? If the system is then OK, your filter length may be a problem.
    Hmm, I guess that is a decently long time. Unfortunately the software doesn't allow for decimals and 1ms was still tripping occasionally.


    UGH, all this is starting to sound like I need to just pony up and buy a new VFD instead of trying to limp this one along. Are there any decent 2.2kw VFDs out there for the money? My spindle still seems to be operating 100% fine.



  8. #20
    Member
    Join Date
    Apr 2016
    Location
    United States
    Posts
    780
    Downloads
    0
    Uploads
    0

    Default Re: Z-axis overshooting sometimes

    Quote Originally Posted by drxlcarfreak View Post
    I was having an EMI issue with my VFD so I run 2ms debounce ignore so the spindle doesn't trip the limits.
    Quote Originally Posted by drxlcarfreak View Post
    T unforunately, I need a version with the debounce option otherwise my vfd is tripping the limit switches.

    Unfortunately the software doesn't allow for decimals and 1ms was still tripping occasionally.

    UGH, all this is starting to sound like I need to just pony up and buy a new VFD instead of trying to limp this one along. Are there any decent 2.2kw VFDs out there for the money? My spindle still seems to be operating 100% fine.
    New information we didn't know before. As you've already identified, you have a noise problem.

    1. Have you been using dust collection, and if so, do you have it grounded with a bare braided ground wire in the hoses. If you are using PVC anywhere in the system, the static buildup goes way up. I have typical 4" clear hoses and some PVC in my system. The amount of static buildup was incredible. If I got anywhere near the hose with the dust collector on, my hair stood on end. I had limit switches tripping from static within less than a minute of running time. Properly grounded (to earth ground) and the problem went away. I have a ground wire inside the PVC and wrapped abound the outside. Ditto for the hose running from the PVC to the Z axis.

    2. Mactech suggested EMI filter on your incoming power line to the VFD to reduce noise. With the problem you are having, it makes sense to try it. The filters are pretty cheap. I admit that I don't have one, but I'm not having problems with too much noise.

    3. What wire are you using from your VFD to the spindle? Is it grounded at both ends? Improper wire (it needs to have a braided shield) and improper grounding can/will cause noise issues. Igus makes some great cable for VFDs. I'm using IGUS CF6, which is flexible and well shielded.

    4. Is the wiring to your limit switches shielded? If not, it probably should be.

    I wouldn't be too quick to blame the VFD for your noise issues. I believe virtually all VFDs will put out noise. Some just put out more than others. If you are using an HY VFD (one of the more popular Chinese VFDs), you should be able to get the noise down to a usable level. Even if you buy one of the very best/most expensive VFDs, you could still have noise issues. The trick is to filter out/shield out/ground out the noise.

    If you want a lower noise VFD, you might consider an Hitachi model. The WJ200 is popular and gets good r=eviews, although it costs a lot more than the Chinese models. In any event, if you don't take the precautions noted above, you many not solve your limit switch tripping problem.

    Maybe mactech54 will stop by with some pearls of wisdom. He knows way more about VFDs than most of us.

    I still think your over-travel issue sounds mechanical.

    Gary


    The Old Man and the C -----NC


  9. #21
    Member
    Join Date
    Jan 2005
    Location
    USA
    Posts
    10309
    Downloads
    0
    Uploads
    0

    Default Re: Z-axis overshooting sometimes

    Quote Originally Posted by drxlcarfreak View Post
    That's true, and I don't see these motors losing position at all with their strength and the speeds I have dropped them to.

    True as well, unforunately, I need a version with the debounce option otherwise my vfd is tripping the limit switches.



    Hmm, I guess that is a decently long time. Unfortunately the software doesn't allow for decimals and 1ms was still tripping occasionally.


    UGH, all this is starting to sound like I need to just pony up and buy a new VFD instead of trying to limp this one along. Are there any decent 2.2kw VFDs out there for the money? My spindle still seems to be operating 100% fine.
    Another VFD is not going to change anything they all make EMI, if it is not wired correctly will play havoc with your whole system, a good test if you have everything wired correct is no debounce is needed, so need some photos of your wiring

    Do you have a input power filter in front of the VFD Drive this can help a lot with the EMI here is one that I think is best suited for your VFD TDK Lambda RSEN 2030L this would be a starting point

    Next is your cable from the VFD Drive to the Spindle 4 core Shielded or Double Shielded is required , and the shields correctly terminated at each end

    Attached Thumbnails Attached Thumbnails Z-axis overshooting sometimes-tdk-lambda-rsen-2030l-power-filter-png  
    Mactec54


  10. #22
    Member
    Join Date
    Jan 2008
    Location
    USA
    Posts
    87
    Downloads
    0
    Uploads
    0

    Default Re: Z-axis overshooting sometimes

    Yes, sorry I forget that not everyone knows what I have been going through and forget to mention everything. The machine was running fine for about 2 years, then all of a sudden I got a weird issue where starting the spindle would occasionally trigger a limit, if I jiggled the spindle cable it would work. I just assumed that the wire for the spindle was too close to the Z axis inductance sensor. Then one day no matter what I did I couldn't get the spindle to turn on at all without triggering a limit. The only way to get the spindle to run without tripping a limit was to unplug everything from the control box minus the spindle, but then again when I did that, it was running in limit override mode because the servos were showing as tripped, so it may very well have still been tripping a limit. It was mentioned on the UCCNC forum that the cheaper chinese VFDs main capacitors dry out quickly and increase the noise. Updating the software to get the debounce parameter fixed it, but I think it possibly has introduced other issues.

    1. I do have dust collection, but with my issues, I use it very rarely, but, I will try grounding it to see if that helps with anything.

    2. I am looking for a decent EMI filter. For a 2.2kw spindle, will a 20amp filter do? Doing a quick calc at single phase 220v, the max it will pull is 10amps, plus losses in the VFD.

    3. I am using shielded and either 18/4 or 16/4 wire. Inside the case is a different story. For some stupid reason I stripped all of the wires to make it easier to terminate the wires so there isnt any shielded cables inside the case. I do plan to get clamp on ferrite rings to try and reduce the EMI in there as well. The grounding I will have to check, but I believe that I wired the ground to the spindle. and the other ground to the case.

    4. Looking at the webpage I got them from (https://www.automationdirect.com/adc..._cables/evc179) again, it doesn't look like it is shielded. Would ferrite rings on either end help out here?

    I only blame the VFD because it seemed to be the root of my other issues forcing me to update the software to an alpha release. Another problem I can think of is I put everything in one case and the breakout boards and directly above the VFD. When wiring it up, I had read to keep everything away from the power supply because it has noise, but heard nothing about a VFD until after everything was wired in. Would moving it out, or maybe adding sheet steel/aluminum between the top of the VFD and the control boards help keep noise separated?

    Z-axis overshooting sometimes-20181110_010351-jpg

    If you think it is mechanical, what other items could it be? Losing steps I totally get in every direction, but gaining steps doesn't make sense to me, at least mechanically wise. The Z sort of made sense since if the coupler is slipping the weight on the acme thread just allows it to keep going into the board when the motor stops. I don't think it was slipping because before I welded the coupler to the rod I had the motor holding steady and used vice grips to try and make the rod slip, I put a ton of force on it and the rod didn't budge. That said, I am totally willing to have my horizons broadened.



  11. #23
    Member
    Join Date
    Jan 2008
    Location
    USA
    Posts
    87
    Downloads
    0
    Uploads
    0

    Default Re: Z-axis overshooting sometimes

    Quote Originally Posted by mactec54 View Post
    Another VFD is not going to change anything they all make EMI, if it is not wired correctly will play havoc with your whole system, a good test if you have everything wired correct is no debounce is needed, so need some photos of your wiring

    Do you have a input power filter in front of the VFD Drive this can help a lot with the EMI here is one that I think is best suited for your VFD TDK Lambda RSEN 2030L this would be a starting point

    Next is your cable from the VFD Drive to the Spindle 4 core Shielded or Double Shielded is required , and the shields correctly terminated at each end
    I was only saying a new VFD based on the assumption from another forum thread that this one is on its way out (https://www.forum.cncdrive.com/viewtopic.php?f=2&t=2092).

    I have no input filter yet. I was planning on picking one of those up as well as some ferrite rings for the VFD output and the stepper motor outputs.

    Sure, here are a bunch of pictures from when I upgraded to servos.

    Z-axis overshooting sometimes-20181101_220300-jpg

    Z-axis overshooting sometimes-20181102_231706-jpg

    Z-axis overshooting sometimes-20181106_210517-jpg

    Z-axis overshooting sometimes-20181106_215453-jpg

    Z-axis overshooting sometimes-20181106_215458-jpg

    Z-axis overshooting sometimes-20181106_215503-jpg

    Z-axis overshooting sometimes-20181107_172302-jpg



  12. #24
    Member
    Join Date
    Jan 2005
    Location
    USA
    Posts
    10309
    Downloads
    0
    Uploads
    0

    Default Re: Z-axis overshooting sometimes

    Quote Originally Posted by drxlcarfreak View Post
    Yes, sorry I forget that not everyone knows what I have been going through and forget to mention everything. The machine was running fine for about 2 years, then all of a sudden I got a weird issue where starting the spindle would occasionally trigger a limit, if I jiggled the spindle cable it would work. I just assumed that the wire for the spindle was too close to the Z axis inductance sensor. Then one day no matter what I did I couldn't get the spindle to turn on at all without triggering a limit. The only way to get the spindle to run without tripping a limit was to unplug everything from the control box minus the spindle, but then again when I did that, it was running in limit override mode because the servos were showing as tripped, so it may very well have still been tripping a limit. It was mentioned on the UCCNC forum that the cheaper chinese VFDs main capacitors dry out quickly and increase the noise. Updating the software to get the debounce parameter fixed it, but I think it possibly has introduced other issues.

    1. I do have dust collection, but with my issues, I use it very rarely, but, I will try grounding it to see if that helps with anything.

    2. I am looking for a decent EMI filter. For a 2.2kw spindle, will a 20amp filter do? Doing a quick calc at single phase 220v, the max it will pull is 10amps, plus losses in the VFD.

    3. I am using shielded and either 18/4 or 16/4 wire. Inside the case is a different story. For some stupid reason I stripped all of the wires to make it easier to terminate the wires so there isnt any shielded cables inside the case. I do plan to get clamp on ferrite rings to try and reduce the EMI in there as well. The grounding I will have to check, but I believe that I wired the ground to the spindle. and the other ground to the case.

    4. Looking at the webpage I got them from (https://www.automationdirect.com/adc..._cables/evc179) again, it doesn't look like it is shielded. Would ferrite rings on either end help out here?

    I only blame the VFD because it seemed to be the root of my other issues forcing me to update the software to an alpha release. Another problem I can think of is I put everything in one case and the breakout boards and directly above the VFD. When wiring it up, I had read to keep everything away from the power supply because it has noise, but heard nothing about a VFD until after everything was wired in. Would moving it out, or maybe adding sheet steel/aluminum between the top of the VFD and the control boards help keep noise separated?

    Z-axis overshooting sometimes-20181110_010351-jpg

    If you think it is mechanical, what other items could it be? Losing steps I totally get in every direction, but gaining steps doesn't make sense to me, at least mechanically wise. The Z sort of made sense since if the coupler is slipping the weight on the acme thread just allows it to keep going into the board when the motor stops. I don't think it was slipping because before I welded the coupler to the rod I had the motor holding steady and used vice grips to try and make the rod slip, I put a ton of force on it and the rod didn't budge. That said, I am totally willing to have my horizons broadened.
    Yes you have a problem, the VFD Drive cable must be shielded all the way to the VFD Drive and the shield clamped to the metal case which would have a Ground connected, Twisting the wires can also help with noise, by your photo you have a lot of problems

    Yes you could put a Aluminum cage around the electronics, it really is a big problem the way it is, the power supply will be nosy as well, you may be able to move the electronics into another smaller cabinet on top

    Ferrite rings are not going to help with your cabinet layout

    Your VFD will be drawing Double the Amps that the spindle max is so you need a 30A EMI filter

    Mactec54


Page 2 of 4 FirstFirst 1234 LastLast

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

Z-axis overshooting sometimes

Z-axis overshooting sometimes

Z-axis overshooting sometimes