Router Crash


Results 1 to 12 of 12

Thread: Router Crash

  1. #1
    Registered
    Join Date
    Jul 2007
    Location
    Canada
    Posts
    139
    Downloads
    0
    Uploads
    0

    Default Router Crash

    I'm sure most of us have at least one crash under our belt, and hopefully we learn something so mistakes aren't repeated again. The first big crash I had was carving a 3D tool path with a 3/4" bit. At first I thought something had gone horribly wrong with the G code as the cutter plunged beyond full depth and then tried to make a cut. I hit the E stop pretty quick, and thankfully there was no permanent damage.

    After some investigation, I found the case was some dirt had got into the ball nut and caused it to seize, and the coupling between the screw and motor slipped and the Z height was off. So I've dismantled, cleaned and reassembled everything, and I've also added pins that go through the coupler and screw so that there is no chance of slippage. That combined with servos that will fault and shut down if they're too far out of position should be enough, right?

    I thought so until another crash happened today. This time I checked everything and there were no sheared pins or any evidence of slippage, but my Z height was 10mm below where it should have been. That's two full rotations out of position that the encoder didn't pick up on, or something else has gone wrong. My VFD shut down the spindle and the x axis faulted so at least I know some of the systems are working, but I can't think of how a servo driven ball screw could get so far out of position. Looking at the G code where it crashed, it was just after a large retract and rapid move, so it likely lost position lifting up, and then plunged too far down into the block of maple I was cutting and just started to move in the x and y when it crashed.

    Has anyone experienced anything similar to this and how did you end up fixing it? I've run the same program before without issues, and I'd really like to let it run unattended, but this crash is making me a bit paranoid.

    Similar Threads:


  2. #2
    Community Moderator Jim Dawson's Avatar
    Join Date
    Dec 2013
    Posts
    5717
    Downloads
    0
    Uploads
    0

    Default Re: Router Crash

    Sounds like you are losing the encoder signal. Could be the encoder, wiring, or board, hard to tell.



  3. #3
    Member ger21's Avatar
    Join Date
    Mar 2003
    Location
    Shelby Township
    Posts
    35538
    Downloads
    1
    Uploads
    0

    Default Re: Router Crash

    A servo should fault if it loses position, so something is definitely wrong.


    And you should NEVER let a machine cutting wood run unattended. There's always a serious risk of fire, even if the machine does nothing wrong. Pieces of wood can break loose, and get caught in places where the collet nut can rub, which can create enough friction to start burning in seconds.
    Machines with vacuum tables are even more dangerous, as the suction pulls the fire under the table. You don't even know it's burning.

    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)


  4. #4
    Registered
    Join Date
    Jul 2007
    Location
    Canada
    Posts
    139
    Downloads
    0
    Uploads
    0

    Default Re: Router Crash

    I've checked the connections for the encoder and everything looks fine, but I'm not sure how I'd test them further without an oscilloscope.

    As far as running unattended, I'm not planning on leaving the shop, but I do like to work on other things nearby, especially when I'm running a 40 minute long program. I keep the wireless controller nearby so I can easily stop it if I heard something or saw smoke, but in this case, the couple seconds it took for me to grab the remote was long enough for the bit to bury itself and trip the emergency shutoff on its own.



  5. #5
    Member ger21's Avatar
    Join Date
    Mar 2003
    Location
    Shelby Township
    Posts
    35538
    Downloads
    1
    Uploads
    0

    Default Re: Router Crash

    Are you sure it's not mechanical? I would expect that the drive would fault if there was an encoder issue?

    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)


  6. #6
    Registered
    Join Date
    Jul 2007
    Location
    Canada
    Posts
    139
    Downloads
    0
    Uploads
    0

    Default Re: Router Crash

    I think I may have discovered an error that could be responsible for the loss of position. Usually a servo fault will show up as an error 002 or 000 on the driver, and requires a power cycle to clear the error. I noticed that I have had a few cases where the system triggers the e-stop, yet the drivers don't show an error, and nothing on the diagnostic page shows anything being triggered. While troubleshooting this problem, I sent the machine to x0 y0 and noticed that it was about 5mm off from where it should be. Is there a bug in UCCNC that could be causing this?



  7. #7
    Member ger21's Avatar
    Join Date
    Mar 2003
    Location
    Shelby Township
    Posts
    35538
    Downloads
    1
    Uploads
    0

    Default Re: Router Crash

    Any time there's an estop during motion UCCNC will lose position. Your steppers are only closed loop to the drive, not the control. You're assuming that after the E-stop, UCCNC knows where your steppers are. You need to Home the machine any time there's an E-Stop.

    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)


  8. #8
    Member
    Join Date
    Jun 2015
    Location
    Sweden
    Posts
    943
    Downloads
    0
    Uploads
    0

    Default Re: Router Crash

    Quote Originally Posted by Andrew22 View Post
    I think I may have discovered an error that could be responsible for the loss of position. Usually a servo fault will show up as an error 002 or 000 on the driver, and requires a power cycle to clear the error. I noticed that I have had a few cases where the system triggers the e-stop, yet the drivers don't show an error, and nothing on the diagnostic page shows anything being triggered. While troubleshooting this problem, I sent the machine to x0 y0 and noticed that it was about 5mm off from where it should be. Is there a bug in UCCNC that could be causing this?
    An e-stop has to stop your axes as soon as possible which is an instant stop.
    If you running steppers then an instant stop can make them to jump out of syncron causing a position error.
    If you running servos and the servo faults causes the servo to stop counting the position and to release the servo motor which will then freely rotate possibly and likely causing a position error.
    My advice is to always home your machine if it was e-stopped while running. And depending on how your drives are wired you may even need to always home your machine even if it was not running when estopped. for example if your control box removes the power from the drives on estop then you will always have to home the machine after estop.
    With UCCNC I advice you to use the Cycle stop button to stop your machine, because that will stop your machine with decceleration and will not remove the enable signal and the charge pump.
    Only use the reset and e-stop buttons if there is an emergency and you have to immediately stop the machine without allowing it to deccelerate.



  9. #9
    Member
    Join Date
    Jun 2015
    Location
    Sweden
    Posts
    943
    Downloads
    0
    Uploads
    0

    Default Re: Router Crash

    And take ger21's advice to never leave your machines alone when they're working. There was a video on youtube for some time with a guy got his house burn down because he left his CNC working in the garage and it made a fire and the house fully burnt down and the security camera got all captured.



  10. #10
    Member
    Join Date
    Jun 2005
    Location
    USA
    Posts
    1723
    Downloads
    0
    Uploads
    0

    Default Re: Router Crash

    You should also check how the servo motor is configured. You can set the amount of following error on most servo systems and if that is set to high the servo can be out of position and not issue a fault. If it was set to a few hundred encoder counts it should have faulted. As other have said after a fault you need to home the machine to ensure everything is in sync again. The only exception to this rule is if you have a servo system with an absolute encoder system, which retains absolutely position by using a battery powered encoder.

    Russ



  11. #11
    Registered
    Join Date
    Jul 2007
    Location
    Canada
    Posts
    139
    Downloads
    0
    Uploads
    0

    Default Re: Router Crash

    I typically use the cycle stop button, however the way my drivers are wired into the UC300 board, if there is a fault from one of the drivers, it triggers a generic e-stop signal to the software. It would be great to have a macro that would show a message saying which driver faulted along with an estop command, but I can't find anything that would let me do that. The servos I'm using are a hybrid servo, essentially a closed loop stepper with a built in encoder. I'll check the following error, but I don't think such a large error could be possible. The encoders are set for around 4000 counter per rev, and an error of 5mm would mean 2000 counts. I'm definitely going to be more vigilante about homing the machine. I usually use a fixed point on the machine for my origin, however in some cases when carving a big chunk of wood, it makes more sense to set the origin at the centre of the workpiece, which would be much more difficult to recalibrate if the machine loses position.



  12. #12
    Member
    Join Date
    May 2005
    Location
    USA
    Posts
    3920
    Downloads
    0
    Uploads
    0

    Default Re: Router Crash

    Some ideas:

    Quote Originally Posted by Andrew22 View Post
    I'm sure most of us have at least one crash under our belt, and hopefully we learn something so mistakes aren't repeated again. The first big crash I had was carving a 3D tool path with a 3/4" bit. At first I thought something had gone horribly wrong with the G code as the cutter plunged beyond full depth and then tried to make a cut. I hit the E stop pretty quick, and thankfully there was no permanent damage.
    Yep crash happen.

    Do you have a feel for how this extended plunge happened? That is do you believe the axis was out of position before the plunge.

    After some investigation, I found the case was some dirt had got into the ball nut and caused it to seize, and the coupling between the screw and motor slipped and the Z height was off. So I've dismantled, cleaned and reassembled everything, and I've also added pins that go through the coupler and screw so that there is no chance of slippage. That combined with servos that will fault and shut down if they're too far out of position should be enough, right?
    Ball screws or linear bearings are funny things as they can lock up and you might not be 100% sure why, One thing you need to determine with a DIY ball screw repair is how they control ball positions and prevent counter rotation. Often the ball screws are assembled with every other ball undersized, so if you take a ball nut apart you really need to sort the balls by size.

    One thing I've run into over the years is linear ball bearing seizing (maybe better called grabbing) in one spot along a long run. This highlights the need to visually inspect the entire length of the ball channels in a screw or linear bearing for damage. It only takes one bad spot to fault your servos or force a coupling to slip. Note this is generally on highly worn parts which means you can visually see bad spots in the ball groves. Newer stuff can also grab but it might be hard to visually see why.

    Mechanical slippage can be hard to verify. One thing you can do is to mark all couplings, pulleys and the like with a dab of paint or some white out ink. Do so in such a way that slippage is easy to see so what you might do is to run a thin line of paint from a shaft to a coupling, You can use paint pens for this but the white out dabbers make the job a bit easier, This idea isn't really mine, I got it form some Japanese machines we have, they go overboard marking literally everything. You end up seeing a line of paint from a bolt head, over the side, across any washers to the machine frame. Basically anything that can be turned resulting in looseness gets marked on the machines, On yours I'd only worry about the motion segment.
    I thought so until another crash happened today. This time I checked everything and there were no sheared pins or any evidence of slippage, but my Z height was 10mm below where it should have been. That's two full rotations out of position that the encoder didn't pick up on, or something else has gone wrong. My VFD shut down the spindle and the x axis faulted so at least I know some of the systems are working, but I can't think of how a servo driven ball screw could get so far out of position. Looking at the G code where it crashed, it was just after a large retract and rapid move, so it likely lost position lifting up, and then plunged too far down into the block of maple I was cutting and just started to move in the x and y when it crashed.
    One thing to do here is to set your acceleration bit lower and then dry cycle the machine, I'd go so far as too craft a program specifically to stress the Z axis. Also stand near by and listen, if there is a stepper problem you should hear some radical noise from the steppers. Steppers can make some really gross noises when the drive shaft locks up but the driver continues to try to turn the motor.
    Has anyone experienced anything similar to this and how did you end up fixing it? I've run the same program before without issues, and I'd really like to let it run unattended, but this crash is making me a bit paranoid.
    There is a good possibility that you can get the Z axis to fault by simply dry cycling the machine in a loop. This should help with finding the cause and creating a resolution. As for unattended operation I don't think that is a good idea either.



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

Router Crash

Router Crash