Welcome to Our Community

Some features disabled for guests. Register Today.

XYZ probe open builds grbl error 9 and alarm 4

Discussion in 'CNC Mills/Routers' started by AaronCnc, Feb 28, 2021.

  1. AaronCnc

    AaronCnc New
    Builder

    Joined:
    Sep 12, 2020
    Messages:
    14
    Likes Received:
    3
    Hey guys,

    I am experiencing an issue when I try to use my XYZ probe that I purchased from open builds.

    Here's my setup:
    - Open builds nema 23 stepper motors four of them. X, y, y1, and z.
    - Oppenbuilds Black box.
    - Open builds 24 volts power supply.
    - I am running them all on a 1510 work bee.

    - My cam/cad software is v carve pro.
    - I am using the open builds controller.

    My machine is calibrated and fully functioning.

    But I am running into two codes when I attempt to use my XYZ probe.

    - The first code is an error 9 sent from grbl when I attempt to use the probe.

    - The second which happeneds simultaneously is an alarm for probe error code.

    I am using the XYZ probe as instructed by placing the probe on the corner of my workpiece, and hovering the bit with the magnet attached to the bit over the open circle in the XYZ plate. And in the open builds controller I am selecting the XYZ probe from open builds, then selecting the action to run the probe.

    This is the point at which I run into the error code and alarm.

    Some things I did notice, before I even select the probe icon on the controller, green lights are appearing under the XYZ probe as soon as I attach the magnet to the bit.

    When I select run the probe series, the gantry moves towards the center of the XYZ probe. But the z axis never drops. And the controller sends the grbl error 9 code and the alarm 4 message.

    I have two photos attached to show the codes. And some photos of my grbl settings.

    Hopefully this explanation is detailed enough for troubleshooting solutions.

    Thank you!
     

    Attached Files:

    #1 AaronCnc, Feb 28, 2021
    Last edited: Feb 28, 2021
  2. sharmstr

    sharmstr OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Mar 23, 2018
    Messages:
    2,030
    Likes Received:
    1,428
    The green lights should only light when there is contact between the plate and the magnet/bit which completes the circuit. Check your wiring.

    If you go into the troubleshooting tab, you'll see that your probe says "on". It should be off.

    The error message is telling you that as well. Though you'll only understand it if you have experience with it :)
     
  3. Peter Van Der Walt

    Peter Van Der Walt OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Mar 1, 2017
    Messages:
    13,753
    Likes Received:
    4,070
    Ground fault loop. Spindle is making a connection to DC GND by mistake, providing an alternative path for the probe signal. Also very dangerous. See XYZ Probe Plus with VFDs and other ground-loop inducing components
     
    sharmstr likes this.
  4. AaronCnc

    AaronCnc New
    Builder

    Joined:
    Sep 12, 2020
    Messages:
    14
    Likes Received:
    3
    Thank you for the information.

    I have read the linked forum and have come up with a couple of ideas.

    Note* I am running a China VFD water cooled spindle. 4 wire connection. I know for a fact, that is wired properly.

    1) I don't have my cnc aluminum frame grounded.
    - could this be the cause for the ground loop? If I were to run a #8 copper wire from my cnc frame to the exterior ground rod outside, would this solve the problem?

    2) would it be wise to run a test where I leave my VFD off and try running a test to see if I still have a ground loop or would that be a waste of time?

    3) based on the above information provided, the issue lies within the spindle and not the black box or XYZ probe? I followed the wiring for the XYZ probe based on the information provided in the open builds video on YouTube. So I am almost certain the black box, meanwell 24v power supply, nema 23 stepper motors, and hard limit switches are wired correctly.
     
  5. Peter Van Der Walt

    Peter Van Der Walt OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Mar 1, 2017
    Messages:
    13,753
    Likes Received:
    4,070
    With the VFD disconnected, in turn, from mains, and from the controller wiring, will help you pinpoint where the short is yes.
    If its neither of those, then it is via the frame/limit switch/something else. Whatever you disconnect that makes the green light go away, is the key to look at


    Usually yes, but could be a ground loop through the PC power supply, or a stray wire, or a AC earth, etc - the mission at hand is to figure out what

    Not the issue at hand. Magnet connected to spindle > via spindle/spindle via frame or spindle via VFD is completing the circuit to between SIG and GND on the Probe
     
    sharmstr likes this.
  6. AaronCnc

    AaronCnc New
    Builder

    Joined:
    Sep 12, 2020
    Messages:
    14
    Likes Received:
    3

    What would the solution be if it is in fact my VFD? How can I resolve the issue?
     
  7. Peter Van Der Walt

    Peter Van Der Walt OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Mar 1, 2017
    Messages:
    13,753
    Likes Received:
    4,070
    Lets find what it is, then we'll work on that :) easier to chase down real rabbitholes, than theoretical ones.
    Start doing some elimination testing.

    There are workarounds if the VFD is wired safely and we can't find the loop - but I'd rather not let you take the shortcut until we've checked the VFD
     
    sharmstr likes this.
  8. AaronCnc

    AaronCnc New
    Builder

    Joined:
    Sep 12, 2020
    Messages:
    14
    Likes Received:
    3

    Sounds good! I'll test it tonight and send my results!
     
  9. AaronCnc

    AaronCnc New
    Builder

    Joined:
    Sep 12, 2020
    Messages:
    14
    Likes Received:
    3
    I believe the issue has been isolated!

    I'll be as descriptive as possible:
    As stated above, the VFD seemed to be the issue. So I started with that part of my build. As it appears, with having the VFD unplugged from power and having every other element of the cnc plugged in and in the on position, I conducted my XYZ probe test.

    Following the instructions, and the XYZ probe worked perfectly.

    Okay, so, now I know it's the VFD. To test that for certain it is the VFD only, I ran a series of tests with the micro meter, and found that all other elements are running as expected. Checked wiring and all connections. All were without flaw.

    So I pointed my attention to the VFD and spindle. My spindle has 4 wires, that link to an insert that has only one insert direction. That 4 wire insert series was put together by using an electrical engineering diagram. I am positive that it is done correctly.

    So I pointed my attention to the VFD where the four wires are attach to the VFD.

    I decided to run four tests: safely of course! While the XYZ probe was attached with a magnet to the spindle, with all components on. I removed one wire out of the VFD at a time.

    This was my conclusion. The three power wires changed noting. But when I pulled the green, ground. The green light on the XYZ probe turned off. So I plugged all the wires back into the VFD, all except the green ground, (note*: the green ground wire for the spindle shared the same port as the green earth for the VFD. This seemed to be an important detail, as the power supply cord for the VFD shared the same port as the spindle wire.) Before I touched the machine. I did a micro meter ac and dc test to ensure there was no current flowing through my machine.

    There was nothing. So I ran the XYZ probe test and... It ran flawlessly!

    Q: Now that I am here. I would like some information about this issue. From your knowledge, what could this mean, why would the green ground for the spindle create a ground loop?

    - my theory; the spindle being connected to the green earth for the VFD is creating a loop because the computer, is sharing the same ground, and thus creating a loop. Not conclusive, but seems probable.

    Q: do you think it is safe to run the spindle without having the ground connected to the VFD? (Note* the ground for the power to the VFD is still inserted.)

    Q: what are some things I should keep in mind as I move forward with this issue.

    Alternative: I suppose, as an alternative, I could keep the VFD off while I set my XYZ probe, after setting zero, plug my VFD back in and run my gcode. But that seems less of a solution and more of a headache. My goal is to have my spindle connected to my black box so I can control it from the open builds controller. But I'll work on that once I have my XYZ issue resolved.

    Thanks team!
     

    Attached Files:

  10. JustinTime

    JustinTime Veteran
    Builder

    Joined:
    Dec 18, 2013
    Messages:
    756
    Likes Received:
    252
    Come on!!! Everybody knows it's the dog's fault! :D

    I have a similar VFD. My spindle is not water cooled but I don't think that that matters. I have a three wire, shielded, cable that connects the VFD and the spindle. I have the shield, which is the same as the ground wire, connected to the main electric system ground wire (what is called in other countries 'earth') but NOT to the spindle for just the reason that you experienced, that the probe is acting up since it needs to be connected to the computer, low voltage, electric system's ground.
    I don't think that the spindle needs a ground wire for safety since, I think, VFD is acting as an isolator.
     
  11. Peter Van Der Walt

    Peter Van Der Walt OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Mar 1, 2017
    Messages:
    13,753
    Likes Received:
    4,070
    Plausable, but:
    Earth and GND are different nets: Earth (mains earth, the one shared by VFD and PC) should be isolated from DC GND (DC Negative) - so your PC may be the source of connecting AC EARTH to DC GND - in which case a USB Isolation like: XYZ Probe Plus with VFDs and other ground-loop inducing components would help break the loop. Some ATX PSUs bypasses AC earth to DC GND (not ideal)

    You'll want to read the link in the last point of section 2.8 of docs:blackbox:faq-emi [OpenBuilds Documentation] (it links out to motioncontroltips.com) - that IN DEPTH explains why not having an earthed spindle, can end up killing your spindle bearings (; (not a joke - bearing currents can cause bearing-race fluting: See Grounding Motor Shafts to Suit VF Drives ) and also a tonne of other problems, including increased EMI.

    Checkout Problems VFDs cause and cable types that help solve them - lengthy, but amazing read.
    I'd rather suggest isolating DC GND then from the PC (if the PC is the source after all) and keeping the Earth on the Spindle
     
  12. Peter Van Der Walt

    Peter Van Der Walt OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Mar 1, 2017
    Messages:
    13,753
    Likes Received:
    4,070
  13. JustinTime

    JustinTime Veteran
    Builder

    Joined:
    Dec 18, 2013
    Messages:
    756
    Likes Received:
    252
  14. AaronCnc

    AaronCnc New
    Builder

    Joined:
    Sep 12, 2020
    Messages:
    14
    Likes Received:
    3
    Thanks for the information.

    As of now. I have ordered a usb voltage isolator from Amazon. Will be in on Friday, I should have an update this weekend.

    My goal is to connect the usb voltage isolator to the computer which is linked to my black box. Run a test Making sure the XYZ probe runs fine.

    Then run a test with my ground on the spindle, making sure the system runs with the spindle attached to the ground on the VFD.

    I'll let you know!
     
    Peter Van Der Walt likes this.

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice