Welcome to Our Community

Some features disabled for guests. Register Today.

*WARNING* Incorrectly wired Blackbox fried my laptop (Need Help)

Discussion in 'Controller Boards' started by archalien, Jul 13, 2019.

  1. archalien

    archalien New
    Builder

    Joined:
    Jun 1, 2018
    Messages:
    6
    Likes Received:
    2
    I was manually jogging and (Sphinx55) Z crashed into the Zmax/Limit Switch and continued to go, my laptop issued a warning "Surge detected on USB" but before I could do anything, my laptop(on 50% battery power) went hard off, power button doesnt respond, charging light wont light up when plugged into power DEAD.

    Aside from now owning a fubar laptop, how is the USB not protected from this type of issue???
    Thismake NO sense, is this an oversight in design, have I made some other egregious error.

    I tried looking or usb extensions with inline surge/fuse because I dont want to burn up another 3k laptop, but only found 1 product for datacenters that may? work.

    Is this an appropriate topic, a defective board? Should I be contacting OB support directly on this?

    Thanks
    Bryan
     
  2. archalien

    archalien New
    Builder

    Joined:
    Jun 1, 2018
    Messages:
    6
    Likes Received:
    2
    I found the/a culprit for usb power surge, improperly wired endstops.

    Even after unplugging, resetting/clearing from the Win 10 dialog box,I had to uninstall the usb drivers from device manager and then let them reinstall, before the error would clear from the OS (it persisted reboots otherwise). Not sure how a miswired endstop is not electrically isolated from the usb? It didnt kill my other laptop, so Im still wondering if the Z crash was the culprit.
     
    Peter Van Der Walt likes this.
  3. Peter Van Der Walt

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

    Joined:
    Mar 1, 2017
    Messages:
    14,024
    Likes Received:
    4,117
    Sorry to hear about your misfortunes. But alas, thats why it's important to check all wiring.

    When the controller is powered from 24v the USB power is isolated, was yours powered on at the time?

    The Limit signal pins also has protection (reverse diode prevents any current flowing back)

    So we did our best protection wise and no other reports of this nature. Plenty of Incorrectly wired endstops on here and support - usually prevents board from booting Grbl as the Atmega doesnt get power when its shorting.

    But would like to find out what was wired wrong, see if another vector can be fortified... What exactly did you wire wrong? Mustve been something totally odd and unexpected?

    If you just had the limit wired wrong, worst case would short 5v to gnd which wouldnt ruin a USB port as they have overcurrent protection. (also needs a few other things in place anyway like 24v off, jumper on 5v, wrong switch wiring (use our plug in Xtension Limit PCBs Xtension Limit Switch Kit - safer, no need to self wire, one less thing to go wrong) all at once

    So i am not 100% sure it was that... Check all wiring especially any 24v wires

    In terms of protection: self powered USB Hub ought to do...
     
    Matthew Bates likes this.
  4. archalien

    archalien New
    Builder

    Joined:
    Jun 1, 2018
    Messages:
    6
    Likes Received:
    2
    Thanks for the reply @Peter Van Der Walt

    So the wiring issues stems from 3 things, the sphinx build vid assumes a different switch, the alternate switch wiring guide was not referenced iirc, and Im an idiot and tried to go too fast/too far ahead and made assumptions on the switch setup as I was fairly familiar with most of the required mechanical and electrical tasks and tried to optimize the build where I thought I could, ala hooking up both end connectors in the harness at the same time, which had me placing a 2 pin connector on the 2-wire (box end) of the switch harness. I then, assuming low voltage and sufficient safeties, on a normally open connection plugged into the #1 and 2 pins on the box again assuming the third pin was a hot or signal type lead for other switch types but fully expecting they would just not operate correctly but no catastrophic failures would occur with a limit miswired in its own port. So yeah ultimately it was totally my fault in the moment.

    As I was trouble shooting homing issues (see my other thread) I discovered my wiring folley from the alt wiring guide, switched to proper 3 pin terminals, corrected a couple other non-related errors in cfg and everything worked as expected.

    Back to the laptop. I have good news for me, not as good for the box. I was able to recover the laptop, it required pulling/reconnecting the cmos and main batteries. Fixing bios time/bitlocker/etc I was able to boot back into OS, correct the USB driver corruption and it was back to normal. However, whatever was the cause, the black box was culprit, and only because of my non-cnc experience in industry was I able research and recover what was an otherwise completely dead laptop.

    I can tell you this, the only thing that was miswired where the 3 endstops, the x and x where never activated, and the z was activated when the z crashed on the endstop, it was properly wired for normally open on the switch but connected to #1 and #2 pins on the box instead on 1 and 3 as it should have been. It was running on 24v at the time of the crash, all jumpers had not been touched, and since once I resolved all other issues with above mentioned corrections it is behaving exactly as expected so I dont think it shipped with a misconfigured jumper.

    I am immensely relieved my 3k laptop isnt toast, and Im am not being critical here in any way in any attempt to extort or degrade.

    My concern now remains, that it happened to me, it could/will happen to someone else, they may not be fortunate enough to be able to recover their laptops, or they may become unrecoverable.

    The recovery solution I found specifically came from research on recovery from power surges over usb. It appears that what ever happened, the box was able to overcome the laptops(2018 Lenovo X1 Yoga) overcurrent protection and soft brick it at the bios level. I know great effort was put into place to protect against such issues, it seems there is an edge case that is producing this undesired result. I can get by in electrical, but it is not my strength, Im not sure how this happens, but it did, and will likely happen again as its not like my box died during such issue and possibly others may follow my mistakes as well.

    if anyone wants to try (hard)crashing an axis using the large nemas with the activated endstop miswired as I did, and test the usb line without sacrificing a pricey laptop to see if this can be repro'd

    If you have any follow-up questions trying to isolate/correct this, post here and Ill try to respond.
    Huge thanks!!!
    -Bryan
     
  5. Peter Van Der Walt

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

    Joined:
    Mar 1, 2017
    Messages:
    14,024
    Likes Received:
    4,117
    I don't forsee any reason for fear, as this is quite a popular mistake people make and USB ports, by specification, should be safe against (USB spec says motherboard should have a selft resetting Polyfuse or similar onboard). Sure we could add another layer but isnt that just compensating for a particular PC whose ports are not designed to the USB spec.
    See section 7.2.1.2.1 http://www.poweredusb.org/pdf/usb20.pdf


    Of the many user's we've helped who did make the mistake, the usual reply comes back "I fixed the wiring and all is back to normal" - so something was still different about your incident, no other reports but yours yet. So yeah, as you mention, must be some kind of an odd edge case that probably even IBM wasnt planning for (;

    To address endstop Wiring mistakes we opted to provide users with this option already: Xtension Limit Switch Kit (Clearly labelled, cannot short circuit, and 1:1 connector pinout) so shouldnt happen much anymore.

    Thank you for the feedback though, we will keep it in mind for sure, either for future improvements, or if anyone else ever stumbles upon the same, at least have one known baseline to work from.
     

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