1. Got a question or need help troubleshooting? Post to the troubleshooting forum or Search the forums!

Auto-leveling just... quit working

Discussion in 'Troubleshooting' started by Park A., Dec 23, 2017.

  1. Jerry RoBo 3D

    Jerry RoBo 3D Administrator
    Staff Member

    Joined:
    Oct 28, 2013
    Messages:
    139
    Likes Received:
    126
    Park, thank you for sharing that video!

    It does appear that the sensor is not working. We can do some quick checks to verify that. Right after you pushed print the bed homed and then rose up to the nozzle. It should stop when the sensor senses the bed. I suggest you attempt to interrupt this process with a piece of white paper. As soon as you press print and the bed homes, place a piece of white paper under the sensor and see if you can get it to respond. That will at least give us some more info to work with.

    If you are able to trigger the sensor with the white paper, start the process over again, and this time please lay the paper on the print bed and see if it will sense the bed, now with the paper on the bed.

    If none of that works, then we need to trouble shoot the sensor further (bad cable, loose connection, etc.). If one of those tests do work, we probably just need to replace the sensor as its diodes and may have been poorly aligned during its production.

    Please let me know what you find out. Thank you for your patience and your willingness to help find the source of the problem!

    I'll be available all weekend. I don't have any big plans to ring in 2018.

    Happy New Year BTW!
     
  2. Park A.

    Park A. New Member

    Joined:
    Nov 2, 2017
    Messages:
    27
    Likes Received:
    15
    Happy New Year to you, too!

    No Joy on the paper test.

    Of note, tho, this concept has functioned recently, days ago, before the auto-leveling became problematic -- during the sequence, as the print bed was raising, I reached into the build area to reposition the bed's ribbon cable and the IR sensor registered my hand. I simply reinitiated the process, was careful to not intervene, and ended up with a successful print.

    I have no doubt this R2 was solid when it shipped out, which is likely why it passed your QC process. It seems its only when these things arrive at the Bermuda Triangle of maker spaces (my home) that things go haywire. :(
     
    mark tomlinson likes this.
  3. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    Shipping issues have been a plague on 3D printers since day 1 ... the R1 had them too (as do MANY other vendors, I have had failures for at least two others due to similar issues). Heck, my C2 arrived with a rail knocked out of a press-fit on a steel frame. Go figure.

    It is one of the biggest issues for non-kit printers. The build-it-yourself kits are more flat-packed and generally arrive OK
     
  4. Jerry RoBo 3D

    Jerry RoBo 3D Administrator
    Staff Member

    Joined:
    Oct 28, 2013
    Messages:
    139
    Likes Received:
    126
    I honestly don't know why an IR sensor would go bad. Could be a bad solder job. A component could have fell off the PCB. A pin on one of the wires could be loose. Some other things I'm sure, could go wrong as well.

    Park, would you mind removing your IR sensor and posting a few close-up pictures of it if you get a chance?

    Have a warm and safe weekend everyone!
     
  5. Park A.

    Park A. New Member

    Joined:
    Nov 2, 2017
    Messages:
    27
    Likes Received:
    15
    Gladly! I'll have those pics for you this evening.

    When I switched it out I looked at it, and it is what seems to be a fairly straight forward board; what's most worrisome is that TWO sensors have gone out in the same machine. I'm afraid simply replacing the sensor would only be treating a symptom of a greater problem - a less obvious root cause.

    I also checked the wiring, connectors, and pins, everything seems solid.
     
  6. Park A.

    Park A. New Member

    Joined:
    Nov 2, 2017
    Messages:
    27
    Likes Received:
    15
    Hope these help.

    20171231_000319.jpg 20171230_235905.jpg 20171230_235752.jpg 20171230_235658.jpg
     
    jim3Dbot and mark tomlinson like this.
  7. Jerry RoBo 3D

    Jerry RoBo 3D Administrator
    Staff Member

    Joined:
    Oct 28, 2013
    Messages:
    139
    Likes Received:
    126
    Great pictures Park., thanks for posting! Everything looks to be in order.

    Do both the IR sensors you have used on the R2 work on the other machine you have at the shop?

    I had a really strange issue on Friday that I want to mention. An R2 unit I was printing with thought the IR sensor was always triggering. So it would go through the homing sequence incredibly fast and drive the bed down. I finally figured out that the main cable to the extruder assembly was loose. It was plugged in and snapped down all the way. You may have already ensured that your cable was plugged in correctly but I wanted to mention it in case you had not yet. Can you also check the connection down by the main board in the bottom of the machine as well?

    Can you also please flash the firmware on the machine, for good measure? Please take a look at this tutorial to see how that is easily done: https://help.robo3d.com/hc/en-us/articles/115000722572-How-to-update-R2-firmware

    If we know the sensors work on the other machine, then we can rule out the sensor being the problem.

    I wonder if there is a loose connection in the extruder assembly wires. There is a secondary ribbon cable that connects the "Uptown" PCB to the "Downtown" PCB. It's possible that this ribbon cable is not totally plugged in as well. The Uptown plug is easy to see by removing the plastic cover at the top. The downtown plug is fairly easy to see by loosening the 4 screws that hold the assembly together. Find one screw in each corner of the top surface of the assembly. One screw is a little hard to access because the motor is sitting above it. I use a skinny PH1 screwdriver to get at that one. If you do end up checking this and are having too much trouble getting those screws loose, I apologize. We can find another way. Just let me know.

    Have a safe night everyone!
     
    supercazzola likes this.
  8. Park A.

    Park A. New Member

    Joined:
    Nov 2, 2017
    Messages:
    27
    Likes Received:
    15
    The sensors do not work on the R2 at work. But no one has looked at connections or messed with the firmware there.

    Your Friday issue sounds like what mine was doing when all this started. Then everything came full circle - from always firing, to never firing.

    No issues with checking the visible connections again, and I'll get to digging in there with a head lamp so I can check the not-so-visible connections, too. Once the connections are verified I'll re-flash the original .hex file, and reboot the printer with fingers crossed. More to follow.

    I'm just at the house tonight, so I'll be hangin at Club Robo Forum dot com tonight... where all the cool kids are rocking out.
     
  9. Park A.

    Park A. New Member

    Joined:
    Nov 2, 2017
    Messages:
    27
    Likes Received:
    15
    HAPPY NEW YEAR!!

    No change to the printer.

    I was really hoping to find a root cause. Checked every ribbon cable, connection, and pin I could find; I re-flashed and updated the firmware, walked through the z-offset and bed calibration wizards. . . and this print bed still tries to push through the nozzle. :mad::(

    I have visions of an Office Space printer moment, complete with Louisville Sluggers and gangster rap.
    99_Edp.jpg
     
    #29 Park A., Jan 1, 2018
    Last edited: Jan 1, 2018
  10. Park A.

    Park A. New Member

    Joined:
    Nov 2, 2017
    Messages:
    27
    Likes Received:
    15
    Here's whats next:

    My intention is to take these boards down to naked. I can pull them, de-solder/solder-wick all connections, and test every pin and pad for continuity and resistance. This might help identify a possible power input/output problem somewhere AND eliminate any cold solders that might be causing data to stumble.

    Thoughts?
     
  11. Jerry RoBo 3D

    Jerry RoBo 3D Administrator
    Staff Member

    Joined:
    Oct 28, 2013
    Messages:
    139
    Likes Received:
    126
    Hi Park,

    That's not a bad idea. I will send you two new IR sensors tomorrow so we at least have something on its way to you.

    Will you also solder the components back on once you have striped them clean?

    Were you able to test for continuity to the IR board all the way to the main electronics board. You would need to probe from the 3 wire cable to the end of the main ribbon cable. You will probably need to attach a paper clip to the end of one of your probes so it can reach down into the ribbon cable female ends. If you have already tested that, then the problem has to be the IR sensor.

    The question then becomes, how is it going bad? I wonder how we can test the individual components of the board?

    There is firmware on the IR sensor board. I wonder if that's getting deleted somehow.
     
    mark tomlinson likes this.
  12. Geof

    Geof Volunteer Moderator
    Staff Member

    Joined:
    Nov 9, 2015
    Messages:
    6,757
    Likes Received:
    2,339
    Jerry, will his new sensors come with the Wiring to the downtown board? Might be a good idea, could be short in the connections frying the IR (or worse in the uptown or downtown board).
     
  13. Geof

    Geof Volunteer Moderator
    Staff Member

    Joined:
    Nov 9, 2015
    Messages:
    6,757
    Likes Received:
    2,339
    @jim3Dbot did an amazing IR design for the R1 series. He may be able to give some insight in how or what to check, but I'm not sure if he has any information on the Robo one, so it may be generic "how it works" info.
     
    jim3Dbot likes this.
  14. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338

    That just .... seems like over-engineering :)
    Frankly, if this ever happens to mine, I am go to see about swapping it to a simple mechanical switch :)
     
  15. Geof

    Geof Volunteer Moderator
    Staff Member

    Joined:
    Nov 9, 2015
    Messages:
    6,757
    Likes Received:
    2,339
    Just remove all togather. My Ultimaker homes at the bottom and you set the "distance" from the nozzle via a wizard (this is an Ultimaker 2+ not a 3). I see no reason why the code couldn't be changed to do the same on the R2 since it also has a "z max" mechanical switch :D
     
  16. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    That would require a firmware change, so not quite as practical. Also driving the bed to Z Max would make the homing sequence unbearably slow. But, you are correct, it would eliminate the need for the IR sensor.
     
  17. Geof

    Geof Volunteer Moderator
    Staff Member

    Joined:
    Nov 9, 2015
    Messages:
    6,757
    Likes Received:
    2,339
    That would likely be the fastest part of the 3D Printing process :D

    Ultimaker method: bed is on the bottom after a print, homes x, y and z, heats and the bed lifts to start the print, print starts. No biggie :D.
     
  18. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    You could duplicate that functionality using the OctoPrint scripts driving the bed to Z max then lifting it 5 mm before stopping. Then homing to Z Max. But I want my 3D "Printer" to work like my Epson Laser and pretty much start printing a few seconds after I tell it to print. It is a "printer" after all.
    ;););)
     
  19. Geof

    Geof Volunteer Moderator
    Staff Member

    Joined:
    Nov 9, 2015
    Messages:
    6,757
    Likes Received:
    2,339
    You stop that! lol

    but in all seriousness it takes longer to heat than it would to home without the IR :D just mentioning an option :D (the route I will go when mine dies likely)
     
  20. jim3Dbot

    jim3Dbot Active Member

    Joined:
    Jun 1, 2015
    Messages:
    246
    Likes Received:
    124
    Peace & Love for 2018........Just caught & read this thread, (Geof Tag)..........Sooooo, Here's my take.........First Luv my R1, worked right outta the box, maybe just lucky.......... @Park A. Thanks for the pics of the R2, never seen one before...Noticed Robo shrunk the cap between the IR's & photo-transistor in value & physicals, hmmmm.......Boards look clean & well assembled SMT's.........Sooooo

    After replacing your sensor in the home skillets' Robo with the working schools sensor you have found now either Robo works......generally an open connection isn't going to 'take out' both sensors. There is no silkscreen on the sensors 3 pin header to indicate the +5v, Ground, and output pins......one would assume that Robo kept what was on the original design, with Ground in the center & Vcc, (5v) on one end....thus reversing the connector would not zap the board....I understand at the sensor the polarized connector cannot be swapped but, what about the ramps end?

    Since the sensor has all it needs on board, firmware & electronics, a safe & simple solution would be to apply +5v and ground to the correct to pins of the sensor, let the output pin with no wire attached. Forget the Robo, leave it in the other room.......I have no way of knowing if Robo revised the Digital/Analog code, but approx 3 seconds after power is applied you should see the LED blink either 2 or 4 times............make like a Robo and slowly lower the sensor to the table top..........it should trigger....doesn't matter the color of the table. Do this FIRST on the new sensors. Try on old ones as well. Keep in touch bro & good luck..........Jimmy
     
    Park A. and WheresWaldo like this.

Share This Page