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

Unhandled communication error

Discussion in 'Troubleshooting' started by supercazzola, Jan 6, 2018.

  1. supercazzola

    supercazzola Active Member

    Joined:
    Jun 1, 2017
    Messages:
    424
    Likes Received:
    111
    Any idea how to get around this ?


    The was an unhandled error while talking to the printer. Due to that Octoprint disconnected. Error: Thermal Runaway, system stopped! Heater_ID: 0
     
  2. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    Yea, that is a thermistor disconnect/failure/glitch.
    Were you heating the bed?
     
  3. supercazzola

    supercazzola Active Member

    Joined:
    Jun 1, 2017
    Messages:
    424
    Likes Received:
    111

    In middle of print. Bed 70C hotend 200


    Sent from my iPad using Tapatalk Pro
     
  4. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    I am not sure which is Heater_ID: 0
    It may be the hotend (I would assume 0 is the hotend an 1 is the bed, but I could be wrong on that).
     
  5. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    In order to keep processing time to a minimum Marlin does not automagically send every temperature reading to any display device. OctoPrint doesn't request every temperature reading either for the same reason. So a momentary glitch may never appear in the temperature graph.
     
    mark tomlinson likes this.
  6. supercazzola

    supercazzola Active Member

    Joined:
    Jun 1, 2017
    Messages:
    424
    Likes Received:
    111
    Yeah, but to me , the word “runaway” indicates a pattern of increasing beyond some caution or warning value. A momentary glitch hopefully would not cause it to poop on itself.

    I switched firmware and it went away on the same gcode. So not sure...




    Sent from my iPad using Tapatalk Pro
     
  7. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    You should see the discussions around that on the marlin mail list :)
     
    Geof likes this.
  8. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    I agree the algorithm they are employing is ... insufficient, but I am not motivated enough to code it for them.
     
    Geof likes this.
  9. Geof

    Geof Volunteer Moderator
    Staff Member

    Joined:
    Nov 9, 2015
    Messages:
    6,757
    Likes Received:
    2,339
    Please? :p :D

    (motivation from me to please FIX IT FOR THEM lol!)

    I'd love to see real errors, not generic nonsense :D
     
    mark tomlinson likes this.
  10. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    <mini-rant>
    No, I was in there hanging with the devs and even had done a pull with intentions to commit, but ... that 'group' was too disorganized.
    I think they have improved somewhat, but most open source projects of size need a strong leadership to stay 'on point' and they didn't have it. More like a zoo. At this point, 1.0 works on some of my printers and 1.1 works on the others (except for the one using repetier firmware) and you know my whole deal with firmware -- if the printer is working, don't touch it. The different approaches to auto leveling they have taken with 1.1 and later are good steps (but oh! the infighting around them) but once you manually level the beds you are ... well, done. Most of mine no longer even invoke the autoleveling (G29) for that very reason.
    </mini-rant>

    I admire the fact that @WheresWaldo tolerates them enough to try and package their releases up in an actual consumable form. I left the lists a long time back :)

    These days I just want the printer to print reliably and accurately. I don't have time for the drama.
     
    supercazzola likes this.
  11. Geof

    Geof Volunteer Moderator
    Staff Member

    Joined:
    Nov 9, 2015
    Messages:
    6,757
    Likes Received:
    2,339
    Thats my feelings on firmware in general as well :D. Not broke leave it be. Sometimes there are hints of things but my R1+ machines would still be rockin the original firmware had the replacement beds not been so wavy.
     
  12. supercazzola

    supercazzola Active Member

    Joined:
    Jun 1, 2017
    Messages:
    424
    Likes Received:
    111
    I’d like to just learn enough to add the Mesh or UBL to my R2. But it seems we are blessed that someone has done that already - thank you!
    But if still like to build the knowledge of how it all compiled, and what he did so if I want a change, I can do it.





    Sent from my iPhone using Tapatalk Pro
     
    mark tomlinson likes this.
  13. Vince Westin

    Vince Westin New Member

    Joined:
    Sep 29, 2017
    Messages:
    5
    Likes Received:
    0
    I have a robotics team printing many large/complex parts. Even with a 0.8mm nozzle, these are running 9+ hours (some are 24 hours). I have had several of these errors around the bed temperature that appear to be both sensor errors and I don't care - once we are more than 15 or so layers in, the bed heat is fairly meaningless to the print quality. Is there any way to have the Merlin code ignore bed temperature errors once the print is above some number of layers?
     
  14. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    Merlin is simply not built that way. Too many ways to burn down you space. but you could disable the thermal protection for the bed by looking for and commenting out this line in Configuration.h
    Code:
    #define THERMAL_PROTECTION_BED     // Enable thermal protection for the heated bed
    But if your communication error is the result of something else, then this really won't help.
     
    supercazzola likes this.
  15. Vince Westin

    Vince Westin New Member

    Joined:
    Sep 29, 2017
    Messages:
    5
    Likes Received:
    0
    Thanks for the idea. I am trying not to reload the code, as the printer has been otherwise stable (so I like to leave things alone).

    I am playing with the 'tweak at Z' extension in Cura to just turn the bed heat off after the first 10mm. I do not think bed heat helps much above that anyway (yes, it can still help adhesion, but if the part needs that much adhesion at that point I am in trouble). If it really turns the bed heat off, then it will stop checking for low thermistor values (I hope).

    I have had to be very aggressive with the leveling screw in the back right (the front left is almost disconnected while the back right is at a physical stop). I may need to make sure I am not bending the temperature sense part of the cable too hard.
     
  16. Vince Westin

    Vince Westin New Member

    Joined:
    Sep 29, 2017
    Messages:
    5
    Likes Received:
    0
    It turns out this was a self-inflicted problem. It appears I have my extruder putting out a bit too much plastic (multiplier is a hair above 1 for historical reasons). The part I have been trying to print is ~9" tall, and has a ~5" bar that cuts between faces on the part. This bar, running across at ~45 degrees, seems to be accumulating extra height as the print progresses. It gets higher than the bottom of the nozzle, and so the nozzle will push on it. It pushes on it hard enough to tip the build plate up from the magnets on one side - not enough to dislodge it, but enough that if that is on the sides or the back it will break the connection between the build plate and the printer long enough for a timeout on the bed thermistor. And so the part dies.

    I happened to turn the part 90 degrees (Z axis) for a different view as it printed, and I could watch it tipping the part a bit. By pure luck, the rotation make the tip happen at the front of the plate, so the connections at the back remained and the 17 hour print was able to finish.

    I am working to adjust my extrusion multiplier. The tweak at Z hack should allow this to still print in any orientation, since once I set the bed temperature to zero the check for the thermistor should also stop (without power to the bed, there is no danger of a runaway).

    Thanks for the assistance on this. I hope this will help the next builder who comes across this challenge. And this may have also been the problem supercazzola was having at the start of this thread (part lifting the bed edge). It is unfortunate that there are no sensors to help the printer give a better explanation of what went wrong. It is challenging to see this in the time lapse videos, since things appear to move rapidly anyway (and in my case the error was only showing on a tiny part of the active print area).
     

Share This Page