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

Solved Thermal Runaway

Discussion in 'Troubleshooting' started by cdsl810, Feb 29, 2016.

  1. cdsl810

    cdsl810 Member

    Joined:
    Feb 5, 2016
    Messages:
    46
    Likes Received:
    14
    I just got this error for the first time ever. The print halted and the extruder and bed heater turned off. So, at the very least, this error triggered the proper protection. But what causes this? I'm running the RC3 firmware on a R1+.

    Of course this only happened because I actually NEED to get 8-10 hours of prints finished tonight... haha

    Any insight on the error is appreciated.
     
  2. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    A few of possible causes; loose thermistor wiring, parts fan pointing at heater block, improper PID tuning.

    I would PID tune first then upload the new values via firmware. The ones in our version of RC3 are specifically from my machine.
     
  3. cdsl810

    cdsl810 Member

    Joined:
    Feb 5, 2016
    Messages:
    46
    Likes Received:
    14
    I copied my PID settings over to RC3... but... I will re-tune the PID to eliminate that possibility. I have to mention that I have the Printed Solid stud thermistor installed in the E3D and that I did experience another type of thermal dropout error with that thing already. The wiring is soldered and runs all the way back to the RAMPS board, so I'll also check to see if the threads have backed out or something. I used thermal paste on the threads when I first installed the thermistor because I didn't want any air gap in there.

    I guess I was hoping that it was a known fixable glitch in the code instead of a hardware thing.
     
  4. cdsl810

    cdsl810 Member

    Joined:
    Feb 5, 2016
    Messages:
    46
    Likes Received:
    14
    It turns out that overcooling was the cause. It only happens very close to the build platform. My dual fan cooling is too much airflow close to the glass and the air curls back up toward the nozzle instead of down across the layer. I either have to keep the fans off (even at 20%) until the 5th layer or so--or only use one fan. I chose the one fan option for now. It also lets me see and get at the nozzle during a print.
     
    braxton likes this.
  5. cdsl810

    cdsl810 Member

    Joined:
    Feb 5, 2016
    Messages:
    46
    Likes Received:
    14
    OK. I guess that wasn't it. There are no loose connections for sure--everything is soldered and there are no connectors between the thermistor/heater and the RAMPS board. One thing I noticed is that the cooling fins in on the E3D are hot to the touch-and the fan is running full speed (it's even a 40mm fan instead of the 30mm stock). I thought I remember reading that the cold side never got hot to the touch. I guess I can switch back to the stock 30mm fan and see if there is an issue with the airflow of the 40mm.

    The only other clue I have is that when the error triggered the nozzle temperature was 238, but set at 245. This happened 4 hours into a 5 hour print.
     
  6. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    Did you retest PID values?
     
  7. cdsl810

    cdsl810 Member

    Joined:
    Feb 5, 2016
    Messages:
    46
    Likes Received:
    14
    I updated the PID values in the firmware and then uploaded.

    I am trying the autotune once more right now with the parts fan on full speed. I will try those values and see if the nozzle can hold a steady temp. I also read that the print speed can affect the E3D temperature, because the hotend is capable of higher flow rates. Apparently, it can deliver filament at such an increased rate that it will draw heat out of the block. The Hexagon is much more restrictive and does not have this issue.

    I will run a test print and post my findings.
     
  8. cdsl810

    cdsl810 Member

    Joined:
    Feb 5, 2016
    Messages:
    46
    Likes Received:
    14
    After re-tuning the PID with the fan on, the print failed with a THERMAL RUNAWAY error in exactly the same spot as the previous 2 failed prints. Something is causing my temperature at the nozzle to drop 5 degrees and not recover in that portion of the print. The layer cooling fan was running at 20%, and had been running at that rate for 40 layers with no problems. I have printed 10 of this same part before with no issues... with the dual fan setup and the same settings. The ONLY thing that is different this time around is the firmware (RC3 BUGFIX version). I guess it's time to dig into the code and see what's different. The print quality is perfect, and all the slicer settings are just right.

    EDIT: I FOUND THE PROBLEM... just have to test some more and verify. There are code differences in Configuration_adv.h that theoretically will cause this problem on our machines. The thermal protection period and hysteresis were way too short. I set them all to the same settings I found in RC3. Here goes nothin'...

    I will report back, hopefully with the final answer to this symptom.
     
  9. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    Then it might actually be your heater cartridge as I am using the exact same build with no changes to Configuration_adv.h and not seeing this issue at all and have done some very long prints with the BugFix branch (over 12 hours).
     
  10. cdsl810

    cdsl810 Member

    Joined:
    Feb 5, 2016
    Messages:
    46
    Likes Received:
    14
    Well, with the Configuration_adv.h values copied directly from the RC3... I had success. The temp did drop in that portion of the print, but the error was not triggered. If you look at the thermal settings in the BugFix Configuration_adv.h they are very different than the ones in RC3. RC3 has them at 60 seconds/10 degrees... the BugFix is 4-16 seconds/4 degrees.

    Anyway, my problem has finally been solved.
     
  11. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    I did not mean to suggest that changing Configuration_adv.h wouldn't alleviate the situation, just saying that it might not actually solve the underlying problem of the heater not holding the proper temp all the way through the print. It's kind of like the kid who stuck his finger in the hole of a dike, it may have stopped the leak but the hole is still there.

    Heater carts are cheap and perhaps you might have one that is working in a nonlinear fashion. The narrow range was put in as a safety measure to prevent catastrophic failures from causing catastrophic damage. Anyway, I am glad you solved your issue.
     
  12. cdsl810

    cdsl810 Member

    Joined:
    Feb 5, 2016
    Messages:
    46
    Likes Received:
    14
    Gotcha. You have already helped me and many others a great deal with our ROBO setups. I greatly appreciate your prompt and wise responses. I will certainly keep a close watch on the heater.
     
  13. braxton

    braxton New Member

    Joined:
    Jan 15, 2020
    Messages:
    1
    Likes Received:
    0
    You are very much right about the fans pushing too much air up into the exstruder cooling it down. Thank you very much, this problem was driving me nuts. Now im sane again.
     
  14. BrooklynBay

    BrooklynBay Active Member

    Joined:
    Apr 7, 2018
    Messages:
    452
    Likes Received:
    50
    The stock extruder fan is 25mm. I also have the upgraded bracket with a 40mm fan. The issue you're describing is similar to what I was experiencing about a year ago. I kept having these temperature fluctuations, and noticed that I could cause a thermal shut down error just by blowing on the hot end. That was how sensitive it was. After testing, replacing, and experimenting for a few days I figured out what was causing this issue. The power supply drifted out of spec, and the voltage was too low. I turned up the set screw adjustment, and the problem was solved. Check your power supply with a voltmeter.
     

Share This Page