Getting heater runaway message

S550Stang 2018-11-26 11:58:16 UTC #1

All of a sudden got a heater runaway message. This seems to be the bed heater. I have tried manually step up the temp and it seems to work fine up until about 45-50c then it gives me the error. I changed the thermister and ran a PID cal. The one thing I did prior was resolder my bed wires (adjusted cable routing) and I will check those tonight. I am running cura on Ubuntu and for manual control pronterface .

S550Stang 2018-11-28 01:23:53 UTC #2

Ran some tests after:
checked solderjoints on ramps
solderjoints on heated bed
checked tested 2 thermisters.
reconnected power wires from 12vdc
reconnected board to heater wires and sensors

Below is a series of heated bed tests. Any help in diagnosing what i am seeing is helpful. Or any more tests I should perform?

1 heated extruder to 220 no issue
2. heated bed 30c
shut off extruder heat
3. heat bed 40c
4. heat bed to 50
thermal runaway
reset
5. heated bed to 50
6. cool to 48
7. heat bed 50
8. heat bed 55
runaway
9 heat bed 55
10 heat bed 60
runaway @55
reset
11 heat bed 60
12 add glass top to bed
cool 51
13 heat bed 60
thermal runaway
14 heat bed 60
15 heat bed 62
16 heat bed 65
fail @62.2

Wyvern 2018-11-28 04:26:52 UTC #3

Will it fail when you simply select preheat PLA or ABS? I ask because I accidentally turned up nozzle temp via control on the LCD instead of TUNE while it was pre-heating before a print and once again when it was printing.

Almost instantly it threw a runaway error on my FT-5 however, if I used TUNE it worked fine- again I never print VIA USB,

Are you positive your thermister is firmly attached to bed? does it take a long time to heat? I know one failure mode is if the machine thinks that it's taking longer than expected to warm up, or if it overshoots.

Did you check your PSU output during load?

I'm still new to PID tuning printers (but I have drone background with gyro PID tuning)

S550Stang 2018-11-28 11:46:49 UTC #4

Wyvern:

Will it fail when you simply select preheat PLA or ABS? I ask because I accidentally turned up nozzle temp via control on the LCD instead of TUNE while it was pre-heating before a print and once again when it was printing. I have not tried this, however my tests were done with no LCD connected, this is a i3 2020 printer

Almost instantly it threw a runaway error on my FT-5 however, if I used TUNE it worked fine- again I never print VIA USB,

Are you positive your thermister is firmly attached to bed? does it take a long time to heat? I know one failure mode is if the machine thinks that it's taking longer than expected to warm up, or if it overshoots.yes, i confirmed thermister was attached to the bed and changed it out as well thinking it was faulty.

Did you check your PSU output during load?
**yes, however the meter I was using was cheap and depending how i connected the leads gave various readings, I will retest with my professional meter **
I'm still new to PID tuning printers (but I have drone background with gyro PID tuning)

not sure what gyro is, I did auto tune and I am new to it also Thanks for the input.

carefudundae 2018-11-28 16:31:46 UTC #5

There are some safety features built into marlin to prevent thermal runaway, fire, etc. It errors when either it doesn't see the temp go up when you tell it to, or by drifting away from the set point too much (can be signs of a detached thermistor or control board/psu issue).

I'm curious how much time it takes to heat up to a given temperature... Do you have a pronterface temp plot screenshot you can post? Is it a nice gradual increase until it errors out? or is it jumping around? is it taking 10+ mins to heat up just a little bit? a plot of the temp when heating from room temp to 50C for example might help a lot in determining if its a firmware or a hardware issue.

S550Stang 2018-11-28 19:11:52 UTC #6

i have pronterface installed which I use the terminal version (test input output) for testing and manual g-code control. It does not have that feature. I am running Ubuntu. I havnt got Repartier to work or Pronterface with the GUI, only Cura and it has no graphical output. I will have to test/time manually.

carefudundae 2018-11-28 19:26:47 UTC #7

ok, gotcha... just a ballpark value might help. 20-50C in under 5min? at least 10 min?
by the way, I don't have a 2020 prusa, but this issue doesn't seem specific to that printer... though the bed heating time/power would be, especially if this is 12V. I'd still think that it should get to 50C in 5min or so. Also, I don't think its best to have PID for the bed heater, instead use bang-bang (I think this is set in firmware configuration_adv.h).

S550Stang 2018-11-29 12:26:28 UTC #8

Thanks, for the suggestions. I was actually looking at bang bang the other day. The only thing I havnt checked, or changed (that I recall) is the firmware. I may reload the last stable version that I recall using on another laptop. Not totally ruling out a bad piece of hardware, I will first check firmware...

S550Stang 2018-12-04 12:11:36 UTC #9

Update. I am not sure, unless someone can convince me, that this is hardware related.
I can heat bed without thermal runaway without issue and the bed will maintain temperature. The runaway message happens when, for example, I heat the bed to 50c and it reaches that temp and then set bedtemp to 55.. Thermal runaway only after a few seconds. Unfortunately I am not getting much response from FB.

Thanks!

Wyvern 2018-12-04 15:00:32 UTC #10

some type of hadware/software glitch. My FT5 using filament change on prepare will shut down the extruder and soft-lock the printer-but only sometimes.

S550Stang 2018-12-04 18:50:15 UTC #11

I will meet with The Bobs, and have them "fix the glitch".

S550Stang 2018-12-04 21:00:27 UTC #12

I will prob just buy an new ramps, arduino, and a few thermisters.. cant hurt to have spares.

jefbed 2018-12-05 06:33:55 UTC #13

According to the comments in Marlin's Configuration_adv.h, you should increase THERMAL_PROTECTION_HYSTERESIS and/or THERMAL_PROTECTION_PERIOD if you get false positives for "Thermal Runaway".

S550Stang 2018-12-05 19:31:49 UTC #14

ty for the response. i will look at that as well. so far its working fine... as long as I dont heat the bed with 2 different temperatures, without a reset in between.

jefbed 2018-12-07 05:43:11 UTC #15

Correction, I meant THERMAL PROTECTION_BED_HYSTERESIS and THERMAL_PROTECTION_BED_PERIOD.
Ahh, yeah, that would be a problem if increasing temperature, especially when over 80C, and can be helped by modifying those two configuration definitions. I typically only decrease temperature after the first layer slightly. Increasing the THERMAL_PROTECTION_BED_HYSTERESIS to 10 and THERMAL_PROTECTION_BED_PERIOD to 90 would help on a heater that slows down at the high end like the Mk2a one.

S550Stang 2018-12-11 13:05:35 UTC #16

Its back... thought I had it solved... happened on a PLA print just after the raft printed. bed temp 70, nozzle 250... WTF!!!!! Will look into jefbed's solution tonight..

csorrows 2018-12-12 07:27:38 UTC #17

Your bed is too hot for PLA. It should be around 60c.

Your nozzle temp is way too high. Most PLAs print around 200c.

I know this does not have to do with your issue, but it will cause future issues such as anything above 240c will destroy the PTFE liner inside the hotend.

S550Stang 2018-12-12 14:47:48 UTC #18

sorry, I meant ABS not PLA .

I'm a bit confused, why would Folgertech claim the i3 2020 prints ABS if the hot end ptfe is destroyed at 240c? Or are they claiming it will print ABS with modifications (I'm thinking that should be mentioned if that were the case).

Back to the issue.....
Hoping someone can run this test for me on their 2020. Using manual gcode, heat the bed to 50c. once reached. heat the bed to 60.. if I do this without a reset in between i get thermal runaway error.... I am also posting this in FB also.

Thought it might have to do with a cooling fan, not totally ruled out..

Wyvern 2018-12-12 16:58:03 UTC #19

I have printed at 240 for 18 hours with no real degradation of the liner.

I am upgrading the print head right now to a MK10 and using high-temp capricorn PTFE, which has been said to have a working temp of 250C and a peak usable temp of 300C, it takes two minutes and $0.10 to change a tube, which makes it able to print most everything but PC and high temp nylon (but then you also need a hood and filament drier on board)

I don't like all metal hotends, retraction and temps get really touchy when printing anything low temp, and while the E3D V6 works well as a all-metal hotend (I have one on my delta) you will lose over 25mm (or more) build space.

I'm also kicking the ramps 1.4 to the curb and installing a Duet 2 as the Ramps setup works, but when you have multiple items printing or very complex models it struggles to keep up, I get juddering artifacts when printing 10 items at once.

It's also worth mentioning that I installed a $10 mosfet tube for the bed heater, this offloads the current in the board.

carefudundae 2018-12-12 18:40:55 UTC #20

i almost exclusive print ABS on my FT-5 and its usually at 235C, so no issues there...

I still think this is about time and firmware settings. Did you look into my previous post/reply:

I think its a time issue, not going from 50-60... try going from 50-53, then 53-56, etc. the firmware requires it to be within range of that within a given period of time, etc. before it will throw an error. How long does it take to get the error when going from 50-60 and what temp did it reach?
Again, all of this is with the caveat that I don't own an i3, but the firmware is essentially the same.