Battling Heater Faults. Could use a hand.

Havoc340

Ok, I have thrown out nearly a spool of filament, and this is getting old.

The other day, I was printing a long print over night, and noticed an unusual burning smell in the morning. It was something new I haven't smelled before and I really couldn't figure out what the smell was, or where it was coming from. Sniffing up close... I couldn't find it. I've been cautiously babysitting my printer during shorter runs, and everything has been working fine...

I had a print just stop, and seen a print cancelled warning due to heater fault. I figured that had to be related to the smell. I happily swapped my heater cartridge. While I was at it, I installed my PT100 I have been putting off.

So with a new heater, heater block, nozzle and heat break I felt confident that my problem was solved. That is until I had another long print fail for heater fault. I need these parts, and cannot babysit a 20 hr print, so I have no choice to run them over night. (They are at work, not in a residential building).

I have changed the heater, again, investigated my wiring and connectors, and everything seems to check out.

I am not running a fan at all (this seems to be a common issue).
I have ran the autotune a couple times now.
Firmware Electronics: Duet WiFi 1.02 or later + DueX5
Firmware Version: 1.21 (2018-03-21)
WiFi Server Version: 1.21
Web Interface Version: 1.21

This last time it faulted while I was nearby. I was able to snag a screenshot after I resumed printing.
alt text
What should I look into now?

Havoc340

https://ibb.co/gFTkSMv
I have no idea how to upload images, on my list of things I don't know.

nophead

The graph looks like heater lost its connection and regained it rather than a sensor fault.

If its a moving head machine how are you getting wires to the head? Moving wiring tends to break unless you take precautions to ensure a minimum bend radius compatible with the wire.

Havoc340

The regained connection is a result of me resetting the heater fault, turning the heater back on, and resuming the print.

The wires are ran through a cable chain. The only connector for the heater wires is an XT30 connector, near the heater. The wires are flowing nicely, and nothing is sharply bent other than the heavier wires going into the cartridge itself (they have to be).

I will attempt to get a couple pictures.

Havoc340

https://ibb.co/PrzC6Lf
https://ibb.co/Cvz2vRG
https://ibb.co/7NkZWxJ

This is the wire routing.

deckingman

@havoc340 I agree with @nophead that it looks like a heater fault, rather than a sensor fault. Sensor failures or sensor siring faults tend to show as a near instant change in temperature but that graph is commensurate with the slow change you get from a heater fault. The fact that you've changed the heater cartridge would indicate that its most likely a wiring fault but it's impossible that it could be failing mosfet on the board itself.

Suggest you change the cable and use a different heater channel on the board if you have one available. Preferably change one thing at a time but if you are desperate to get a print done, do both together.

Oh, one other thing to check - is the PSU voltage good? Quite often if a PSU is failing and giving out a low voltage, the heater is the first thing to suffer. If it;s the PSU, that might explain the smell??

Havoc340

Thanks for the help guys.

I had tested continuity of the heater wires once while tapping and shaking the cable chain and the rest of the harness, and they checked out. I decided to check them again, but this time while moving the carriage around. There was a dead spot towards the back and middle of travel, where I would loose contuinity.

I've replaced the heater wires and hopefully that is that.

I was using that red reinforced silicone heater that comes on the heaters. I have never had that stuff burn before, but I am thinking that has to have been what that smell was. I know what motors, and circuit boards, and regular wires, and plastic smell like when they are burning. But not that stuff. Maybe now I do.

nophead

This is why I use flat ribbon cables on my 3D printers with lots of wires in parallel to handle the current. The bend radius you need for thicker wires is surprisingly large.

3D printers tend to make a very large number of moves compared to other CNC machines. I have thick cables in cable chains on my router but, being mainly 2D, it simply doesn't have the mileage my 3D printers get.