Spastic hotend temps

[Vprints]

As you can see in the pics, my hotend has suddenly gone spastic.

It printed fine yesterday.

Today, I:

1) Using the printer control interface in S3D, I set the hotend to 215 to swap out filaments (PLA to PLA)
2) Retracted the first one
3) Fed in and extruded the second one
4) Tried to print

Hot end starts reporting wild fluctuations once it gets around 160/170C -- some of which, if they're accurate, would be a Bad Thing (I've got a V3b.)

Tried rebooting, printing off SD via Octoprint, rebooting & printing locally over USB from Octoprint, & rebooting & printing over USB via S3D [insert panic here]

Foto 3.JPG

S3D inducing panic

(984.62 KiB) Downloaded 469 times

Is my thermistor gone? (Already sent a note to MakerGear support, but figured I'd ask the hive mind too while I wait...)

[jsc]

Previous hive mind effort: viewtopic.php?f=7&t=2086

[Vprints]

Thanks, but I was hoping for a more magical answer.

Assuming the thermistor is the short metal rod with a single lead? From that connector, I'm getting things in the 125-126 Ohm range (it gradually increases.) Pulled out of the hot end, it was around 111 -- but, again, slowly creeping up.

Not an engineer, and I have no idea if that's good or bad, but I'm guessing based on previous guesses that's bad?

[ednisley]

short metal rod with a single lead?

Mmmm, there should be two hair-fine wires emerging from the thermistor.

If you're measuring around 100 Ω rather than 100 kΩ (that's 100 kilo-ohms), then you've found the problem: those two wires are shorted (perhaps twisted?) together. If you can find that spot and un-short the wires, then the thermistor should begin working again.

There's no magic involved: when you suddenly see rapid, erratic temperature variations, then the thermistor wires are either shorted together (invalid high temperatures) or broken (invalid low temperatures).

[Vprints]

Thanks Ed.

Just rechecked, and there is in fact a tiny little K on the screen (lighting in my basement isn't ideal for my aging eyes.) Fairly stable at 133.6k.

And yes, of course it's two wires, but on mine (built Jan 2015) they're contained within a single insulating sleeve.

So: I take it that means the thermistor itself is probably OK, and I need to check the rest of the cabling / look for dust etc. on the RAMbo?

[therenegademaster]

This is weird.. I just came to the forum now because I'm having this exact problem. I was printing things pretty frequently without problems until today, when I observed the same wildly fluctuating temperature readings. It first happened when one print didn't stick to the build plate, and I tried to reprint. So the extruder was already hot, and I was trying to print again. The temperature reading would jump up to 240, then immediately down to 215, then 236.. just all over the place. I shut it down and decided to restart and try again later.

Just now, starting from room temperature, the hotend showed a normal linear increase in temperature from room temp (14 degrees C) until about 80 or 90, when it started jumping around 140.. 120.. etc. Mid-print just now it was at 224 when it's set to 215.

Could it be a software issue?

[triddle]

Most likely both of these issues are problems with the thermistor wires. Perhaps the connection is OK until the thermistor heats up, or maybe the connection is OK unless you wiggle the wires a certain way. I have seen this before and it was fixed by replacing the thermistor. I wouldn't even look at software or anything else until I replaced the thermistor.

I think I have also figured out why this happens. The thermistor wires are pretty small. I have a bad habit of letting the hot end dangle by the wires when I take the filament drive loose. I won't be doing that any more.

[ednisley]

Could it be a software issue?

No, no, a thousand times: NO! [grin]

[ednisley]

I need to check the rest of the cabling

That's about it; unless you're living in an iron foundry, conductive dust won't be an issue.

At some point along the cable, the two wires from the thermistor to the RAMBo are just barely touching. The "short circuit" won't be exactly zero ohms, because the normal thermistor resistance is (as you've measured) around 100 kΩ... even the slightest grazing contact between the wires will be lower than that.

Unplug the cable from the RAMBo and measure the resistance looking into that connector; jamming two snippets of hookup wire into the connector and using alligator clips will free up your hands. When you get a stable reading from that lashup, systematically and gently poke / pull / push / nudge / twist the cable all the way from that connector to the thermistor on the hot end.

With a bit of luck, you'll find where the resistance abruptly drops: that's the problem.

Based on the observation that "replacing the thermistor" seems to cure all of the intermittent high temperature problems reported so far, I predict you'll find the short at the spot where the two hair-fine wires exit the thermistor's potting epoxy at the small steel cylinder inside the hot end. If someone inadvertently twisted the thermistor cylinder at the heatshrink tubing around the wires while building the hot end, those wires crossed and have been getting cozy at high temperature, pressing their insulation out of the way, ever since.

Perhaps keeping a spare thermistor on the shelf would be a Good Idea...

[Vprints]

Wiring looks OK (at room temp, anyway.) Definitely sounds like the thermistor, particularly since it really only manifests itself when hot.

At any rate, I already got a response from Makergear; they're sending me a replacement thermistor under warranty. While we'd all certainly prefer to never have to contact tech support, I have to say their support reputation was one of the key reasons I purchased an M2 -- and my limited experience says their reputation is warranted.

(Thinking I'll probably throw a spare into my next purchase too, just to be safe...)