[Prusa i2] My printer is a thermistor mass-killer

A moment ago next thermistor has been killed by my Prusa i2. It's its third victim in the past 2 months. Does anyone has any idea what's happening?

The last thermistor was the original one delivered with a J-Head Mk V-BV. The previous one was Epcos B57560G104F and the first one was an unidentified, compatible¹ element.

While I may never be sure about configuration with the previous two, as I was not the one who was building firmware, with the last one I'm certain what values are set. The lookup table was selected properly — tested with a thermocouple. So exceeding termperature is not an option. PID configuration was fine too, probably even slowing down heating too early, so overshooting set temperature and going too high is — again — excluded as the source of the problem. No shorts were possible in both this and the previous case — the cables are not damaged and near the heater they were well separated. The ADC is working right (tested with headbed's thermistor and a potentiometer).

So… I'm puzzled. Puzzled and pissed off a bit, because I'm actually spending more time on waiting for new thermistors than on printing.

Related hardware spec:
- Hotend: J-Head Mk V-BV
- A 5.6Ω resistor (delivered with the J-Head)
- RAMPS 1.4
- Arduino Mega 2560
- Power: 12V from an old PC PSU

Also: can anyone suggest a good parts distributor from Poland (or from EU but shipping to PL) that sells thermistors compatible with J-Head Mk V-BV that have temperature tables already created, does sell to private persons and don't have insane part prices or even less sane shipping costs? I would like to avoid chinese suppliers on eBay for a part that both requires some precision and safety depends on it.
____
¹ "Compatible" as in "it had similar dimensions and its lookup tables roughly matched Epcos B57560G104F".

I too had a couple of thermistors fail in sequence on one of my printers (as in 48 hours apart). The third one seems to be holding on, for the moment. I am using the extremely inexpensive 3950 type (they cost like $0.20 each).

Similarly to you, I am curious about the failure mode of these devices. I found this article which seems to hint that failures could be induced by different thermal properties of the different materials used in the manufacturing of these devices: [www.designnews.com]

While I am not able to use this information at this stage, I will be looking into this issue further if I get more thermistor failures - something I expect to happen in fact. Let's see how your next thermistor goes and mine too!

is it possible the thermistor leads are sorting on the heater block

Edited 1 time(s). Last edit at 05/17/2015 05:04AM by jinx.

Quote
jinx
is it possible the thermistor leads are sorting on the heater block

"...shorting inside...", you mean?

It is possible, but that should not kill the thermistors, as the current is limited by a 4.7k resistor in the RAMPS in my case.

Semitec 104GT-2 (the one shipped with the J-Head) has operating temperature range -50°C–300°C. For EPCOS it's -55°C–300°C. I doubt it has ever been exposed to temperatures under -50°C, much less over 300°C. The Semitec one was delivered from USA, so I can imagine it could have been transported in an unheated plane and go waaaay under minus few tenths centigrade, bringing it close to the operating limits. But this do not apply to EPCOS¹. And storage range is usually wider than operating range.

Shorting is not an option in the last one, for sure. The wires had isolation all the way to the RAMPS, including few centimeters of silicon isolation directly near the thermistor. And the silicon tubes were pushed well towards the thermistor by heatshrinks. I've checked this after removing the faulty thermistor. No connection to either ground or +12V, and between thermistor wires. The last one would not be harmful too — it would just give bad readings, but the thermistor would be fine.
____
¹ Until one of the major european distributors is using airmail for transporting their parts (doubtful) and doesn't care about storage conditions of what they're selling.

I have to second the idea of actually figuring out what is going on here. Are they opening up, shorting, going off-curve? I had mine open up but it was because the thermistor wire broke at the solder connection to the feed wire. The printer shut down on a MINTEMP fault. I would expect a MAXTEMP fault if the element shorted. The other possibility that I have in the back of my head is a "gray market" part.

In my case all thermistors stopped conducting (>2MΩ resistance), so MINTEMP was triggered.

If you are confident in your crimping/soldering and installation of the thermistor, and have an open circuit, it may not be the thermistor. after a few hundred hours of printing with an e3d thermistor, I kept getting MINTEMP, and got through 3 thermistors before realising that the fault wasn't the thermistors, but a break in the cable going the the motherboard ( Megatronics) caused by hundreds and hundreds of hours of flexing back and forth. ( I reused the wiring when changing from a mendel-parts hotend to the e3d).

Edited 1 time(s). Last edit at 05/21/2015 05:32PM by Aggresive.

I was testing the thermistors with an ohmometer. Thermistors themselves, taken out of the machine.

Edited 1 time(s). Last edit at 05/23/2015 02:50PM by crazyExtruder.