Hot-End Failsafe...?

I'm new to Reprap, slowly collecting the Bill of Materials for a Prusa I3.

I can't help but notice people having problems with thermal run-away on their hot ends due to thermistor failure.

I would imagine one could have a watchdog timer in the code that would notice if the hot end stops cycling for a given period and assume the worst and shut down.

Or design somewhere above the hot end a thermostat.... Take Cantherm PN:

F20A16005ACFA06E

for example...

Or a backup thermistor?

Forgive me, I haven't built one yet so I don't know what I don't know. Just kicking around ideas.

Edited 1 time(s). Last edit at 12/05/2012 01:20PM by dshay.

Complete thermistor failure isn't usually an issue, the firmwares usually shutdown in open or closed circuit conditions.
The bigger issue is when a thermister comes loose from the hot end or worse creeps out.
In the first the thermister continues to read slightly above room temperature and the heater stays on, you could probably detect this relatively trivially in software
In the second case the thermistor looks like it's reading constant temperature, but as it's contact with the hotend gradually reduces, the delta between it and the actual hotend temperature increases. Detecting this in software with any real reliability would be IMO hard and prone to false positives.

I've had exactly 1 thermal runaway, on my first build I did not secure the thermistor sufficiently well, over the course of the first week it creeped out of the HT silicon I used to secure it and eventually the PEEK part of my hotend melted.
It's really not a common issue, and IME it's almost always the result of poor assembly rather than component failure.

Auto shutdown when the temperature fails is the nice option in the firmware, some of the firmwares like Marlin, Teacup etc uses this option

which firmware and electronics are you using ?

Check the thermistor is broken by checking the resistance between the two wires