Heater Fault debug facility

c310 · 7 Aug 2018, 09:19

for debug issues i suggest to split (if possible)
heater issues from termistor issues

by the way is there any theoretical or practical chance that smatreffector (broken or crazy) is causing heater faults?

dc42 · 8 Jul 2018, 11:20

@c310 said in Heater Fault debug facility:

for debug issues i suggest to split (if possible)
heater issues from termistor issues

The firmware can't always tell the difference. For example, if the temperature rises too slowly when heating is commanded, it could be caused by either a thermistor fault or a heater fault.

by the way is there any theoretical or practical chance that smatreffector (broken or crazy) is causing heater faults?

Yes, if the heater or thermistor connector isn't properly soldered.

Barracuda72 · 8 Jul 2018, 11:27

What would help a lot would be a screenshot of the moment when the error occurs. For example, if the temperature drops by 30° within a second in the event of an error, it cannot be a Heater error, but must be a problem with the thermistor. because NO heatbed can drop 30° in that time. If theres a heater prob, the temp would fall more slowly.
a menue point in the right mouse menue on the bed and tool heater to reset it would be nice.

c310 · 9 Jul 2018, 10:37

in last case fault was issued by high temperature i think. pt100 is up to 500 degrees
but RRF M143 command limits temp to 262 degrees and i was trying to print at 290.

when tool was hot, and no print job - no errors so far. once i start print - heater fault immediately.
i think it is RRF bug. it should not allow 290 deg in such a config.

to find the issue i removed all cables from the duet board, and connected heater and PT100 directly without heatsink ... took me 2 hrs to find

i have added that to wiki on heater fault troubleshooting

r3cvr · 9 Jul 2018, 18:25

With default settings, I was getting a heater fault whenever the model fan came on, as it would cause a dip in extruder temperature until the PID caught up. Had to greatly relax the fault detection. Should detect on temperature direction instead of anomaly from set temp?

dc42 · 10 Jul 2018, 07:15

@r3cvr said in Heater Fault debug facility:

With default settings, I was getting a heater fault whenever the model fan came on, as it would cause a dip in extruder temperature until the PID caught up. Had to greatly relax the fault detection. Should detect on temperature direction instead of anomaly from set temp?

A falling temperature may be caused by the heater or the thermistor falling out of the heater block. Either could cause a fire.

dragonn · 10 Jul 2018, 07:53

@r3cvr and what is wrong with that? One step of setting up a new printer is always do a PID tune.

r3cvr · 11 Jul 2018, 02:34

@dc42 said in Heater Fault debug facility:

A falling temperature may be caused by the heater or the thermistor falling out of the heater block. Either could cause a fire.

But false positives lead to effectively disabling the feature?

Is there any way to measure the resistance/power consumption of the heater? Could try and use the resistance as a secondary temp measurement and set thresholds for disabling. I'm assuming most heaters have some positive coefficient of resistance.

@dragonn said in Heater Fault debug facility:

@r3cvr and what is wrong with that? One step of setting up a new printer is always do a PID tune.

Print fails with no extrusion after heater fault shuts down heater. That's how my first dozen or so print attempts on DuetWifi went. I actually have several PID tunes, done at different model fan settings. No surprise it comes up with different parameters. Doesn't help this issue that much.

dragonn · 11 Jul 2018, 05:48

@r3cvr They are no "false positives" when dealing with "thermal runaway". If it triggers then is something wrong with you hardware setup. It can by that you part cooling fan is to strong, it can be that you part cooling tunnel is wrong and blows on the heater block.

First of all, I suggest isolating you block, even warping it with kapton tape helps much. And E3D silicon socks are just amazing.

Don't blame software for problems with you hardware. 99% of users don't have any problems with current thermal runaway implementation in DuetWifi.

dc42 · 11 Jul 2018, 15:31

@r3cvr said in Heater Fault debug facility:

Is there any way to measure the resistance/power consumption of the heater? Could try and use the resistance as a secondary temp measurement and set thresholds for disabling. I'm assuming most heaters have some positive coefficient of resistance.

Heaters usually use nichrome wire, which has a very low temperature coefficient of resistance, making it hard to use the element as a temperature sensor.