Extruder Heat dropping

[CBruner]

Hi, I'm trying to print with pet, first layer 240, other layers 245. The print starts off well, temp is at 240, then 245 until about 10 minutes into the print. Then it drops to 236, then 233, then 232 and then the firmware halts.
I've tried a few different times, and closed the doors to the trident, but it always seems to happen at the same place. Any ideas what might be going on?

 

[claudermilk]

Sounds like a bad thermistor or heater. Will it PID tune? Have you just set it to a temp and let it heat soak for a while?

 

[NoGuru]

Yeah like Claudermilk said, probably a bad wire, or thermistor or heater.

 

[CBruner]

So it makes sense that the thermistor/wire/heater cuts out at the same place consistently? (I really don't want to take it apart again). Is there a way of verifying the problem without taking things apart?

 

[Le0n]

If you're lucky, the problem is in the toolhead side crimps. You can check the connectors on the thermistor and hotend without any kind of major disassembly. See if any crimps came loose and need replacing.

 

[CBruner]

Hmm, I made one last attempt. used the Prusaslicer latest alpha, and it seems to be working. At least it's past the layer that was always failing. slicer bug?

 

[Indeed]

It could also be caused by high filament flow.
It's easy to exceed what the heater can keep up with, using a standard 40W heater element and any one or more of: high print speed, thick layers, large nozzle, huge part cooling airflow.

If the flow is just above the max capacity of the heater at the selected fan speed, then after a couple of layers it would be expected to see the temperature drop despite heater pwm at 100%.
Because the slicer usually starts with a slower print speed on the first layer and then increases it for each layer, for the first maybe four or five layers.
Just like it does with the part cooling fan speed. It's to improve bed adhesion and reduce warping I think.

There is a setting in prusaslicer to limit the flow called "volumetric flow" or some such, but since I use cura I'm not yet blessed with a similar setting.
So I have to manually make sure the flow won't exceed 25 mm³/s, and that the heater block has a silicone sock, when I print PLA using a very powerful part cooling fan.
Or my weak heater won't be able to keep up.

A thicker, home-made silicone sock on the heater block may help with the heat loss due to the part cooling fan.
I'm going to try it out some day. Or maybe buy a 100W heater. Or both.

 

[vukrook]

My two cents; might help someone these days, though it's been a while since previous post.

Shortcut to the point:
40W heater is not covering heat curve expected by Klipper.

Story:
The issue first occurred as ambient temperatures dropped below, say, 10°C. To wit, the printer is allocated in a non-heated room.

The initial "summer" PID calibration went through without a hassle.
Newly made "autumn" PID Cal. went (for now surprisingly) ok, too. Here are the reasons, why:
a) PLA and PETG get printed at 212°C and 224°C, respectively. (PID Cal. referred to the higher temperature.)
b) The calibration was made omitting to run the print-cooling fan.

Starting in late autumn, nozzle temperature dropped as much as 5K or more, and Klipper halted the printer. This happened mainly with models where long straight extrusion lines were produced, or if a sudden speed increase came into play, e.g. after first layer.

The situation was confusing at first, for I did all printing in the enclosure, and during cold days filaments were getting some local pre-heat before entering the enclosure.
However, I ended up with "winter" settings limiting the fan speed, and lowering print speeds. Later on I learned the latter can be achieved more efficiently by setting the volumetric flow way more down than "summer values" of 24mm3/s.

Printing TPU at cca 228°C with volum. flow limited to 11mm3/s still worked fine.

Issues arose with TPE printed at 240°C.
I went curious how Kp, Ki and Kd would change whilst simulating the working ambient, ie. bed at temp, print-cooling fan on.
Failure occurred during the second heat-cool cycle.
Having fan lowered to half, the failure occurred at one of subsequent cycles.
With fan off - success.

At this point suspicion feel upon the heat element - that it cannot deliver enough energy to the nozzle.
After exchange to a 60W piece, PID Calibration, some Kp + Ki and Kd tweaking, the observations are:
a) no failures during PID calibration using various fan speeds
b) much steeper heat curve
c) slight overshoot / oscillation before print (will be cured, hopefully)
d) printing PETG + PLA at max. volumetric speed
e) no need to pre-heat filaments

It would have been nice if the printer package equipped with a high-flow nozzle already contained a heat element more powerful than 40W.