Solved underextrusion, but only on the front side of every print

[BigBadWolf]

Printer Model

Voron 2.4

Extruder Type

Clockwork 2

Cooling Type

Stealthburner

Hey guys,

I got another pretty strange problem with my Voron 2.4 with a Stealthburner. The extruder skippes and makes clicking-noises, but only when in front of a centered print. The backside of the layers is printed fine, as you can see in the pictures. The testprint seen was printed in PETG in the middle of the bed.

The problem exists even when not printing: the printhead extrudes only when in the "back half" of the printer, in the front it will only click but not extrude any filament. The higher the gantry moves, the far back the problem remains. Small objects can be printed in the back half of the printbed, tall objects will get a "broken front", worse and worse as the gantry moves higher.

Everything I tried so far:
- replaced the reverse bowden tube with a fresh capricorn (to lower friction)
- removed the filament runout sensor
- switched filament (ABS, ABS+, ASA, PETG, PLA, PLA+)
- rebuild the extruder, swapped the extruder gears and stepper motor
- removed all custom settings from the Klipper config: Pressure Advance, Input Shaper, all set to stock-settings
- checked all wire crimps and the mainboard connectors for the extruder/hotend
- checked nozzle-temperature empty and while extruding
- checked belt-tension and lubed all moving parts
- replaced the nozzle

Nothing made a difference, the extrusion remains bad and the extruder starts clicking, when the printhead comes to the front. The Rapido hot-end melts every material like a charm, when the filament is pushed through by hand - so this does not seem to be the problem either.

My question: what am I missing? Does someone have a similar problem or an idea what to check next?

 

[NoGuru]

Any chance of getting a short video of the issue?
I mean you describe it well enough, and sounds like you already did most things I that would recommend with the exception of the Capricorn. I find it to be too tight on the ID of the tubing.

 

[Aeronut]

If your setup uses cable chains, my first suspect would be a broken wire somewhere.

 

[BigBadWolf]

I made a short video, the focus is a little bit off, but you can hear the clicking extruder around 0:47 Sek. into the video:

I already replaced the Capricorn with a new 4mm OD 3mm ID bowden tube, the issue remained the same - but it feels slightly more easy to pull the filament by hand.

I use cablechains with the old silicone wires, so a broken wire is totaly possible - but with a 100% repeatable problem I did not take it into consideration. Should I?

It would be a pain to check all the printhead wires or replace the whole cable tree, just to find out it's a software problem

Other Ideas are welcome, but i will start checking prices for PTFE wire.

 

[NoGuru]

Since its clicking in the front I swear its the filament being pulled or stuck.

Cut a length like 500mm and just run a print with the filament going right in the extruder without the PTFE and sitting free. Make sense?

 

[BigBadWolf]

Makes sense. I have tried three different size samples from three drifferent PLA Spools. Now I know, it's not caused by the filament-supply. The clicking was even better to hear:

When it is caused by a broken wire, it could be the extruder stepper or maybe the thermistor - any other ideas?

 

[BigBadWolf]

I got another view on things by testing manual extrusion. If the toolhead is positioned in the front, it won't extrude. So i tested the "minimal difference" between the working and not working position:

On this position, the extrusion works fine:

Send: GET_POSITION
Recv: // mcu: stepper_x:-14495 stepper_y:14437 stepper_z:16177 stepper_z1:16177 stepper_z2:16177 stepper_z3:16177
Recv: // stepper: stepper_x:251.000000 stepper_y:111.000000 stepper_z:49.999500 stepper_z1:49.999500 stepper_z2:49.999500 stepper_z3:49.999500
Recv: // kinematic: X:181.000000 Y:70.000000 Z:49.999500
Recv: // toolhead: X:181.000000 Y:70.000000 Z:50.000000 E:290.000000
Recv: // gcode: X:181.000000 Y:70.000000 Z:50.000000 E:290.000000
Recv: // gcode base: X:0.000000 Y:0.000000 Z:0.000000 E:0.000000
Recv: // gcode homing: X:0.000000 Y:0.000000 Z:0.000000
Recv: ok

But 10mm more to the front of the printer, the extruder completely changes sound and does not push any filament:

Send: GET_POSITION
Recv: // mcu: stepper_x:-15295 stepper_y:15237 stepper_z:16177 stepper_z1:16177 stepper_z2:16177 stepper_z3:16177
Recv: // stepper: stepper_x:241.000000 stepper_y:121.000000 stepper_z:49.999500 stepper_z1:49.999500 stepper_z2:49.999500 stepper_z3:49.999500
Recv: // kinematic: X:181.000000 Y:60.000000 Z:49.999500
Recv: // toolhead: X:181.000000 Y:60.000000 Z:50.000000 E:380.000000
Recv: // gcode: X:181.000000 Y:60.000000 Z:50.000000 E:380.000000
Recv: // gcode base: X:0.000000 Y:0.000000 Z:0.000000 E:0.000000
Recv: // gcode homing: X:0.000000 Y:0.000000 Z:0.000000
Recv: ok

While experimenting with these positions, I manually produced two errors:

TMC ‘extruder’ reports error: GSTAT: 00000002 drv_err=1 ( ErrorShutdown!)
and
TMC ‘extruder’ reports error: DRV_STATUS: 40100060

Which strongly indicate an dead driver, I assume.

 

[NoGuru]

Did you try and extrude just a piece of filament not attached to the spool in the forward location?

Changing Driver should be quick and easy.

 

[BigBadWolf]

Got it!

It wasn't the driver, switching that out didn't do a thing. But one of the wires of the extruder motor is broken somewhere. I just bypassed the cable-chains with a quickly crimped short "mini loom" and suddenly it extrudes like a charm wherever it is.

Another wire seems to be broken too: The thermistor sometimes shows a jittering temperature, which jumps way to quick for the rapido to really change temperatures.

Well, I always liked the idea of an umbilical CAN bus upgrade. It's about time.

Thanks alot for your input!

 

[Aeronut]

Pleased that you managed to find the problem. The cable chains seem to be an almost endless cause of problems. Some of the problems are surely caused by unsuitable wire types being used, but I'm very glad that I changed to CANbus and umbilical, only the Z chain to worry about now!