Settings Based Periodic Under-Extrusion

[burtoogle 513]

14 minutes ago, BasW said:

What do you mean by project file?

 

The file that you get when you save the Cura project (file menu, Save...)

[burtoogle 513]

Hey @BasW, looking at your original image above that shows the really bad under-extrusion. It occurs to me that the spacing of the light/dark regions is such that it could be explained by a cyclic variation in filament diameter. You have tried printing using a different filament?

[Snickerdoodle 0]

As an update, because I haven't posted in this thread in a bit, Cura is still giving me this same issue with some prints, no matter how many times I reprint or re-slice in Cura.

 

Example: https://i.redd.it/ud40tsspo7921.png (Excuse the not-perfectly-tuned Slic3r settings, it's just there as a control to show that my printer can do this with other slicers just fine)

 

It's always at a set location for every layer. I have M400 as a start gcode command, so it's not a printer buffer issue.

Edited January 29, 2019 by Snickerdoodle

[BillieRuben 0]

@smartavionics

here is the gcode an project file for the pink rail (images 3,4,5,6,8,9 in my OP) I turned down the speed to 40 and retraction to 0.6 in a later version and it's a little better, but even at 20m/s with an aluminium extruder arrangement and microswiss all metal hot end on my CR10s5 I still see this effect on my models when using any Cura above 3.4.1, and I still get this on my wanhao duplicator i3 mini no matter what I do. 😞

CCR10S5_Hornet -Rail -5pcs change reaction.3mf

CCR10S5_Hornet -Rail -5pcs change reaction.gcode

Edited January 30, 2019 by BillieRuben

[burtoogle 513]

Hello @BillieRuben, looking at your gcode I see that it is altering the jerk for every move. Do you still get the same problem occurring if you disable the jerk control?

[BillieRuben 0]

I'll try tonight! Thank you! 

[BasW 0]

Hi!

I installed Slic3r and tried the same print and guess what...

Different slicer, same problem (sorry Cura for blaming you)!

Besides, I tried using a different red filament (although this filament is from the same brand as the black one) using the Slic3r software and again the periodic under extrusion is present (although less visible).

 

Edit:

Inspecting the pictures: the under extrusion regions of the black triangle are almost at the opposite location w.r.t. the red triangle. However, both triangles are printed on the exact same spot on the bed..

 

Honestly, I am running out of solutions.. Has anyone an idea how to solve this?

What I tried up to this point:

- Downgrading to Cura 3.5.1

- Using Slic3r instead of Cura

- Using a different filament

- Leveling the bed (a lot of times)

- Cleaning my hot end

- Trying different hot end temperatures (190-210 C)

 

To give a little bit more background information:

I am using a modified Anet A8 printer. I upgraded to a bowden extruder set up, upgraded the firmware to Marlin, use grid bed leveling and 1.75mm PLA with a 0.4mm nozzle. 

 

Could it be that there is some fault in the software that is controlling the extruder? Or that my extruder stepper motor is faulty?

 

Thanks!

Edited January 30, 2019 by BasW

[gr5 2,167]

Well if this happened on a UM2 with the original black feeder I would say this is a common problem.  Feel the filament going through the feeder when you print the bottom layer.  If it jumps suddenly backwards that will tell you a lot - it will tell you that there is no problem with the software but the stepper motor can't push hard enough.  You could then test the feeder strength.  A UMO, UM2, UM2+, UM3, and S5 can all do 5kg of pushing/pulling on the filament.  So you could see how much force it takes to pull the filament out of the feeder.  The newer printers above can do more like 7kg of pushing/pulling the filament.

[BasW 0]

Hi @gr5, thanks for your response!

Are to refering to the 'clicking' extruder problem? I encountered this problem before, but this time no clicking sound is present. 

I will look into the extruder rotation tomorrow, I will keep you posted!

 

 

[BillieRuben 0]

On 1/31/2019 at 3:46 AM, smartavionics said:

Hello @BillieRuben, looking at your gcode I see that it is altering the jerk for every move. Do you still get the same problem occurring if you disable the jerk control?

Solved!!

 

@smartavionics, I could kiss you! I've been struggling with this issue for many moons and it's really weighed on me. You're amazing! Thank you thank you! 

 

The new prints are looking a million times better. I can now get to troubleshooting the rest of the issues that I can now see thanks to you clearing up the main one. I'm thrilled. Thank you! 

 

 

https//imgur.com/a/pHR9XdG

[BasW 0]

@BillieRuben good to hear your problem is solved! 

Is your solution to disable jerk control?

 

Thanks!

[burtoogle 513]

Hi @BillieRuben, I'm glad to hear the problem has been solved. I do think that frigging the acceleration and jerking the jerk (so to speak) often causes more problems than it solves. Personally, I try to avoid changing accel/jerk for different parts of the print.

[BillieRuben 0]

Yes indeed @BasW! 

 

@smartavionics

I will definitely be leaving the jerks be from now on. Haha! 

[BasW 0]

Hi guys!

This morning, I did a somewhat more systematic research to find the cause of my period under extrusion problem.

Using Cura 3.6.0, I systematically changed one parameter in the slicing software while keeping the others constant. The results:

1. Disabling jerk control: similar results as using the Slic3r software.

2. Printing on left side of the bed instead of the middle: worse results.

3. Increasing wall line count, to check whether the problem only occurs during the infill or also during wall printing: the problem also occurs during wall printing.

 

Thinking about these results, I slowly changed by ideas about a software fault to a hardware one. Since the problem showed a periodic behavior, I came to the conclusion that the extruder could be the problem, since this one is rotating (and can therefore introduce a periodical error). So I marked a spot on my extruder gear and watched it rotating and, although minimal, I think my extruder gear was rotating more slowly in the under extrusion regions. 

So, I replaced my extruder gear with a spare one and guess what: problem solved!

 

Apparently:

1. My extruder gear was not perfectly circular, which caused period variation in the pressure on the filament while pushing it through, or

2. Since my extruder gear only has one fixing bolt and the diameter of the hole in the gear may be a little larger than the diameter of the stepper motor axis, the fixing bolt pushes it away from the axis on one side, resulting in a non-concentric fit. This, again, causes a period variation in the pressure on the filament while pushing it through.

 

I did a quick research regarding extruder gears, and I have a question about them:

Some of the gears have a kind of ditch on the side in which the filament is supposed to fit, other are competely flat. What is the difference and which one does a better job in your opinion?

 

Thank you all for helping me solving this problem, I really really appreciate it!

And, I am sorry for blaming you, Cura...