Hi fellow rappers,

Today I just realized this upon my prints :



That cube is single walled, 3x3x3cm.

I was printing it using my prusa i3-styled printer, with 0.2mm layer. I'm not really sure what it is. My Z thread rod is Tr8*8, it has 8mm pitch. Consequently, true wobbling only occurs with a period of 8mm, and banding only occurs then I'm using layer height not a multiple of 0.04mm.
That image shows neither theoretical wobble nor banding that I foretold. Being more weird is that this artefact only starts to show several mm up from the bottom.
Does anyone know that it is?

Thank you.

Edited 1 time(s). Last edit at 10/18/2015 11:59AM by Vashikovich.

Do you use auto bed leveling? In that case, does the banding occur if you disable it?
Or do you use a geared extruder? In that case, is the banding the same if you print with infill?

I just printed it with different layer height; from left to right : 0.1, 0.2, 0.3 mm.
It seems the banding is not mechanical related but rather software related because the banding starts sooner and denser in lower layer height.
From 0.1 to 0.2, the banding start layer seems doubled and the banding density is also halved. But from 0.2 to 0.3 it also oddly doubled and halved (should be just 3/2).

I've checked the g-code and I found nothing wrong. Every run on any side on every layer roughly extrudes 1.59mm of filament.
I still have no clues on this...
Anyone?

If the part starts to wobble because it isn't firmly attached to the bed the wobble will get worse the taller the part gets. Are they staying stuck to the bed? Can you see the part wobbling as it prints?

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Can you try put some kind of light and see the layers? from what I know on banding, on specific layer the plastic somewhat 'stretched' or something, I don't know if this helps but worth a shot

Quote
cristian
Do you use auto bed leveling? In that case, does the banding occur if you disable it?
Or do you use a geared extruder? In that case, is the banding the same if you print with infill?

I don't use neither auto bed levelling nor geared extruder.

Quote
the_digital_dentist
If the part starts to wobble because it isn't firmly attached to the bed the wobble will get worse the taller the part gets. Are they staying stuck to the bed? Can you see the part wobbling as it prints?

It was firmly attached.

Quote
sarf2k4
Can you try put some kind of light and see the layers? from what I know on banding, on specific layer the plastic somewhat 'stretched' or something, I don't know if this helps but worth a shot

I don't quite understand what you mean by the plastic stretching.
I've examined how the plastic laid down during prints, and on that banding, the plastic seems to be underextruded for the whole perimeter on that layers. oh, and btw, i printed it using spiral vase, so that underextrusion spans for 4 layers (4 distinct Z positions) for the whole perimeter. the G-code seems perfectly fine, i've examined it that it doesn't have change in E, X, nor Y value, it's perfectly constant end consistent.

Heated bed?

Quote
cristian
Heated bed?

yes. I was suspecting temperature, but the temp graph seems very stable.

Were print speeds the same for all three?

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Quote
Vashikovich

Quote
cristian
Heated bed?

yes. I was suspecting temperature, but the temp graph seems very stable.

Can you see a correlation between the banding and the times when the heating of the bed is turned on/off?

One of the things that happens slowly is printer temperature rising- I don't mean the bed or nozzle, but the air around it and the rest of the machine. Is the belt tension changing with temperature? Is the machine enclosed? What is the ambient temperature?

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

what is the size of your z axis screw/leadscrew/threaded rods? if possible some specs of the rod might help

Usually banding occurs when you tried to use non-optimal layer height paired with the leadscrew/threaded rods, I'm sure you already know this

If not about non-optimal layer height, what about previous prints using other filament brands?

I just did another test and here is the temp graph :



cristian pointed it out right. See those spikes on bottom right? The banding, i.e., underextrusion, happened exactly when the heated bed was turned on for a moment.
It was as if all the power channelled for the bed leaving the extruder motor under powered.
Is it because the power supply weak?

EDIT : It wasn't the extruder getting less power when the bed is on. I've seen it with my own naked eye that the bed flexes downward quite much when it is energized, and flexes back to its original height when off. How can you explain that physically..?

Edited 2 time(s). Last edit at 10/18/2015 09:19AM by Vashikovich.

Switching the bed on would create a magnetic field, but I don't know which y-carrier part would be "attractive" enough to drop the bed.
Your bed isn't running on PID, but bang-bang, right? That's the point, where I would start troubleshooting.
-Olaf

PS: Just in case the bed isn't the cause for the banding, I'd put my money on the heatsink getting too hot. The banding happens to be an "almost_jam" which causes the extruder to skip steps.
That would explain, why it starts later in the print and is also related to layerheight.

Edited 1 time(s). Last edit at 10/18/2015 09:47AM by o_lampe.

O_Lampe: I just fixed this problem after about 2 months of "eBay and the forums". Realized that all my prints had random layers that were (the only way I can describe it) under-smushed, and so further in from the perimeter. I removed the tape from the heatbed, and when it went into a power cycle (more on this later), you could see the nozzle-bed gap increase by half a millimeter. Anyway, I remembered that I thought I was using PID, so I went into the firmware and found one little line
#define PID_BED_HEAT
Or something like that; but it was commented out.
I brought it back in, and ran a m303 for PID autotune, and then all my test cubes and 3DBenchys were perfect.

But, due to the consistencies of the lines on Vashikdvich's prints, I wold look to Z wobble or the lack of sufficient micro stepping; (not sure quite how it would work, but if your motors have a very low resolution it is possible that it cannot choose a perfect position to move the Z axis to, resulting in the bands.)

Hope something helps!


Edited 2 time(s). Last edit at 10/18/2015 09:51AM by Montiey.

---

Master Tinkerer

You're right, I almost forgot about lorentz force. The only things that are ferromagnetic along the Y assembly are smooth rods, threaded rods, and bearing; but they are on the sides. I think the magnetic force is strong to bend 2mm aluminium clad.
Looking at the PCB bed heater, the copper trace has a snaking pattern rather than spiraling, so, any magnetic field generated should cancel out, hence it's not lorentz force...

By the way, this event led me into migrating from laying the PLA directly on top of aluminium with some layer of stick glue into paper tape with the heated be turned of.
The banding is now a goner, along with elephant's foot and, of course, a portion of my electricity bill

Thank's all for all of your advices!