Irregular layers print quality issue

I am finishing the build of my first self-designed and built a 3d printer, which is massively inspired by Hypercube Evolution.
I am tuning and tweaking it and one of the issues I have encountered are some sort of inconsistencies in layers. I don't know if I am not expecting too much from this machine as this is mostly only visible when outlines are placed directly above each other, forming a vertical wall, and when a sharp light is shone from above, but this is driving me nuts! My first thought was retractions, as this is a Bowden system. But retractions between 3-10mm made no difference.
Started to print objects in single wall mode as I believe this is allowing for a constant pressure of filament in a Bowden system, and this made no difference.

What I did/tried so far:
* bought titan extruder from e3d to eliminate extruder skipping steps etc.
* changed retractions speeds between 10mm/s to 100mm/s
* shortened Bowden tube to 40cm
* tried pressure advance
* slowed down
* different filaments
* different temps
* and some more I don't remember anymore

Some of it is probably to Z wobble or something like that, but this would leave repeating pattern I believe, visible repeating itself in another print, but here it does not.
Here are some pics of single-walled test I was doing. Please help me, before I will torch this printer and shoot myself.

Unbelieveable! I have been dealing with the same issues using white ABS on prints that are similar shape!

In my case, the problem occurs with a volcano heater block with a 0.8 mm nozzle, and my printer has a direct extruder and belt lifted Z axis.

I checked about a 100 things and made about 50 test prints to try to isolate the cause, and I still can't say for sure what it is, but I switched out the V6 hot end for a Chinese hexagon with a 0.4 mm nozzle, and prints have returned to their former awesomeness.

I don't think there's anything special about the hot-end, except that it fits into the Titan extruder more securely. I suspect some of the problem may have been caused by minor wobble of the hot-end in the Titan. At some point I'll put the volcano on it and see if it still has the problem with the larger nozzle size.

One thing I never tried that could be a cause for such a problem is if the Z axis acceleration is set too high, it might be causing missed steps that would result in variable layer thickness. I don't think there's much reason to have the Z axis acceleration set above 100-200 mm/sec^2. Do you use DRV8825 drivers? They are known to skip usteps at low speeds.

Since you're running a Bowden set-up, I say make sure that the hot-end mount is rock solid- you don't want the Bowden tube tugging on the hot end to cause any deflection of the nozzle.

I have examined quality prints from 4 other machines, and if I light them just right, they all have some Z banding to one degree or another. I think some very small amount of banding is unavoidable in FDM printing, but the amount in your photos is definitely more than should be considered acceptable.

---

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

Whoa I had the EXACT same issue with my home built printer, I never got round to solving the issue because upon realising that wood is quite flammable, I scrapped the whole thing (It was made entirely out of wood).

Like you, I tried a bowden setup, a Titan extruder, slower accellerations, etc.

Even though I never solved the problem, I do have some theories, so maybe you can check the following:

-When making printing moves, is the entire printer moving? My wooden printer actually wobbled lol.
-Are all your bearings and axes aligned well? My printer had some misaligned bearings that would screech at certain moves, that would certainly introduce vibrations to the print!
-Is the movement of your bed smooth when driving it up and down? Slight misalignment in bearings here will again result in sticky and jittery operation.
-Are your timing belts tight?
-I'm assuming from your post that you made a CoreXY printer, are you certain the belt tension is the same for both loops?

Don't give up!

That's actually a very good few ideas there, food for thoughts. Thank you, guys.

When building Z axis I have realised that smoothness of movement is not ideal, but I had to choose between play in the bearings and smooth motion. Decided that play in bearings will cause more harm, so linear motion had some glitchy points.
I am right now printing two 300mm long walls, one on the front of the build platform, one on the back. The idea is if this issue is caused by some sort of inconsistencies in bed movement the defects will be exactly the same on both prints...

I also have a new theory: I noticed that my belts stretch with time. Problem is, the stretch is uneven on top and the bottom of the belt... And I use toothed idlers... So it could be those teeth on the belt misalign with the teeth on the idler. But then I would expect this misalignment to repeat itself on every layer, so it wouldn't be visible...

What kind of belts? How do they get stretched unevenly?

Pictures of the machine might help...

---

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

gt2 6mm polyurethane belt. Well, I threw away that old twisted one where it was clearly visible, but I replicated it on some of the photos how it looked like on some leftover I had.
And some photos, in google photos album for better resolution...

Photos of the machine

I briefly tried steel core belt on the XY stage, and when I realized how much more force it took just to make the belt bend around the pulleys, I took it off and went back to glass core belt. Steel core belt needs relatively large diameter pulleys and in a coreXY stage the belts are going to be flexing a lot around the pulleys and the steel wires may break allowing the belt to stretch under tension. I don't know that that's what is causing your print quality issue right now, but I'd say try glass core belt for the XY stage.

Can you tell if the banding is small shifting of the layers or does it look more like a variable layer thickness issue? It helps to examine the prints under a microscope.

---

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

Yeah, I think I will look for some better belt+idlers solution. As I said, it started as Hypercube Evolution, but I was not sure about steel core belt on 12mm idlers. I was expecting that belt will actually deform on those idlers, but I can't see this happening after a month or so of prints. What I am worried about is stretching as each loop is almost 2.5m long... The task for today was to look for new belt solution... However, inspired by tips here I think I am finally getting closer to the problem...

Through printing same parts in different places on the bed in one go, I think I have eliminated uneven layer height as a cause. Patterns on those parts didn't match so I think Z axis is not the culprit and must be caused by X/Y. Printed the same wall placed on the bed diagonally, so only one motor is moving at once, and hurray, I had much more even layers. A curious thing is I have repeated successful prints on both diagonals... Will continue testing once I will replace these cheap toothed idlers with 225zz bearings...

Yeah, microscope.... One day....

Edited 1 time(s). Last edit at 05/28/2018 01:49PM by hedeon.

did you ever fix this? anything you could share to help me? i'm desperate !

Edited 1 time(s). Last edit at 11/13/2019 11:29AM by AmyTheBun.