Hi everyone,
Are these lines caused by ‘Z-wobble’?
I printed it at 0.2mm with calibrated flow at 195 °C. The flexible coupling has a little opening between the rod and the stepper motor (from another topic about Z-wobble) to reduce Z-wobble.
It’s now not so heavy anymore as before but still it’s there. I installed this Z-axis upgrade from h60 on Thingiverse also.
Is there a way to fix it?
Regards,
Dylan
Not really a Z-wobble as the borders are straight. Could be variations in the filament feed. Or some play. Or the X-Y belts too tight or the motor belts not tight enough.
I shall try to make the X and Y belts a little bit more lose since they are very tight. I will also check the motor belts.
Dylan
If adjusting the belts does not help check the screws that hold the pulleys on the motors and shafts.
I adjust the pulleys some of them were a little bit looser then the rest, almost unremarkable…
The belts were OK. I started a test print and then I saw that I had Z-wobble.
I removed the Z-stable from Thingiverse and the problem is GONE!
Thanks for your help everyone 
Cheers,
Dylan
The problem is back 
In a couple of weeks I’m going to upgrade the extruder to an E3D v6 (it’s on it’s way), at the same time I will remove the belts and clean up the gliders.
I guess there’s some junk in the gliders that cause those random wobble… Is that possible?
Dylan
Having similar issues getting it randomly on prints,
I’m replacing my pulleys/belts and iglidurs next week when I have moved house so will let you know if that solves the issue, it may help close in on what causes it
There can be several issues causing horizontal banding. If it’s going all around the part - i.e. a whole thinner layer, often followed by a thicker layer - then it’s probably Z-axis related. If it’s starting mid-layer and especially always on the same side of the part, then X/Y-axis.
There are several things than can cause Z-axis related problems: The threaded rod, the vertical linear bearings and software settings. And some things may look like z-axis related problems. Like a partially clogged nozzle or a feeder problem, either by filament stuck on the spool or contaminated pulley.
I did several tests the last two days to fix this problem. This was after I cleaned the nozzle (it was worse before):
The starting point after cold pulling ashes caused by the 210°C out of the nozzle. (The cubes are 25mm by the way)
After changing all settings related to the z-axis in Cura to be multiples of 0.0075mm. The reason is that one step of the z-axis is 0.0075mm (200 steps, 1.5mm per revolution). Cura doesn’t actually know that. Having a value that can not be divided by 0.0075 results in one step more or one step less, i.e. one layer slightly over-extruded and one layer slightly under-extruded. It’s not so extreme, if a sligtly over-extruded layer is followed by a normal one, but still noticable.
Part No.2 doesn’t look actually better, but on the corner there was now a regular pattern, exactly every 1.5mm (one revolution of the z-axis threaded rod)
After aligning the stepper motor to be in line with the nut in the carriage. I did this by adding a z-axis stabilizer on top without the bearing and measure, how the rod moved when going up and down. I needed a spacer of 0.6mm between case and motor mount to eliminate any movement of the rod in the y-direction
After adding a z-stabilizer on top … didn’t do anything really.
(not shown - don’t know where it went): The threaded rod also moved slightly left and right. The reason was that the mounts on the case were not parallel and slightly too close to each other. On the carriage the rod-distance was 111.65mm, top and bottom around 111.3
I also noticed with the threaded rod removed that the linear bearings had a really harsh movement and were binding so much, that at some positions I didn’t even have to hold the build-plattform. It was stuck and I had to actually add a little pressure to make it move down.
So I sat down and took the whole z-axis apart. The main problem now was, that the bearings on each side were not perfectly aligned, which is no wonder, because it’s almost impossible if done the way, the manual suggests. This would not even matter, if the bearings were good. The ones in the set are mediocre at best. I have one good one, two “kinda okay” ones and the last one is really bad, but I noticed that it’s almost okay in one direction, but binds a lot in the other, so now it binds on going up - which I don’t really care about.
Took me about an hour to align everything and another 30 minutes to get it back into the case. The way I did it, was to leave the bearing mounted to the carrier (the metal part, not the plastic with the nut for the threaded rod) of the build platform with the screws tightened as hard as possible with the plastic clamps.
Took quite some time and cursing, but was well worth it. (Still ordered better bearings, in case the problem reappears)
The two round ones are after that procedure. The banding is caused by manually changing the extrusion-rate, to find the sweet spot. That’s because I noticed that the white PLA needs actually more extrusion than the black one I used before. This caused an amplification of the z-axis problems.
The top 20 mm of the bigger cylinder are the settings I’m using now.
The lefte Trooper-Buddha in the background was after the test-cubes, the right one after the cylinders. (My computer decided to restart, while I was printing via USB - very nice, thank you Windows Update).
@ Lukeatkin
Keep me updated!
@ Sverenja
So if I understand it correctly, the problem is related to the Z-axis bearings?
When I am going to install my e3Dv6 I will take everything apart and clean the X and Y axis rods and bearing throughly and check the Z-axis threated rod distances.
Can you post a screenshot of your Cura settings please?
Which better bearing have you ordered (Velleman delivers bearings equivalent with those cheap LM10UU bearings)? Maybe I can replace them immediately?
Cheers,
Dylan
EDIT: changed LM8UU -> LM10UU
Hello Dylan,
yes I think at least a significant part of the problem is related to the z-axis. If you realign the bearings, make sure to fasten the screws connecting the bearing clamps to the metal bed holder first. These are the difficult ones, as tightening these can lead to a slight rotation of the clamps leading to the bearings not being in line with each other. The screws on the plastic carrier in the back are easier, because tightening these with a hex key while only holding the nut does not lead to skewing of the braces.
By the way the bearings in the back are LM10UU. These are the ones I ordered: http://www.cncshop.at/index.php?a=3564
I do not know yet whether they are really better, but never had bad quality parts from that shop before, so I hope for the best. If they don’t work out, I will exchange them for iglidur bearings.
Additionally I’m going to stabilize the threaded rod using pillow bearings, so it doesn’t sit on the motor anymore.
Here are my Cura settings:
This is what it looks like straight out of the printer right now:
Cheers
Your prints are great!
I think you’re right about the fixation of the bed and the bearings, that seems most possible…
However you’re Cura settings related to the pitch of the threated rod could also improve the print quality a little bit while you’re printing small critical parts.
Cheers,
Dylan
Those IGUS bearing in plastic are they capable of holding the bed or not?
I know Velleman uses them in the X and Y axis and they are really smooth and don’t make a lot of noise. However I don’t know if they can handle the weight of the bed?
If it would be possible I will order 4 of them right away 
I am also removing the aluminum plate and replace it with the custom heated bed from Marlak which also reduce the weight
Dylan
I would probably go with the drylin R RJUM-01 (aluminium sleeve with plastic core), as those have a higher rating for Fmax, just to be on the safe side.
But the RJMP-01 (like those in the X/Y axis) are actually enough, as I wouldn’t call the load dynamic. I mean, nobody is jumping around on that bed. 
I will try them out and let you know
Dylan
So I installed the IGUS RJ4JP-01-10 since they are identical to the original LM10UU and the difference is HUGE!
The original bearing were indeed broken since I found some little balls on the table after changing the bearings.
My Z-axis also doesn’t wobble any more even without a Z-stabilizer. So the LM10UU were even more evil 
Thanks to IGUS to deliver them as a sample since I didn’t know if they would fit and work. Now I know 
They are great! I will never use again those LMxxUU bearings!
EDIT: I replaced the IGUS one again for new LM10UU bearings since I had issues with the building platform after a couple of months.
Cheers,
Dylan
A test print:
If you compare it with the print in the first post you can see it’s a huge difference!
Dylan
I’m using nophead’s Z couplers already with rubber tubing. Is there a special way that i’m supposed to be aligning my couplings, or do i just tighten them on? I’m almost done printing out my second set of rod constraints, so hopefully they make a difference
I’m not aware of those couplings. Can you post a link?
Dylan
