25mm calibration cube horizontal banding

Good evening folks… so i have this (less now) on my makerfarm prusa 12" but im starting to see it now on my taz 5
this printer has less than 20 hours on it, and this is at 35mm/s so not fast. .2mm layers.

First pic is one side (2-3 of the sides look like this)

This pic is one other side, where it isnt quite as bad.

anyone else having these types of issues? id really like to get them resolved.

That’s not normal, I would definitely check with support about that since it’s a new OEM machine. Because the ridges look like they are the same spacing as the z-screw pitch, the screw may have been damaged in shipping and is precessing a little in the bearings.

Precessing might not be the right word, I think this dude puts it better:


I’m guessing if you put a straight screw in it would go away.

Well according to this guy, there is no truly straight screw haha.
With the rod confined on both ends is there any way to tell which screw may be not straight? or just replace them both? or RMA the machine?
ill wait to see what support says.

There was a post recently where a guy got rid of the z wobble by eliminating the bearings on the top and bottom of the screws. This way, the screw will wobble and not the x and y axis. The only problem is that the entire weight of the x axis and printhead are carried completely by the z motors.

I did see that post… given the screws can freely spin inside the bearing now means the motors are carrying the weight anyway.
Its interesting, because i dont have this wobble on my makerfarm even though one of my z screws is REALLY bent, i mean at the top the circle it runs is like an inch… but like you said, i think because there is nothing to transfer the motion to, it doesnt affect the print.

here are a couple of shots from square calibration thing.

This project might be of interest to y9ou. https://forum.lulzbot.com/t/adjustable-heat-bed-glass-mount/38/1

I expect one more minor revision to the lower Z wobble eliminator mount to retain the bearings better, but the unit does work as is.

You know on second look, I can’t decide if that’s the same pitch as the screw. What type of filament are you using? Is it a fairly consistent diameter?

I’m wondering if it’s a bad filament tolerance issue.

…straigt-er screw:)

wouldnt that cause the issue to be on all sides, not 3 out of 4 very consistently?
its hatchbox 3.00mm white ABS. Its quite consistent. I think its the best filament for the money

So support requested I RMA my taz, which i did… I just got the new one today…
First test print was this cube (used the included ABS, but it wasnt enough to finish it, so its smaller)
The banding is still there… so my question to you is.

The first layer (stopped it a few times to measure) is .3mm. I have layer height at .2, and 150% first layer… so that should equate.
However, id doesnt seem overly smushed, but the following few layers you can clearly see are being squeezed. Less and less as it gets to layer 10 or so.
Will this induce horizontal banding as the part builds?
The resolution is good. I have the extrusion multiplier etc set right, i have no artifacting etc.
If i wanted to completely eliminate that bottom layer pushing out (first 8-10 or so) my first layer would be tall enough that it may not adhere.

That’s more like it. All I’m seeing is layer lines, and they look within the dimensional accuracy specs of the machine. I don’t think the flare on the bottom will induce screw-related wobble. After a few layers, prints tend to sort themselves out and the other related defects go with it. Even in the extreme cases.

There are a few different things. Having a little bit more plastic coming through the nozzle makes it easier to get the first layer to stick, but it’s not required. Maybe try a shorter first layer and keep an extra close eye on the nozzle-bed clearance, it would be a quick test. From your pic, it looks like you might have had a tiny warp on one corner, so maybe try printing a few at a time so the layer time is longer for cooling, or slow the print speed down on the first layer for super-adhesion.

Happy printing.

So i just finished another cube… kept adjusting that bottom layer to get a better smoosh but not so much as in that last pic… and pretty much got it dialed in. I think on most rep raps there will always be some banding when held in certain light, i think ive come to that realization.

So this is a new one i did in pETG… its .1mm layer height… 200% size, so 50mm x 50mm
What i find odd is the layer banding is clearly visible but the overall quality of the print is great… its smooth, and looks great. just certain angles expose those banding effects… usually when i see banding its physical, you can run your fingernail across it. in this case I cant. its smooth.

Thats normal. Hold it next to your leadscrews and look at the pattern compared to the threads.

i will do that… so ok… fact is, i have to live with it… quality is good, i can do that.
However, i do notice on my flashforge creator pro that i dont have these banding issues. Roughly the same lead screw is used to raise and lower the platform (lower it during printing) wondering why it doesnt show up on that one.
thank you

There are things you can do to get rid of it, Those things involve modifications to your printer though.

The Flashforge printer uses a single leadscrew raising bed design. The X axis isn’t attached to the Z axis at all, and the single leadscrew is much closer to the two bearing rods. It may also have an anti wobble built in already. The Taz design has that minor banding due to the placement of the leadscrew and the bearing, and the slight flex in the metal plate. If you look closely at your Z bearings on the Taz, you will probably find a 0.01-ish gap that the bearing can move on that rod. The rod is also able to flex a little, particularily in the middle. The metal plate can also flex slightly, and the leadscrew, being in between the bearing rod and the X axis mount, can axtually exert slightly more leverage than the retentive force of the bearing. There is also a tiny amount of backlash present in the Taz leadscrew nut. All of those things can tend to result in that pattern, It’s very shallow, hard to feel, and usually only visible at certain angles or light conditions.

Check out the development forum for some ideas on how to get rid of it.

piercet thanks for the detailed info… I have read up on a few people actually removing the bearings to allow the slight bend in the Z rod to be not transferred to the x axis carriage… I have also read where folks dont recomend doing that :slight_smile:

I am not at all above modding my printer. THe way I feel about it is that if it takes $200 in mods to improve an already $2200 printer, im all for it.

Removing the upper bearing just trades one source of potential error for another. For example, here is my horrible and exagerated for effect drawing of what can happen when you remove that bearing. Without the bearing, the leadscrew can depart from it’s centerline, particularily as you get near the top of a tall print. Once that happens, the leadscrew can effectivly become shorter than it would be if it were constrained due to the geometry. It’s basically the right triangle equation, As the angle out from the centerline increases, the leadscrew basically shifts to being the hypotenuse of the triangle: Even though it remains the same length, the height it can reach varies with the angle, which in turn leads to thicker or thinner layers. The leadscrew to motor coupler, which is basically a spring with metal clamps on either end, can also now move up or down without being constrained to the top and bottom bearing, which also tends to increase the potential Z error. The result is usually a lumpy surface with hills and valleys at various points and can lead to nozzle impacts and fillament grinding in some cases. It will however relieve some of the banding effect. I just feel it’s the wrong way to do so.

so given there are a ton of people saying they have fixed it one way or another, what would be your preferred method for fixing it?

My thoughts on the z banding by removing the bearing deal is that you will be removing banding and most potential issues up until 70-100% of the max height of the printer… so most of the prints you print will never reach the height at which that slight top z line wobble becomes an issue… I had a bent rod on my makerfarm and it never became an issue because i never printed about 4 inches.

is that accurate? or am i being dumb? haha

These two projects are how I fixed it on my Taz:

https://forum.lulzbot.com/t/adjustable-heat-bed-glass-mount/38/1 - Anti Z wobble Leadscrew Nut Mount

https://forum.lulzbot.com/t/adjustable-heat-bed-glass-mount/38/1 - Adding an openbuild extrusion Z axis.

The removing the bearing trick can work, many people have success with it, You’ll never see that occuring on one of my printers though. I’ve experemented with it, and while it improves that outer layer, it gives you other issues that most people identify as extruder calibration issues, but aren’t.