Inconsistent Layers, would appreciate some feedback

aluminium

TLDR: I'm fairly new to 3d printing and would like to improve on this https://imgur.com/a/R8NrPRo

Some more background: I messed around with 3d printers at a local FabLab and was convinced to start to build my own by somebody there.

So my first own 3d printer is a self-built one. I built a Hypercube (https://www.thingiverse.com/thing:1752766) which was slightly modified (a bit taller and some more metal parts instead of 3d printed). It's basically a very sturdy framed CoreXY bowden printer.

It's been printing reasonably (as seen here https://i.imgur.com/tmc50F4.jpg) and the Duet WiFi i have is a joy to work with. Initially i had a bunch of tweaking to do but it seems one of the last things i need to work out is this inconsistent layer problem that ive linked before. Lots of my fellow hobbyists say that the quality i'm getting is good but i'm a bit of a perfectionist and i would like to try to improve.

I slice with the latest version of cura, i've did the math on what kinds of z steps i can make, i've experimented with jerk, acceleration, i put in a geared nema 17 at some point to try and increase the step resolution (did help a tiny bit but not much), i've tried several brands of PLA, calibrated e-steps, etc. I feel like i'm getting closer to the root of the issue but there is still something preventing 'perfect' prints.

I compared my own little towers (as i show in the picture album initially) to a similar one printed on a Ultimaker 3E with effectively the same slicing settings and while that too has a slight inconsistency its a lot less than my own.

Maybe this is a lot to ask and take in but i would greatly appreciate any help feedback or guides. I'm starting to feel at sort of a loss here.
Happy to supply more details if need be but there could be so many causes that i'm not sure what to attach

Thanks a bunch

Software Information
Firmware Name: RepRapFirmware for Duet 2 WiFi/Ethernet
Firmware Electronics: Duet WiFi 1.02 or later
Firmware Version: 2.01(RTOS) (2018-07-26b2)
WiFi Server Version: 1.21
Web Interface Version: 1.22.3
Cura Engine 3.6.0

Hardware Info
Printer: Hypercube
Hotend: e3dv6 clone
Extruder: Bowden Type (https://tinyurl.com/y96n3xg8)
Nema 17: Nanotec ST4118L1804-A
PSU: 24v 25A Meanwell (way overkill but i had this available)

burtoogle

At one time, my printer started printing similarly inconsistent layers that looked quite like your images. The problem turned out to be a bad thermistor in the hotend. I changed the thermistor and the prints looked great again.

Scachi

Not sure what style of threaded z rods your printer uses, but this is what I have tried...
I had no clue that his issue seems to be related to the Z axis (overlooking the obvious) so I just started at one point doing changes and checking the results.

The prints from my printer (based on this one http://www.3d-proto.de/index.php?p=project_3ddrucker#3ddrucker) are having the same issues right from the beginning.

I tried a lot of things to get rid of this issue as I have no clue what was causing this.
Using different belts, different smooth rods, different bearings and belt wheels (non printed one) for the belt guides at the corners, 2xZ threaded rods, 3xZ threaded rods with 3 motors, two motors and a single one (based on the detailed mod post at thingiverse by Phaedrux https://www.thingiverse.com/thing:2651649)

I reassembled it a few times hoping that I will get rid of the issue somehow.
The design/alignment of the belts had some flaws. It was not possible to align them correctly like they should have been according to some post I found explaining the corexy mechanims (as long the page is correct) https://drmrehorst.blogspot.com/2018/08/corexy-mechanism-layout-and-belt.html
But fixing the alignment issues didn't fix the layer height problem.
Tuning the heaters several times with different settings (fansd on/off,limited..) didn't help at all in my case.

My diy set of the printer came with some printed parts that were having the same issues on a closer look. They were printer with a bigger nozzle and higher layer height so my first thought was that it is just how thicker layers look like.

I recognized that the layer pattern was matching the pitch of my leadscrews.
I had an anycubic i3 ( prusa i3 style) printer before the core xy one and I had no issues with the layers there.
As I had disassembled the anycubic and it was using trapezoidal rods I removed the m8x1.5 rods from the core xy and mounted the trapezoidal ones with the bronze nuts (or whatever material this is) . Now the layers are looking fine like they did on the anycubic cartesian one.
I ordered a third trapezoidal rod already, waiting for it to arrive so I can go back to the 3z , single motor version again and see if it still is working that good as it is right now with 2 Z threaded rods driven by a single motor via belt.

My guess is that the original m8 rods of the printer diy set and the somewhat long metal nuts aren't working that well in combination causing the Inconsistent layer height.
But there might be a better/easier way to solve this issue than changing the m8x1.5 threaded rods to trapezoidal ones.

bartolomeus

@aluminium I have the same issue. From some angles the print is almost perfect. But when you shine some light on the object from different angles, it looks inconsistent. I have tried everything you mentioned as well. I am going to try a new temp sensor next. After that I am almost out of options. Last things I can think of is stepper motor and/or filament quality

briskspirit

Have the same problem after upgrade to Duet and still can't solve it yet...

Phaedrux

Sounds similar to the odyssey I went through. https://forum.duet3d.com/topic/6618/solved-help-me-solve-some-z-banding

In my case I think the culprit was pulleys that were out of round.

mrehorstdmd

It's not unusual to have some roughness in straight sides of prints when viewed under harsh lighting, but in your print it extends to the corners as well. To me that implies a precision problem which might be the result of loose parts such as a wiggly extruder carriage or hot-end. I would start at the heater block and try wiggling it, then move to the extruder and work your way through the whole coreXY mechanism. Be sure to check the belt clamp(s) on the extruder carriage. It could also be that the bed is wobbling on the Z axis guide rails, so check there too. You're looking for very small movement which you might not be able to see but might be able to feel or hear.

This is what should be achievable:

alt text

Edgars Batna

I see from the images that the layers do not shift left or right but have varying thickness. I had a similar problem and this is caused by varying hot-end pressure. You need to find the root cause of the variation, which may include anything in FDM printers. Consistent layers are only produced if the nozzle is consistently exactly one layer above previous layer and if the hot-end and extruder can consistently produce the same pressure so that the filament fills the expected layer area.