Beginning of new line is curling / not adhering

Filament 3D Printer Technical Help Creality Ender 5 Plus
1

The beginning of most lines are not adhering to the bed. Layers on top of the initial are doing the same for many layers. I thought maybe bed temp but not sure. See photo.
I’m using a 0.6mm nozzle (tried several new).
15-20mm/s initial layer (glue stick on bed - thin or thick, dry or fresh)
Euro PETG 215-230 hot end (no change with temperature, higher temp = too much oozing).
Bed temp 50 or 60 (higher layers do the same so bed temp doesn’t seem to be critical).
Several areas on the bed (not location dependent).
Layer height 0.3mm (initial 0.2-0.35 - no change).

Creality Ender 5 Plus
Micro Swiss Direct Drive (tried many retract distance and speed settings)
Micro Swiss all metal hot end.
V2.2.1 silent board.
Bed is level (no warp) and using ABL.
No fan first few layers.

Any suggestions?
Thanks
R

LineCurl

2

I don’t consider my self an expert at this, but the first thing I’d do is lower the nozzle slightly to make sure the first layer gets “squished” onto the bed.

Second, there is an option in Cura to specify the initial layer line width. I use 120%.

Third, consider raising the nozzle temperature by 5-10C.

3

Thanks. I have been working on these.
Lower and higher nozzle.
Lower and higher temp.
110-130% first layer line width.

I’m thinking it may be with my extruder.
It seems the "un"retraction is late as the new line starts.
I have tried different distances and speeds as well.
I did have some success with extruder extra prime amount but concerned I was going down a rabbit hole.
Just about burned through a spool trying to troubleshoot

I appreciate your thoughts.

R

4

You are not close enough to the bed.
When I print with PETG, nozzle 230, 70 bed and fan off first 3 layers, then only 30% after that.
Never retract over 2 with a Micro Swiss all metal hotend.

5

Thanks Logan5,
That was my first thought as well.
The bed adhesion is good elsewhere and only seems to have this problem after some retractions (not all) and the issue persists on subsequent layers.
Retraction has always been 1.0mm or less (from stringing calibration test). No change with this issue.
Fan is off for first several layers then tried 25% to 40%. Issue persists.
For the first layer, do you recommend lower than 0.2? 0.15 perhaps?
Could this be an issue with my model? I don’t often see this on other prints.
Thanks
R

6

I agree, not close enough to the bed. Maybe some adhesive glue as well.

7

Jason, I’m using glue stick as noted. I’ve tried 0.2 from the bed with a 0.6mm nozzle. No change. Do you recommend 0.15? Sounds too thin to me. And what about the layers on top of the first layer? They are doing the same thing. I’m going in circles here. Frustrating.

8

Your photo doesn’t show layers on top of layers separating and curling, If you look closely, it’s actually beside it - curling off the bed. I still say lower it. Squish it into the bed.

9

Yes. I stopped the print to get a photo of this first line but there are many layers doing this. I’ll try lowering to 0.15. Has anyone used this for the first layer thickness on 0.6 nozzle?

10

@RascalHoudi if the didn’t help. I found similar with a 0.8 nozzle. I had to increase the temperature a lot. The volume and speed of the nozzle extrudes at means higher temps of the hot end to compensate. I found if I was normally printing @ 215 first layer 210 others, I needed to increase temps to 240/235. Some say more than this.

11

Update:
No change. Lowered the nozzle, still curls. The rest of the line is flat/squished and adhering to the bed. Could this be a priming issue in the nozzle? Too much? Too little?

Also, it seems to only happen in the X direction. Rotated the model and still shows on X direction and in more than one location on the bed.

@kitedemon I’ll try some higher temps. Will try other materials next.

12

Rather than reslice with a different first layer height, I’d suggest using a small negative z-offset on the printer itself. If you trammed (leveled) then bed with a feeler made from a piece of paper 0.1 mm thick (.004" a typical piece of paper), and did not compress the bed springs while tramming, then the printer’s “0” position is set that distance above the bed. I generally give the printer a z-offset to compensate.
Secondly, especially with PETG, a slower re-prime, with a bit of extra prime, helps me a lot. With a direct drive extruder, you won’t need to be as extreme as I, but I’ve been using a reprime speed down around 12mm/sec, and extra prime volume that roughly matches the volume extruded for 1mm of line. This lets the nozzle “push off” from the beginning of a line, rather than pulling on it.

13

Thanks. I was working on tuning in a reprime when I found out my z vref was too low when the bed dropped a couple of mm. Most of the curl is gone and adhesion is restored. Still fine tuning but pleased with the results so far. I will get back to the reprime soon.

14

Hi. Can you please tell me how you were able to resolve this curling problem? I read from your last post on the topic : “… tuning in a reprime when I found out my z vref was too low when the bed dropped a couple of mm. …”. I am not sure what ‘reprime’ means and also what the "z vref is too low, means.
In my case, I changed from a 0.4mm nozzle to a 0.6mm nozzle when this problem occurred. The 0.4mm worked fine but not the 0.6mm.

15

Hi Yves,

Sorry, it looks like this will be a long one. Just giving you some areas to explore. If you are new to printing then please start with the basics. See link below.

Quick summary… set the z distance to the bed, manually level the bed, use Automatic Bed Levelling (ABL) if you have it and reduce the print speed.

I’ve changed and fine tuned several things since my last post. I will give you my 2 cents worth for my Ender 5 Plus but these may or may not be helpful for your set up. I’ve been printing for 2 years and I like to tinker with the machine so it is no longer stock. I use direct drive, not bowden. I don’t know what machine you have and I have no experience on any other machines.

I suspect there is a first layer issue as there can be dimensional differences between nozzles but also filament flow factors and other settings when using larger nozzles (in my experience, I’m still new at this). There’s a link at the bottom to help with some of theses.

First, by reprime I’m referring to extra extrude during de-retraction. This seems to help with larger nozzles and it depends on your set up (bowden / direct drive). It extrudes a little extra filament before the print head starts to move. It didn’t solve my initial problem with curling and this is a personal preference setting that I continue to fiddle with. My prints are not fine detail as I generally make mechanical parts with some clearance built-in to allow for assembly / installation.

I suspect your nozzle gap to the bed is too large. I would start there. The Vref is here as a short explanation but the default currents should be OK however not always set correctly at the factory.

The first and biggest solution for me was with my Z-axis current to the Z stepper motor. In my post I stated that I observed the bed drop as I was troubleshooting. The Vref is a setting on the control boards that will increase or decrease the current to the stepper motor (not as simple as it sounds). This requires accessing components on your control board and I don’t recommend it if you don’t have electronics/electrical experience (my career). My bed is heavy and more current was needed to hold the bed in place for each layer. There are many videos on YouTube addressing this and many articles to read. The calculations are different depending on which drivers you are using (at the time I had a Creality board with TMC2208 drivers). I’ve upgraded my main board where I use firmware to adjust the current to the stepper motors(less fiddling). I increased the current just enough to stabilize the bed.

More important is your first layer. Yes, I know, you’ve heard this many times before but it is critical.
Depending on your machine, of course, there are some steps to this. I first use a block that I printed to level the left and right Z lead screw bed supports (the block is easier than measuring each time). This is a very coarse bed levelling. Then I use the Marlin probe wizard (I use BLTouch) to carefully set the nozzle distance to the bed (0.1mm) and save that setting to the EEPROM. Your set up may have a different name for this. I have a set of feeler gauges with a 0.1mm gauge. Then I manually level the bed using the feeler gauge. I go around the bed at least twice. I no longer use a piece of paper (typically 0.1mm thick) because there is too much variation for my liking. I then go back to the probe wizard once more to fine tune. Saving my changes of course.

Next, for the curling or gap at the beginning of a line, is setting your print speed. (The accelerations for X and Y defaults should be fine.) What can happen is the printing speed can be too high for the nozzle size and the print head accelerates to get to the printing speed. As the filament begins to extrude the acceleration will move the print head to the desired speed. If the print speed is too high the filament doesn’t have enough time to reach the bed or previous layer as it accelerates and shows as a gap in the print or curling. I tuned my max acceleration and max print speed using the link below. Seems OK at this time.

Recently I reset the EEPROM back to factory and started fresh. This has been useful because my learning curve lead to me having some settings changed that I should not have changed. I use Prusa Slicer and have taken a great deal of time to understand the settings and use those instead of firmware settings whenever possible. I also use custom G-code to test firmware changes without saving to the EEPROM and getting into that mess again. I have also taken a great deal of time to understand Marlin and document all of my changes so I know the defaults and effects of my changes.

I have tried setting up Linear Advance but I’m still learning and it appears there is little benefit for my set up at this time.

I use the setup website provided my Michael from Teaching Tech on YouTube. A great place to start although you won’t find an exact solution to the “curling” problem I had. His site is https://teachingtechyt.github.io/calibration.html

Best of luck with this interesting hobby. Some fiddling required.

Rascal Houdi
AOMU (Actual Operating Mechanism Unknown)

16

HI, @Yves Welcome to the forum. Glad you found us.

If you don’t mind can I ask what symptoms you are seeing personally? Sometimes the problem gets a little muddy if it’s just like the last one. Love troubleshooting hard problems and just wanted to hear from you what the symptoms are.

17

Hi Jason.

Thank you for contacting me.

The problem that I am facing is this ‘funny’ curl that appears at the beginning of each new line (see photo below).

I tried many combinations of variables but cannot manage to make that curl go away. From what I can see, the filament coming out of the nozzle appears to stick to the nozzle for a while then lets go once a certain length of filament sticks to the bed.

This only started to happen once I moved from a 0.4mm to a 0.6mm nozzle.

I printed successfully for the past year using the 0.4mm nozzle (PLA, Nylon+CF, PETG, ASA) but I wanted to experiment with a larger nozzle - it is all about learning - and that is when this appeared (a.k.a ‘The curl’).

Here is my system:

Ender 5S;

32 bits motherboard (BTT SKR-mini-E3-V2.0);

Capricorn bowden tube;

BL Touch;

Glass plate on bed;

Octoprint (1.8.7);

Using CURA slicer (5.2.1);

Marvin Firmware (2.0.9.3);

All metal hot end (micro-swiss);

Both 0.4 and 0.6 nozzles are hardened steel nozzles;

Using the same roll of PLA for this nozzle switch.

In terms of debug, I did the following:

I ran the bed mesh (in octoprint);

performed the Z-offset and stored the values - also ran Z offset wizard;

Perform the extruder z-step calibration;

changed the following variables:

  • nozzle temperatures (190-240C; steps of 5C);

  • bed temperature (55 - 70C; steps of 5C);

  • Layer height (currently at 0.3mm);

  • line width at 0.6mm;

  • Initial layer line width from 100% to 125%);

  • Print speed (from 10 mm/s to 50 mm/s);

  • travel speed 150 mm/s;

  • retraction distance (5mm and 6.5mm);

  • retraction speed (45 mm/s)

  • retraction Extra Prime Amount 0 and -0.216 mm cube;

Best Regards,

Yves

photo
photo1920×2560 308 KB

18

what I am seeing there is the offset is too high, basically, you are not squishing the filament into the bed hard enough. when the head pulls back it pulls the filament back off the bed with it. and it’s leaving you with a funny little curl.

Your temps and such seem close enough.
Are you using any kind of an adhesive on the bed or just clean glass?

19

Hi Jason. I reverted back to my 0.4mm hardened steel nozzle and I am now seeing the same/similar curl. I am trying to debug that now. I have not used any adhesive on the glass. I just cleaned it to make sure that it is as clean as possible. Also, while back with the 0.4mm nozzle, I am reducing the height of the Z offset using the wizard. Using a 0.1mm feeler gauge (following Rascal’s approach), I obtain a Z-offset of -2.300mm. Using the Z-offset wizard, I am now at -2.490mm.

Cheers,

Yves

20

sounds like you are on the right track, I know every manufacturer says “you don’t need glue on my bed”. On a personal note, If I need glue to make it stick I just use glue. Its all water soluble and easily cleaned later.