Question Irregular First Layer

[wmattias]

Printer Model

V0.1 with E3Dv6/Orbiter2.0/After-Sherpa

Extruder Type

Orbiter

Cooling Type

Other

Wondering if it’s normal to have a slightly irregular first layer on the V0.1.

I’ve got a pattern roughly traversing the X axis on my first layer. See pictures. I’ve ruled out the basics (Dirty build plate, uneven bed, uneven build plate, …) and my speculation is that it’s the printer head that lifts slightly on Z over that area.

Note also that the squish (in the banding region) is different depending on which direction the print head is going. I.e. every other squishes better, and every other line bulges up a bit.

Could this be related to the play in the MGN7 X-Axis?

Anyone else who has had a similar issue?

(Nevermind the lift in the lower left corner, I pulled the sheet up a bit there).

 

[LoadMaster7]

Did you calibrate your extruder? If the bed is level, it looks like over extrusion to me. If you have calibrated, flow needs to be dialed back.

 

[NoGuru]

Are you using a Bed Mesh?

 

[Laserbea4k43]

yeah either over or under extrusion, i think

 

[wmattias]

Did you calibrate your extruder? If the bed is level, it looks like over extrusion to me. If you have calibrated, flow needs to be dialed back.

yeah either over or under extrusion, i think

Thanks for the tip! - However, the extruder is calibrated: Both for length (i.e. 1m == 1m) as well as the Extrusion Multiplier (Flow). I get great results on the first and top layers if I print in the "middle" of the printer area, it's on the front side (Y around 10-40mm) I get this bulge that traverses X. I'm submitting a picture of the Elli's EM cube results I have as "evidence"

Are you using a Bed Mesh?

No, this is a stock V0.1 (i.e. no probe) so there's no possibility to do a bed mesh.

I did some more investigation, and to me it seems like it must be the play in the X-Rail MGN7 that is the culprit. There's some play, and I'm able to rock the printhead just slightly back/forward with minimal force. Looking into it, it seems that the pressure from the PTFE filament guide shifts the head from a "back" to a "forward" position when moving on the Y-axis - and around the same area where there's a irregularity in the first-layer.

How tight are your X-Rails? If you push/pull on the PTFE filament guide (or the top of the toolhead), are you able to rock it back/forth a little bit?

I'm tossing in the "First Layer Squish Test" STL I'm using also in case somebody wants to give it a spin. Note that it's designed for the 0.3mm layer height I prefer to use (I'm using a 0.4 Nozzle though).

------------------- ANALYSIS PICTURES BELOW ---------------------

Flow/Extrusion Calibration on Ellis Tuning Guide EM Cube (30x30x3mm) => IMHO this is well tuned.
I tried to use the harshest light I could: Shining a LED flashlight from the side and taking the picture from the most unfavorable angle.



Here is a better picture picture of the print irregularity/artifact (See bottom part of the print area - ignore the "turd" in the center)



First Layer "Seethrough on white background"
Note how the print is being printed too high off the bed in the lower left corner, but too squished in the right side center (you can see slight ridges).
Also note the lower left, where the print lines stick together in pairs (more on this in the next picture).
Center part of the print area I would call "well tuned first layer"



Below: Closeup of the lower left corner of the print area.
Note how every other line has been printed with a slightly too high Z, while the other direction is printed with a better squish.

 

[wmattias]

Here's a video of me pushing/pulling the printhead.

Note that I'm using no force at all, just gently pushing. All the play is in the MGN7 rail, not in the Beam or Printhead.

Do you guys/gals have similar play in your MGN7 X-Rail?

 

[Laserbea4k43]

One thing that caught me off guard was that my bed wasn’t hot enough. Setting it to 108 barely got me to 105. I had to push it to 110….
is that pla? I haven’t observed that much flex in my toolhead…not sure what would need to be tightened on your setup, if even at all, since I’m running a 2.4

 

[LoadMaster7]

That does seem to be excess play. Posting a photo of your printer would be of help.

I'm wondering if the toolhead is being pulled by cabling or something else at the outer areas that show layer issues. If true, it would not show as an issue in the center, only at the outer areas of the printbed.

 

[wmattias]

One thing that caught me off guard was that my bed wasn’t hot enough. Setting it to 108 barely got me to 105. I had to push it to 110….
is that pla? I haven’t observed that much flex in my toolhead…not sure what would need to be tightened on your setup, if even at all, since I’m running a 2.4

Yeah, that's just PLA. I can get the bed up to 110 (i.e. when printing PC) (Verified with IR + Temp Probe), but use 55 for PLA.

That does seem to be excess play. Posting a photo of your printer would be of help.

I'm wondering if the toolhead is being pulled by cabling or something else at the outer areas that show layer issues. If true, it would not show as an issue in the center, only at the outer areas of the printbed.

Attaching some photos of my printer here.

Yeah, I think that's the correct diagnosis: The X Linear Rail I has too much play, and the PTFE guide tube is pulling/pushing on the toolhead causing the irregularity. I've verified this also:
- If I move the toolhead to the back the PTFE pushes the top of the toolhead forward.
- If i move the toolhead to the front the PFTE pulls the top of toolhead backwards.
- The shift in position happens at the same Y position where the irregularity is seen.

I guess that the resolution is to reduce that play, seems that there are 3 main routes and 1 improvised to go:
1. Get a new MGN7H X-Rail with less play in the X-Rotation direction (i.e. Higher Quality and/or Higher Preload).
2. Do the MGN9C mod (or this) to get less X-rotation play: i.e. Use a MGN9C that is stiffer and can resist X-rotation better. I'm not sure which Preload to go for though.
3. Do the Dual X-Rail MGN7H mod: More work and more weight, but I guess this would take the Linear Rail X-rotation play completely out of the equation since we would be using 2 contact points on X-rotation for stability (not just relying on the Linear Rail being stiff enough in X-rotation).
4. Adjust the PTFE tube length so it has a constant pull (or push) on the print head. A bit fickle solution but maybe it could work. Not sure if it’s even possible though.


Not sure which route to go though - anyone have recommendations or experience with either? Especially around which Preload level to choose if doing the single MGN7H or MGN9C rail variant.

As a sidenote: Seems like the newer LDO V0 kits ship specifically with a Misumi X-Rail. I'm assuming this is to combat the X-rotation play. Would be interesting to know people's experience with that Linear Real - and what preload is used.

 

[NoGuru]

I think I would go MGN9. But the run is so short Duel 7's would be fine too.
I don't think you can get constant pressure with the PTFE. You could have a constant in one direction but the amount of pressure will vary.

 

[TinnT]

Have you seen this video from Nero3d -

I watched it yesterday and this looks a lot like what he explained in the video.

 

[wmattias]

Have you seen this video from Nero3d -

I watched it yesterday and this looks a lot like what he explained in the video.

Yeah, that’s a good video to keep in mind. The bed mesh picture resembles the pattern above (although mine is slanting on one side). My main suspect is the linear rail not being tight enough though - will get my hands on some new linear rail and report back.

 

[Jozi]

Did you resolve this in the end? I found I had the same problem and got a Hiwinn linear rail as replacement (to the LDO one). I haven't printed enough with it to conclude anything. What I also did was feed the wiring loom a little more through so it doesn't pull as hard and the filament tube is also long enough that it rubs the top panel.

 

[wmattias]

Did you resolve this in the end? I found I had the same problem and got a Hiwinn linear rail as replacement (to the LDO one). I haven't printed enough with it to conclude anything. What I also did was feed the wiring loom a little more through so it doesn't pull as hard and the filament tube is also long enough that it rubs the top panel.

Hi!

Yeah, managed to fix it! The Linear Rail was at fault - it was too loose in itself (i.e. preload too low).
I wouldn't say the quality of the rail is low, i'd rather claim the preload was too low.

I wanted to go with a HiWin, but couldn't get hold of one quickly so ended up ordering the following MGN7H 150mm rail from AliExpress:

https://www.aliexpress.com/item/32806622073.html

It is significantly tighter, and has very little play. Consequently it doesn't run as "freely" as the Vivedino i replaced but the play is nonexistent.

 

[FkYeahScience]

Here's a video of me pushing/pulling the printhead.

Note that I'm using no force at all, just gently pushing. All the play is in the MGN7 rail, not in the Beam or Printhead.

Do you guys/gals have similar play in your MGN7 X-Rail?

@wmattias could you please provide a before-after-comparison? I seem to have the same issue with my v0.1

thanks

 

[wmattias]

Hi!

Yeah, managed to fix it! The Linear Rail was at fault - it was too loose in itself (i.e. preload too low).
I wouldn't say the quality of the rail is low, i'd rather claim the preload was too low.

I wanted to go with a HiWin, but couldn't get hold of one quickly so ended up ordering the following MGN7H 150mm rail from AliExpress:

https://www.aliexpress.com/item/32806622073.html

It is significantly tighter, and has very little play. Consequently it doesn't run as "freely" as the Vivedino i replaced but the play is nonexistent.

Sure!

Before:
Have a look at my post from
Jan 30 where I’m holding up the sheet. There you can clearly see that there is an issue on the lower part of the printed sheet. Also there is a video that shows the play when you gently push the print head forward/backward.

After:
Look at my post from Apr 5. No artificats can be seen - it’s just a cleanly/evenly printed sheet. No video of the play posted, but the play is gone.

 

[FkYeahScience]

Sure!

Before:
Have a look at my post from
Jan 30 where I’m holding up the sheet. There you can clearly see that there is an issue on the lower part of the printed sheet. Also there is a video that shows the play when you gently push the print head forward/backward.

After:
Look at my post from Apr 5. No artificats can be seen - it’s just a cleanly/evenly printed sheet. No video of the play posted, but the play is gone.

I'm sorry, I meant a comparison of the play in your motion system / x rail, how much play there is - now that this is fixed.

 

[wmattias]

I'm sorry, I meant a comparison of the play in your motion system / x rail, how much play there is - now that this is fixed.

There’s zero play from the linear rail now. Before there was a very noticeable play.

Sure, if I push with some force the nozzle will move a bit but you can see that’s because there is some overall flex in the total X-Carriage assembly.

Before I changed the rail there was a very clear play that was clearly from the rail itself.

Video after changing to the new tighter rail:

 

[FkYeahScience]

I did the same now, the new rail with preload solved it for me, too.

All carriages in the formbot kit seem to have no preload (as it seems) and have play on all relative axis (slide block vs rail).
On x and z the double-mounted rails negate most of it, for x I had to order a new one (even the cheap one from fermiolabs solved this issue for me)

 

[dikkekletser]

Oh man this is my issue too, I've been racking my brain about this exact issue and I figured it might have been related to that play indeed. Have not had time to test anything so good to see it resolved this for you. I will be ordering some new rails aswel then .

Thank you!