Strange first layer problem  

  RSS
TT
 TT
(@tt-2)
Active Member

Two weeks ago, I assembled an MK3S and passed all self test and xyz calibration, but got stuck in a strange first-layer printing problem.

After completing the First Layer Calibration, Bed level test, the result is not bad (of course, after making the bed flat as possible with 8 springs).

Bed test

However, when I trying to print real stuff, a relatively stable imbalance problem appeared on the first layer  and this issue appeared in almost the similar position in each test.

first layer issue

After re-installing the X and Y axis and the print head, making sure that every step was carried out in accordance with the installation instructions, the same problem was still reproduced stably, and I completely lost my direction.

What should I do next?

Posted : 21/04/2021 3:25 pm
Dan Rogers
(@dan-rogers)
Honorable Member

What springs?  There are no springs ...

 

Posted : 21/04/2021 4:13 pm
Dan Rogers
(@dan-rogers)
Honorable Member

Aside from that - I am seeing under extrusion, and a pretty bad first layer.  Have you enabled 7x7 bed mesh leveling?  If you modified with some kind of leveling mod (springs?) then I'd be concerned more with the bed shaking and wobbling from having a sprung bed.  

Enable 7x7, and learn about live Z.  Calibration is just a setup step.  You will need to adjusts Z for each sheet, and each type of filament, sometimes each brand, each color.

Posted : 21/04/2021 4:16 pm
TT
 TT
(@tt-2)
Active Member

Yes, it comes with nine 6mm isolation columns. The first layer of printing problems occurred at the beginning with isolation columns, so I replaced the isolation column with springs as leveling mod and leveled the bed to eliminate the problem caused by uneven heat bed.

Hot bed after leveling:

and the numbers:

0.22917 0.22454 0.21593 0.20333 0.18676 0.16620 0.14167
0.20898 0.21921 0.22165 0.21630 0.20316 0.18223 0.15352
0.19426 0.21402 0.22387 0.22380 0.21381 0.19390 0.16407
0.18500 0.20898 0.22259 0.22583 0.21870 0.20120 0.17333
0.18120 0.20408 0.21782 0.22241 0.21785 0.20415 0.18130
0.18287 0.19933 0.20955 0.21352 0.21124 0.20273 0.18796
0.19000 0.19472 0.19778 0.19917 0.19889 0.19694 0.19333

 

Posted : 21/04/2021 4:27 pm
TT
 TT
(@tt-2)
Active Member

@dan-rogers

Yes I have enabled 7x7, and adjusted live Z  many times for each sheet and each filament from beginning.

This post was modified 4 weeks ago by TT
Posted : 21/04/2021 4:30 pm
Diem
 Diem
(@diem)
Honorable Member

@tt-2

Yes, it comes with nine 6mm isolation columns. The first layer of printing problems occurred at the beginning with isolation columns, so I replaced the isolation column with springs as leveling mod and leveled the bed to eliminate the problem caused by uneven heat bed.

This is a new build?  Not a rebuild?

If the new heatbed is uneven then first remove it and recheck it with a straightedge; is it flat now?

If so reassemble with the spacers, no springs, taking extra care over the tightening sequence making sure not to distort by overtightening.

If not then photograph the bend, contact Prua support by logging in to your shop account and opening a chat session.  Show them the evidence and arrange a replacement; probably under warranty.  Reassemble as above, no springs.

Design and print something that requires 9 springs...

Cheerio,

 

 

Posted : 21/04/2021 8:50 pm
TT
 TT
(@tt-2)
Active Member
Posted by: @diem

@tt-2

Yes, it comes with nine 6mm isolation columns. The first layer of printing problems occurred at the beginning with isolation columns, so I replaced the isolation column with springs as leveling mod and leveled the bed to eliminate the problem caused by uneven heat bed.

This is a new build?  Not a rebuild?

If the new heatbed is uneven then first remove it and recheck it with a straightedge; is it flat now?

If so reassemble with the spacers, no springs, taking extra care over the tightening sequence making sure not to distort by overtightening.

If not then photograph the bend, contact Prua support by logging in to your shop account and opening a chat session.  Show them the evidence and arrange a replacement; probably under warranty.  Reassemble as above, no springs.

Design and print something that requires 9 springs...

Cheerio,

 

 

Thank you for your suggestion.

I will remove the springs, replace them with the original isolation columns, re-adjust Living Z, print the test models again, and upload the picture.

By the way, how to measure the flatness of the bed? It is difficult to tell by visual inspection.

Posted : 22/04/2021 2:51 am
Diem
 Diem
(@diem)
Honorable Member

@tt-2

By the way, how to measure the flatness of the bed?

Hold a straight edge, a ruler, against the bed and check for gaps; repeat for the main axes and corner to corner. 

It is difficult to tell by visual inspection.

If it's difficult to tell by this method then it's probably close enough fo mesh-bed levelling to eliminate.

If you need a photo to show Prusa do the job against the light so the gap(s) show up well.

Do this with the bed alone, unmounted, to check whether the part itself is distorted or your mounting procedure is the problem.

Cheerio,

 

 

Posted : 22/04/2021 9:12 am
TT
 TT
(@tt-2)
Active Member

Has it been resolved?

After re-assembled the original isolation columns, the mesh leveling showed that the bed was tilted to the right side seriously.

In order to invest the issue, I printed three 0.8mm spacer and added them under the isolation column on the right. The mesh leveling results are as follows:

0.15750 -0.08167 -0.11417 -0.07333 -0.11750 -0.09583 0.05250
0.05750 -0.07750 -0.16667 -0.17000 -0.12833 -0.06417 -0.01167
0.07000 -0.07667 -0.16583 -0.13875 -0.10583 -0.02167 0.05000
0.14250 -0.03833 -0.15083 -0.11333 -0.06042 0.02333 0.14167
0.11417 -0.03417 -0.11417 -0.08542 -0.04583 0.05500 0.14250
0.13833 0.00250 -0.08000 -0.06833 -0.00250 0.06917 0.17083
0.20750 0.01417 -0.07083 -0.02833 0.01583 0.11167 0.24250

Compared with the previous point of view, although it is not flat enough, the values are not always positive, especially the middle height.

It seems that the issue is caused by the unevenness of the heat bed, but why does mesh leveling not compensate for the unevenness?

Posted : 22/04/2021 11:19 am
TT
 TT
(@tt-2)
Active Member
Posted by: @diem

@tt-2

By the way, how to measure the flatness of the bed?

Hold a straight edge, a ruler, against the bed and check for gaps; repeat for the main axes and corner to corner. 

It is difficult to tell by visual inspection.

If it's difficult to tell by this method then it's probably close enough fo mesh-bed levelling to eliminate.

If you need a photo to show Prusa do the job against the light so the gap(s) show up well.

Do this with the bed alone, unmounted, to check whether the part itself is distorted or your mounting procedure is the problem.

Cheerio,

 

 

Thanks for your suggestion, I will try it later.

Posted : 22/04/2021 11:24 am
TT
 TT
(@tt-2)
Active Member

I successfully printed some things this half day, and it looks good, I will do more tests in the next few days.

But what I am confused about is why mesh bed leveling work abnormally when the points mesured by MBL are higher than 0mm or the center is more than 0.2mm, but  the bed is nearly flat? Or is there another reason?

In short, thank you all for your help. 😘 

Posted : 22/04/2021 2:08 pm
Diem
 Diem
(@diem)
Honorable Member

But what I am confused about is why mesh bed leveling work abnormally when the points mesured by MBL are higher than 0mm or the center is more than 0.2mm, but the bed is nearly flat?

If you are getting some good prints then it *is* working, at least, it is in those parts of the bed you used.

You need to check the flatness of the bed off the printer.  If it is OK then check the Y carriage, if that's OK look at the alignment of the linear bearings, then the smooth rods and so on.

The object is to get the printer working before too many mods, otherwise you risk voiding the warranty.

Cheerio,

 

 

This post was modified 3 weeks ago by Diem
Posted : 22/04/2021 8:25 pm
Dan Rogers
(@dan-rogers)
Honorable Member

Springs.  There aren't NO springs.  That may be your whole issue.

Posted : 22/04/2021 8:27 pm
TT
 TT
(@tt-2)
Active Member

Springs.  There aren't NO springs.  That may be your whole issue.

Not all because of the use of springs. The reason why I tried the spring mod is because of the problem mentioned above. The original isolation column cannot level the hot bed, and the "bed level correct" and 7x7 are of no avail. The maximum difference between the height of the bed is greater than 1mm. I originally thought that mbl could compensate, but it turned out. The problem described above has occurred.

In order to eliminate the problems that may be caused by the unevenness of the heat bed, I searched for a physical solution for leveling the hot bed and found the spring solution. Then I tried to use the spring to level the hot bed and got a good flatness, but The problem is still.

I have repeatedly removed and re-used the isolation column many times, but the problem remains, until this time I don’t know why, I re-used the isolation column and compensated with a gasket on the right side, and it works magically.

I suspect that in all the leveling operations tried before, the MBL value of the center point is too high (greater than +0.2mm), but this time for some reason, after installing the isolation column back, the center point is slightly lower (-0.11333mm), maybe Is this the key? The height difference between the center points is as small as possible? It looks like this.

This post was modified 3 weeks ago by TT
Posted : 23/04/2021 3:11 am
Diem
 Diem
(@diem)
Honorable Member

Is the bed flat when isolated?   Is the Y carriage flat?

Cheerio,

Posted : 24/04/2021 1:06 am
TT
 TT
(@tt-2)
Active Member
Posted by: @diem

Is the bed flat when isolated?   Is the Y carriage flat?

Cheerio,

No, The bed is not flat when isolated. It's why I decide to try the springs.

I am not sure whether the Y carriage is flat enough, but from the MBL value, the Y-axis direction deviation is not large, but the X-axis direction has a relatively large deviation, and the right side is about 1mm higher than the left side. It can be seen from the heat bed leveling diagram (by the "Bed Visualizer" a plugin of OctoPrint) I posted earlier.

Posted : 24/04/2021 2:03 am
Diem
 Diem
(@diem)
Honorable Member

No, The bed is not flat when isolated.

Photograph the bend. Hold a straight edge against it with light coming from behind to demonstrate the problem.  Contact Prusa support by logging in to your shop account and opening a chat session (bottom right in most browsers, some ad blockers interfere so switch them off temporarily if you can't find chat). Show Prusa the evidence and arrange a replacement.

Prusa are very good at warranty issues.

Cheerio,

Posted : 24/04/2021 7:48 am
Share:

Please Login or Register