Darker layers always on the same layers

[baxi 2]

1 minute ago, UbuntuBirdy said:

Could it be the height where the heat from the bed no longer reaches the layers but the fans are at 100% (perhaps the fans reach at exact this layer 100% and the temp at the nozzle drops down until the heat element can correct the temp) and their air is reflected from the bed so the layer is cooled twice?

 

Thx for the idea.

I found this shift in all distances to the bed, from 2-3mm to 40mm highest. So I would guess this is not the reason. Esp. in the benchy I can move the layer shift up and down by setting different top layer numbers.

[baxi 2]

Finished now the slow print, speed 40/20, Acceleration 300, Jerk 7.

No change, the line shift is still there.

[redslifer 2]

This is the (well, one of the multiple reasons) I abandoned cura. I never found a way to print a benchy without the split on that section, all the benchies I tried separated on that layer like if they were printed separately.

This never happened with any other slicer, and I don't use jerk and acceleration either actually, or at least I tried both ways before.

This along with the bugs that corrupts cura 4.0 machines and profiles at every restart (very common bug since 4.0, I read about it everyday...) made it impossible to use Cura for any purpose.

 

This said.. the split happens on the first "top" layer, that's where the material on the top doesn't bond at all with the one under it, it's not like it partially melts in some points, it just doesn't on any point. As I said, they seem printed separately.

 

Edited May 6, 2019 by redslifer

[neotko 1,417]

On 5/6/2019 at 2:16 AM, baxi said:

Finished now the slow print, speed 40/20, Acceleration 300, Jerk 7.

No change, the line shift is still there.

 

The slower the acceleration and jerk the bigger the issue. Assuming you have a um2 it should be 3000/20

 

Try to print it rotated 45 degrees. If the problem goes away I would triple check also x/y alignment of the axis

[Smithy 1,145]

I see similar issues on some objects and I also did some tests some time ago. But I cannot confirm it is a slicer issue, I sliced it with Cura and S3D, both with the same result. It also depends on the color of the filament how much you see.

 

For me, I did several calibrations on the x/y axles, but it was not the problem. I still think it is some kind of over/under extrusion in combination with speed and temperature. But I stopped investigating deeper, so I really appreciate the test series from baxi, which give a great overview of the problem.

[baxi 2]

6 hours ago, neotko said:

 

The slower the acceleration and jerk the bigger the issue. Assuming you have a um2 it should be 3000/20

 

Try to print it rotated 45 degrees. If the problem goes away I would triple check also x/y alignment of the axis

Thx for your ideas.

Maybe I will try higher acc later.

 

I already turned the print with no change in the shift. So it's not an axis problem.

[gr5 2,224]

It would be nice to see more photos of the shift - especially ones that show the worst shift.

 

One possibility is that you have "play" or "backlash" (look it up in wikipedia). With printers that have noticable backlash if you print the same layer clockwise versus counter clockwise you can get shifts like you seem to describe.

 

I would check to see if the shifted layers are CW and the unshifted layers are CCW (or opposite).

 

Many times people have reported a bug in cura versus some other slicer only to find out that it was a hardware issue and the only difference in slicers was some minor speed difference, CW versus CCW, "wipe distance" or other minor thing.  Sometimes a tiny change that shouldn't affect the print does affect the print.

[baxi 2]

How to figure out if it's printed CW or CCW?

[gr5 2,224]

Cura will tell you.  In layer view go to a "good" layer with the vertical slider.  Then use the horizontal slider to see the order of printing.  Then repeat on a "bad" layer.

 

You can do this with one of your projects that you have saved and then reslice it before doing this.  Or if you have the gcode file you can load that into cura and do the same thing.

 

There are other gcode viewers also like repetier host, that show you each gcode and then highlight them in red on the model.  But for purposes of CW versus CCW, cura is most likely the best option by far.

 

[baxi 2]

Ah, there is no horizontal slider....

 

Btw. i have the Linux version.

Edited May 7, 2019 by baxi

[weisheng1998 0]

I am having the exactly same problem on the benchy , constant layer shift in front one line width at that layer... already have solution for that?hmmm

Edited April 4, 2020 by weisheng1998
attach picture

[v0g3l 0]

@baxi Could you find a solution for this problem at last. I always had this issues as well, finally found this thread. I cannot get rid off it, hoping you can answer me something.

 

This issue appears in all prints I do, as well, for ages.

 

Thanks,

Hendrik

 

[WhatWouldAnEngineerDo 0]

This is an old thread, but a continuing problem.  For anyone researching this issue, I recommend this article: https://help.prusa3d.com/en/article/the-benchy-hull-line_124745