Weird layer shift

[klapper2 0]

I have tried printing this part on 2 different S5s and sliced it multiple times. Unfortunately the only orientation where it will fit on the bed is standing like this. For some reason at the same height on every attempt the layers seem to shift on the ends but not the center of the print. I am using pretty much default settings in Cura and have breakaway filament for the support. I also tried printing some flat towers in the corners of the bed to see if the issue was just because the part is way in the corner of the bed but they came out fine. Not sure what the issue could be.

[Slashee_the_Cow 541]

If you could share the Cura project file (.3mf, set it up then go to File > Save Project) so we can see your settings that would help.

 

Your problem isn't layer shift though, that's when one of the printer's motors skips a step or two and from that point on the rest of the print is out of alignment with what's already been printed.

 

It looks more like it's either ringing (which is an artefact from printing too fast) or your nozzle is partially clogged (so it can't always push out the right amount of filament).

[gr5 2,295]

The top corners are being pulled inwards.  The top flat area is creating a very strong pulling force pulling the top two corners towards the middle.

 

Let me back up.  When you print PLA (not every plastic but most plastics), it comes out of the nozzle sticking to itself like snot - like a mucus.  Like a liquid rubber band.  As you lay in the top cover of that part it is pulling inwards because in the first few milliseconds after it comes out of the nozzle it's already shifting.  So the problem starts where the top of that "box" is - just the last few mm of the part.  This inward pull is very strong with that large surface area.

 

I don't know what the solution is exactly. Maybe less fan for the "top of the box" layers?  Put the fan at 2%.  Plus cover the printer?  I'm not sure that will help much but getting the air inside the printer to 35 or 40C will help quite abit.

 

Normally one fixes this kind of thing with more support holding things in place but you don't really have any more space outside the part.  It's a hack but you could design in having that top layer lean outwards a bit by about 3mm or whatever amount is shrinking.  Is there room on the bed to move outwards another 3mm on each end (or whatever the shrinkage amount is)??

 

Another common solution that I assume won't work for you is to put lots of holes in that "top of the box" to reduce how much pulling force there is.  But I assume you need that "air tight" as this looks like some kind of funnel.

 

It may seem crazy to add in printer shrinkage issues in your model but this is how they do it for injection molding - except usually the engineer who does the tweaking is located at the factory and is different engineer who designs the part.  They even have to tweak 90 degree corners in injection molding to be at a different angle because they know corners will change by a few degrees.

[Slashee_the_Cow 541]

1 hour ago, gr5 said:

Let me back up.  When you print PLA (not every plastic but most plastics), it comes out of the nozzle sticking to itself like snot - like a mucus.  Like a liquid rubber band.  As you lay in the top cover of that part it is pulling inwards because in the first few milliseconds after it comes out of the nozzle it's already shifting.  So the problem starts where the top of that "box" is - just the last few mm of the part.  This inward pull is very strong with that large surface area.

 

I don't know what the solution is exactly. Maybe less fan for the "top of the box" layers?  Put the fan at 2%.  Plus cover the printer?  I'm not sure that will help much but getting the air inside the printer to 35 or 40C will help quite abit.

I've had problems with stuff being pulled. Well it's less a problem with stuff being pulled and more a problem that Creality think Creality quality printers that can accelerate at 4000mm/s² or higher are a good idea. Not sure how fast an S5 can accelerate but I'm sure the quality is a case of "you get what you pay for"... and I didn't pay for much.

 

Err... anyway, you could try reducing the acceleration. Will make your print take longer but don't forget Slashee's Golden Rule™: Slow print > bad print. There is such a thing as acceleration too low - at least for the very precarious thing I was printing upright at the time - but for a test for the truly paranoid, maybe limit it to 500mm/s² - including travels. Some of the warping I was getting from stuff being pulled was the print head zooming away on a travel move like an F1 car at the start of a race.

 

On the bright side, you probably also don't have to deal with Creality crap like the printer ignoring the acceleration values in the gcode and just going as fast as it wanted to (I had to limit the acceleration in the printer's interface to actually put a cap on that).

[klapper2 0]

On 1/24/2025 at 12:04 AM, gr5 said:

The top corners are being pulled inwards.  The top flat area is creating a very strong pulling force pulling the top two corners towards the middle.

 

Let me back up.  When you print PLA (not every plastic but most plastics), it comes out of the nozzle sticking to itself like snot - like a mucus.  Like a liquid rubber band.  As you lay in the top cover of that part it is pulling inwards because in the first few milliseconds after it comes out of the nozzle it's already shifting.  So the problem starts where the top of that "box" is - just the last few mm of the part.  This inward pull is very strong with that large surface area.

 

I don't know what the solution is exactly. Maybe less fan for the "top of the box" layers?  Put the fan at 2%.  Plus cover the printer?  I'm not sure that will help much but getting the air inside the printer to 35 or 40C will help quite abit.

 

Normally one fixes this kind of thing with more support holding things in place but you don't really have any more space outside the part.  It's a hack but you could design in having that top layer lean outwards a bit by about 3mm or whatever amount is shrinking.  Is there room on the bed to move outwards another 3mm on each end (or whatever the shrinkage amount is)??

 

Another common solution that I assume won't work for you is to put lots of holes in that "top of the box" to reduce how much pulling force there is.  But I assume you need that "air tight" as this looks like some kind of funnel.

 

It may seem crazy to add in printer shrinkage issues in your model but this is how they do it for injection molding - except usually the engineer who does the tweaking is located at the factory and is different engineer who designs the part.  They even have to tweak 90 degree corners in injection molding to be at a different angle because they know corners will change by a few degrees.

Sorry I should've elaborated more. I aborted the print when the issue started, but the problem is not on the very top layers, its the top 20-30 layers .

[klapper2 0]

On 1/23/2025 at 7:52 PM, Slashee_the_Cow said:

If you could share the Cura project file (.3mf, set it up then go to File > Save Project) so we can see your settings that would help.

This has been sliced 4 different times, I think the only setting that were changed were infill% and support (between tree and regular supports) 

UMS5_Router Vacuum Port.3mf

[gr5 2,295]

Let me clarify.

 

 

Let's call the teal area the "roof" and the red area the "floor".

 

Because these are large and solid (no holes) and horizontal and don't have a lot of walls to hold them apart (there is that back wall) I would expect problems on layers that include these.

 

I would expect pulling forces to pull in the left and right edges but only where the roof and floor layers occur.  Maybe this is not your issue - I don't know.

 

Also I would expect another issue on the layers above the teal roof: you have a long "pencil" like structure - it's long and thin like a house beam or like a pencil.  I would expect that also to pull inwards messing up layers above the roof.

 

All of these errors I would attribute to shrinkage - I tried to describe the process above and it has to do with "liquid rubber bands of filament".  You are less likely to have problems with the "floor" because you have lots of support under that floor.

 

Now in some of your photos I also see what looks like irregular Z movement where a single layer sticks in or out.  That could be unrelated to shrinkage - that could be caused by a dirty Z screw.  If you ahve a dirty z screw and you print a tower and then place the tower next to your other prints - do the single layers line up?  If so then it's a dirty z screw and has nothing to do with your model or your settings.

[Torgeir 287]

Hi.

 

When I saw this "huge" model, I've first wondered -will this be able to print in an UM2E+?..

Actually, if I riced it up and "dived" it about "Z to go -4 mm down" it actually could go in..

 

Here's what's look like:

 

So, -why not try to print it the same way in the much bigger UMS5?

Looks like this:

Sure, you might improve this support issue in the beginning -but heat expansion would be less of a problem.

But as Gr5 said it is needed to use a top cover to keep inside temperature stable.

 

Anyway, just another way to see it.

 

Thanks

Torgeir