"Sedimentary layers" in z-axis

[choco 0]

Hi,

I'm printing a chocolate mold, specifically this bunny mold http://www.thingiverse.com/thing:17507.

I found that by changed the orientation of the print to ensure that the mold itself was defined purely by the perimeter of the shape and I got a HUGE improvement in the quality of the mold.

Unfortunately, once every inch or so I have these strange z-axis layers that are out of place and stick into the mold space. They are of varying thicknesses. I've attached some photos. They show the front and back of the same piece with the sedimentary layering clearly visible.

Can anyone suggest what might be causing this?

Thanks!!

Choco.

 

[gr5 2,072]

It's probably either increased extrusion or errors in the z height. lol. Both of these are probably hard to find. I assume the darker bands are where there is more material? It's probably less visible when the light isn't passing through. Since it's a mold you could sand it a bit to get rid of these ridges.

Z:

Did you maybe put a weight on the build platform at one point and then later remove it? lol. Seems unlikely. Is there something inhibiting the z movement? A rope, a wire, a sticky spot? I guess I recommend greasing the lead screw again with the green stuff.

Extrusion:

Well... Did you change the print speed during the print? Probably not but if you did that might do it. Or temperature at the head (higher temps let the plastic flow better). Or the filament may have varying thickness. Because there are no gaps - no possibility of stringing - I'd print this very hot - 240C would be great. With the fan on high.

speed:

There is a "minimal layer time" feature in the "cool" menu. If you are hitting this limit then each layer will print at a different speed which would explain everything. If you are even slightly musical you can hear the different speeds on each layer - each layer will be at a particular speed. Typically every other layer is different speed due to different direction (amount) of infill. This is a pretty important feature so if you are hitting this limit you should probably just print slower. I found that 5 seconds per layer is plenty of time to let it cool at .2mm layers with fan on. Also you will probably get better quality at .2mm layer height than .1mm layer height. At least I usually do contrary to what you would think.

Printing at 240C will also help with speed issues so that the pressure in the head is similar at slow print speeds versus faster print speeds.

Don't print at 240C for other parts that have gaps where the head needs to move across to another part of the print because at 240C you will get bad stringing.

 

[Daid 304]

Are those solid layers? That could be the "interface layers" problem people are talking about. The beta version of Cura has solved this, so you might want to check that one out.

 

[choco 0]

Thanks gr5 and Daid for taking the time to reply. I really appreciate it.

I'll give your suggestions a go and post some updated photos.

Cheers,

Choco

 

[choco 0]

Hi again,

Writing back with an update on progress since trying the suggestions from the forum...

I have re-greased the z axis with the green stuff, upgraded to the latest version of Cura (nice updates Daid) and tried printing at 240 degrees rather than 180 as I usually do.

I'm only printing at 50mm/s, so I don't think I'm hitting the configurable 5s limit. I guess I could have inconsistent diameter filament, but don't have any calipers to measure it. I'm printing at .2mm rather than .1 and have selected "Duplicate Outlines" in the Quality settings of Cura. I've chosen this because it's the outlines that form the shape of the mold.

The attached photos show that I'm still seeing the z-axis layers. The one on the right is 240 degrees, the one on the left is 180 degrees.

 

This photo shows more detail of the 240 degrees print. The vertical lines you can see are just the shape of the 20% infill being seen through the transparent plastic and don't impact my mold at all. It's the seemingly random horizontal lines that are causing the problem.

 

While the suggested changes haven't solved the problem, they do seem to have mitigated it somewhat. The lines are still clearly visible but are less obvious to the touch and should impact the mold less. So in some ways, it's now mainly a matter of aesthetics.

 

If anyone has any further ideas, I'd love to hear about them.

 

BTW, this forum is great.

 

Thanks,

Choco

 

[gr5 2,072]

It's hard to tell but it appears that if your filament wasn't transparent you might not even notice this.

Because it happens at different places on the same printed part I think it is most likely filament related.

Alternatively it could be temp related - some people have a loose wire to their thermocouple and it reads the wrong value sometimes and the temp can drift up or down by 20 degrees even though the display is showing a consistent temp. If it's the later problem it will get worse and eventually you will get a temperature failure reported from Marlin.

I'm going to sick with: filament quality.

Regarding duplicate outlines (aka skin) - that feature isn't very useful. It does both passes identically (but moving up a half step). Because if it didn't it might collide with the infill which is twice as thick in the z direction. So you don't really get a better surface yet it takes longer.