Eliminating a stubborn line on one of my prints

[Bryan Murphy]

Hi, it’s me again! 😀

I continue to push my boundaries and have run into yet another problem. I’m trying to print this vase (Elegoo PETG-GF, Elegoo Slicer) and I consistently get a visible line towards the bottom of the print: https://www.printables.com/model/1105898-sand-palms-vase

When I look at the Line Type, Speed, Temperature, and Flow Rate views, everything looks normal:

However, when I look at the Fan Speed and Layer Time views, the line becomes visible:

Taking a closer look at the offending layers, I can see this coincides with the internal bridging that support the inside of the vase:  

If you look closely at the Fan Speed view, you can see that the speed drops on the external walls for the offending layers. You can also see that those walls are not used for inside bridging purposes:

Now I understand why layer time shoots up, but I don’t understand what is going on with the fan speed.  The fan speed hovers around 40% for most of the print but it drops down to the 10% range for the outer wall on the layers where the internal bridging occurs.

Why isn’t the fan speed for the outer wall consistent with the other layers?

I haven’t quite mastered OrcaSlicer’s fan settings.  I don’t find them intuitive.  Here’s my current settings for reference:

Thanks,

Bryan

[Sophie]

The fan speed for the outer walls isnt consistent with the other layers because the slicer takes into consideration for the entire layer. 
When your layer time takes longer than 30 seconds, the fan speed is set to be no more than 20% for that entire layer. The algorithm is thinking the layer is already taking a long time to complete and therefore cooling down on its own, and so doesnt need additional cooling.... in this instance its taking too long! and cooling too much!

Have you ever had a print pause for a period of time, whether thats a filament run out, or power outage, or blocked nozzle? Usually, even with a heated build plate, when you resume you will always have the dreaded ring where the plastic has cooled and shrunk enough from thermal contraction that there is a slight mismatch as the new layer goes down. It feels like thats essentially whats happening here?

The bridging could be causing a problem, whereby the settings are configured to slow down while it bridges - but even if you turned cooling and slow down off for bridging, you may still well see the ring appear when it completes the top layers of the base, because of the amount of time that takes to complete that as well. 

This is my take on whats going on here anyway - others might have some other suggestions

Depending on how fussed you are about the inside bottom of the vase being smooth - you could try adding a modifier cylinder that turns Top shells to 0, which will allow the actual top lip of the vase to be finished nicely, while keeping the top surface of the base just as infil and therefore restricting/minimising the layer time. 

I've used Bambu Studio but seeing as Orca Slicer is essentially a carbon copy of that - I imagine the method is the same.

It might be worth just doing a test print of the section you're having trouble with to save filament and time (drop the model below the surface of the buildplate a good 10-15 layers before the artifact appears and see what happens.)

[Bryan Murphy]

I think that makes sense.  I had how the thresholds worked backwards in my head.  If I change both speeds to 60% the fan speed is locked in across the whole print except for overhangs and bridging which ramp up to 100%.

I'm going to try a quick test print to see what happens.  If this gets rid of the line I'll print the whole thing tonight.

--
You received this message because you are subscribed to the Google Groups "3D Printing Tips and Tricks" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 3d-printing-tips--...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/b646d6dc-c465-450f-b459-a955c9fc5b24n%40googlegroups.com.

[Sophie]

If you're printing glass fiber PET then you ought to be using less fan to minimise warpage and shrinkage.

I would set your max/min to 20% and turn off 100% fan on overhangs. 

[Bryan Murphy]

Thanks!  I'll keep that in mind for future iterations.  I already kicked off the print with the current settings.  I like the quality of the test prints well enough with the current settings, at least for the vase it gives it a  nice rough but not too rough texture.

To view this discussion visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/ca1a8a1c-994a-4b43-b477-8667f42c1625n%40googlegroups.com.

[Sophie]

It's certainly some heavy duty filament you got running there for an ornamental object :) but if it gives off the effect you're after then worth it :)

[Sophie]

I wonder if you could get the same or similar effect with a white stone pla, or white marble pla with fuzzy skin on. Would be interesting to know.

[Bryan Murphy]

Probably!  It's all experimentation right now: what can I do with the new printer that I couldn't do with the old one.  The Elegoo PETG-GF was less than $20.  I'm definitely being a bit more judicious with the more expensive higher quality filaments.  My roll of Polymaker Fiberoligy PETG-CF is quite a bit more expensive so I'm saving that one for more purposeful prints.

To view this discussion visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/dd0bd2c3-a300-4061-ac62-276ea9d06b86n%40googlegroups.com.

[Bryan Murphy]

No luck. Same outcome.  I'm going to take a break from this problem tonight.  Maybe I'll try again tomorrow with less cooling or maybe it's time to accept that PETG isn't going to give me the price+quality+characteristics I'm searching for.

[Sean]

Hey Bryan,

Sorry to get slightly off topic, but I believe you and I are the only Centauri Carbon owners in the group and I thought you might find this video helpful. This fellow has some suggestions that seem to help overall print quality quite a bit for the ECC. Or at least seemingly in the examples he's showing. I'm definitely going to look into implementing these strategies in my printing processes.

https://youtu.be/mxjMyfFBUZ0?si=D8-3pCNBqeIpGihM

[Bryan Murphy]

This is great and I love this guy's content.  Thanks!

To view this discussion visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/3eec8fd1-3649-4910-87ce-ceeedb2ba127n%40googlegroups.com.

[Bryan Murphy]

3D Jake just posted a video about this exact problem: https://www.youtube.com/watch?v=03trZJWz7c4

[Bryan Murphy]

It's interesting.  The first layer where bridging starts goes all the way to the edge in a few places, but the layer on top of that does not.  I wonder if different or more infill would give it less need to stretch all the way to the outer walls.

[Bryan Murphy]

Rectilinear @ 30% infill (was Gyroid at 5%) for comparison.  This would add an extra 50g and 2-3 hours of print time.  I think I need to run some experiments tomorrow and see what impact this has in practice.  It doesn't have that red bridging layer at all, it goes straight to the blue layer.

[Bryan Murphy]

This is a better picture.  It is the red layer, I had to switch to the same view as the earlier pictures (layer time).

[Sophie]

You could indeed increase infill and turn bridging off if you really need that smooth surface inside. In addition maybe add another top layer count from 2 to 3? (I can't remember what it was at)

The layer time for that layer is ALWAYS going to be more, because you're throwing more plastic down. It's just how much more time you want to control.

And really you only need a more dense infill layer as you get closer to that surface. I haven't worked it out in bambu studio yet, but I remember in simplify 3d and maybe cura, you could have progressive infill, whereby it would vary the density of the infill based on what it was supporting. 

I'm sure the same thing could be achieved using modifier blocks and commands. So you could make the section below that surface have an increased infill and then once that bit has done make it go back to 5% to do the walls. Hopefully saving time and material. 

I would also say that 30% could be overkill? Maybe 15-20% is all you need? I haven't gone into the model this time to see what the impact is, just throwing out some suggestions off the top of my head for ya.

One other extremely farfetched thing you could do, is say no top layer (as I said in a previous post) but then 3d print a thin top layer separately, and then when it gets to a certain heght, add in a pause where you very quickly drop the separately printed top layer (include a bit of glue if you like) and then continue printing. The pause to add that would be way quicker than it printing it. And you would still have your surface.

But I feel like this is more than just finding a way around the problem...it feels like you're on a mission to understand the relationships between layer time, bridging, impact of infill density on the algorithm for bridging and the want to be able to just print this vase without issue or without having to do "special" things...

To that I say - why don't ya grab yaself some more kind material, like pla or PETG :)

[Bryan Murphy]

I did play around with the infill settings a bit last night.  It depends on the infil but 15% wasn't enough to get separation from the edge.  It was somewhere in the 20%-30% range for most settings I tried in order to get a clean gap. 

Originally I had low percent infill, 3 walls and 5 top/bottom layers.  Going for strength, but I might be able to sacrifice some of the layers and go with 2 walls if I end up using more infill to resolve this problem.

I like the separate piece idea, but at the end of the day I'm doing this to (a) learn and (b) see if I can one shot a clean print.  If I can one shot a clean print I'd feel more confident that I could sell a part like this.  I want to avoid too much manual labor.  I will try other filament types when I have a chance.  Unfortunately I'm quite busy so I don't get to dedicate as much time to this as I would prefer.

--

You received this message because you are subscribed to the Google Groups "3D Printing Tips and Tricks" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 3d-printing-tips--...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/6f0304f7-09c4-4118-9d40-3694a61305ffn%40googlegroups.com.

[Bryan Murphy]

It looks like it's going to work.  I'm well past the point where the line shows up and it looks good.  I ended up going with 15% rectilinear infill.  I'm usually not a big fan of rectilinear but it was the one that gave the most even fill at the lowest percent, gave me a good clear gap, and it didn't add anything to my print time.  I also went with 2 walls and 3 layers to save some time.

I'm letting the print finish, it should be ready for some photography this evening.

[Bryan Murphy]

I take that back. It's still visible it just requires more light to see. I'm going to let the print finish and move on to a different filament.  

[Sophie]

Ah that's annoying that you were nearly there...

I'll take a look this evening too see if it's a model issue or if it's specific to your settings. 

[Bryan Murphy]

Well I'm glad I printed the whole thing. I was able to prove it's not even close to waterproof with these settings. I'll probably use it as a small nice looking trash bin instead of a vase. If I try again I'll need to revisit the number of walls and work on the seam quality.  

To view this discussion visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/3944975a-dafa-4196-90e7-0bd9df10b148n%40googlegroups.com.

[Sophie]

well...yep - there are a whole bunch of other things to learn about making prints watertight... whether that's changing the slicer settings to print watertight or post processing with an epoxy/vapour smoothing (abs)/acrylic varnish

But you said you wanted to be able to do minimum work in order to sell these right? so post processing is probably not for you...

As a kind of side note...I take it you're the owner of this file? Or at least have a licence to print this and sell it? Only I noticed it had a non commercial license on it....

[Sophie]

I just tried a real quick print to see if it was a model problem....I only printed the first inch and a half using some old PLA that I don't mind wasting.

I had complete water tightness, and no artifacts (in this small section anyway)

Some settings outside of default I used in Bambu Slicer:

Initial layer flow 1.2

0.8mm line width for all setting variables

Spiral mode and all the mandatory settings that come with that (aka vase mode)

Random seam position

Of course I realise each printer is different and this just works for me, but at least it works so its not a model thing.

If I had one I would print this in a 0.6mm nozzle and make the line width 1.2 and the initial flow rate 1.4 to give the vase more rigidity.

Total estimated time for this print (including a 5 minute calibration warm up) 2hrs 37min, and total filament usage would be 82.27g

Point to note - by using spiral mode completely negates the need for infill, bridging, or the slow top layer of the base which was happening for you some way up from the bottom.

So - it's something to consider I'm sure. I don't have any PET GF on me right now to test, but so long as its not getting hot (anything above 45-50degsC start to worry) PLA would be fine.

[Bryan Murphy]

To quote myself:

"I could sell a part like this"

I'm definitely aware of the licensing restrictions.  I just found this piece would make a nice test piece for learning purposes.  If I do start selling something I intend to fully model my own so nobody can make any claims on them.

To view this discussion visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/66bb6d95-3eb2-4447-a4d7-7aa51867abb1n%40googlegroups.com.

[Bryan Murphy]

I'm not surprised it's working a little better with those compromises.  If I were to productize something like this I would likely go with a .6 or .8 nozzle.  I'm stuck with the .4 for now (I'll have a .6 soon).  I wouldn't trust a vase print though, too flimsy.

I stayed in a hotel last month and I noticed the trash cans and a cup holder in the bathroom were 3d printed.  I thought they looked good which is why I'm thinking about selling stuff like this.  I'm under no illusion I would be creating any big competitive business but it would be nice to have something for the local farmers market or some school fundraisers, things like that.

To view this discussion visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/72e49b18-9ed7-45f6-9355-c1c2ffaae96dn%40googlegroups.com.

[3D Printing Tips and Tricks]

Sophie

You stated that you used vase mode…

There’s no seams in vase mode. The toolpath spirals upward with no end/start points.

Regardless… Looking at your pics… it looks a bit thick for vase mode.

[Sophie]

Ah fair enough Bryan, you never know what people do and don't know, so no offence meant. And I'd be keen to see your handy work should you get the design cap on, not to copy, just because I enjoy seeing.

You know it might be worth trialling vase mode with a .6mm nozzle, just to see! You might be surprised at the strength especially if you adjust flow rate and line width!

And Andy, yep there are "hardly any" seams, not "no seams". The first couple of layers have seams, and then when it goes to print the wall, that's when it gets to seamless.

And yes! It does look thick! On purpose! I adjusted the line width to 0.8mm! This allows the plastic to smoosh (add technical term here, but I'm using smoosh for now) together increasing the water tightness probability. 

Maybe I should do one at .4 width to show the difference. 

I forgot to mention, layer height was set to 0.2mm, in case anyone was wondering...

[Sophie]

Hopefully you can see the difference between .4 and .8 line width.

Surprisingly, the 0.4 line width version (which by the way also had just 1.0 filament flow for initial layer), was still watertight! (at this height of around 1.5")

But you can see from the gif just how flimsy and flexible it is, if I printed the whole thing you'd only have to pick it up in the middle and I imagine it would flex and split just like that! Not ideal. 

But given how much stiffer the 0.8 line width version feels just with a 0.4mm nozzle, yea I reckon it would be great on a 0.6 or 0.8mm nozzle with a double nozzle size line width. 

[Sophie]

[Kurt]

Sophie - I concur with Andy - Vase Model printing means a single continuous line - and thus should have no seams. Looking at your pics - I could not discern the seams.

Will admit - I'm also surprised it holds water and does NOT leak at all. My suggestion - let it sit overnight with water in it. It really is common that 3D Printed items have microscopic holes that do eventually leak - and I would think this would be more the case with Vase model - since it's only a single wall shelled object...

-K

[Kurt]

Darn - my bad Sophie - I didn't read ALL of your prior reply - as I now DID indeed see you replied back to Andy's comments. Also, interesting you are using 0.8mm layer width - yet only a 0.2mm layer height. Since, with a bigger nozzle - I figured you'd print faster with thicker layer heights. But, then I suspect you can't go as wide as 0.8 layer line width if you increase layer height too much more. 

-K

[Bryan Murphy]

Ah that squish!  Definitely not the functionality I'm looking for right now. 

I decided I'm going to design my own vase that takes the line into consideration as part of the model.  I can start with something simple, prove it works, and then add flare later.  That will get me past this flaw and I can instead focus on waterproofing and optimizing a final print.

--
You received this message because you are subscribed to the Google Groups "3D Printing Tips and Tricks" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 3d-printing-tips--...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/4bb7bc07-abcc-41ad-a62f-6fa4e583ab46n%40googlegroups.com.