Why does S3D print the tiniest paths first? Always lifts

[swbluto]

It matters because the filament on the bed is still attached to the filament in the extruder and so it provides a "peeling force" as the extruder moves away, and if the mass/collective-adhesion of the filament on the bed isn't enough to counter this "peeling force", it tends to lift. If the extruder had some way to cleanly "cut" the path away from the extruder's filament, this wouldn't be a problem.

Anyhoo, successfully modified (By app) the code to print the outer first and then inner, and calc'd the path lengths on the inners and sorted them. Just need to produce the output for inners based on the sorted path lengths (Easier said then done, lol.).

[jimc]

yes exactly. its not stuck to the bed well enough.

[swbluto]

yes exactly. its not stuck to the bed well enough.

Because it's not laying down enough to begin with as the path is so short. So we agree.

Every other path is adhering perfectly well, no visible signs of "not adhering well enough".

[jimc]

no, it shouldnt matter the amount. the amt of plastic doesnt make the adhesion better. how its laid down and the bed prep makes the adhesion. your printer should be able to move all over and put tiny little dots all over and if the bed is prepped right and the z gap is calibrated then they should all stick.

[swbluto]

Adhesion = surface-area

Less filament = less surface area = less adhesion

If("extruder pulling force">adhesion) lift.

See the issue? The amount of filament/surface area matters A LOT.

No amount of tweaking is going to get you a magical infinite adhesiveness.

Anyway, happy to report I've reversed the outer and inner and have the inner paths sorted by length, longest paths first. Output passes automatic and manual review. Let's see if it prints!

[swbluto]

It's alive! Oh, such a good feeling see your own code in action and working perfectly as intended. That said, I can't help but feel that "gap fill" is a waste on the bottom layer. It seems to lay down negligible filament, and it's constant traveling from point to point only seems to create pocks in the laid down filament. It doesn't affect the layer too functionally much (I'm sure the finished product will be fine), it just doesn't look nearly as perfect as it does before gap fill starts.

Anyway, here's some C# code that shows the bottom layers structure for those curious. (Original has inner before outer)

outcon.AddRange(L1_skirt);
outcon.AddRange(L1_outerper);
outcon.AddRange(L1_innerper);
outcon.AddRange(L1_gapfill);
outcon.AddRange(L1_solidlayer);

EDIT:

Just got done printing! The bottom layer looks PERFECT! Like, holy crap, way better than the original monstrosity of a bottom layer. Also, changed it to sort all the outer and inner perimeters by path length; prints them ALL from longest to shortest, regardless if they're "outer" or "inner". Noticed some prints didn't have a clearly well-defined "inner" section, so this was the solution. Just group all the paths together and sort.

[ednisley]

the filament on the bed is still attached to the filament in the extruder and so it provides a "peeling force" as the extruder moves away

That's what retraction is all about. After the nozzle finishes laying down a trace (or a dot), retraction breaks the connection between the nozzle and the plastic on the platform. When the nozzle moves away, there's barely a hair connecting it to the platform.

Works for me, anyhow:
http://softsolder.com/2015/04/02/miniat ... hain-mail/



The bottom bars of each link have two concentric loops, with the inner ones printed first in no particular order across the platform. A couple hundred of those bars came off without a hitch in PETG.

A few thousand more, large enough for a pair of infill dots in each bar that also stuck with no problem, done in PLA:
http://softsolder.com/2014/12/05/3d-pri ... -mm-sheet/



So I think the real problems comes down to getting the platform alignment, adhesion, and retraction right, rather than twiddling the printing order.

On the other paw, anything that makes 3D printing less fussy counts as a Good Thing!

[swbluto]

I should probably add I'm printing ABS on bare glass. It sticks well enough, but not nearly as sticky as PLA, and definitely not nearly as well as duct tape.

I do it for the shine / lack of clean up, but it's a 'lil tricky to get right.

[Matt_Sharkey]

I should probably add I'm printing ABS on bare glass. It sticks well enough, but not nearly as sticky as PLA, and definitely not nearly as well as duct tape.

I do it for the shine / lack of clean up, but it's a 'lil tricky to get right.

well there's your problem. Use hair spray.

I print on bare glass too, but i print with PETG, which has no problem with bare glass in my experience.

[swbluto]

I should probably add I'm printing ABS on bare glass. It sticks well enough, but not nearly as sticky as PLA, and definitely not nearly as well as duct tape.

I do it for the shine / lack of clean up, but it's a 'lil tricky to get right.

well there's your problem. Use hair spray.

I print on bare glass too, but i print with PETG, which has no problem with bare glass in my experience.

Hairspray doesn't work in my experience. Then again, I'm using a pump-spray version. It has all the plasticizers as the aerosol version does, but maybe the propellants are important, too?