weird side effect from retraction

Hi guys,
yesterday I started to print an ice sculpture from transparent PETG. I wanted it to be as translucent as possible, so I sliced it in spiral/vase mode with single wall width ( 0.5mm )
It turned out pretty nice, but as soon as retraction was used, the print showed lots of tiny air bubbles. ( air got sucked in the nozzle? )
Retraction was set to 4.5mm for an e3d clone with 500mm Bowden tube. #see "angel_R45"

I started to play around with different Retraction and Z-lift setting. A delta can make use of a much bigger z-lift than Prusas or other Cartesian printers.
See pictures named angel_Rx-Zx to evaluate the outcome.

It seems, that less retraction eliminates the bubbles and more z-lift ( 6mm! ) keeps the stringing fuzzy and thin. ( easy to clean up )
Now only the tips of the wings are a bit ugly. ( very slow print speed and tiny moves )
-Olaf

Edited 1 time(s). Last edit at 09/13/2015 04:44AM by o_lampe.

I have noticed I get a lot of blobbing if I don't use retraction when printing PETG (leaky e3d v5 clone), but as you say it introduces small bubbles in the print.

how much retraction did you end up with?

I was perhaps thinking that some extra start-length after retraction would help as well.

I noticed most retraction prblems disappear if the retraction speed is lowered.
Finding the sweet spot between a clean line and a little blob is time consuming though.

I ended up at 2mm retraction at 80mm/s and 3000mm/s^2 acceleration.
I haven´t touched retraction speed yet, but I´m willing to try. The best string free result was at 8mm retraction, but with lots of bubbles and some tracks seemed to be ripped away by retraction.

Maybe I start from there and reduce speed, it´s a rainy day anyways
-Olaf

PS: all files were sliced with Cura and it annoyed me, that the z-lift and retraction are not processed at the same time. Also moving to the new location and re-feeding the filament could be done simultaneously. Does Slic3r have such an option?

Edited 1 time(s). Last edit at 09/13/2015 11:08AM by o_lampe.

For a bowden try this to find the right retraction:
Extrude a bit of filament and mark it somewhere above the extruder.
Release the motor or pressure plate.
Measure how much the distance from the extruder to the mark has increased - this is your minimum retraction.
From there you can play 2mm up or down to find the sweet spot.

What you want with the retraction is stop filament from flowing through the nozzle.
The bowden tube acts like a spring, so you need more retraction.
Also the speed for the retraction should always be a bit slower than what you can extrude without problems.
Otherwise all you do is eating filament...

Quote
Downunder35m
Also the speed for the retraction should always be a bit slower than what you can extrude without problems.
Otherwise all you do is eating filament...

I don't quite follow you here.
I usually retract with 70-80mm/s at an acceleration of 3600mm/s²

That is much faster than I can extrude, if I try to extrude that fast it starts skipping steps.

I could probably go way higher in that speed setting if only I feed the motor more current, without eating any filament.
However I personally don't want to feed the motor any more current since I have a geared extruder and I want it to start skipping steps instead of grinding into the filament in the event of a jam or the occasional overextrusion.

Edited 1 time(s). Last edit at 09/14/2015 12:41AM by Koenig.

I agree with Koenig,
the spring effect of the Bowden tube helps to retract with high speed.
I made some tests with slow retraction and it turned out, I get same good results with only 2mm retraction at 40mm/s when I set z-lift to zero. ( Had been 0.2mm at least before )
Seems the partypooper was the z-lift, although I had a hiccup on one of the wings caused by zero z-lift...

You can´t have it all
-Olaf

I've been playing around some earlier trying to get that smooth surface of PETG, and I've also found that lowering the temp or aiming for slight overextrusion helps a little.
I guess both ways actually does the same, raise the pressure in the nozzle.

But this certainly makes a huge difference.

Good find I'll have to say, and thanks for sharing.

This with 0 Z-lift tends to give other problems though.

The fine stringing that always is there with PETG tends to accumulate on the nozzle to the point that it leaves a small blob somewhere.
With Z-lift enabled this mostly happend so it would leave the blob right outside a perimeter so that the nozzle on the next pass just gently pushed it aside,
but with no lift it leaves the blob right in the path, which then knocks the part loose or knocks the effector loose (magnetic) on the next pass.

I'm printing parts for my new E3D Lite6 now wich I fill fit with a 0.35mm nozzle, hoping that the slightly smaller nozzle will create the higher pressure required the bubble free PETG printing.

The prints I´ve shown were made with a 0.5mm nozzle. Their results with PLA were already bad, but with PETG it´s a nightmare.
I guess with a 0.4mm nozzle, the problems wouldn't occur and I wouldn´t have learned the lesson.

I was thinking about adding obsolete parts to the design. Close enough to the wings, to make the nozzle jump to it without retraction. The obsolete parts would gather the blobs and strings, but the original design would stay clean. If only the slicer would be able to slice it the right way...
-Olaf
Such a slicer option exists for dual extrusion, called "Abstreichturm" in german Cura