Really tiny little details with 0.4 nozzle (S3D, UMO+)

[neotko 1,417]

I just found out, after searching on google and only found one post about this, I don't think people use this, but it's really interesting.

Check the photo. A photo of something I'm printing now. The 'square' think was printed at standard 0.4 and the round one it's also on a 0.4 nozzle but making Simplify3D print it like a 0.26 nozzle.

This it's a over-specs test, since S3D says that 1.2 ratio it's the maximum recommended. But this rocks, I'm able to print (sorry can't show the full image) texts that are WAY to small for a standard 0.4 nozzle to print by default.

[rigs 735]

nice

 

only found one post about this

Link ?

thank

[neotko 1,417]

nice

 

only found one post about this

Link ?

thank

 

This was what made me check it further:

"You can use a lower layer height or larger extrusion width if you wish, it will work fine. The slicing software automatically calculates the appropriate volume to extrude based on the settings you choose. There is no hard lower limit on layer height - it is limited by your ability to keep flow consistent at very low flowrates. Some reprappers have printed layers as small as 5 micron - 0.005mm!"

At: http://reprap.org/wiki/Triffid_Hunter%27s_Calibration_Guide#Slicer_settings

I suppose the gear of the umo/umo+ allows the very little constant extrusions. Would be nice if someone can check this on the um2 direct (I mean gear-less) extruder.

Edited July 29, 2015 by Guest

[rigs 735]

Community confirmed this quote ?

 

Best results are obtained when layer height < 80% of nozzle diameter, and extrusion width >= nozzle diameter.

I'm curious to have the advice of others...

[luisito 12]

This feature is commonly used to set the width account for extruded material being wider than the nozzle itself, wich is common for most printer/filament setups. Setting the filament width lower than the nozzle itself would tight the steps between lines, which would produce the effect you are showing in this picture. I would expect to have some overextruding with this setting but it seems it is not the case. What i cannot understand is why this setting can add any detail to the part. Do you have any more detailed picture ?

[neotko 1,417]

I do, but can't show it's a production test. I'll try to download some generic calibration file and print it with one and the other setting.

Anyhow 0.26 it's overdoing it, the text shows but it's not 'good quality'. But the advantage somehow it's that the slicer process parts of the design that otherwise would show as 'braille' if you get what I mean. Also I just wanted to 'try' the effect. The overextruding does not show, it's true that on one line texts the nozzle can't do magic, but it's really close doing thinner lines, probably a 0.32 it's perfectly doable. I'll try to get some time to make a test but I posted this just to show the option to others, I think one of the worse thinks about simplify3d is the huge lack of documentation about every option.

A side note. That img it's with cheap'o'pla orange (it sucks so bad, but it's nice to burn prototypes). The diameter changes on every meter, it's mad.

Edited July 29, 2015 by Guest

[neotko 1,417]

Ok the print finished (It was a 3h print on the side).

My settings where:

0.22 layer, 2 perimeter shells 70mm/s. Orange-crappy-pla from bq (tinny little bit better than other bad pla's)

The little 'cactus' measures 7.7mm by 10mm

Here's the shot of the slicer path

Ofc it can't do magic, but not bad at all. And that was a 70mm/s

I did check one of the letters and one was actually 0.59 and that according to the slicer it's a 2 pass line, so it was able to do 2 lines of at least 0.28

[patrick181 0]

Thanks for posting this considering there is no trail demo of Simplify3D to justify a $150 gamble. ;-(

[neotko 1,417]

Before you say that this images show 'underextrusion'. Yes they do. The filament changes from 1.72 to 1.84 on a meter and it sucks but for this test that's what I had in hand.

First image setting at 0.39 second 0.30. Speed 50mm/s 0.22 layers. Fans 94% on at layer 8. 3 top 3 bottom layers and 2 perimeters. To make a propper full test this needs a reliable filament but be my guest since I don't have much time this coming weeks. Oh also 10% infill for thinwalls option was on. That off should clean a lot the details. I suppose for this configuration a 5% would suffice.

Printed this file

https://www.youmagine.com/designs/test-your-3d-printer

Photo 1 0.39

Photo 2 0.30

My own conclusion it's that for big thinks it can make thinks a bit worse but for me it's perfect to make a separated process for the parts that need finner details and leave the big blocks as they are. Good think it's that the tinn walls are really thin.

Edited July 29, 2015 by Guest

[neotko 1,417]

Well I bough simplify3d 10months ago when I had a horrible xyzdavinci. Cura it's nice. But right now as today s3d it's just way better. Supports selectable a great range of options for dual extruder tunning, this option I didn't even know about that could be use for this and specially since 3.0 the best user interface on any slicer. The down side, it needs a stl fixed properly.

But with 3.0 you can group stls move them as one, duplicate fast and specially the posibility of making areas or objects at more detail, fast, slower. Not just only at a Z but the full stl or part of it. It take time to use all the settings but when you prepare a job, keep it saved (it saves the stls, position, process settings).

Oh and my favorite, clean surfaces. Retract when crossing empty areas works superb to have a clean surface or bottom. The photos up here where done with that option off btw so it shows strings. But I was going for speed over quality.

Edited July 29, 2015 by Guest

[rigs 735]

You want to see the 3d printing = CURA

You want to understand the 3D printing = S3D

[IRobertI 520]

I included it in one of my guides a while back in a way. In that case I used it to create "better" fill in some cases:

http://support.3dverkstan.se/article/30-getting-better-prints#solid-layer-fill

Edited July 30, 2015 by Guest

[neotko 1,417]

This should have been a plug to calculate the extrusion width since your post. I remember that from your guide but I remember when I saw it I only though that getting the trouble of matching the flow/nozzle to be perfect could be a bit of a problem. Someone could get on board a make a ¿simple? plugging adjustment for cura. I'm sure for some stuff this could be helpfull. Well at least it it's for me now since I'm printing lots of texts and some fonts are a pain to print on 3D.

Edit: One think, I would really love to see this test on an UM2. When I saw my gear moving soo slowly on a 0.22 layer-50mm/s I really was thinking.. "If this it's how slowly it's moving to print a fake 0.30, and this it's a 1:6 gear, how hyperslow will move the UM2 gear-less feeder?". Just curiosity because I extrude more than UM2 users since I use 1.75 filament, to move a bigger 2.85 it would need even less movement. I mean for real, if you watch the gear rotate it was moving slooooww. (I know the um2 has 0.9 400steps, but that's just the double of the umo motor and umo has 1:6 gear that um2 compensate in precision and specially torque but can't do real revolutions as a 1:6... i suppose fakemicrosteps fix that?)... Just thinking out loud.

Edited July 30, 2015 by Guest

[neotko 1,417]

This morning I just had an idea, I never thought before (silly me XD)

Using this method of making the nozzle smaller, I just though that since S3D let you define the extruder, even if they are for the same tool, you could make infill/supports/rafts etc, thinner or bigger. As you can see on the printed photos, the area where the nozzle width change seems to make more 'effect' it's on the perimeters. So using this idea you can effectible make the outside of the object look more detailed while keeping the infill with a normal (or bigger ofc) size.

This it's how the slicer shows what would do. Note that I exaggerated the effect of the outside lines changing it to 0.10

[neotko 1,417]

Loaded some Faberdashery and printed a robot at 0.07. Infill 0.40 Perimeters 0.28. 40mm/s 192C

I think a more realistic maximum might be 0.32-0.35 but still, it does print the stuff. (Don't mind the overhangs, my head fan design it's still far from finished).

[rigs 735]

Hey neotko

This is another approach

Do you think it is possible to do that with S3d ?

[media=3959][/media]

Sorry for french language...

nozzle : 0.6 mm

1 shell : 0.1 mm

2 shell : 0,2 mm

infill : 0,4 mm

Crazy or serious ?

Edited July 31, 2015 by Guest

[xisle 34]

Crazy or serious ?

 

Looks interesting and probably viable with vertical walls, but curved surfaces on the Z axis would be a problem with that method.

[neotko 1,417]

The way the line on the img extrudes it I don't think that s3d can do that. It could do 2 perimeters width but you would need to make a separed object with the edges expanded that size empty and another process with the object with one perimeter size and other infill width. A lot of work since s3d doesn't have a setting for inner perimeters. You can change with this method the widths of infill/perimeters/supports/primtowers/oozeshields but the way it extrudes it it's the clasic first one think then the next (as far as I know).

And that reminds me that I need to test big infill supports with the width small and less extrusion flow. It might make them better (or worse).

Edited July 31, 2015 by Guest

[rigs 735]

Ok, Thank for your answers.

[frederiekpascal 193]

hey @neotko, I got an UM2 and a well configured S3D (thanks to you!). Which tests exactly do you want me to do?