First try with Slic3r - what settings to use for Orca 0.2

Hi,

I have an Orca 0.2 (V9 extr. 1.75mm fil & .35 mm nozzle); until now I am using the skeinforge with the standard settings (slow-profile) proposed by Camiel from mendel-parts.
What are the ideal settings to use with slic3r for my above Orca ?
I was planning to use the following : (default + changes for 1.75 mm & .35 nozzle - but I am unsure about the layer-height to use ?)

thx,

Chris


--- config.ini ---
bottom_layer_speed_ratio = 0.3
bridge_flow_ratio = 1
bridge_speed = 60
duplicate_distance = 6
duplicate_x = 1
duplicate_y = 1
end_gcode = M104 S0 ; turn off temperature\nG28 X0 ; home X axis\nM84 ; disable motors
extrusion_axis = E
extrusion_multiplier = 1
extrusion_width_ratio = 0
filament_diameter = 1.75
fill_angle = 45
fill_density = 0.4
fill_pattern = rectilinear
first_layer_height_ratio = 1
g0 = 0
gcode_arcs = 0
gcode_comments = 0
infill_every_layers = 1
infill_speed = 60
layer_height = 0.25
nozzle_diameter = 0.35
perimeter_speed = 30
perimeters = 3
print_center = 90,90
retract_before_travel = 2
retract_length = 1
retract_lift = 0
retract_restart_extra = 0
retract_speed = 30
rotate = 0
scale = 1
skirt_distance = 6
skirt_height = 1
skirts = 1
small_perimeter_speed = 30
solid_fill_pattern = rectilinear
solid_infill_speed = 60
solid_layers = 3
start_gcode = G28 ; home all axes
temperature = 255
travel_speed = 130
use_relative_e_distances = 0
z_offset = 0

Start with the defaults and only change the ones marked in bold. Then you may change the others, one at time and only if you understand what they do and what do you want to achieve. Choose the layer height that you want to print with. For a 0.35 nozzle you may want to print at 0.10 or 0.20.

ok, I'll give it a go - I'll start with layer-height of .20 - I thought that having a nozzle of .35 would require a layer-height of .35 ? How is it possible that it can be less ? I assume that the extruded filament comes out round and probable flattens a little bit, but 30% to 60% seems extreme ?

thx

Cvanaalst Wrote:
-------------------------------------------------------
> I thought that having a
> nozzle of .35 would require a layer-height of .35

If you did that the layers would not stick together, each filament would attempt to sit on top of the other and not be bonded together, they layers need to squash-down so you have a high surface area of the last layer to bond with the next. With a 0.35 nozzle the highest layer height you should use is about 0.25, you should really be using 0.2 and also try out 0.1

My tutorial on Slic3r may be of some interest to you - Here

let us know how you get on.

Rich.

---

[richrap.blogspot.com]

Well actually I print 0.35 layers with a 0.35 nozzle. The nozzle size is irrelevant. The important value is the size of the filament extruded into free air. The layer height must be less than or equal to 0.8 times that.

---

[www.hydraraptor.blogspot.com]

And for a 0.5mm nozzle, what is the maximum layer height?

---

Frédéric

Is this really true? I use almost exclusively 0.35 layer height with my 0.35mm arcol nozzle, and the layers bond very well, making the parts very strong.

Maybe I should recalibrate my extruder...

---

--
-Nudel
Blog with RepRap Comic

nophead Wrote:
-------------------------------------------------------
> Well actually I print 0.35 layers with a 0.35
> nozzle. The nozzle size is irrelevant. The
> important value is the size of the filament
> extruded into free air. The layer height must be
> less than or equal to 0.8 times that.

Thanks, I must try that, I have always printed with layers less than my nozzle size, but I know that my 0.5mm nozzles give me 0.54mm filament into free air (with PLA), I have not measured the free air filament from my 0.35mm nozzles.

Do you see extrusion temperature altering the size of free air extrusion? And any difference with types of plastic ABS/PLA and brands/mixes?

---

[richrap.blogspot.com]

I am currently printing a hollow box created via Slic3r with layer-height 0.2 and although it's finished yet, it looks very good - MUCH better than the versions I've printed with Skeinforge - I am impressed !
I'll test a few other layer-heights later on !

Almost every single parameter affects the free air extruded filament diameter, unfortunately.

---

www.Fablicator.com

The die swell seems to depend on the plastic viscosity and the pressure: -

ABS swells more than PLA.
Faster extrusion makes it swell more.
Lower temperature leads to more pressure and it swells more.
Smaller nozzles give more swell as a percentage.

Yes different makes and even different colours have different die swell. I found lapis blue from Fabberdashery had massive die swell compared to firetruck red.

---

[www.hydraraptor.blogspot.com]

Nudel Wrote:
-------------------------------------------------------
> Is this really true? I use almost exclusively 0.35
> layer height with my 0.35mm arcol nozzle, and the
> layers bond very well, making the parts very
> strong.
>
> Maybe I should recalibrate my extruder...

I'm going to experiment with my free-air extrusion sizes as per Nophead's comments. I have been doing a lot of very low layer printing over the last week and looking at the resultant extruded walls under a microscope, I'm now very interested in seeing what a 0.35mm nozzle and 0.35mm layer looks like under the microscope.

Mind you I'm starting to wonder again if Slic3r needs the nozzle size entered or the free-air extrusion size. I expect everyone is entering the nozzle size. it's trying to calculate WOT with the wrong values...

Maybe this is why Sound keeps getting reports of too much and too little infill

---

[richrap.blogspot.com]

Neither the nozzle size or the die swell affects the volumetric calculation, it just sets the limits of the values you can use for the height and the width of the filament.

When the filament swells as it comes out it also gets shorter, so the volume is the same.

Edited 1 time(s). Last edit at 01/06/2012 03:29PM by nophead.

---

[www.hydraraptor.blogspot.com]

I find I have to tweak each colour and type of filament by tweaking the diameter if I want to get the correct infill.

With green PLA from supply3DPLA I have to set the diameter about 0.2mm above the actual measured diameter.

With pink PLA from diamond age I have to set the diameter about 0.1mm below.

Both filaments measure ~2.75mm, all other settings in slic3r are exactly the same, so I guess it has to be from swelling?

---

--
-Nudel
Blog with RepRap Comic

I print everything with exactly the same settings, and with almost exactly the same results (as far as infill goes, anyway)

You probably have harder vs softer filament and are using a spring tensioner for the extruder? The harder filament will sit further out on the pinch-wheel so more will be extruded.

As has been said before, the swell does not affect total volume, only how much the filament needs to be stretched to span gaps effectively.

---

www.Fablicator.com

Thanks Andrew, I run with the springs under very high tension, so that makes sense.

I should be able to measure the difference if I calibrate each filament then. I must try to do that someday.

---

--
-Nudel
Blog with RepRap Comic

nophead Wrote:
-------------------------------------------------------
> The die swell seems to depend on the plastic
> viscosity and the pressure: -
>
> ABS swells more than PLA.

> Faster extrusion makes it swell more.
> Lower temperature leads to more pressure and it
> swells more.

We are well off-topic but that all poses quite a number of questions and worries about the way in which we print.
What speed of extrusion do you then decide to use to measure free-air filament size? The perimeter speed or infill speed?

> Smaller nozzles give more swell as a percentage.

That makes sense, so we must have an absolute minimum size we could use in 3D printing then.

>
> Yes different makes and even different colours
> have different die swell. I found lapis blue from
> Fabberdashery had massive die swell compared to
> firetruck red.
I wonder why they were so different, I don't think they use many (any?) other additives other than the colour pigments.

So unless we print all parts (outline, infill, inner perimeters) of the object at a constant same speed then the optimal (compatible) layer height is going to change depending on the speed? how much of a problem is this likely to be? I often print inner shells at 120mm/sec+ and then the perimeters at 60mm/sec
and for some hollow objects infill at 100+ and sometimes the perimeters very slow at 35mm/sec

I have noticed that I'm getting better results if I settle on more similar constant speeds when experimenting at very low layer heights.

But I'm doomed - I also change colours constantly and have different temperatures for colours and makes and different temperatures for different speeds and I know some of the PLA I use has other plastic's mixed in it with it (PE).

In a hobby full of variables maybe only individual experiments can give best constants... So I think maybe I'm still safer sticking to layers lower than my nozzle size with all the other variables I have going on.

---

[richrap.blogspot.com]

richrap Wrote:
-------------------------------------------------------

> What speed of extrusion do you then decide to use
> to measure free-air filament size? The perimeter
> speed or infill speed?

Perimeter, as it is slower, giving less die swell, and die swell sets the maximum filament cross section.

>
> > Smaller nozzles give more swell as a
> percentage.
>
> That makes sense, so we must have an absolute
> minimum size we could use in 3D printing then.

Yes smaller nozzles give diminishing returns so that would be a logical conclusion.

> I wonder why they were so different, I don't think
> they use many (any?) other additives other than
> the colour pigments.

It surprised me as well. I had to up the temperature to get the die swell down but that was too hot for the red as became liquid.

>
> So unless we print all parts (outline, infill,
> inner perimeters) of the object at a constant same
> speed then the optimal (compatible) layer height
> is going to change depending on the speed? how
> much of a problem is this likely to be?

It only sets the maximum so you just have to measure it at the slowest speed. The reason you can't extrude bigger than the die swell is that you have to compress it in length rather than stretching it. That makes it squirm and sag when doing overhangs and infill.

> I often
> print inner shells at 120mm/sec+ and then the
> perimeters at 60mm/sec
> and for some hollow objects infill at 100+ and
> sometimes the perimeters very slow at 35mm/sec

I stick to much lower speeds and consistent speeds and mainly use natural plastic, so it isn't a big problem for me.

>
> I have noticed that I'm getting better results if
> I settle on more similar constant speeds when
> experimenting at very low layer heights.
>
> But I'm doomed - I also change colours constantly
> and have different temperatures for colours and
> makes and different temperatures for different
> speeds and I know some of the PLA I use has other
> plastic's mixed in it with it (PE).

That would be an odd mix.

---

[www.hydraraptor.blogspot.com]

We're looking into this too much in my opinion.

As long as your layer thickness is 75% or less of your nozzle diameter you are practically guaranteed success.

The swell only needs to be figured out if you are really concerned with making a print with huge layer thicknesses (at which point you may as well have a bigger nozzle)

---

www.Fablicator.com

The 0.8 x the die swell value gives the layer height that is most productive for a particular nozzle and also the least stretch. As you go smaller you get more corner cutting and so more hole shrinkage. So in some terms it is the magic optimum value.

You can use lower layer heights if course but you can't go too much less in width or the filament will snap.

When using PLA there is not much difference between nozzle diameter and the die swollen size, but for ABS and HDPE it is significant.

Edited 3 time(s). Last edit at 01/06/2012 06:38PM by nophead.

---

[www.hydraraptor.blogspot.com]

My filament in air with my 0.5 nozzle is 0.7. 0.8*0.7 = 0.56 optimal layer height if I am understanding what you're saying. I'm not even sure my printer is capable of printing anything other than a mess of noodles at that large a layer height.

The important thing is the cross sectional area of your filament must be less than a 0.7mm circle, so it depends on W/T.
I use the maths here: [hydraraptor.blogspot.com] which actually results in 0.76 x rather than 0.8 x. That is for W/T = 1.5, if you use a bigger width then your maximum layer height is less.

I am not sure what maths Skienforge, SFACT and Slic3r use. A little bit of stretch is good for bridging so I generally use 0.48mm with a 0.5mm nozzle. The die swell is about 0.7mm and I can make objects with 0.5mm layers, but the bridging is not good. Also the nozzle tends to get smaller over time until I ream it so there has to be some margin.

So it depends on the maths the slicers use, but the bottom line is the cross sectional area of the filament can't be more than the die swell area as the filament will start squirming. In practice it should be a little less to pull tight over bridges and sparse infill.

---

[www.hydraraptor.blogspot.com]