Feature request: "double wall vase"

[358Eki]

I have been using VASE-feature some different purposes and I like the simplicity (and lack of "not extruding movenents" and retractions. Quite some practical cases a single extrusion widht vase wall is too weak or flex. With transparent plastics the separate infill between two walls also affects optical quality of vase-type-printing.

Would it be possible to add two wall thickness to vase print as an option: 1.bottom layer height (solid or with infill) 2.vase feature start after selectable bottom layers height 3.optional second vase wall -build change is done after first vase wall build round by moving one extrusion width to inward (for shorter perimeter?) and lowering z with one build turn layer height and spiraling againg sorhter perimeter. After shorter spiral it would come to same level and can continue with the first wall.

I can imagine that this would make vase wall tougher and also increase light reflections inside and between transparent and translucent plastics (despite the place of wall changes will make a visible seam there). I would make simple to design and faster to generate useful stronger boxes and vessels (without need to think about nozzle width, layer height and extrusion widht to accomplish 2 perimeter width wall without infill AND with perfect perimeter adhesion).

[Lodran]

If you're going to go as far as double wall, you might as abstract the problem completely, and allow any even number of walls.

[qharley]

I think this defeats the purpose of vase mode (spiral printing to create no seams)

You can get the same effect by just printing with 0 infill, 5 solid layers at the bottom, on 0 solid layers at the top, You can the set the thickness of the wall by the amount of perimeters.

On Fri, Feb 21, 2014 at 1:27 AM, Lodran notifications@github.com wrote:

If you're going to go as far as double wall, you might as abstract the problem completely, and allow any even number of walls.

Reply to this email directly or view it on GitHubhttps://github.com/alexrj/Slic3r/issues/1797#issuecomment-35682324 .

[358Eki]

With "double (tripple, quadrupple,...) vase wall" there is no real seam (stopping current extrusion and starting a new one) - rather a crossing of perimeters. It will be likely more ridig than a "real seam". The crossing is also hidden between the walls so there is likely no blobs or other artifacts as there is with stopping and starting of extrusion with unperfect machinery and conditions.
It would be also more productive way with extuding plastic than not extruding plastic. That is the main thing what this is all about.

"Multiple wall vase mode" is closest description I can now think about what I´m proposing. In vase mode there is no predictable information of wall thickness other than "one perimeter width" that is depended individual machine and its settings.

With multiple wall vase I see practical use between 1-1.5-2mm wall thicknessess. With 0.5mm nozzle this is something over one perimeter and less than four. Just the area where I have got a lot of troubles and time used with settings and printing with not so perfect machines and not so perfect models.

Many times I have been in situation where raft, support material, brim and infill is just perfect but part itself is awful because dealing with unperfect model (some variation in dimensions between two surfaces.) I agree that its close with a "normal way" with several perimeters - but there I have not seen or found a way to get rid of another wall surface.

If I make a comparsion for other manufacturing technology this would be like using a one side open molds where one surface is all that counts and other one is less important and can tolerate manufacturing inaccuracy and high tolerance with wall thickness... I can not make that decision with .stl models though. I need to know exact layer height and extrusion widht while modeling to make it work.

For example someone has modeled a perfect part but with a wall thickness in model does not suit you for some reason. You want to get rid of another surface or predefined measure between the surfaces - other possibility is to go and start to remodel the .stl and that is not nice - also where to get right measures for the model every time...

[nophead]

@358Eki "-build change is done after first vase wall build round by moving one extrusion width to inward (for shorter perimeter?) and lowering z with one build turn layer height and spiraling againg sorhter perimeter. After shorter spiral it would come to same level and can continue with the first wall."

How can you lower z one layer height while being adjacent to the outer wall. The side of your nozzle will collide with it.

[qharley]

Nophead is right. The nozzle of the hotend is a tiny hole in a flat piece of metal. It will iron down the first spiral in the second movement.

To hide seams using the no-fill method, you could randomize layer starting points.

Another way to increase vase mode strength is to use a "larger" nozzle size. You can trick slic3r into extruding more plastic by specifying a larger nozzle size, and making the the layer hight half of the increased value. Note that this trick is probably only useful in vase mode... It might cause mess during normal printing.

[qharley]

Which makes me wonder... Is there a way to increase the extruded width without being nannied into a width smaller or equal to the nozzle width?

[nophead]

You can go bigger then the nozzle width as long as the cross sectional area is less than the die swell area so that it is not being compressed lengthways.

On 21 February 2014 20:09, Quentin Harley notifications@github.com wrote:

Which makes me wonder... Is there a way to increase the extruded width without being nannied into a width smaller or equal to the nozzle width?

Reply to this email directly or view it on GitHubhttps://github.com/alexrj/Slic3r/issues/1797#issuecomment-35768348 .

[qharley]

The speed of the head during extrusion should take care of that right? As long as the extruded with is not larger than the diameter of the the flat tip of the nozzle of course

[358Eki]

If you model a cylinder tube let say 50/30x20 hight and slice it using VASE mode it makes bottom as specified. It seems to make also both walls of tube: the in and out without infill. There is real seams coming because alternating print between the walls. For me it seems that walls could be in fact spirals but turning in opposite directions AND spiral is cut for the change of wall to built. I can imagine that if I can make a just right tube wall thickness (extrusion widht x2) it might do the thing. Its hard to do because I do not know exact extrusion width or settings when modeling.

Here is the parts and picture of my "simulation" of double wall VASE http://www.thingiverse.com/thing:255151/#files Also a screen shot from Gcode viewer there and two videos about regular perimeter print with retractions and same part with vase-feature. Clearly there is no pauses in extrusion eighter fast nonextruding movements. Wall is tuff enough for many uses and seam is quite nice without artefacts. Also dimensions are fine. (I have another blobs from USB though). Print is fast without any retractions and unneeded and unwanted movements. Most simple and valuable feature of this if I don´t have to edit the design file like to make it hollow...

[nophead]

@qharley, The die swell area is a fundamental limit on extrusion area,regardless of speed (other than the fact higher flow rate gives bigger die swell). If you try to extrude a cross section bigger it is like trying to compress elastic. It will squirm and make a wiggly line.

[whosawhatsis]

@qharley You can specify a larger extrusion width in print settings > advanced. What Slic3r won't let you do is specify a layer height greater than your nozzle diameter. I've never had Slic3r complain about extrusion widths significantly greater than the nozzle diameter.

[qharley]

@whosawhatsis I will give that another try then. I remember being annoyed at setting a wide perimeter width, and being ignored... That was some time ago though

@nophead you see this a lot at the lower layers of a print if the z offset is not set correctly. Nice description of why it happens...

[358Eki]

nophead commented 8 days ago

@358Eki "-build change is done after first vase wall build round by moving one extrusion width to inward (for shorter perimeter?) and lowering z with one build turn layer height and spiraling againg sorhter perimeter. After shorter spiral it would come to same level and can continue with the first wall." How can you lower z one layer height while being adjacent to the outer wall. The side of your nozzle will collide with it.

So you mean that spiral build would be impossible here (with incremental z level increase). Maybe these z-increment count could be lowered to 0. So its vase without spiral z-build. Still there is no retractions or other measures that stops and starts plastic flow. Extrusion width and its multiples is used as a build basis for wall thickness. External walls should be accurate - in order to do that hole dimensions or internal wall measures might have to be ignored or give them a tolerance to closest multiple of extrusion widht.

At present slicer 1.0.0RC3 I can -place number of perimeters -ignore infill by parameter 0 I can´t ignore perimeters of holes or internal walls if there is any in the model / wall thickness is specified in the model.

I can ignore these internal walls with VASE-mode, but I can´t have multiple perimeters - right?

So my question/suggestion should be to have "something similar" with VASE-mode with multiple perimeters without needing to modify and fill up the hollow stl-model.
Even if it does not have spiraling z-build (if its difficult to realize without nozzle´s real outside diameters specified).

Correct question/proposal would be then to get possibility to specify different perimeters more accurate and "getting rid of internal perimeters": external perimeter count = x infill = 0 internal perimeter count = 0

Basis might work as the same as VASE-mode just taking account of external perimeter. Z-level increments during perimeter work is lowered to 0 (ignoring spiraling build if it is hard to define in practice).

I don´t know how external and internal perimeters are determined and separated in VASE-mode, but I guess that calculating all external perimeters areas in one layer and ignoring all but the perimeter with the largest area could be one solution...?

Should I close this Feature request: "double wall vase" #1797 and place a new and better specified feature request like "Vertical shells: getting rid of internal perimeters"? "Vertical shells specified separately as "internal" and "external count " in a same way as solid layers are specified on "top" and bottom" count.

[358Eki]

Sorry, I dont know why some of the text appear in bold and larger size. This was not my intention...

[358Eki]

I´ll close this and will open a new "Feature Request" with better description...