Logo by Pauldelbrot - Contribute your own Logo!

END OF AN ERA, FRACTALFORUMS.COM IS CONTINUED ON FRACTALFORUMS.ORG

it was a great time but no longer maintainable by c.Kleinhuis contact him for any data retrieval,
thanks and see you perhaps in 10 years again

this forum will stay online for reference
News: Follow us on Twitter
 
*
Welcome, Guest. Please login or register. April 19, 2024, 03:08:11 AM


Login with username, password and session length


The All New FractalForums is now in Public Beta Testing! Visit FractalForums.org and check it out!


Pages: [1]   Go Down
  Print  
Share this topic on DiggShare this topic on FacebookShare this topic on GoogleShare this topic on RedditShare this topic on StumbleUponShare this topic on Twitter
Author Topic: 3d printing directly from 2d slices or voxel objects  (Read 10611 times)
0 Members and 1 Guest are viewing this topic.
ansr23
Forums Freshman
**
Posts: 10


« on: December 13, 2015, 05:59:22 PM »

Hello fellow fractaliens, alien

I have been trying to print 3d fractals on an Ultimaker 2 printer. I am using the workflow developed by Teena described in this thread:
http://www.fractalforums.com/tutorials/voxel-image-stack-and-then-what/

Currently my current method is:
•   Generate fractal slices using voxelstack in Mandelbulb 3d
•   Combine slices to create obj mesh using Fiji (method described in earlier thread)
•   Reduce mesh in Meshlab using quadric edge collapse decimation to 1,000,000 polygons – results in a file size around 50mb
•   Export stl
•   Open stl in Rhino. Delete all interior “bubbles” or extraneous geometry trapped within the fractal so there is only one “watertight” mesh remaining.
•   Import stl into Cura (ultimaker slicer) or similar slicing software
•   Export gcode

This is where I am running into issues. Many of the slicer softwares seem to hang or crash trying to calculate g-code for any files over 50mb. Because of the complexity of the fractal mesh (ie lots of voids, openings etc.), many of them require support for overhangs that need to be generated by the slicer software. Calculating how to build this support material is resource intensive and seems to be what is causing problems

I realize many of you who have 3d printed are using shapeways or similar services. My understanding it that they use commercial quality printers that have dissolvable support material, whereas with the smaller printers (makerbots, ultimaker) support material is built like scaffolding out of the same plastic and is then manually removed. Has anybody had any success printing on these printers?

I am wondering if anyone has had any success in developing g-code directly from the 2d slices exported from voxelstack. Is there a way to completely circumvent having to combine the voxel slices into a mesh and back into g-code. It seems a cumbersome route that results in a huge loss of detail. In other words can we simply 3d print layers of the 2d slices without having to recombine them into a 3d mesh that is then re-sliced. I have come across a few Voxel based slicers like Voxelator which can import voxel formats. Such as .vdb, svx (shapeways format). But I haven’t yet come across a method of converting 2s slices into one of these formats which I am certain would be a massive but lossless file. Has anyone tried this method?
I realize that the voxelstack method I am using is 4 years old now and I am curious to see if there has been any progress with alternative methods.
Logged
KRAFTWERK
Global Moderator
Fractal Senior
******
Posts: 1439


Virtual Surreality


WWW
« Reply #1 on: December 13, 2015, 07:28:50 PM »

This has been a dream for me (and also for bib) since we started out 3D printing fractals, sadly none of us has found a method.
Interested to hear if any one else has...
Logged

3dickulus
Global Moderator
Fractal Senior
******
Posts: 1558



WWW
« Reply #2 on: December 13, 2015, 11:04:49 PM »

completely speculating here as I don't have a 3D printer to play with but...

it seems to me that the hardware functionality is to lay down layers essentially one pixel thick.
can you send data 1 layer at a time instead of translating into mesh?

I think that the mesh route is not applicable without a whole lot of computing resource behind it to make it work and because of the nature of the hardware functionality it should be un-needed ie: the mesh is ultimately converted into slices.

Upon acquiring a 3D printer to play with, the first thing I would do is read the manual, then read the technical manual, then dig into the driver specification and finally code up a slice-only driver if I found that the machine didn't have one already but is capable of accepting layer data.
just my 2 cents worth, I can't wait to get my hands on one wink
Logged

Resistance is fertile...
You will be illuminated!

                            #B^] https://en.wikibooks.org/wiki/Fractals/fragmentarium
KRAFTWERK
Global Moderator
Fractal Senior
******
Posts: 1439


Virtual Surreality


WWW
« Reply #3 on: December 14, 2015, 08:51:55 AM »


Upon acquiring a 3D printer to play with, the first thing I would do is read the manual, then read the technical manual, then dig into the driver specification and finally code up a slice-only driver if I found that the machine didn't have one already but is capable of accepting layer data.
just my 2 cents worth, I can't wait to get my hands on one wink


I hope you get one for Christmas!  afro

It would definitely be the most efficient way to 3D print, less data needs to be fed to the printer at once, this would make it faster to print very detailed big prints, which is one of my goals wink
Logged

3dickulus
Global Moderator
Fractal Senior
******
Posts: 1558



WWW
« Reply #4 on: December 15, 2015, 06:22:05 AM »

these beasts must have a "magic number" or some way to set a raw data mode,
 factory testing must entail going from the simplest calls up to the point where all parts are functioning and working together.
look at the printer language spec "g"?
see if there is a command to encapsulate raw XY data and try sending it a single image.
Logged

Resistance is fertile...
You will be illuminated!

                            #B^] https://en.wikibooks.org/wiki/Fractals/fragmentarium
ansr23
Forums Freshman
**
Posts: 10


« Reply #5 on: December 16, 2015, 05:38:51 AM »

gcode can easily be generated one layer at a time with free software like Inkscape or others, and these layers could conceivably be grafted together layer by layer. It is a simple language to understand and edit with any text editor.

However the raster image generated by voxelstack must be vectorized first because that's how print paths (or cut paths for cnc) are generated, resulting in a smoothing of the shape and a loss of detail(which could be ok, if a smooth shape is desired) Also this does not account for the support material required for overhanging material. These would also need to be designed manually.

Another method would be to to generate shallower sections using the previous worksflow(voxelstack>fiji>meshlab>cura)  into gcode and combine these sections, but again there is an issue with the support material because the when calculating the bottom section the slicer does not know if support material is required for above portions because they are not in the same model.

Or I could print these sections individually and glue them together but I wonder if the seams will be obvious. Although this might be a way of making some huge printed models!

looking into this I am becoming intrigued by other potential methods of representing fractals in 3d by layering cut material like paper or by cnc milling

Logged
3dickulus
Global Moderator
Fractal Senior
******
Posts: 1558



WWW
« Reply #6 on: December 16, 2015, 07:47:31 AM »

ok after a little research, vectorizing or meshing is not a problem, lots of software and established methods, the hard part is adding support structure, removing support structures, applying a final finish to the surface unless the surface texture created in the process is what you want, then you just need to make the places where support was attached have the same finish as the rest of the object after they are removed.

I saw one system that uses soluble support medium and one other that doesn't need support structures but it prints silicon only.

could a fractal object be embedded in a generalized support object like a 3D grid?
Logged

Resistance is fertile...
You will be illuminated!

                            #B^] https://en.wikibooks.org/wiki/Fractals/fragmentarium
ansr23
Forums Freshman
**
Posts: 10


« Reply #7 on: December 31, 2015, 06:11:20 AM »

could a fractal object be embedded in a generalized support object like a 3D grid?

This is a promising solution. One challenge is figuring out how the support material interfaces with the print so that it can be easily snapped off without leaving any residue. The Ultimaker slicer, Cura, seems to do this pretty well by 'floating' the overhanging print - it is not a solid bond.

... applying a final finish to the surface unless the surface texture created in the process is what you want,

I have found a few ways of creating a smooth finish for the prints which look promising such as acetone vapour which is demonstrated in this video <a href="https://www.youtube.com/v/h2lm6FuaAWI&rel=1&fs=1&hd=1" target="_blank">https://www.youtube.com/v/h2lm6FuaAWI&rel=1&fs=1&hd=1</a>...although this is for ABS plastic. I am using PLA.


Here are some pics from one of my successful prints. This is a much flatter print in the sense that there are very few intermediate support areas. Please note that i am using the 'old' process with a file that didn't crash generating the gcode file.

This one shows the print in process on the ultimaker bed which is around 200mm x 200mm




Here we can see the support material on the underside of the print being removed. The finish of the underside is pretty good.



And here is the final print with some backlighting. I love how the lighting pulls out some of the interior details.
Logged
Pages: [1]   Go Down
  Print  
 
Jump to:  


Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines

Valid XHTML 1.0! Valid CSS! Dilber MC Theme by HarzeM
Page created in 0.177 seconds with 27 queries. (Pretty URLs adds 0.01s, 2q)