Toolpath Software

Might be a stupid question but I would like to learn how to create (unless there is already on available) a program to create toolpaths in G-code for a concept 3d printer. This isn't for an extruder type tool head but a plotter cutter style with another toolhead for the glue/binder. Need to slice the .stl file and generate toolpaths for my tool heads. Wondering if anyone could give me some leds/links on how to educate myself on this. I'm aware this involves programming but I'm okay with that.

[objects.reprap.org]
[objects.reprap.org]

Paper-LOM?

---

-Sebastien, RepRap.org library gnome.

---

Remember, you're all RepRap developers (once you've joined the super-secret developer mailing list), and the wiki, RepRap.org, [reprap.org] is for everyone and everything!

Ya, that's the basic concept.
Thought maybe changing code in Skeinforge by at the very least removing extruder controls.

We seem to have a few other folk doing LOM.

What's your setup going to be like? What materials, and what mechanicals?

---

-Sebastien, RepRap.org library gnome.

---

Remember, you're all RepRap developers (once you've joined the super-secret developer mailing list), and the wiki, RepRap.org, [reprap.org] is for everyone and everything!

Material: Paper first, aluminum foil for conductive situations(copper too if thing enough) and of course plastic.
The mechanics for starts will just be stepper motors and maybe a combo with dc motors. Thinking a plotter cutter. Binder will probably be a problem...gonna try heat activated glue already on the paper with a heated roller and then a anti-glue for the removal of layers. An air brush might be able to apply some of this or just a good old fashion stick applicator.

Use a dot matrix ink jet type head to spray the glue and then you can use normal paper. Very cheap material you just need the special formula for the glue.

---

Bob Morrison
Wörth am Rhein, Germany
"Luke, use the source!"
BLOG - PHOTOS - Thingiverse

3DSystems, whose founder Charles Hull devised .stl format, also came up with .slc file format, probably with LOM in mind. It's stacks of 2D polylines, the outline of each layer. Like 2D g-code, without having to be written as g-code, because many CAD programs will give profile series through a model that can be directly formatted for the layers.

So then how is .slc file toolpathed. Do I do this in a CAM program? Can't imagine toolpathing each layer.

... have a look at Skeinforge - it's slicing a 3D-STL-object in single slices, outputting them as sequencial SVG-array (SVG=Scalable Vector Graphics fileformat) and generating G-code from the SVG's ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

It can be done in a g-code generator like Skeinforge or any other, the parameters are that if the g-code is compared to subtractive, there's no tool offset or waste passes, if it's compared to additive, there's no inset for extrusion thickness or infill. .slc is described here, [local.wasp.uwa.edu.au], and it can be seen from the description that it's easily formatted as it's easily human-readable, an .slc file could be easily produced even from an Excel .csv file, and almost any CAD program can display the toolpaths, which would appear as a dense wireframe of the model.

Won't Skeinforge create toolpaths based on an extruder for the reprap? I suppose this is where I'll have to modify the code for my parameters.

.slc files ARE toolpath files, because an LOM machine has no tool offset, so all the software has to do is the very simplest post-processor stuff: x,y position. Netfabb generates .slc files from .stls automatically already. Right-click in the window, choose "Export as slices", set the thickness and tolerance parameters and file extension, and off it goes. It also does animation of the file it's generated, so you can load it and watch a build simulation.

Edited 1 time(s). Last edit at 03/16/2010 07:40PM by murrayd.

Awwwwwh sweet...plus I downloaded Netfabb cause I was thinking it would be useful for me. So when I need the machine to load a new layer of paper, I will need to make a header file and include it in the g-code generated? Clearly I don't know enough about programming, just enough to get me by in class, lol.

It seems silly to do foo_1 -> .slc -> foo_2, compared to using vanilla toolpath and CAM stuff.

---

-Sebastien, RepRap.org library gnome.

---

Remember, you're all RepRap developers (once you've joined the super-secret developer mailing list), and the wiki, RepRap.org, [reprap.org] is for everyone and everything!

Where can I get tutorials or help files for Netfabb, the free version....their wiki help is unaccessable....want to learn how to use it...for one, creating g-code from slices.

Netfabb won't help you with that. A slice is a polyline, a series of vertices. When you create g-code, you're drawing that polyline with the machine - but if you've got an extrusion that's .5mm wide and you've got an absolute perimeter outside size, you've got to recalculate the offset path, which is what the g-code refers to. Then the infill passes have to be added as additional paths. If you're milling with g-code, the toolpath is half a tool diameter's width away from the perimeter, and it stops short on internal corners so it leaves a fillet and doesn't cut into the model. A slice is a profile of the model at a discrete z steps, nothing more. The advantage of a slice lies in not having to add the infills, offsets or waste passes for milling that g-code is for. As reprap is a minimalist rapid prototyper, .slc is a minimalist model description format that's compatible with most CAD programs without having to go through g-code interpreters. It can be written as g-code but it has simpler machine considerations per layer, so why bother? I think that even a dot matrix printer's carriage could be set up - so maybe old-style multifold paper would be good for feeding it, along with the cogged tractors and perforated edges. There's a thought I hadn't had before.

... some time ago we had some discussion regarding Skeinforge and LOM-toolpaths - i managed to output some paths for lasercutting a laminated minimug (atached image).

For breaking away the outer volume i made another hack: - if you 'cut' the 3D-object with a boolean subtraction from a bigger block, then you'll receive an inverted object-representation you can directly cut or mill with a LOM-system.

The inner volumes are 'solid' and the outer volumes are hatched with the infill-paths meant as paths for the extruder (now as cutting paths).

For vertical surfaces it's enough for separating the volumes, for bigger horizontal areas you have to insert some finer hatching as aditional cutting paths, so you can separate the horizontal surface-borders too ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Hey Viktor....were you able to get software that could output toolpaths of innner and outer of the slices all at once without manually toolpathing each layer seperately?

I don't know, maybe I can bring in each layer into a CAM program and run a quick toolpath on each layer saving into a folder and then combining the output code into one larger file for the machine to run through. This still seems like a long process. Looking for a program to slice and toolpath for LOM all the layers in one shot.

Netfabb does that. Embed the model in a block, subtract it from the block, and the sliced result is an LOM toolpath. Ten seconds. How long do you want it to take?

Viktor, the waste can be left attached by two or three mm "runners", and sliced to separate in halves, as in this picture. I've drawn the cuts as wider slots for clarity, it's one layer of a sphere, cylinder or cone.

FYI, there's some boatbuilding software out there that does slicing like that, except boatheads call the slices "stations". PolyCAD and Delftship are two, they don't do solids, but they interpret IGES solids as surfaces enclosing a volume, and slice the surfaces. PolyCAD is free but not open source, Delftship grew out of an open source project called freeship, which works much the same. PolyCAD works with nurb surfaces and meshes, Delftship/freeShip works with subdivision surfaces.

Edited 2 time(s). Last edit at 03/17/2010 07:28AM by murrayd.

... i made the complete G-code for all slices with Skeinforge.

For LOM-fabbing i have to implement the laminating-procedure ... best triggered sheetwise when increasing the Z-position ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

How would I change tools for a non-glue section of an inside object for Netfabb to know?
What I'm wondering is how the laminating procedure (subroutine?) is written up and where to place it. Sounds like a code inserted after the new Z-position.

Software stuff is my weak spot.

Hi Dman,

... the only difference between inner and outer voulume is that outer volume-areas are criss-crossed by lines a laser or cutting bit will draw. For this i simply inverted the 3D-object, so Skeinforge handles the outer volume as if it's the area to do infill-paths.

For adding the laminating procedure you'll have to pause after every completed sheet, start the laminator (or do it manually) and then start drawing/cutting the next slice ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Sounds like everything won't run on its own...I'll have to either manually add code or devise an electronic circuit (data bussing...new learned thing in school) do to controlling in order to make everything run smoothly without human interferance.

There will be two tool heads....cutter and binder/anti-binder.
New layer 'layer'....roll of paper/single sheet...probably roll

Thinking paper with heat activated binder so a heated roller can press the new layer...quicker if the machine only lays a layer for anti-binder versus the whole sheet for the binder.
There'll have to again be a subroutine for the cutoff of the new layer from the roll.

I'm looking for engineered parts and I think this method of 3d printing will give close tolerances.

Project for the summer. Figured I can start with the electronics hooked up to the motor and a modeled machine for verification before investing in the structure.

I looked at EMC software for controlling these subroutines but thought there was a all in one package that had settings.

... i have plans for developing a "lathe-LOM" some days (or maybe faster) with some help: - here i roll/glue an endless stripe on a cylinder, so the laminating and linewise cutting with a laser-scanner can run as continuous process until you'll reach the topmost layer ...

But here i need software that can calculate all cutting dots per line for the actual angular position of the cylinder and the height of the surface ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

"Lathe-LOM"...sounds unique....the software makes me think how a digital oscilloscope displays a sine wave by calculating each point over time....sounds like your treading on new ground.

... i had the idea some ten years ago, when i realized, that conventional LOM-fabbing was way to slow

My idea for a speedup was first the elimination of the pauses between the slices ... and second the possibility to apply more identical setups circular around the cylinder, so you can process [N] slices per revolution synchronous, where [N] is the count of laminating/cutting heads around the fabbing cylinder.

The software have to calculate all the cutting dots/lines synchronous, and the fabbing speed is [N]-fold enhanced as you'll stack [N] sheets in a single turn

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

The software's not too much additional complication, Viktor. A slice is a boundary intersection, your proposal additionally intersects each slice with an involute developed from a circle having a diameter which is the thickness of the material. It would be a bit of a bear unless there's constant feedback on the thickness. Unless it's self-correcting, a material thickness error compounds as an angular error.

Victor,
I am confused. If you are cutting the sheets before they are glued to the stack, how do you keep them connected and flowing when you cut all the way around a free standing object? It would seem to me that the final cut pieces will all be separate, like cookies after a cookie cutter. Even if you are leaving the whole thing on the lathe, once the lowest spot has been laid down, you will reach a point where you must cut across the whole width of the paper, and the new end of the roll falls down and fails to build up the last of the model. If you have away around this, patent it quickly!

Mike

---

Team Open Air
Blog Team Open Air
rocket scientists think LIGHTYEARS outside the box!

Peel the waste off the roll instead, which would make undercuts more difficult but feedback positioning easier.

... a laminating-cutting-head is simply a roll pressing a selfadhesive sheet an the cylinder and shortly after a scanning laser-line will burn dots through the sheet representing the cutting lines.

For metallic or ceramic sheets you'll coat or spray the underside with something for soldering/fusing with the surface - either solder for metall or ceramic slip (or waterglass) for ceramics.

... here is a discussion and appended sketch showing how the objects are embedded in the laminated cylinder. In reality you'll have a cylinder of some centimeters diameter in the middle, so the LOM-volume will be a thick-walled tube around ...

... instead of a cutting laser you can use a piercing needle or a drillbit too ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]