[slave1driver]

Hi everyone, I've been doing some research on rapid prototyping technology and was wondering if anyone has had any success in tackling this as a diy project yet. What works, what doesn't, lessons learned, etc. I'm rapidly :lol coming to the conclusion that I MUST build one. Talk me out of it!!!!



ps - Sorry to the mods if this is the wrong forum section- I wasn't sure where this question belonged

[Chris D]

I have not built one of those, but it sounds like fun to me. Which type are you thinking about? There are two types that I am aware of

1) Large container of resin that is hardened with a focused laser of some sort. Seems very hi-tech and difficult as a DIY.

2) The type that works like a 3 dimensional ink-jet printer. This seems to be the easiest approach. I have not looked into it at all, but I would suggest be starting to investigate the commerical machines to see what they are using for print heads or nozzels and what the material of the "ink" is.

From that point on, it is very much like all the other CNCs people are making here - 3 axis etc.

The "CAM" software, generally uses a special file type as well, don't know what it is called though.

This does sound facinating!

Chris

[InspirationTool]

Check out:

http://www.reprap.org/

For what some people are working on in the way of a "self-replicating" plastic extruder.

-Jeff

[chuckknigh]

Over on the DIY-CNC list (yahoo groups) we discussed something similar.

One rather creative idea was to use a hot glue gun to lay down strands/layers of polymer, which should work rather well. The hot glue would melt the next layer down, and bond quite effectively.

Given that a hot glue gun can be had at a dollar store, and a feed mechanism wouldn't be too terribly complicated, this could definitely work!

-- Chuck Knight

P.S. I think they're called STL files.

[slave1driver]

I don't want to do the powder bin, although those are pretty cool. I'd like to be able to fabricate ABS, acrylic, or some other form of resin. This forum seems like a great place to start researching my xyz axis options. ( Yeah- I was thinking of actually caniballizing printer parts, but I think a scratch build would be more fun )

So I guess I need to think about those extrustion nozzles, and what sort of support system to use. Maybe a disolvable wax or something. Once it all comes together (and this is thinking WAY ahead) I can fine tune the "smoothness" of the object if that makes sense.

thanks for the replys everyone. Keep the knowledge comming! I know I'm not the only one thinking about this!

[slave1driver]

Check out:

http://www.reprap.org/

For what some people are working on in the way of a "self-replicating" plastic extruder.

-Jeff

Thats a great site! I can't wait to dig into it more.

[murphy625]

I dont know.. DIY and the word RAPID kinda reminds me of OIL and WATER..

Murphy

[DieGuy]

I saw one on TV from the university of Wisconsin that cut layers of butcher papar with a laser leaving tabs and then the layers were glued together. They found a new use for it in forensics to create a copy of a skull that could be then modeled into a bust.

[BillPSu]

The key to rapid prototyping is the material which is used to make the model and how to apply the material. The materials used vary from paper to wax to powdered or liquid plastic to even metal. All these materials are proprietary and is where the companies supplying RPM machines make their money. Unless you are willing to develop your own material and find a way to accurately apply the material building a DIY rapid prototype machine will be extemely difficult.

[chuckknigh]

Hot glue.

Think about it...it's a cheap, easily available thermoplastic material, and an extruder for it can be had at any dollar store in the US.

Think about it.

-- Chuck Knight

[BillPSu]

The glue gun concept is used in some of the current machines but controlling the size of the extrusion, the rate of deposition, the spacing, and then you will need another material to be used for the support structure which can be sperated from the part. Developing that technology is the hard part about rapid prototyping.

The machine I saw used a small plastic filament .010-.015 diameter and used two different materials one for the support structure and one for the part. Temperature was controlled in the deposition chamber for bonding.

If you are willing to do the development work for the extrusion process go for it!

[Agent739]

That does sound rather fascinating. I was just looking at some syringe needles. So what are you thinking about a pulsed feed or continuous. Is the machine supposed to draw the part layer by layer using CAD

[BillPSu]

The machines build the part layer by layer and use a special file format .stl (stereolithography) Normally using the plastic machine the shell of the part is solid while the internal portion has some air spaces between the lines of plastic. The parts I've seen made with this method have lines which change directions every layer. I believe you will have to have a continuous application of the plastic and probably at the end of application a deceleration of the plastic at the end point to clear the nozzle. A syringe would be a method to apply the plastic however the size would have to be relatively large and the barrel and needle/orifice would have to be heated. A stepper motor would have to control the flow of material.

I have attached a picture of a metal casting pattern which was built using RPM. The original model was modified by adding radii to the model with body putty. This part was made on a powder machine which used a laser to fuse the powder together. This pattern consumed about $300 of powder to make the model.

[slave1driver]

Thanks for the input everyone. After researching a bit I came to the conclusion that I would have to develop the material and the support system as well. Yes- The big trick for this project is going to be developing those two and the deposit system. I was thinking about using extruded wax like the "glue gun" approach for the supports and a heated syringe type of nozzle to deposit fillament. I even saw one of those "cold heat" solder guns at the hardware store yesterday and thought... I wonder if I could have a hole big (small) enough to fit a syringe drilled into that thing...

[cbass]

I was thinking about using extruded wax like the "glue gun" approach for the supports

From what I understand, this support material is no ordinary material. It has to be soluble so that it can be removed (dissolved) with a liquid of some sort without requiring physical access to the support material. This will allow for hollow internal cavities where desired.

This is assuming that internal cavities in the part are required.

Carlo

[slave1driver]

From what I understand, this support material is no ordinary material. It has to be soluble so that it can be removed (dissolved) with a liquid of some sort without requiring physical access to the support material. This will allow for hollow internal cavities where desired.

This is assuming that internal cavities in the part are required.

Carlo

There really isn't a way (that I can think of) to get around making hollow, sealed, internal cavities in an object and not make the support material a permanent part of the object. I know that this is a limitation of a couple of commercially available "powder bin" based machines.

What would be ideal is a material that would breakdown over a short period of time but stay rigid enough during the construction process. Of course, I'm not a chemist so I think I'll just have to live with having either no ability for internal hollows, or have the support be a permanent part of the object. Having said that, because this is just a prototype machine, i'm pretty ok with either of those situations. But I'm open to any ideas!

At the moment the leading canidate for a support material is wax that can be dissolved or melted away in water.

[slave1driver]

There are a couple of opensource cad programs that will create the .stl format files out of a drawing. Blender is an opensource program that is generally an artist's tool, but is gradually becoming a better and better cad tool. It can create the stl files. I'm definitely going to stick to the stl standard if I can.

Did the people that built that part for you address the "inner hollow" question? It may not have even come up based on your needs.

The machines build the part layer by layer and use a special file format .stl (stereolithography) Normally using the plastic machine the shell of the part is solid while the internal portion has some air spaces between the lines of plastic. The parts I've seen made with this method have lines which change directions every layer. I believe you will have to have a continuous application of the plastic and probably at the end of application a deceleration of the plastic at the end point to clear the nozzle. A syringe would be a method to apply the plastic however the size would have to be relatively large and the barrel and needle/orifice would have to be heated. A stepper motor would have to control the flow of material.

I have attached a picture of a metal casting pattern which was built using RPM. The original model was modified by adding radii to the model with body putty. This part was made on a powder machine which used a laser to fuse the powder together. This pattern consumed about $300 of powder to make the model.

[rweatherly]

This looks like a great project.

Regarding the concern about voids in an item, in my experence is that there are very few (if any) normal manufacturing processes that make parts with internal, sealed voids. Most items that require a void inside are made of two or more pieces, and then put together in some manner.

What the support material us usually used for is to support areas that start out not being connected to any other part at the "Z" level you are working at. (as you proceed along the Z axis to build up the part, this area eventually is connected to the main sturcture)

Good luck, and let us know what seems to work.

Richard

[DieGuy]

here is a link with lots of linked information
http://www.cc.utah.edu/~asn8200/rapid.html#ACA

[slave1driver]

Thanks DieGuy- Thats ALOT of info and some good reviews. Should be helpful!

rweatherly- yeah, I've resigned myself to the idea that I internal hollows aren't that important for this project.