Help: Physical model I need to construct.

[Warlord Zsinj]

I'm building a (physical) model for an exhibition, but I've run into a brick wall.

I've modelled exactly what I want to produce in 3D, but I can't work out how I'm going to build it physically.

I thought a few of the brains here might be able to lend a hand.

We start with simple 5*5 crossing planes forming a cube, looking like this:



Then, I twist it 90 degrees on the Z axis, so that it looks like this:



I can build it out of any appropriate material, even paper is fine.
The final size I am aiming for is quite small, about the size you could fit on the palm of your hand.

Ideally I'd like to make it out of one material only. Glue is fine, but where used it would need to be not visible.

Ideas?

I've attached the model as a mesh in obj format. If anyone would like it in another format, just let me know.

[Argh]

Take the shapes, use Rhino's Flatten command, see if you can reduce it to a 2D representation of the material before being twisted. May work.

Or try Pepakura, which was specifically designed to handle the issue you're having- how to realize a model as a 2D object that can be unfolded. Be warned, however, that you need to have an object with the appropriate cuts already designed into the mesh, in this case, because you want it to lay out the objects on paper, ready to "fold". I've used it before, and it's pretty cool.

I really dunno how you're supposed to realize this, from a manufacturing standpoint, though, short of building it with stereolithography.

I'm trying to imagine doing it with paper, and I think that'll be a nightmare, due to the twisting- it'll almost certainly crinkle, probably as you work towards the "outside", and since paper doesn't have any stretch at all, the cuts of the geometry would be incredibly finicky, and would probably be an absolute bear to get correct without destroying it. Gluing it's no biggie- you could use some little bent strips somewhere in the middle where nobody could see to hold it together. I really don't think that's the problem, though.

Very thin-gauge metal would probably work better, I suspect, if you could get it cut on a laser and then formed the pieces mainly by hand, matching the corners to pins located at the appropriate points. You probably don't have the experience with that to do it without extensive help, though. Might want to talk online on the "real CAD" forums, where there are people who routinely have to deal with these issues.

[Warlord Zsinj]

I'm familiar with pepakura; I have been playing around with it to see what I can get out of it, but I'm not being too succesful.

Suppose I get hold of some sort of thin metal plate that can be bent into into shape; how would you suppose I jointed it all?

[ironized]

umm, why not just make the box using half intersections

then twist it and stick it to something.

[Argh]

He's trying to build a sculpture for a fancy design competition. It needs to be physically perfect, if possible.

As for the metal... doing the joins is easy, just spot-weld it in a few places. Very easy to do, if you know how.

The hard part would be to get the cuts made, taking stretching into account- I'm not kidding about finding a shop with a CAD-CAM driven laser, and talking to people who do this for a living.

I don't suppose you've checked into stereolithography in your country? The more I think about this, the more I think it's the right way to go.

The machines for doing that are everywhere nowadays, and buying time on one is probably at least as cheap as doing it with metal- I mean, this is a design piece, it doesn't have to bear any loads, right? And at the scale you're talking, it wouldn't require one of the large machines, which are almost strictly used for industrial applications and you can't get time on them. Here in the USA, you can get time on one for fairly cheap, I know of people who built resin kit molds and stuff using stereolithography-made masters, and at the time, several years ago, the prices weren't bad- large numbers of folks bought them and are willing to sell time. As a student, you might even be able to beg that for free.

You'd have to remodel the shape to be closed surfaces first, but that's easy, with Rhino.

[Warlord Zsinj]

While it's not expressly ruled out, I suspect getting this custom produced, which is indeed possible in my city, would be going against the general gist of the exhibition.

Also, I suspect the price will be well above what I can afford.

[Argh]

While it's not expressly ruled out, I suspect getting this custom produced, which is indeed possible in my city, would be going against the general gist of the exhibition.

Er, why? I mean, there's hardly any method of building this that doesn't require rather sophisticated technology, and you're using a computer to do the design phase... seems rather silly. What would the objections of the judges be? That it's a physically perfect representation of your design? That, and I really have serious doubts about manufacturing this by hand. You picked something with seriously screwy curvature, none of those parts will look anything like what you'd expect, flat, and they are all going to be different

Cost... meh, you're a design student... BEG! Seriously, time on small machines, for a single run, isn't ridiculously expensive- we're talking a couple of hundred bucks or so, not thousands. Moreover, these guys want to do business with you when you graduate.

[ironized]

l2woodcarve

[Warlord Zsinj]

I'd rather change my concept then fork out a couple of hundred bucks, to be honest. I'm not that far along in the concept phase.

I still want to explore a few ways to achieve it using different materials and approaches, though...

[Felix the Cat]

If you're willing to make it a little bit bigger and spend some time on it, high-quality clay might work.

Even better, you could make a clay or ceramic mold, fire it, and then plaster it. Something tells me that making a mold of that would be at least 10x easier than making the thing from scratch, especially if you're willing to do some work by hand after firing/drying/curing/whatever the molded material, as that means you can use thin metal sheets to provide support horizontally (fill each section separately, fire/dry/cure/whatever the material, then join the sections by hand).

[Wolf-In-Exile]

If cost is a problem, but you don't mind some labor, i've a suggestion.

1. Get some reasonably stiff wire. Not too hard that its difficult to bend using your bare hands/fingers, but not too soft either, since you're using it as your base.

2. Make your base using the wires. Depending on what material you're using, you'd want to make the 'bend' before or after applying your 'filler' material.

3. Choose your 'filler' material and apply it to the wire armature.
- Cheapest and easiest to handle is modelling clay (plasticine), since it stays pliable, but it also comes off easily.
- Scuplting clay, like ceramic or Green Stuff used in minis are a little more costly but are more durable as they harden after awhile.
- Foil (for wrapping food to microwave) is another material you might want to consider. Since its relatively cheap and very pliable, you can fit it in the wire armature then bend it.
- Aluminium sheets are the next more durable material, but are more difficult to work as you'll need proper machining tools and you'll probably have to fold the sharp corners so you and others don't get hurt.
You'll want to fold the top and bottom edges of the aluminium strips to the wire armature, or loop it around so both 'sides' are covered.

[Forboding Angel]

Even easier than clay...

For the basis, go get a half gallon tub of that crayola kids molding clay stuff. It isn't clay and doesn't require any firing or anything and stays in molded position. Use chicken wire as a framework, then once hte crayola crap is dried, cover using a more durable material.

Imo that would be the cheapest and easiest way to go about doing this.

Another thought is to get ahold of Lathan Stanley, he is an engineer and knows how to turn models into paper models at least somewhat easily, and considering the complexity of the stuff I've seen him do, this should be relatively easy by comparison.

[Warlord Zsinj]

Thanks for the input guys;

My biggest concern with a number of the options - particularly clay - is getting a decent surface finish. If it looks tatty, then it's not going to be appropriate for a quite 'refined' exhibition. I don't have any serious experience with clay; I don't know how I'd be able to get the inner joints to work appropriately; and then to look reasonably good, too.

The exhibition states that any design must fit within a 350^3mm box.

The ideal solution here would be to make it out of interlocking card. at the moment I'm exploring whether that its actually physically possible.

Alternative solutions I have considered are forming the opposing edges of the curves with a steel wire, and then running hundreds of rubber bands between the two over the height of the wire in order to create the appearance of a solid infill.

[Argh]

Rubber bands won't work, they won't form the curvature required. The farther you stretch them, the more they will resist being shaped into a curve, too. And each one will exert more force- even if you used steel rod, you'd have to use a pretty ridiculous diameter to be strong enough. And then, how are you supposed to hold the rods together to form a structure, without having to do a lot of welding? And the welds would be considerably weaker than the rods, too, or you'd have to drill through some rods. Having done that before... let's just say that it's much easier said than done.

Putty over an armature might work, but that'll be a an absolute beast to make it look completely refined- you'd have to let it cure, then work it very carefully with files, etc., to get it cleaned up. Firing clay over a wire armature usually just results in explosions, too

Hmm... hows about this, though... you could make a mold, and put putty into it, to make a cross-section, then stack it up on the previous one, and distort all of the squares to match the positions of some steel wires used as guides?

I dunno- none of this sounds like it'd result in a very polished-looking result, though. And this isn't the kind of competition you want to bring something that looks shabby to, no matter how cool the concept is.

[Warlord Zsinj]

The rubber bands would be exerting quite a bit of force; but it depends on how strong the end supports are; also, if there were intermediary grids spaced evenly along the height then you would have a stronger frame to act in compression to resist the tensional forces of the bands.

An alternative is to actually use some sort of stretched material...

[Forboding Angel]

350^3mm box

Anyone feel like converting this over to standard?

[lurker]

Yeah, sure. In standard units:

0.350^3m box

[Argh]

350mm == 13 3/4" approx., so a little over a foot on a side. Smaller than a breadbox

[Tribulexrenamed]

3D rapid prototyping printer. The cheapest ones are only about 25k USD in the US.

[Saktoth]

Thats as far as i got, forgive the web-cammery. Its relatively easy to make your base structure using half cut double-notch joints. Presumably you can do this by pre-cutting and then twisting pieces of metal (cutting the joints after the twisting would probably be a nightmare). I really really wouldnt want to try and do this in wood... you might be able to do it with steam bending but i think you'd need a special press... *_*

I think your best bet for a low-tech solution is a wireframe resembling prettymuch what you have in 3d there, and covering it.

You can probably get a relatively good finish using moddeling clay if you smooth it with a modelling tool (IE a stick). Or long strips of paper mache running vertically.