The only way to achieve it by building the grid and twisting would be with a deformable material - metal or perhaps thermoplastic. A material that is bendable but not deformable/ductile (and obviously if it is deformable but is elastic as well) will not be able to achieve the correct shape - unless it is cut out with the 'deformed' shape in mind, rather then the pre-twisted grid shape.
So far I am thinking that the only way to achieve the perfect curve is with tensile material - probably some sort of stretchy fabric which is hung off of a frame. However, my attempts to build an appropriate frame out of metal wire have all failed miserably it's a bugger to get it even near the right shape; and even getting close it looks awful, not smooth and dynamic at all...
Help: Physical model I need to construct.
Moderators: MR.D, Moderators
-
- Imperial Winter Developer
- Posts: 3742
- Joined: 24 Aug 2004, 08:59
Re: Help: Physical model I need to construct.
Trust me... you really don't want to try it with metal, unless you can get it laser-cut and formed by professionals. Thin metal won't retain the shape correctly. Metal thick enough to keep the shape, 28 gauge or so, is too tough to form well by hand. You can bend it, but you're not wanting to just bend it- you need a mathematically-precise curvature. You'd have to make a form and use that as a mold, among other challenges.
A thermoplastic might be a wee bit easier, in terms of forming- just use a heat gun.
However, getting it to the right shape will be a massive pain, and it's going to be utterly awful to deal with as you add more to the object- you can't heat the center bits, but you need to apply heat to form the outer ones. So, you'd have to form them all separately, then bring them together for assembly. Assuming you can mark all of the lines for the slots (you dare not try to glue it all together, that'd have awful results) perfectly, including where the cuts need to be after being formed (which would be really very difficult, even if you did it mathematically)... then it could work. Too many "ifs" and "buts", imo. Moreover, styrene / plasticine sheet isn't terribly cheap.
That said... that crazy Japanese guy whose modeling work I love, Water, uses a vacuum-molding rig he built with a household vacuum and forms to make really complex curvature. If you could build a form of the precise curves necessary, you might be able to pull it off, with a strong box and a shop-vac (and a lot of work). You could make the form out of green foam with wood underneath (for support, since you'd be applying a vacuum force) and put lots of small-diameter holes through it. If this idea sounds like what you want to try next (it's the cheapest option- I assume you can find a shop-vac on your campus somewhere), I'll make a drawing of the rig, if you can't find documentation on the Web.
See why I said to use STL? The form you're trying to make is possible to make with lower-tech methods, but only with the appropriate resources, in terms of skills and tools.
A thermoplastic might be a wee bit easier, in terms of forming- just use a heat gun.
However, getting it to the right shape will be a massive pain, and it's going to be utterly awful to deal with as you add more to the object- you can't heat the center bits, but you need to apply heat to form the outer ones. So, you'd have to form them all separately, then bring them together for assembly. Assuming you can mark all of the lines for the slots (you dare not try to glue it all together, that'd have awful results) perfectly, including where the cuts need to be after being formed (which would be really very difficult, even if you did it mathematically)... then it could work. Too many "ifs" and "buts", imo. Moreover, styrene / plasticine sheet isn't terribly cheap.
That said... that crazy Japanese guy whose modeling work I love, Water, uses a vacuum-molding rig he built with a household vacuum and forms to make really complex curvature. If you could build a form of the precise curves necessary, you might be able to pull it off, with a strong box and a shop-vac (and a lot of work). You could make the form out of green foam with wood underneath (for support, since you'd be applying a vacuum force) and put lots of small-diameter holes through it. If this idea sounds like what you want to try next (it's the cheapest option- I assume you can find a shop-vac on your campus somewhere), I'll make a drawing of the rig, if you can't find documentation on the Web.
See why I said to use STL? The form you're trying to make is possible to make with lower-tech methods, but only with the appropriate resources, in terms of skills and tools.
Re: Help: Physical model I need to construct.
It may be possible to make this out of paper. If I had the model in wings I might be able to unwrap it well unough to make a template.
I suggest make lots of grids up out of drinking straws then stacking them up ontop of each other with a small rotation in each layer.
(Wire mesh would work too.)
Another way would be to make a framework of wire representing the outer edges and weave the inner surfaces from (8) threads from the bottom up.
I suggest make lots of grids up out of drinking straws then stacking them up ontop of each other with a small rotation in each layer.
(Wire mesh would work too.)
Another way would be to make a framework of wire representing the outer edges and weave the inner surfaces from (8) threads from the bottom up.
Re: Help: Physical model I need to construct.
HOLYSHIT WEAVER! HOW'S IT GOIN!
Re: Help: Physical model I need to construct.
I'm not sure this thing is possible to do with sheets or anything like that. Consider that the deformation is inconsistent - the centre of a sheet remains at a constant length, while the edges are stretched, not just twisted. Twisting is always incredibly challenging with materials, except for malleable ones that don't tend to hold a shape well.
Edit: thinking it over, a stretchable fabric and hard wire. Preferably one that does not apply a lot of force when stretched, or it will bend the wire. You build the structure you desire as a single mesh, with a loop running vertically through each intersection. That is, the top view is
where the o's are vertical wires, and the lines are fabric. Of course, the fabric structure would be extremely hard to build. But this way you avoid trying to deform and stretch a hard material.
Then bend each (heavy, tough wire) each wire seperately into the form you desire for the columns.
edit :dammit, code doesn't monospace
Edit: thinking it over, a stretchable fabric and hard wire. Preferably one that does not apply a lot of force when stretched, or it will bend the wire. You build the structure you desire as a single mesh, with a loop running vertically through each intersection. That is, the top view is
Code: Select all
o o
| |
o-o-o-o
| |
o-o-o-o
| |
o o
Then bend each (heavy, tough wire) each wire seperately into the form you desire for the columns.
edit :dammit, code doesn't monospace
Re: Help: Physical model I need to construct.
Pretty good thanks, I do read this thread and the mapping area most days, so I am not so far away.smoth wrote:HOLYSHIT WEAVER! HOW'S IT GOIN!
-
- Imperial Winter Developer
- Posts: 3742
- Joined: 24 Aug 2004, 08:59
Re: Help: Physical model I need to construct.
I also think that a stretchable fabric is the way to go; but in my first attempt to build a frame, it failed miserably. It's actually bloody difficult to construct something out of wire that will hold the tension of fabric (that is, the wire is thick enough), but be able to mold the wire into the correct shape...
Re: Help: Physical model I need to construct.
I think I would cut about a hundred thousand copies of the grid out of cardstock? or something, and rotate/stack/glue em, then brush on some epoxy or something to smooth over the edges.
3-d printing is totally the way to go, for really, but that seems like cheating to me.
3-d printing is totally the way to go, for really, but that seems like cheating to me.
-
- Imperial Winter Developer
- Posts: 3742
- Joined: 24 Aug 2004, 08:59
Re: Help: Physical model I need to construct.
Just for everyone's general interest, after working on the idea, I came to no conclusion that could satisfactorily produce a copy of the structure. There are a few methods that would have worked, but would have taken far too long to be worth it - other methods that might have worked, but had a lot of work that might've looked a bit scrappy and would have taken a long time to do. Either way there was no simple solution.
... Except of course, for altering the idea
... Except of course, for altering the idea

- Attachments
-
- model photo smaller.jpg
- (237.91 KiB) Downloaded 68 times
- Forboding Angel
- Evolution RTS Developer
- Posts: 14673
- Joined: 17 Nov 2005, 02:43
Re: Help: Physical model I need to construct.
Wow, nicely done