Rhino Challange Question

[pyronaught]

This is not something most people would probably want to do very often, which is why I'm having a hard time finding information on how to do it. In Rhino you can create curved surfaces that fit together to form rounded objects like a boat hull, football, balloon etc. and then unroll those surfaces to create flat, 2D panels that can then be used to make the objects in real life. But how would you go about doing the reverse and start with 2D panels and then stitch X number of them together to form a 3D object? For example, a football has four slices of pigskin that are sewn together to make the football. Now lets say instead of four you want to see what five of those slices would look like. Both the diameter and the curvature of the resulting football will not be the same as when only four were used, so Rhino would have to figure out both the diameter and the curvature of the new football based on the edge length of the slices and how many of them are required to close a circular arrangement of them. I have a feeling this functionality doesn't exist, but though I would check if anyone knows how to do it.

[jsc]

I don't think your particular example is possible. The longitudinal curvature of the football is determined solely by the arc length of the side of one of the slices. With that fixed, the slices have to be a certain width in order to get full coverage all the way around. You can't just toss in an identical patch and end up with a football without wrinkles.

Oh wait, I guess it could be a fatter football. So maybe it is possible.

[jimc]

Im not sure on that one but i can tell you where to go to find out. Post this question on the mcneel forum. If a user there doesnt know then someone from mcneel will answer your question.

http://discourse.mcneel.com

[willnewton]

That is some funky functionality. First, you would always have to know the final design of the 3D piece beforehand, such as a football and would simply slice it into five pieces and unroll it. You can't work the other way.

This would be like laying down five sheets of paper and telling the software "build a football". The software needs to know what a football is. You would also have to supply shapes that exactly made a football. The software does not know how to assemble items. Even if you laid down six equal flat squares, you can't tell it to make a cube from those pieces, although you could use them to assemble a cube yourself.

You can do stuff with certain add-ons. Solidworks has sheet metal functions for unrolling, calculating extra materials lost in the corner bending and generating assembly tabs, etc. It can't however, "know" what your intentions are from a flat sheet.

What you are asking for 100% possible with human assistance, but not possible using only the software. It does make me wonder if there is a university artificial intelligence lab working on this though.

Lastly, Rhino's Unroll function is not magical and makes many compromises to flatten 3D curvature. Unrolling a cube is easy and works great when you "roll" it back together. Unrolling a sliced football as your example would work, but when rolled back up generates a football with five flat sides and the seams will not align as the did originally. I have unrolled a set of curled horns in order to make them in paper for a costume and it pushed Rhino beyond its limits as it tried to compensate. With enough tape and patience and swearing, I was able to get a basic horn from my paper pieces, but nothing resembling the virtual horns.

Rhino is good, but not that good.

[Jules]

Don't know that it could be used for connecting "stitched" panels in the other direction. You would have to rotate them and place them where you wanted them.

But i do love a challenge:

To generate a football shape with separate panels, I would use the Revolve function with a simple arc of varying degrees/shapes. The math generates the panels, the arc dictates the shape. And the end result is going to look the same either way when it's assembled. But the difference is, (if you are using it for a pattern of some kind), the individual panels can be smashed flat and turned into a cutting guide. (The full 360° rotation couldn't be smashed out flat when i tried it.)

Just revolve the arc according to the formula: 360°/ number of panels that you want

End result looks the same either way, but the number of panels used varies. Shape of the football can be changed by changing the arc.

[pyronaught]

The unroll surface command only works accurately if the surface is curved in only one dimension. So in the case of a foot ball where the panels are curved in two dimensions, you have to choose to flatten one of the dimensions when generating the 2D panels with unrollsrf. When making blimps (same as making footballs), I flatten the short dimension that would normally have a circular arc to it, which results in the object on the upper right in the image below looking like the image on the upper left with the flat faces.

What I'm trying to do is take one of the 2D panels already generated for a blimp designed to use 8 panels and expand the actual blimp being made by sewing in a 9th panel. The blimp becomes unhinged from the 8 panel 3D model at that point, so I no longer know what it's volume, diameter and profile curve is. I did search the McNeal forum and some boat building forums where people want similar functionality for taking hull panels and going in reverse to find out what the hull will look like. No such functionality exists as far as I can tell though. I did find a manual way of doing it though which gets close enough to use. I used cross sections of the panel to calculate the diameters of various bulkheads along the center line, then used the curve fitting tool along the edges of the bulkhead to get the new profile curve. Once I have the curve I can then sweep it around a circle to produce the hull and have the new 3D model to use for volume calculations.

It can be done in software, it just hasn't been done in Rhino yet. The very expensive specialty software used in the textile industry can do it though.

[pyronaught]

BTW, this project does involve quite a bit of 3D printing. The blimps inflation valve, 2-axis hinged motor mount and fin brackets are all 3D printed.

[Jules]

Cool blimp!

[pyronaught]

Here's another trivial but very useful 3D printing application I came up with today. After getting frustrated trying to find pre-wound bobbins for my industrial sewing machine, I decided to try my hand at using the fixture built into the sewing machine for winding your own bobbins. A bobbin is just a tiny reel of thread that fits inside the sewing machine and works in conjunction with the main spool. The problem is that since they are small, only 21mm diameter in my case, they run out of thread pretty frequently and you are always having to change them out. If you can find them pre-wound cheap enough then you can find a box of 144 for as cheap as $40, thus eliminating the chore of having to wind them yourself. But if you can't find them, or can only find them at a price closer to $80 then you are paying 55 cents for something that takes you about one minute to spool yourself. Then the problem becomes finding the empty spools, and they are often made of metal and sell for $5 each. You really want to have dozens of these things on hand so that you don't have to keep stopping and winding them as you go. You also can't reload the empty spools from the commercially wound bobbins because they collapse when the thread is gone and are designed to keep you from reusing them (and also to make them cheaper to produce, since the end discs are made of paper and not plastic or metal).

So I decided to see if printing my own spools would work, and turns out it works great! It only takes 12 minutes to print each spool, which is a two piece print that snaps together tight enough to not require gluing. I can print 36 sets on one plate in about 3.5 hours. Now I don't have to worry about sourcing these little bastages anymore

[willnewton]

What I'm trying to do is take one of the 2D panels already generated for a blimp designed to use 8 panels and expand the actual blimp being made by sewing in a 9th panel.

. I understand what you are up to now and love the blimp. The seam geometry for 9 panels would change, especially toward the tips. It is a subtle change, but will happen nonetheless.

I do have an idea to add a single panel with no math involved. Gimme a minute to draft it up.