2d pattern to 3d hull

Not sure what the point would be. Your sketch doesn't not seem to fit very well. Spent a few minutes in TouchCAD on it.

http://www.boatdesign.net/forums/attachment.php?attachmentid=94180&stc=1&d=1410192931

 

Sketches are handmade and therefore it is impossible to match the lengths of the adjacent sides of each panel.
It's totally useless to try to do anything with these sketches. Waste of time.

 

these freehand sketches were just for illustration purposes to explain what I am trying to do. I can not share the actual drawings with you. Thanks for the effort though!

al

 

Then my friend, not knowing what information you have, how you wanted us help you ?.
Even though a magical command on a cad program, you would need to know which object or line / lines would have to apply.

 

Tada -I have he solution !

If you measure the edges, and the depths of the panels, its MUCH easier just to plot the polar points and re-create the panels to the exact dimensions.

No wonder no one bothers to invent a 'flat panel puller together program'

 

rwatson, can you explain further your solution? It is not clear to me.

al

 

More information about your method would be appreciated.

"Polar points" - are these related to conical surfaces? Not all developable surfaces are conical surfaces.

 

This problem has a pretty simple solution, but pretty hard to do by hand: All developable pannels have a shared edge with some other pannel. Just divide the shared edges in equal segments and draw 3dfaces using that vertices. Then just join the vertices in the right order (by 3dmoving and 3drotating each one to fit with the others). The 3d volume will grow up easily. The more divisions, the more accurate (and the more time to do it). In fact, that's what the "Unroll" command does in Rhino, just in the reverse order.
I concur with rwatson, there should be some software routine to do this easily.

 

ludesign: can Touchcad handle this issue?

-

 

More correct to say : in segments of equal length.

 

Its a trigonometry question. ( just like Quenquen said ) Every point along the panels can be calculated by its distance and angle referenced to a known starting point.

Once you have those points, they can be converted to 'panels' easily.

I am not even going to attempt the math, it involces Sin, Cosine SinArc and all sorts of horrible brain stretches that someone would have to pay me a lot of money for, so I could pay an impoverished mathematician from the Eastern Bloc to actually do,

but I did start an experimental method in a rhino drawing.

In the process of attempting a Rhino drawing based solution, I looked up Panelling Tools, and downloaded the panelling add-in




It looks like a potential useful tool, but I have run out of time today.

I have attached the sample drawing with a few notes in case someone else can make sense of it.

I will look into it more when time permits

 

Ps - my wifes suggestion was to actually cut out the panels and assemble them, to get the measurements, but we dont do practical things like that in Cad forums

 

quequen and rwatson, the solution is on this way. Keep in mind that you should work on spherical trigonometry, not flat, to be accurate.
The problem is that you need to have a first point of reference (on the boat) that allows to define the position of the vertices of the triangles, and this first point does not exist. If the, at least, tilt of the transom were known that would be possible, with hard work. If the longitudinal profile were known, things would be simple.
In any case, with sketches shown by the OP is impossible to do anything useful.

 

"you need to have a first point of reference (on the boat) that allows to define the position of the vertices of the triangles, and this first point does not exist"

I first worried about that too - but its not so.

The first triangulation to start with, is easier with a transom, but can be done with a Bow Stem too.

Its the relationship between the first two small sections of the side and bottom panels, oriented along the transom sides.

The process is just like Quenquen explained, but I played with the concept in Rhino to get it clear in my mind.

Once you 'chop' a small section of the adjoining bottom and side panels, making sure the narrow edges are equal divisions of each side, its quite easy to use Orient3Pt and Rotate3d, to make them match in such a way that they triangulate or 'brace' each other in the right direction.

Its exactly like they build Girder Bridges. You build up the panels this way with these small "irregular Rhomboids".

You can start at the bow, without a transom too. If the measurements of the file are really small, the rhomboids will be sufficiently accurate to form the very fine angular relationships..

I have attached some rough illustrations, in the Step 2 I cheekily Oriented the whole side panel to the transom and floor, just to illustrate the concept a bit.

Also, on the first Rhomboid, I left the triangular 'Bracing' lines in, to illustrate that these will be different lengths to provide the correct vertical or Z elevation, while the slightly different short edges will force the sections into curves on the X plane.

Sure , the resultant 'girder' will be 'stepped', but it will be easy to get a curve to take out the jaggedness, and recreate the panel.

Its not a job I would do for the experience only

 

Question. How do you ascertain the proper curvature/rocker of the keel panel, as this will affect the dead rise the distance between the ends of the panel (length) and the beam?