Need help on selection of multichine/developpable surfaces software selections

[Laurent]

Hello everybody!

I am only an amateur, not a professional designer (far from it!!).

I am looking for a design software that would work well for developpable surfaces to design hard chine hulls.

I am looking for slim (multihull sailboat) shapes. I would like a software that is not limited in the number of bulkheads or chines. It would need the "generic" functions such as (but not limited to) standard boat design parameters calculation (wetted area, displacement, prismatic coefficient, righting moment when heeled, center of gravity, center of lateral resistance, etc...), standard boatbuilding functions (flat panels geometry, bulkheads geometry with hull thickness removed, stringers placement and development, etc).

Moreover, I would like a software that can cope with transverse symmetry, instead of longitudinal symmetry; yes, this is for a proa.... And still get all the calculations right...

Of course, the design has to be exportable to standard file format, such as dxf for the flat panels. Something ***dedicated*** to developpable surfaces is fine with me; I would rather limit myself to that than "tweak" the design and surfaces to increase "developpability" to an acceptable level. Tortured plywood is not what I am looking for.

Free is good, but it does not have to be free software, as long as the price is not thousands of dollars...

I tried to find a few things on line, but it seems to be either much more powerfull than what I need (Free!ship) or too limited on the number of surfaces and bulkheads and exportability...

In advance, thank you very much for your advices.

Laurent

[yades]

Maxsurf Really Works Nicely !!!

[DCockey]

Rhino3D in Europe is around 1000 Euros, and has no limitations on shape. www.rhino3d.com By itself Rhino has a function which calculates the areas, center of floation, center of buoyancy, etc needed for hydrostatic calculations. This data can be pasted into a spreadsheet to calculate righting moments, etc but a basic knowledge of hydrostatics is required. Orca3D is a boat desing plug-in for Rhino, and Level 1 which includes hydrostatics is $1390. www.orca3d.com

[DavidJ]

I second Rhino.

[Laurent]

Thanks for the input.

I looked at Rhino.

WOW!! This is powerfull stuff. I looked at a few tutorials on line, but more for industrial design. It does seem to be more than I need, but on the other hand, I understand that "simply only developable surfaces"... is actually, on a pure math point of view, not that simple...

Just want to confirm with the Rhino users; is it easy to add this constrain (developable surfaces) to a hard chine design? It seems very easy to generate the chine curves, but can the software actually generate developable surfaces from that, or tell you why it is not developable, from the selected curves? Is it the proper way to design a developable surfaces hull with Rhino?

Thanks again.

Laurent

[DCockey]

Rhino has two different commands for generating developable surfaces from two curves on opposite sides of the surface, Loft and DevSrf. It also has a number of other commands for generating surfaces from various sets of curves which may or may not be close enough to developable for a particular use.

CurvatureAnalysis colors the surface based on the user selected type of curvature. For a developable surface the Guassian curvature should be small. This is a good way to check that the surface is essentially developable.

UnrollSrf unrolls a developable surface to a flat surface.

There are several different methods for generating developable surfaces using Rhino. My preference is to create the "chine" curves, etc and use Loft or DevSrf. In some cases it's best to extend the curves past where the actual surfaces will end, create the developable surfaces and then trim them to the desired size. This will happen with any software.

[Alik]

Originally Posted by DCockey

There are several different methods for generating developable surfaces using Rhino. My preference is to create the "chine" curves, etc and use Loft or DevSrf. In some cases it's best to extend the curves past where the actual surfaces will end, create the developable surfaces and then trim them to the desired size. This will happen with any software.

David, DevSurf produces ugly shapes, specially at bow. We use 'Normal' option, and this can be unroll as it consists of elementary triangles each of those is developable. Thus, surface approximation is done by elementary planes based on triangles.

[DCockey]

Originally Posted by Alik

David, DevSurf produces ugly shapes, specially at bow. We use 'Normal' option, and this can be unroll as it consists of elementary triangles each of those is developable. Thus, surface approximation is done by elementary planes based on triangles.

Disagree. What do you mean by "ugly"? My experience has been that Loft is more likely to result in strange surfaces than DevSrf. I usually use DevSrf after extend the edge curves past the stem and stern and modify the shape of the extensions until I get the surface to coincide with the desired stem shape. Loft with Developable option will put fill in the ends of the curves with triangular sections, and typically there will be a crease between the triangular surface and the adjoining surface.

With both commands where along their length the defining curves are selected can change the results.

Loft with the other options, Loose, Normal, Tight or Straight, doesn't usually form triangles unless there is something unusual about the edge curves.

[Alik]

Originally Posted by DCockey

Disagree. What do you mean by "ugly"? My experience has been that Loft is more likely to result in strange surfaces than DevSrf. I usually use DevSrf after extend the edge curves past the stem and stern and modify the shape of the extensions until I get the surface to coincide with the desired stem shape. Loft with Developable option will put fill in the ends of the curves with triangular sections, and typically there will be a crease between the triangular surface and the adjoining surface.

With both commands where along their length the defining curves are selected can change the results.

Loft with the other options, Loose, Normal, Tight or Straight, doesn't usually form triangles unless there is something unusual about the edge curves.

Surfaces shapes - on deep-V planing hulls conic surfaces (i.e. DevSurf) at bow result lower deadrise and undesirable curvature of buttocks.

Triangles are formed by algorithm of unroll, that is why it can be unrolled.

[DCockey]

Originally Posted by Alik

Surfaces shapes - on deep-V planing hulls conic surfaces (i.e. DevSurf) at bow result lower deadrise and undesirable curvature of buttocks.

Triangles are formed by algorithm of unroll, that is why it can be unrolled.

I assume you are talking about the bottom of a deep-V, defined by the keel and chine curves. DevSrf with the least twist option results in a surface close to the true developable surface through the chine and keel curves, and the chine and keel curves meet at the stem.

For a deep-V bottom, the surface from Loft with Developable option should be close to the DevSrf result.

For a deep-V bottom, the surface from Loft with the Normal option may be closer to what you want, but it will generally not be a strictly developable surface. It may be close enough to developable to "unroll" without too much distortion compared to a true developable surface surface.

Are you using the UnRoll command or doing the unrolling mechanically?

[Alik]

Originally Posted by DCockey

I assume you are talking about the bottom of a deep-V, defined by the keel and chine curves. DevSrf with the least twist option results in a surface close to the true developable surface through the chine and keel curves, and the chine and keel curves meet at the stem.

For a deep-V bottom, the surface from Loft with Developable option should be close to the DevSrf result.

For a deep-V bottom, the surface from Loft with the Normal option may be closer to what you want, but it will generally not be a strictly developable surface. It may be close enough to developable to "unroll" without too much distortion compared to a true developable surface surface.

Are you using the UnRoll command or doing the unrolling mechanically?

Yes I agree they are not strictly developable, I would call it 'technically developable'. But if You look at my sketch above can note that such shape can be made in paper by splitting into triangles and bending at their edges. Thus all vertexes will be on base curves forming this ruled surface. So the approximation of surface only is due to absence of those bending edges. Important that there should be no intrsection of those surfaces.

I am using Unroll command, and alos Expander application for compound curvature. Besides that, I wrote application for AutoCAD that unrolls ruled surfaces... Did it by hand of course when I was a student