Accruate Developable Panels

[lewisboats]

Developability check in F/Dship will tell you where the problems lie...but it depends on the material in question. It was designed for steel so if you are using plywood you can often obtain results that are...in F/Dship...not achievable. Best to look at the total error and edge error in the develop plates section and judge from there. Anything under .005 is probably gettable out of ply of a reasonable thickness with a bit of torture.

[marshmat]

I've been pretty pleased with the Devsrf plugin for Rhino. It produces a nice set of ruling lines to give some idea of how a panel will unroll. I have not yet tested the real-world accuracy of panels created with devsrf and unrollsrf.

Fair lines and realistic curvature are critical. Ideally, the porcupine plots of the surface edges and isocurves should be continuous and smooth; I'll settle for continuous if need be, but discontinuities or zero crossings in the porcupines do not bode well for developable panel accuracy (or hull fairness in general). Keeping point count to a minimum, keeping uniform point weights and avoiding tweaks to the knot vectors (keep them uniform, no duplicate values in the middle) will make it much easier to keep things fair.

[thommie84]

Originally Posted by DavidJ

If the plate isn't developable you can use the "smash" command in Rhino.

Does "smash" also show the strain as it does in Workshop (Maxsurf)?

[marshmat]

Originally Posted by thommie84

Does "smash" also show the strain as it does in Workshop (Maxsurf)?

Smash will tell you the change in area, but it doesn't show the strains. I've also found that for panels that are close to developable, Smash can be inaccurate. I've taken to using Unrollsrf on a true developable panel that shares the two longest edges of the nearly-developable one, then using the shape produced by Smash from the true panel to correct the shorter edges. Roundabout, yes, but it seems to work for things that are just a little bit off developable in a few spots.

[sissi]

have you seen this product line? http://preforms.com/

Originally Posted by sissi

have you seen this product line?

Do you earn your living by spreading this ad around?
Where is that product related to this thread?

[thommie84]

indeed this has nothing to do with predicting development of plates with various software sollutions...

[The Loftsman]

These post,s and so many more all seem to be adverts for this or that software etc,etc. I can develope shell with compound curvature, or with double curvature etc,etc and all because i served a long apprentice ship to learn how to Loft Ship's/Boats, there are no short cuts and it seems to me that most people think you just push a couple of buttons on a keyboard and hey presto all your design challenges are over.

The Loftsman

[David Cockey]

McNeel now recommends using Squish instead of Smash for flattening non-developable surfaces. Squish has a number of options for how to stretch or compress the surface during the flattening, and can provide a graphical representation of where the maximum distortion occured.

http://wiki.mcneel.com/labs/advancedflattening

Squish needs to be downloaded; its not included in the standard Rhino 4.

[DCockey]

Originally Posted by The Loftsman

These post,s and so many more all seem to be adverts for this or that software etc,etc. I can develope shell with compound curvature, or with double curvature etc,etc and all because i served a long apprentice ship to learn how to Loft Ship's/Boats, there are no short cuts and it seems to me that most people think you just push a couple of buttons on a keyboard and hey presto all your design challenges are over.

The Loftsman

What assumptions about stretch and compression do you use when developing a shell with compound/double curvature?

[The Loftsman]

This i consider to be an art, and not as yet for general consumption of freinds i dont really know yet?