I visited a local manufacturer of small aluminum power boats, and their process seems a bit clunky but my naval experience is limited to carbon fiber things and model boats, so I’m not sure what would work better, with or without Rhino, so I’m looking for ideas.
They use DELFTship to do the hull design, but then they bring it in to OnShape for production detailing. And to make that work they have to export from DELFTship with the hull pieces flattened out, OnShape can’t add parametric “sheetmetal features” to curved surfaces, so that’s the end goal here nevermind that I personally question providing flattened “sheetmetal features” for someone else to feed directly into their CNC fabrication processes as that’s a bit of a black art. So they have to detail the model in this weird semi-exploded state. Just brute-force manually dealing with all of this in Rhino sounds easier to me but I wouldn’t actually recommend that. What do others do for this?
I spent 20+ years doing this kind of thing in Rhino, which i believe, is a lot cheaper than Delft Ship. I dont know the other app, but Rhino is perfectly suited to the entire process with one caveat: the non-flat parts MUST be developable, meaning they derive from conic shapes, so they can be successfully unrolled and provide a reliable flat panel that, when wrapped onto the frames will meet up with adjacent panels, without the need for trimming.
If the shell of the hull is Devlofted, you can have complete confidence in the unrolled parts.
Yeah this sounds so much easier in native Rhino. FWIW I’ve used the D.Loft plugin on some projects where developable surfaces are required, it seemed useful at the time but it’s been years since I used it. Do they want/need matching rivet holes? Or are these welded structures?
My projects were all welded, but if rivet holes are part of the developed surfaces, these can be output with the unrolled surface, as can marking lines for thansverses, longitudinals and any surfaces that meet up with the shell.
Thanks for the replies, these are welded yes. It seems like just doing this in Rhino should be simpler too, but they obviously really think it’s important to use parametric sheet metal features for all the CNC-bent structural details. Are they… kidding themselves that entering k factors for bends is worth their clunky process?
These developable surfaces are not formed. The plates come off the plasma or water table, are placed against the erected frames and bent against those frames using come-alongs. Im not sure how parametrics would be beneficial in those cases.
We did make use of a press brake for forming structural members. This was done to avoid the need to weld a flange to a web to create a stiffener along the edge.
In these cases, Rhinoscript tools were used to calculate the location of the bend line for placing the plate in the press brake and these bend lines were marked on the part along with instructions for bending (eg: bend up 3/8" IR x 90 deg). Im afraid the code for those tools is lost to antiquity.
Not knowing the client, but it sorta sounds to me like they’ve convinced themselves that they need DELFTship to loft a hull, and all the other issues are downstream of that idea.
DelftShip is less expensive than Rhino. There is a free version and a “Pro” version for 160 euros. Both include hydrostatic calculations with the Pro version having more capability but not as much as the Rhino plugin Orca3D. What is expensive for DelftShip are the “extensions” though most are primarily of interest for large vessel design. I looked at DelftShip about fifeteen years ago and it uses sub-division modeling rather than NURBS.
Conic surfaces are developable but not all developable surfaces are conic surfaces. Non-conic developable surfaces can be created in Rhino using DevLoft or DevSrf.
I assume these parts are for the framing, consoles, and similar rather than the hull surface, and have small radius bends. Depending on the accuracy needed for assembly the effect of thickness on bending (k-factor) will need to be accounted for somehow by somebody.