Variation of the cylinder molding method.

Hi folks, ive thinking about a method of building a slender, multihull type hull... well a variation of the cylinder molding method really...

Im not sure if anyone has done this before, but why does the cylinder mold only have a transverse curve? Why cant we put some rocker into the mold whilst we are at it, keeping the same mold frame sections and simply offsetting the frames to get the designed rocker. Once the frames are setup, go ahead and transverse strip plank the mold in thin ply or batten it out first depending of the plank widths necessary. Finish the mold to a nice surface with duratec or similar...

Then to build a pair of cat hulls or tri amas, its a matter of laying up 4 panels, preferably a sandwich panel construction via infusion. All 4 panels come off the same mold without changing it, simply reverse the direction of the panel on the mold, ie 2 facing left 2 facing right. The surface pattern shape is cut from the molded panels and 2 panels are joined along the keel and stem...

Will this work?

Can anyone see a problem with this idea?

 

I think some of the Kurt Hughes boats have some small amount of longti curve in the cylinder mold. What your describing might be like a "constant camber" female tool. A couple of months back there was a mold up for sale on ebay- bit vague but it was similar, could have been a double ended tool maybe for sodebo or castorama, it was at central coast NSW, assuming Somersby/Boatspeed Performance. There's been some built like with flat panel topsides & custom shoe similar to what your building, I think ENZA utilised this approach & would probably allow better distribution of buoyancy.

Jeff.

 

You can make a panel with a constant camber and produce the rocker by cutting a curve in the bottom edge, adding flare to the sides or a combination of both.

 

This is also how the rocker is created with a cylinder mold method.

 

I understand this already, however torturing of ply wood into final shape still requires compounding the panel. Thin Ply has enough stretch in itself, but if you tried this with a sandwich composite panel, you would not be able to torture it hardly at all, the panels are just too stiff... So the mold needs a bit of the longitudinal built in...

 

Hey Groper, I think somebody on yahoo or over at steamradio forum asked Kurt Hughes about this recently. IIRC, he said he thought it would work, but hadn't tried it. I might have this all wrong, so you might e-mail him for a more detailed answer. Let us know what you find out if you do.

 

No sandwich composite panel (ie stiff) will bend over a cylinder mold. The point of the cylinder mold is to build up a stiff hull from many layers of floppy, flimsy plywood. You take 3 layers of 3mm and drape it over the cylinder mold like noodles. When the epoxy hardens, you have a very stiff panel.

If you are taking a cylinder mold and adding the full longitudinal curve in, instead of the very slight one Kurt designs to get you part way there with plywood, you basically just have a male form for half a hull. No advantage over any other way to build it.

 

I think.... in a second guessing kind of way!!! that where Groper is going with these thoughts is to apply the process of "Constant Camber" to a composite foam/balsa sandwich construction using a highly finished female tool to mold the hull skins. There's a saving in basic tooling costs as the shape is repeatable to either construct the full length tool on a plug at a percentage of the total hull length or repeatable on a short tool & joining on the job, the former would be preferable but would rely on a fairly high degree of accuracy in "extruding" or slip forming the panels to full length over a few laminations, in theory it could be possible to slip form the hull panels on a short tool also but some remedial work might be required at the "joins". Here's some audio that represents the process using timber http://www.smalltrimarans.com/r/John-Marples-on-Constant-Camber-Audio.html

The greatest saving I believe are to be found in surface finishing due to the large square meters in a multi build & "smart" application of infusion tecnique as Groper has been doing in his current build, something like this could be a cool application

Regards from Jeff

 

I dont think your quite understanding what im laying down... the cylinder mold would be a female type (i understand kurts plywood method is via a male mold) and it would be surfaced with a plywood skin then surface faired and finished with a tooling coating. This would then be a double ended female mold to layup or infuse sandwich panels on. You only need this one mold to pull all 4 halves of a pair of catamaran hulls (or tri amas)... so your making the panels with curves already built in, you dont need to torture them after there done, only join them along the keel and stem.

The profiles of the hull skins would be transposed onto the mold when you lay them up, but the mold is a bit bigger than the profiles so it can be used in reverse to get the left and right sides of the hulls without having to dismantle the mold and rebuild it in reverse like a frame and batten half mold typically used with transverse foam strip planking... The other problem with transverse foam planking is you can only laminate 1 side of the panel at a time, whereas a surfaced mold you can infuse both sides at once...

So the end result is you can infuse both sides of the panel at once, cutting time and costs, only have to assemble the mold once cutting time, have a female molded exterior surface - so you only have to fair the mold once, but pull 4 finished (gelcoated if you wish) panels from it...

 

So this only works for symmetrical double ended hulls. It is a really difficult limitation to get around.

 

No, the cylinder mold (in this instance) is larger than the profile cut out of the projected half hull panel, so that the panels position can be moved around on the mold somewhat, and as i mentioned earlier, its direction on the mold can also be reversed.

Then, as with kurts cylinder molding method, the panels profile in combination with its position on the mold and direction, will determine its final shape - which could be very much assymetric fore and aft as i would want it to be for a modern multihull with a high prismatic and transom stern...

 

Ok, the misunderstanding is that you use "Kurt" and "Cylinder Mold."

Your process has nothing in common with his, so it was a little confusing. Thanks.

 

Well yes it does have something in common, the full size patterns are cut from a panel thats made on the mold, then taped together along the keel and stem...

The difficult part is designing the whole thing so that it all comes together neatly and the curvatures work together, left and right fore and aft... not all shapes would be possible, but im sure some very good multihull shapes with workable volume distributions would be... That said, it might be rather difficult to determine / design the hull shape, in order for it to be successfully built this way... but i *think* i have method of determining the mold and hull shape together so that it works out... i guess a small model might be in order to prove the idea?

 

If you listen to the J Marples audio he does a small scale timber mold & then molds fiberglass skins off it to test & prove the shapes.http://www.smalltrimarans.com/r/John-Marples-on-Constant-Camber-Audio.html