Tortured composite panels?

Wouldn't you risk delaminating the single skin when you bend it? The bond is only as strong as the foam and it seems to me anything other than a mild bend would be difficult.

 

Well, as long as you operate within the limits of the stuff you are okay. There are a few guys who don't like any tortured panel business.

Take a panel 10' long. Limit bending to 5% at the center. 5% of 5' is only 3", but that is far less than what the panel can do. Also, when the panels come off the table; it takes about a week for a full cure, so getting a 'hot' panel to the jig also helps...they are a bit gooey 24 hours after the pour

my math is all wrong here, so don't pay attention other than to know it is wrong!!

When we moved the fully glassed both sides Skoota main hull panels to the jig, we were really careful and a bit terrified of the wind. I never heard any noise and I always turned the panel vertical before moving them

Using the same percentages falls apart. The Skoota panels close to the bow and went to four feet midships, so we had panel bends of 2 feet in say 16'. The panels in the front @fairing were a little flat, but I expected a bit of trouble.

Someone must be capable of understanding the elongation calculation better. It has to do with arcs versus straight lines, most likely.

 

Common sense tells you that for any sheet material to conform to a compound curve, that material must be able to stretch, and a GRP sheet will not stretch much at all, regardless of whether attached to a core on one side. Ditto plywood.

 

Freeship / delftship has a function that shows the maximum stretch of a compound surface if you unroll it onto a panel.
I believe this is rather complicated and is done by keeping the lengths of the edges the same and then "relaxing" the geometry so that stretch is minimized over the surface.
I wonder is this needs to be calculated different for stretch vs compression and how much of either plywood does. And if a fiberglass foam panel can only stretch.

For a developable surface panel it should be more easy: If you'd bend a panel onto a cylinder and the center of the foam would stay the same length, the stretching would be circumference of a larger circle divided by the original circle:

U = 2 * Pi * R
U2 =2 * Pi * (R+W) / R
Stretch = U2 / U
Stretch = 2 * PI * (R+W) / (2 * PI * R)
Stretch = 1 + W/R

So bending a 20mm foam panel onto a 10m radius cylinder would just be 1.001 stretch or 0.1% elongation.
Is that right? Seems very little. But this would explain why even double laminated foam core can bend enough.

 

Double laminated foam bends OK, not as well as single laminated, but the confusion seems to arise when people imagine that something that bends easily, will follow a compound curve. A sheet of paper bends easily, but has limited ability to conform to a compound curvature

 

Scruff,

Same as delaminating / breaking a piece of plywood.
If you don't do it right it will break, regardless of what "it" is.
Why are you making this more important for a piece of laminate?
Laminate is generally more consistant than a piece of multiply wood with all the woods internal defects.

Fallguy,

There is not a material in existance that won't bend in 3D. But it is very difficult to predict the interaction of geometry and external bend forces to see what the internal "breaking" forces are generated in the panel.
Some materials are so stiff and have fairly low breaking forces that the bends that you can get become pretty small.
I couldn't get some aerospace analysts to accurately predict failure in a foam sandwich panel so we just tested it in a practical sense. Then it survived so much abuse that even the analysts admitted their analysis was flawed.
That was very unusual.

Mr. efficiency,

Common sense rarely describes accurately what is known in engineering.
Composite aircraft wings bend in 3 dimensions at once due to normal flight loads.
They are designed for a minimum life of 8K hours with loads to 9Gs. Fighter aircraft.
And that is panels up to greater than 1/2".
GRP panels stretch significantly and repeatably and there are engineering numbers to consult. I don't remember an exact number so I'm not going to get to lying.
Have you ever seen a blade for a windmill going down the road. See the big curve in the tip. That straightens out under load. And that is a GRP panel.
One of these days you really out to learn some facts instead of just spouting opinion.

 

I would have thought those "numbers" are such that a sheet of grp has, in practical terms, very limited ability to conform to anything other than the mildest compound curve. As for sexy fighter aircraft, that is just irrelevant, a sheet of GRP can be twisted in 3D. But you won't make a dish out of it.

 

Yeah that's the difference, bending a sandwich panel in one direction creates minimal stretch of the fibers, but a compound curvature requires something like making a dish out of a flat circle. Which is significantly more stretch.

But as shown with KSS it can be done with the right lever forces and creates quite good results without even spoiling the gelcoated and finished surface.

I'd be curious about Fallguy's dinghy!

 

In order for a preglassed panel to be "faster", the build has to fit a set of pretty specific parameters. Otherwise the advantage is relative, and can not be generalized. There are a few stages of building a hull: lofting, building a jig, creating the skin, fairing and painting. Best time saving with flat panels would be to have a cnc cut kit from factory made gelcoated panels. This you simply drop into a cnc cut jig, tape the seams and only have to fair and paint this small areas. If you have to make the panels yourself, you have to make the table, cut the foam to patterns, apply glass, etc. It's not that much faster then building a strongback and hanging and glassing in place. If you can not use gelcoated panels, you still have to fair and paint some. This means that fairing and sanding becomes the one area where the difference in speed can be obvious, but this is entirely builder dependent. Experience, tools and workforce size are the really important things. An experienced person with a powered flexible sander and a variable speed polisher can achive more in day then an amateur with a manual longboard in a week. Same goes if you have a team swinging a 10ft manual longboard, they will achieve "fair and smooth" much sooner then the amateur with his 2ft board.

As for "shape compromising", that is entirely up to the designer. Chined boats are coming back into fashion, and boats buildt entirely from developed surfaces can perform just as well (and in some conditions maybe better) then round ones. The designer needs some experience in order to know how much a specific layup will bend safely, and more importantly he needs to design in such a way as to make best use of the building method.

 

I once needed a kayak seat that had a panel curved about 15cm deep in 60cm width. I chose to try some 15mm foam I had been given. When I tried to bend the foam alone, it cracked on the outside. I let it go, laminated glass on the outside, and tried again. It worked fine. That specific foam failed more easily in tension than compression. With the skin on the outside, the foam was mainly in compression, though not so much that it started buckling on the inside.

My impression is that the plywood plans involve bending mostly along one axis, and only a little along the other. That is true for the narrow two-panel hulls built for multihulls:
https://www.smalltridesign.com/Trimaran-Articles/trimaran_article_pictures/95.jpg

Most broader hulls I have seen have a tortured section with deep transverse bend and little longitudinal bend, and where the beam requires strong longitudinal bending, the plans go for a hard chine, like this:
https://c2.staticflickr.com/4/3838/32009239394_8de2051d7d_n.jpg
https://c2.staticflickr.com/4/3888/32850322862_55b10b3236.jpg

 

the dinghy is a Mertens design, all dev surfaces

the idea is to laminate 6 oz glass on the inside of the 12mm panels, glue the boat up on a male jig, glass the outside and finish, flip it over, install seats

I have vac bagged 6 oz before and it has tended to pinhole even under 10" hg, my system won't go lower, so I would probably hand laminate the 6 oz. The exterior would get 12 oz woth perhaps 6 oz reinforcements. Then the inside gets 12 oz and perhaps 6 oz reinforcements.

The designer, iirc, had done the 6 oz outside first, but it seems a cross purpose to me.

Anyhow, it is all written down, but this is my memory. I had some of the forum members help me review the schedule a bit because I want the boat super light. I also might do it in carbon to shave some weight.

It is a pram bow.

 

I think those darts pretty much sidestep the whole issue of bending in two dimensions. The sides of the hull are bent in only one dimension, and in the bilges, the cuts mean you have a lot of conical sections at just slightly different angles, so that it looks like bending in two dimensions, especially after some sanding. Kelsall says no strength fibres are being cut. Denney came to the conclusion that fibres that matter to strength are being cut.

 

Interesting, I didn't know Denney came to that conclusion? I thought he wanted easier build method.

From what I've read I believe (I'm not an expert) I DON'T think fibers are being cut or torn. And at least from the pictures it looks like you have proper compound curvature. You're read about it being a scam if that wasn't the case.
And this makes sense to me because fiberglass can be stretched almost 5% which is a lot. So those darts are only in the foam and aid in guiding the bending process, but the form is ultimately defined by where the edges of the panels are constrained to. The rest of the fiberglass laminate then has to stretch (not compress) to fit that form and that leads to compound curvature with pre-tensioned fibers. The laminate can then only assume the desired compound form and spreads the tension. This should only minimally compromise the toughness of the panels. I mean it works in a similar way for plywood, it should be easier to do with a single fiberglass.

But I also prefer Denney's intelligent infusion design with developable surfaces. It's faster and KSS is rather specialized advantage.

I'm really just trying to explain these processes to myself and reiterating them to better understand them. I am a bit curious if people really doubt KSS works? I mean there are pictures and a manual and existing successfully build boats.

 

I think he wrote that recently, but I don't remember where.

From Kelsall's description:

I interpret that as cutting the outside laminate. I am guessing, but could not find a clear statement on the matter, that the outside skin is sanded and laminated with a second layer along the whole area below the bilge.

I don't doubt that boats have been built that way. I just didn't think it is what I am asking about. Kelsall confirms that. From Catamarans - Kelsall Catamarans - KSS Shaping http://kelsall.com/UniqueKSS/KSSShaping.htm

 

I think "dart cuts" refer to a triangular profile cut into the foam, not the laminate. Relief cuts so the foam creates a kind of cushion around which the laminate stretches.
Think of stretching an elastic membrane over a foam pillow. You would cut the foam to change the shape of the resulting form but it would still tend to be smooth.
If you'd cut fibers you'd immediately get creasing or tearing or buckling and a very messy shape. And you'd need fairing, which KSS claims not to need.

He does mention that more glass is applied to the bilge area. But that would be logical to increase toughness. The KSS torturing wouldn't work if the laminate is too think because the forces would be too big. So you might have to add laminate to keel.