Compound plywood bend in stitch&glue construction?

Hey guys, I have an emergency situation where I potentially need to design and build a 4.5m (14.75') stitch&glue boat for this summer because my other ongoing build failed. Trying to keep this simple, I am opting for the stitch&glue construction with double floor and some bulkheads. I plan to use 4mm birch plywood for all panels, and 9mm for transom and bulkheads. Finally, I will add 1 layer of fiberglass or carbon fiber for extra strength and stiffness.

To give these 4mm panels more form rigidity, I want to introduce some compound bend into each of them. I know from experience that this makes even thin plywood ridiculously strong. However, I have never done that in stitch&glue construction, and I'm not sure how easy/difficult it would be to force the plywood into that compound bend. Obviously, almost any shape can be achieved with enough framing, but I want to keep framing to a minimum to keep weight low. I know that just by stitching the panels they won't get into any compound curves, but I hope that stitching them to the curved bulkheads will force them into shape.

These are a few pictures showing the intended shape of the hull and the construction. Forgive crude screenshots from CAD, I didn't have the time to make proper lines plan yet, I've only been at this design for a few hours.







(Blue lines separate different plywood panels)

And the intended framing, spaced every ~640mm (25"). What you see below is what I plan to stitch together before I start epoxying and taping the seams. Ignore the bow, I didn't design the stem yet.



And with the cockpit panels on (these should be installed last):



I am still refining the shape, so please ignore a few weird bits here and there, lots of stuff is missing, this is just for the general idea.

Now, I found some sources saying that for 4mm plywood the minimum bend radius (MBR) is around 300mm / 12". Any more than this requires steaming or laminating from multiple thinner veneers. So I was very careful in this design not to go below 350mm / 14" MBR.

However, I assume these numbers are for single bend direction, not compound bend. Every panel will receive some compound bend. This one is the most extreme (exactly 350mm MBR):



Software tells me that the maximum stretch/compression for this panel to go from flat to this compound shape is 0.1%. I am not sure if that is a lot.

So, two questions:

1) If this amount of bend realistic to achieve in stitch&glue construction with the framing I depicted? Will the plywood conform to these shapes evenly?
2) In general, what is the methodology / guideline to determine how far I can push compound bending?

I would very much like to put this to a test, but the problem is that it's quite difficult to test this at smaller scale or just some section of the boat. I do plan to build a few jigs to see how far I can push that 4mm plywood, but some prior knowledge would help a lot. Thanks!

 

I think you are headed for a failure.I can't visualise the flare into the wide part of the topsides being compatible with the fine bow and the plan view.I would suggest you look at th Magnum series of International Moths to see how a flare can be incorporated.This might be helpful: Make mine a Magnum https://www.yachtsandyachting.com/news/276282/Make-mine-a-Magnum International Moth Lowriders UK – Dedicated to non-foiling International Moths https://intmothlowriders.org/ . the distinct flare panel could and probably should be a part of the proposed design.

 

Thank you, I think I see what you mean, but I'm not yet sure why you say that the flare into the wide part of the topsides is not compatible with the fine bow? To my eye my top panel seems pretty much the same as in those links (going from transom to bow), except that they have that panel flat in one direction while mine has some inwards camber. Is that camber the issue that you meant?

 

While waiting for your reply, I played around with that Magnum hull type in my CAD, tried to replicate it with some changes to terminate the topside profile earlier (it just looks really ugly when it extends right up to the bow IMHO):





The article says that Magnum 6s were made using tortured plywood. I couldn't find any further info, but I wonder, perhaps they used the same method as this kayak construction?



I played around with that idea in CAD for the hull above, and it certainly seems doable on the screen, here is a flat pattern analysis




Unfortunately my CAD doesn't have the best analysis tools for stress analysis for flat patterns on compound curved surfaces, it only shows stretch and compression, but I can't find many references in literature on what percentage would be acceptable for plywood. I suspect these percentages might not mean anything useful. If the hull could be made this way, maybe with a few ribs to give it more rigidity, then it would probably be even better than stitch&glue construction.

But in any case, my previous hull or this one, or any other, the questions remain the same - how do I judge how much compound bend I can put into that plywood, and how do I validate whether it is build-able before I commit to it?

 

I don't think you can achieve those shapes with the stitch & glue method, I'm almost certain you won't be able to. Why not try thin aveneers, at +45-45°?

 

You are referring to the first post, right? Because of the compound curves? That is not at all achievable with stitch&glue then?

As for veneers, well, I'd need to build molds for that, which is orders of magnitude more complicated than stitch&glue, plus there are no good sources of veneers in my country... I want to keep this project as simple as possible, while having a smooth chine hull (if it can be done). I guess it can be approximated with more plywood panels that would provide smaller angle between each pair, but I was wondering if the panels can be bent in the other direction as well, like in that first post. Not necessarily as much, but at least some. I'm open to alternatives though, just exploring options right now.

 

I think the design in post #4 is worth pursuing,but I have some doubts abut the curvature of the bilge being achievable.The Magnums had three panels below the flare to the best of my recollection and you may have to do the same if the gores on the topsides are to get anywhere near closing.I would remind you that not all plywood is the same and you should avoid any that isn't made of three equal veneers while adhering to a maximum of 3mm thickness.You will need to build over some form of mould and it needn't be too complicated-but don't be afraid to modify it as you go along if it seems necessary.Do also remember that the Moths used a stayed rig with structure to link the load points and you may find it useful to insert a solid stem say 30mm fore and aft to keep the bow tidy.I would suggest the first step ought to be getting hold of a sheet of 1.5mm ply and building the largest scale model that can be produced from it as you will learn a lot that way.I also suggest that you develop the design to incorporate internal structure and devote some time to considering how to locate and bond it.for a few hours work,you went a long way.

 

Sophisticated and hydrodesigned shapes and s&g is a bad idea.4mm birch is hard to bend. Foam and simple female mold made from cheap plywood. Hand lamination inside + vacc from the outside is the way to go.

 

I see... I want to avoid molds at all costs. So what I get from what you said is that compound curves are not at all possible with S&G, correct? As for the design in #4 post, I was thinking about 4mm plywood, but could probably do 3mm too if I can get it (would definitely require serious reinforcement though). And yes, 3 veneers. But why do you say this would need a mould? The guy in the kayak video did it without the mould. I imagine that inserting some pre-shaped bulkheads would help adjust the shape to exactly what was designed (kind of acting like a mould, if that makes sense). Or it could make it worse by disturbing the natural shape that result from that bend. See, I don't know what I'm talking about. I do not know how to figure out how plywood behaves when bent like this. Your advice with scale model and 1.5mm ply makes sense, but building them takes a lot of time, and it's not so easy to make a change in hull shape and know whether it will still be build-able. I mean, I will make a scale model anyway, but I'd like to know some general principles to adhere to when designing such shapes so I don't need as many physical scale models... I understand that you probably judge this from experience, but since I lack that, I'd like to go by the numbers... As much as possible, that is.

Or, the alternative is to forgo smooth chines, and do something like the design in my #1 post but without the compound curvature, this should be doable, right? Just for a sanity check.






(Damn it looks ugly without these sweet round panels...)

So in my mind I can either go for this, which won't be as smooth and probably heavier because of more panels and more seam epoxy, or the single-sheet bend method like in post #4, IF I can do it without moulds (just with some bulkhead framing, which will still be necessary to support the double floor) and IF I can figure out how to predict the behavior of the ply to be able to at least get into the ballpark in CAD before I commit to any scale models... And whether the final shape will match what I drew in CAD exactly. You said "don't be afraid to modify it as you go along if it seems necessary", that bit kind of scares me that I may end up with a very different shape than I intended.

 

In Summary,if you want a complicated shape you will have to build a detailed mould to hold parts in place and to allow force to be applied but if simplicity is the goal,the more recent shapewould be easy-except for the flare panel.Something of the same eneral shape might be achievable but I wouldn't advise cutting the final sheer shape until after assembling the panels.Straightening out the concave forward section might make things a little easier.Add some bulkheads wherethe mast will be,at the aft end of the bow tank and where the helmsman will be sitting and make sure you have at least three buoyancy tanks.With that part in place you will be able to add lightening holes where needed and drainage holes for the same reason.I suggest you make a provisional attempt at a deck surfaceas itwill tie the sides together.

How is the mast going to be supported?

 

Ok, that makes sense to me, but the kayak guy ()demonstrated that it's possible to do a complicated shape even with a mould, so I suppose it is possible in principle, right? If he got that shape, it means other similar shapes should be achievable as well, and the question then becomes what are the rules and limits in designing it, i.e. how far it can be pushed before you have to switch to a mould construction?

Okay, but what is difficult about that flare panel? Seems to me that it has no more curvature than other panels, and no compound curvature, just a little more twist over it's entire length. What issues do you forsee here?

I originally planned for a mast partner bridge, something like this



Just not as fancy (just a single mast hole)

And thank you for other advice... Sorry for being so interrogating, just want to gather as much info as I can

 

The flare panel will try to assume the shape of least resistance.If you want it to pass through specific points,you will have to put something there to hold it,maybe only temporary,maybe permanent.The canoe guy had enough of a mould to create the shapes he was after,but how many attempts did it take to learn what might be needed?The shape o the flared wings in post #11 would be really easy to build from 4mm plywood.I would have doubts about the sheerline staying in place with an unstayed mast as there isn't a great deal of material in the rail or a wide deck panel to resist the forces imposed,distributing the rig load into the structure is a big consideration.

 

That's already getting closer to something feasible.

 

I see... Well, I don't know if that canoe guy actually had to make multiple prototypes, or perhaps, like I'm seeking, had some way of at least making a rough estimation of how plywood will behave when attempted to be bent into that shape. I mean, there's gotta be some way to simulate that, right? Sure, plywood is not isotropic, two plies going one way and middle one going the other, but that can be accounted for. How do I figure out that shape of least resistance when one or two edges are partially constrained by stitching to another panel?

Sorry for being so persistent, but these things might be obvious for you because of your experience, whereas I'd don't have that, and I want to intimately understand what works and what doesn't (and how it would behave if attempted), so that I can converge on something possible on the paper before building any scale models or the real thing.

So to further clarify, when designing any boat (not necessarily the one I showed here as an example), what are the rules to follow to ensure that it can be done with stitch&glue or forming (like that canoe guy) without moulds? How do I know if I crossed the line and the plywood will not behave like I've drawn it in CAD? How do I tell what natural shape it will assume when I stitch such panels together? Let's not focus on this specific example I made here, I want to figure out the rules/guidelines and how to predict plywood behavior so that I don't have to bother you guys asking to validate every single design I make

 

I don't think "rules" have been written for tortured ply.You have to do your own learning and as I posted earlier,what works with one batch of material may or may not work with the next batch.I have no idea how many prototypes flo-mo went through on his learning path,but I guarantee there were some disappointments along the way to the nice designs.Some models can be made much more quickly than the real thing and willsave time and frustration.