Slippery O.2: Building techniques discussion

What this is about

Almost everyone who has a boat is thinking about their next one. This is the second of a series of notes documenting my thoughts as I conceptualise my next boat. The previous note is here. If you haven't already read it, I suggest you read it first, because this one will then make more sense.

A bit about me

Different people are different. Some people have amazing patience and persistence and can take on a project which will occupy all their available spare time for years and carry it through to a conclusion. I'm not like that. Some people get at least as much fun from working on their boats as from sailing them. I'm not like that either.

So when I've finished thinking and get down to building, the main structure has to be reasonably quick to build; I'll hire an experienced friend to work with me, and aim to get two hulls built in three months working full time. So I need to end up with a design which is not fiddly or unduly time consuming to build.

Finally, although I count myself reasonably well off, I'm not rich and I'm not going to be. The project has to be driven by a budget. This note is fundamentally about building techniques with a view to rapid, reasonably economical construction.

How design thinking has changed since Version 0.1

In the previous note I described a boat with (among other features) quite radically assymetric hulls and no daggerboards. The reason for writing these notes is to get feedback from other people, and I've had a lot of feedback. And one of the things I've been persuaded is that the relatively deep section assymetric hulls are not going to give me the sailing characteristics I want; that they won't resist leeway as much as I hope and they won't damp hobby-horsing. Also, significantly, the simplest and cheapest building techniques do not lend themselves to controlled asymmetry.

My present thinking, then, is to go for more conventional hulls with a narrow veed forefoot, a more or less semi-circular master section and more U shaped sections aft; and to go for daggerboards, appreciating that these are going to mean I can't go charging across shoals with gay abandon the way I do now.

Candidate construction methods

In discussing the costs of different techniques I'm focussing on the cost of materials for building a hull skin, unsheathed. This is because all the techniques I'm discussing will require roughly similar internal joinery and decking, and all will require sheathing. The strip plank composite technique wil also require sheathing on the inside, further increasing the relative cost of this option.

Tortured (or 'developed') ply

Tortured ply is a technique derived from birch bark canoes and widely used in day racing catamarans, notably in in A class boats and Tornados. A nice sequence of photos illustrating the technique (and the sort of hull it works well with) is here.

Essentially plywood panels are scarfed to the length of the desired hull, and the two sides of the hull are cut out of the flat panel. Then they are stiched along the keel line, propped open to a predetermined angle, and glued with an epoxy fillet of a predetermined width. This obviously just gives a V bottomed hull. You then force the upper edges of the panels together, forcing the sides to take on compound curvature. While the hull is forced into shape you spile and fit bulkheads, gunwhales and other internal structure to hold the shape.

It's very quick and simple. It only works for very long, narrow hulls, but that's OK because I want very long, narrow hulls.

The problem is this technique doesn't really scale, and hulls the size I want are pretty much at the extreme limit for the technique. Furthermore, as the thickness of the skin increases, the force required to induce curvature in the panels increases and the risk of the panels splitting increases. Also, the exact repeatability of the technique is not that good - it's unlikly that two hulls will be identical. I'm not too worried about that.

If the technique works at all it will be very quick and cheap; the materials for one lower hull skin (the tricky bit) nine metres long will cost about £400; topside extensions to bring it up to a reasonable moulded depth will cost another £200, but these extensions would only be added if the lower hull was satisfactory. This means I could afford to have several failures, and still have a quick, cheap project.

However there's a real risk that I can't make the technique work at all. What I'm thinking of doing to assess it is two stage. The first is to make some half-scale model hulls out of 3mm 'doorskin' ply to get a feeling of the sort of shapes I can achieve. The second is to make a one metre long full scale section at the point of tightest radius - somewhere in the run aft. If these tests work the technique has a lot going for it. However, a lot of experienced people are saying to me that it won't work.

One of the things that increases the likelihood of failure is that a standard 1.2 metre wide plywood sheet, when bent up into a hull side, is going to produce a moulded depth of less than 900mm, which doesn't provide usable accommodation. Usually, in tortured ply boats, a deck is used to hold the gunwhales together, but I can't do that if I'm to get accommodation into it. I'm going to have to extend the hull sides upwards, and that risks being a point of weakness. The hull is also going to be less than 900mm wide, and I need a passage way at least 600mm wide to allow someone to move about in the accommodation. So some very well thought out structure is going to be needed if a tortured ply hull is going to be habitable.

Finally, it isn't realistic to expect that you can torture ply thicker than 6mm, which is pretty thin stuff for lower hull skin on a boat of this size. Careful thought will have to be given in the sheathing to achieve maximum impact and abrasion resistance. However, savings made on the hull skin and moulds (very little mould work is needed in this technique) means that it would be economic to use more exotic layup for the sheathing.

Cylinder moulding

Cylinder moulding is a technigue promoted by Kurt Hughes which builds on tortured ply, but allows much larger hulls to be built.

The name 'cylinder moulding' is something of a misnomer. The mould isn't a cylinder and it isn't a section of a cylinder. Instead, it's an extrusion of a curve which is roughly half the centre section of the desired hull, the length of the desired hull. The mould is relatively simple to make, because the sections are all identical. Flat sheets of thin plywood are then layed on the mould at least two deep and with the butts staggered, and epoxied under vacuum to make a curved panel. The technique is described in detail here.

From there on the process is identical to tortured ply, but the force required to induce curvature into the panels is much less because they are already pre curved (although the precurve is simple and not compound). Also, because you can make up the panels to the size you require, the issue of the stresses being imposed at a particular - inconvenient - point in the structure does not apply, and the interior structure can be much less complex.

As with tortured ply the technique is only suitable for long narrow hulls and it's unlikely that hulls will be exactly identical.

The material cost of a 9 metre hull skin in cylinder mould is likely to be around £800, and, additionally, there is a mould to make (once) which will use probably £200 worth of materials. However, the risk of failure is very much lower. Futhermore, this cost is for the full skin up to deck level, and there will be savings in the internal structure.

Strip plank composite

Strip plank composite is the method of choice for most custom boatbuilding at present. It allows a very wide range of hull shapes to be built, and is by traditional boat building standards fairly simple and quick.

A mould is set up. It may be either male or female, and some exceptionally interesting boats have been made by building hull halves in separate female moulds and subsequently joining the halves along the keel line.

Narrow planks of core material are fitted onto the mould and epoxied to one another, edge to edge. It's unnecessary to spile and fit each plank, as the narrow planks are sufficiently flexible to curve in both dimensions. Obviously, to make a compound curved hull, some fitting is needed. The technique is illustrated here.

Once the mould has been planked the hull is sheathed on the exposed side. It must then be removed from the mould and sheathed on the other side. This is less easy, because inevitably excess epoxy will have exuded in places where, because of the mould, it could not be cleaned up whilst still wet, and consequently this side of the hull will need a lot of sanding and fairing before the sheating can be applied.

Unlike the tortured ply and cylinder mould techniques discussed above, the hulls produced should be more or less identical, and a much wider range of hull shapes is possible. The downside is that a lot more work is involved.

The other downside of this technique, as far as I am concerned, is cost. To produce a light hull with this technique requires a low density core material - the planks. A number of synthetic foam planking materials are available, but they are expensive. The most suitable timber for the job is Western Red Cedar, which has an exceptionally low specific gravity of about 0.34 but good rot resistance and good straight grain. Unfortunately, it, too, is expensive. The cedar alone for a single hull skin would come to around £2000; on top of that you need to allow another 500 for epoxy and for the additional sheathing, and at least another £200 (once) for the quite complex mould. Of course, other timbers could be used, but timbers with similar rot resistance and strength to cedar but lower price tend to be at least twice as heavy.

Cold moulding

Cold moulding was the custom boatbuilding method of choice from the nineteen sixties until the nineteen eighties. As with strip plank composite, a wide range of hull shapes is possible. Essentially a mould is made similar to that needed for strip plank composite, and a skin of laminates is applied diagonally across the mould. A second skin is applied at right angles to the first and glued to it, and a third skin either at right angles to the second or 'planked' fore and aft. Each individual piece of laminate must be individually spiled and cut. The technique is illustrated here.

The final product ia effectively a compound curved piece of plywood exactly the intended shape of the hull. The cost would be in the region of £1800 for a hull skin.

Constant camber

Constant camber is a hybrid between cold moulding and cylinder moulding. A mould is set up which, instead of being the shape of the intended hull, is a section of the surface of an object of revolution. Layers of laminate are layed on this as with cold moulding, but although each piece has to be shaped, each piece has to be shaped in exactly the same way (because it's a surface of revolution) so they don't need to be individually spiled and fitted. You end up with compound curved - dished - sheets of plywood.

The technique only works for long, narrow hulls, and the range of shapes that can be made are limited. Critically from my point of view, it isn't possible to make a hull that is V section forward and U section aft, which is what I want to avoid hobby horsing.


If tortured ply will work, it will produce the fastest, cheapest build. But there is a real risk that it won't work. Also, I've yet to find anyone with real experience of the technique interested in being a consultant on this project, so it would be very much a case of going it alone. My thinking at present is work on a design for a tortured ply hull until I am happy with the structure, and then to build some models to assess shape, including at least one half scale model and at least one full scale model of a one metre long section of hull.

If that is satisfactory it will be worth taking the risk of going for a full scale tortured hull. If the first is satisfactory then obviously you go on to the second. But it seems to me that the tortured hull will be so much quicker and cheaper than any other technique that even if you get as far as a complete tortured hull before aborting and switching to a different technique.

If tortured ply fails I'm left with a difficult choice. Cylinder moulding is undoubtedly the next easiest technique, but the only designer actively promoting this technique seems to be Kurt Hughes, and, let us say, his thoughts on sailing boat aesthetics are different from mine. I don't feel very confident of going into cylinder moulding without an experienced design consultant to refer to.

Strip plank composite would be substantially - at least three times - more expensive for the basic structure. On the other hand it is easy and predictable, and there are many designers who have experience with the technique, including both Dick Newick and Malcom Tennant, whose boats I admire greatly.

give me feedback on this page // show other people's feedback on this page