Ok, I understand
Cookie Notice: This site uses cookies. For more information, please see our privacy policy.

The Second coracle.

Second Coracle (Fourth built at Frome Shed.) Martin Bacon
My first coracle was moderately successful. It took three goes to get the seat right, ending up
with a woven one made of nylon webbing, as the hard seats were just too uncomfortable. I used
it once a week in the River Frome with the Frome Canoe Club. It took a while to learn to steer
and to make adequate progress, but it became apparent it wasn’t really suitable for several
reasons. One is that it lacked freeboard and sat too deep in the water due to the small base.
This meant constant minor collisions with the rocks that come close to the surface in that river.
Another was that as it was fat and short it required quite a lot of effort to achieve any useful
speed. The third problem was that the strength is concentrated around the base and the rim,
and kayaks collided with it above the waterline. Though it survived, eventually a rib was going to
crack.
New design.
I therefore decided to make a new coracle better suited to very shallow rivers which can run
moderately fast. This meant a flatter bottom (which reduces draught), a longer and narrower
design (though with a bow end wide enough for feet and to carry loads) and much stronger
sides. The shape I had in mind, in fact, was more like a Teifi coracle than an Ironbridge one, but
with a less flattened front.
One object of designing a coracle is to provide a good, stable camera platform. The Mk. 1 is
quite good, but the intention is that the Mk. 2 is stiffer. This may mean it tends to slap when it
rocks and it may be less good when the wind is stirring up the water. The higher freeboard
should make it more secure, but more leewardly. We will see.
First thoughts
Having seen the wicker coracle built at Frome Shed, my first idea was to use wicker but to cover
it with a cow hide, which is very strong indeed and traditional. This is the kind of coracle that the
Irish and Icelanders built, sometimes in quite large sizes.
There were two problems: the first was that I am too clumsy to use wicker, and the second was
that I was unable to obtain a suitable scraped cowhide, and I suspected that home tanning
would not go down well, especially the collecting urine bit.
After several iterations of ideas I decided to use the modern equivalent of hide - Kevlar. It is very
strong, and with a little practice easy to work with using suitable epoxy resin. My first attempt at
a design involved an interior frame outlined with thin strips, over which a layer of Kevlar would
go with suitable reinforcement.
A better idea then occurred to me. This was to make an internal mould over which a Kevlar skin
would be placed. Then, the interior mould would be removed and replaced with wicker strips to
hold the skin sufficiently rigid. This looked like being strong and light, and because the wicker
would be fitted into an existing shape my lack of weaving skill would be much less obvious.
Construction
I built the parts of the mould at the Shed. However, the Shed move went slower than hoped and
there was nowhere to store the completed mould, so I had to continue at home. Fortunately this
part of the operation was quite clean and no power tools were needed, in fact I built the mould in
a spare room.
The first pictures show the mould in construction
Note the top, bottom and the widest centre wale are all made by jigsawing out a single sheet of
ply.
Once the top, bottom and rubs were in place a central band was fitted into the cutouts shown.
Then thin strips of 1.8mm ply offcuts were used to build up the shape of the mould. I decided to
cover the strips in paper to form a sort of papier-maché. This meant that rather than a perfectly
smooth hull, the position of the rubs would be visible, and behind these would go the wicker
strips. This would help locate the wicker and maintain a semi-traditional look. In the next picture
you can see that shortly after I started the strips, I tested out the paper idea so if it didn’t work I
could do something else without too much rework.
It turned out little intermediate stringers were needed to hold the shape. The middle wale is in
two parts screwed together so it can be removed.
It turned out three layers of paper were
needed for sufficient strength. Because I had a
limited number of clamps, I worked so that I
could put paper on the last section while the
glue of the next section was setting. This
speeded things up considerably.
The next stage was to fit two boards to hold the mould above deck while the Kevlar skin was
fitted. The outer layer of paper was coated with release agent - first a layer of PVA adhesive to
make it water resistant, then a layer of polyvinyl alcohol.
The rough Kevlar skin. The design makes the bottom slightly recessed, and though it cannot be
seen in this picture there is a layer of carbon fibre (385gsm) in the recess with the 300gsm
Kevlar layer on the sides and a bottom of dense weave 400gsm Kevlar for bash resistance. The
epoxy used was Easycomposites EL2 with slow hardener, which is very easy to use and
solvent free so there is little smell (though I do use protective gear including FFP3S mask).
Further thoughts
At this point the Kevlar skin was left for two weeks to harden while I thought about next steps.
Once I removed the mould from the inside, I discovered the skin was more rigid than I had
expected, but that it was not thick enough that collisions with sharp objects might not crack the
epoxy.
I decide to go whole hog on modern technology and, instead of a wicker frame, line the inside
with a layer of carbon fibre. This was partly because I had just picked up a few square metres of
carbon second grade and half price, and as it wouldn’t be on the outside I decided it was good
enough.
This rather horrible picture shows what I did next. The piece of plywood intended to form a
sheer strake (cut from the larger piece of the mould) was attached to the rim so that the shape
would be held while the carbon pieces were fitted and glued. There were 18 of them altogether,
provided with overlap wherever I wanted strength reinforcement. At the point it is incomplete
and you can see where the top pieces have to be assembled after the sheer strake is removed.
The sheer strake was later covered in Kevlar on the bottom (mounting pads) and Diolen on the
top to give impact resistance.
This actually all proved rather easy to do. The downside was the disgusting process of sanding
the inside to get reasonable smoothness; after each sanding new epoxy had to be layered on to
fill gaps. Finally, after testing for strength, a layer of Kevlar was added to the bottom with some
additional reinforcement in the foot and seat region.
The result was that when the sheer strake and the seat were fitted, no middle support for the
seat seemed to be needed. If necessary, a pad of closed cell foam will be used.
This is somewhat further forward. The sheer strake is reinforced at the bow, and is attached to
the skin with stainless steel right angle brackets pop riveted into place. The seat is attached with
stainless brackets. A carbon fibre tape rim covers the rivet heads and also provides protection
against rubbing. The white strips are Diolen used to cover the joins between areas of
reinforcement, as sanding these down too far would reduce strength.
In the next stage strong channel section is put onto the inside of the sheer strake. This provides
a comfortable grip and protects the edge of the wood from collisions with paddles etc. Note the
four eyes which act as tiedowns for a roofrack etc. These too are secured to the skin at the
thickest point with stainless brackets.
I liked the look of the visible Kevlar reinforcement and decided to varnish it rather than paint it.
This is the finished product. The skin is painted with two coats of Zinsser black, as is the interior
except for the exposed Kevlar, which has two coats of EL2 and then 2 coats of yacht varnish.
You can see four recessed mounting pads for the roofrack. These are made of layers of Kevlar
held together with EBT+ filler, providing a resilient but very strong surface. The edge of the skin
is above the sheer strake round most of the hull but it is reduced at the mounting pads and at
the bow. This is so that there is no sharp edge for the paddle to hit.
The seat is made by wrapping carbon fibre around two hardwood poles. The carbon fibre is on
the underside; then there is a layer of closed cell foam and finally the whole thing is wrapped in
the kind of fabric used for boat cushions. It is extremely light but very strong. The all up weight is
12kg, which is a lot for a coracle, but remember it is designed to coexist with kayaks in a river
with shallow rocks. You can turn it upside down and jump up and down on it without damage.
It is a myth that a teardrop shape provides good streamlining, but the wetted area should be
smaller than the previous coracle and hopefully it will be a bit faster.
Unfortunately at this point everything ground to a halt due to coronavirus, and when it will get on
the water I have no idea