Retractable skeg

My first wooden kayak, Geyrfugl, has shown a definite tendency to be blown off course when paddled by its intended paddler, who is very light, leading to a lot of windage and not much boat in the water. She is fine with a heavier paddler (and would probably be OK with a lot of cargo, too). My second boat was also being built as a low volume kayak, and whilst I intended to paddle her, it also seemed likely that smaller members of the family would get to use her at sometime. Since almost all of the typical "British sea kayak" designs come with a skeg, and I had a lot of (positive) experience using one, I decided to build a skeg for my Hybrid Cormorant. In fact, during her maiden paddle, before the skeg was fitted in its slot, the boat proved to be very strongly tracking, and I began to wonder if I had wasted a lot of time and effort on this part of the construction. However, using the boat in bigger waves and windy conditions, and especially when paddling downwind or surfing, has shown the skeg to be a valuable addition.

There are a number of possible skeg configurations out there, such as the one sold by Chesapeake Light Craft for their kits. British boats tend to have either the NDK style of skeg (which is pulled up by a string and cleat arrangement opposing a bungie-cord which tries to deploy the skeg) or the type fitted by Valley Canoe Products, North Shore and others, which is controlled by a stainless steel cable running in a small diameter nylon tube. I've never liked the NDK style, and the CLC style involves a big deck penetration which I don't like, as much of the work must be done after the deck is attached to the hull.

Skeg control - click for bigger (10k) imageSo I went for a skeg very similar to the one I already use Skeg deployed - click for bigger (21k) image which is fitted to my North Shore Mistral boat, operated by a slider control next to the cockpit, and dropping down as a triangular shape near the stern.

Skeg box parts - click for bigger (29k) imageThe skeg box was made from two pieces of 4mm plywood (glassed both sides with 4 oz glass), spaced apart by 6mm ply strips. The skeg blade was also 4mm ply, again glassed both sides. The thickened epoxy used to assemble the box added a little extra spacing, so that the skeg blade has maybe half a millimetre clearance each side within the skeg box. This is a lot less than the clearance seen in commercial skeg designs, and I hoped that my careful assembly would ensure that this was not a problem. Keeping the clearance low would mean that there was little space for water to circulate when the skeg was up, which would avoid creating a lot of turbulence and drag.

Skeg showing pivot - click for bigger (17k) imageThe pivot would be a piece of stainless steel bar cut off from a chunk I found lying around - this is quite a large (12 mm) diameter so that the radius of the cutout in the skeg blade should not produce too much stress concentration as would be expected with a small diameter pivot. The bar was cut 14 mm long, so that it wouldSkeg box view up inside - click for bigger (8k) image penetrate both skeg box walls, but be covered by glass and epoxy at the sides. If I was doing this again, I might be tempted to cover the ends of the pivot with a small piece of ply epoxied onto the box.

Skeg slot cut in keel - click for bigger (7k) image Skeg box fitted inside hull - click for bigger (10k) image The slot for the skeg box was cut in the keel of the hull after the inside was glassed and the box itself had been assembled. This was quite scary since the cuts must be made at a steep angle to the surface of the hull, keeping the saw blade vertical. I cut the slot undersized and then sanded it out to size using coarse grit paper wrapped round a small piece of wood. This produced a very thin edge, prone to splintering, but as this would later be covered by glass and epoxy (and inevitably some fairing compound), this was not a problem. The box was fitted with a few blobs of hot melt glue, and held in the centre of the boat by a temporary brace.

Skeg slot taped over - click for bigger (9k) imageWith the skeg box temporarily fixed in position, I taped over the outside to stop epoxy running down, then poured a runny mix of epoxy and micro-balloons into Skeg box fitted inside hull - click for bigger (12k) image the groove either side of the box inside the hull. This epoxy mix flowed nicely into the sides of the box and formed a good fillet in the tight V-shapes each side. This is essentially the same sort of epoxy mix I'd use in an end-pour.

Skeg box top edge taping - click for bigger (18k) image The top edge of the skeg box had been epoxy sealed, but now needed a layer of glass to protect it. This was easier to do after the box was in place, and was achieved by wetting out a piece of one inch tape, covering it in a plastic film and clamping it so that the tape wrapped down the sides of the box (the poor man's vacuum bagging approach:). The edges of the skeg box which stood proud at the keel were then sanded fair with the hull, and any gaps where the runny epoxy mix had not filled from inside the hull were filled with a thicker fairing mix. When the outside of the hull was glassed, the open skeg slot was simply glassed right over with both layers of glass (and was not actually cut open again until after the first launch)

Shaping the skeg control slot - click for bigger (17k) image With the skeg box fitted, work could now shift to completing the hull glassing and building the deck, which, being stripped, was a much longer process than building the hull. In parallel with the deck stripping, I made the components for the sliding hand control. First, I cut a square piece of sycamore from a straight branch fallen from one of our own trees. I then routed a deep groove into the centre of this, and sanded the bottom out by using sand paper wrapped round a piece of dowel. Then with this piece in the vice, I used a block plane to take down the corners to produce a piece of Shaping the skeg control slot - click for bigger (12k) image channel with a side thickness slightly more than that of the deck strips. I sealed both surfaces with epoxy, but decided that glassing the inside was going to be too difficult. This was an error - the epoxy sealing was not sufficient and I now have to remember to varnish the inside of this channel rather frequently on the finished boat. The actual slider was cut from a piece of walnut (which didn't turn out as dark as I had hoped). There are two holes drilled in this along the axis of the control. One is to fit on a thin stainless steel rod to keep the slider in place, the other contains a brass cable clamp which locks onto the control cable via two small screws which are accessed by two more small holes drilled in the top of the slider. The brass cable clamp was out of my scrap box - from an old broken electrical fitting.

fitting the skeg control recess - click for bigger (11k) image The ends of the slider recess were filled in with marine ply, also drilled with holes to take the stainless steel bar and the control tube. A hole was cut in the deck (after glassing both sides) slightly smaller than the inside of the assembled recess. This was sanded to remove all sharp corners and epoxy sealed. The recess was then glued to the inside of the deck with thickened epoxy made into a fillet, which was then covered with glass tape. After a certain amount of sanding off projecting bits of glass fibre and a further epoxy fill coat (the control would be quite near my knee inside the boat, so I wanted a good finish), the remaining assembly would wait until after the deck had been fixed to the hull (and indeed, until after the maiden voyage).

After the kayak was basically seaworthy, with hull and deck assembled, cockpit complete and bulkheads and hatches installed, the final job was to install the control tube and slider control, then cut open the outside of the skeg slot and fit the skeg itself, which had meanwhile been attached to a fitting the skeg control recess - click for bigger (11k) image length of stiff 4mm stainless steel wire obtained from my local boat chandlers. I emphasise "stiff" here. There are two grades of s/s wire used for rigging - one is a right-handed laid bundle of cables which are themselves left-handed laid bundles of single strands. This is very flexible so it can pass round pulleys and the like and not what is needed for a skeg control which must "push" without kinking. The other grade is a left-laid bundle of seven thicker strands and is intended for standing wires. This does the job nicely.

The control tube itself is a standard VCP nylon skeg control tube obtained from my local sea kayak shop. First I had to drill out a suitable sized hole in the top of the skeg box - there was a gap in the filler strip for this, so all I actually had to do was cut out a small amount of the glass tape. Getting at the skeg box was easy, since I had built in a small Kajak Sport hatch far aft above the skeg box to provide access to this part of the cargo space - this makes loading and unloading the kayak through the aft bulkhead hatch a lot easier and avoids losing small items permanently behind the skeg box ! Drilling a hole in the aft bulkhead was a little more awkward, since I wanted the hole as close as possible to the hull, but didn't want to accidentally damage the glassing on the inside of the hull. The tube was then fed through this hole, along the inside of the sheer, and down into the skeg box, held in place along the inside of the hull by a few blobs of holt-melt.

The tube was carefully cut to length and the other end inserted into the hole in the slider recess. At this point, the stainless steel guide rod was also inserted into the end of the recess, threaded onto the slider knob, and passed through the other end of the recess, where a second piece of tube was also added, to allow room for the end of the control wire to move. The rod and both pieces of tube were then fixed in place with thickened epoxy, being careful to seal all the drilled holes and the end of the short tube. I had also epoxied over the end of the control wire to be sure that it would neither fray in use, nor scratch the inside of the tube as it was fed through. The end of the wire was fed into the skeg box, carefully inserted into the tube and pushed until it popped out in the slider recess. It was then fed through the cable clamp in the slider knob, and fed into the other piece of tube. At the appropriate point as the amount of spare cable at the skeg end disappeared, the skeg blade was hooked over its pivot so that when the cable was completely absorbed into the boat, the skeg was in the retracted position. The position of the slider knob was adjusted and the cable clamp tightened. The position of the slider knob was chosen so that when pushed as far aft as the recess allows, the top edge of the skeg is still just within the box, so that the skeg is not too vulnerable to breakage if it is caught a sideways blow on a rock.

Problems - well, there were bound to be one or two. Firstly, the slider knob is just too tight a fit on the thin stainless steel rod which guides it along the slider recess. This makes it a bit prone to jamming, requiring a bit more care to operate than ideal. Given how the thing was put together, fixing this is difficult, but I have lived with it for two seasons and reckon I can live with it permanently, or at least until the control has to be dismantled for some other maintenance job...

Shaping the skeg control slot - click for bigger (11k) image Secondly, the narrowness of the skeg box was continued right to the hull of the boat, and this made it extremely prone to getting jammed by sand on beach launches. I tended to test the skeg as soon as I had got through any surf, but this then meant a surf landing and another launch from sand to fix it. In the course of unjamming the skeg several times, using a borrowed knife, I damaged the edge of the skeg box. The rather thin edges also took damage from my rock-hopping games when the skeg was up, to the point where bare wood was exposed. Obviously this needed urgent repair. I sanded the edges down to a much more rounded outline with a lower profile than the rest of the keel. I also sanded out the inside of the skeg box near the keel, so that the the skeg retrieval loop - click for bigger (35k) image internal shape was somewhat flared. I then added glass tape to both sides of the slot, standing proud above the wooden edge, and filled the angle between the wood and the edge of the glass tape with thickened epoxy. I also drilled out two holes into the lower edge of the skeg itself, and inserted a bent piece of stainless steel rod to form a small loop into which can be inserted a piece of wire or similar to get a grip on the skeg and pull it down if jammed. As it happens, the changes to the shape of the skeg slot have meant that it has not jammed since making these repairs.

the skeg retracted - after slot repair - click for bigger (29k) image With the skeg retracted, the keel profile is not now quite so smooth, but the skeg jams much less readily, can be unjammed easily, and there is still no great open slot to cause turbulence and drag, so this is basically the shape I would aim for in building another skeg system.



© A.E.R.Waddington, 2003
Contact the author: mailbox <andrew>
on site <pennine.demon.co.uk>