June '07

43. The Longitudinal Floor Timbers

In the plan there are four longitudinal floor timbers which pass, in the original scheme, from the front of bulkhead E to the back of bulkhead C, and then further on to the level of the front end of the footwells. They intersect the existing athwartships floors with lap joints, and they are discontinuous at the D and C bulkheads, ie. they do not pass through the bulkheads, but continue in front of them and behind them. They divide the bilge area under the cockpit into five strips, the central one of which houses the propeller shaft and shaft seal forward of D. The outer two sections on each side will be accessible only if the sole is removed, so these compartments need to be well ventilated and drained with limber holes.

The dimensions given for these pieces of clear Oregon are 18 mm. x 75 mm. However, the outer ones at least will need to be cut down to narrower widths than that, because the athwartships floors are not 75 mm. deep at that level. It would have been a good idea to have cut the joints in the transverse members before they were glued into the boat, but it is too late for that now. My thinking at the time was that I could not be certain that the joints would align if they were cut prior to fitting, but, of course, they could have been marked and cut after fitting but prior to gluing.

So now I have to cut them in situ, which is easy enough for the intermediate timbers, but requires a bit of chisel work with the ones attached to bulkheads.

One move which will make things easier is to remove bulkhead D before cutting any of the joints. It has already been decided that the bulkhead has to go, so it might as well happen now. There will be a lot more room to manoeuvre after that, and it will allow the option of leaving the longitudinal floor timbers in one piece, instead of interrupting them at the bulkhead D level.

Bulkhead D has been removed, and is now being used as a temporary sole.

After its removal, which happily did not result in the boat falling apart, I was a little surprised and disappointed to see that there was a height discrepancy between the floor timbers on the forward and aft surfaces of the bulkhead. It was a uniform 3 mm. It will be easier to build up the four floors behind the former bulkhead than to shave down the ones in front, and that can be done with strips laid between the longitudinal timbers, so there needs be no delay in cutting the lap joints. I just have to remember to cut the ones in the athwartships floors 3 mm. shallower than will be required.

The 3 mm. discrepancy.

The centre line of the athwartships floors is marked first, and the location of the inner two longitudinal floors is determined. Their position is not specified in the plan, but a they appear to be centred about 145 mm. from the midline. The next set of longitudinal floors is 290 mm. from the first set, leaving a variable distance between the outer set and the hull. Marks are therefore made at the 145 mm. and 435 mm. points on both sides of centre, and a straight piece of timber, cut to the distance between the front of the rear seat and the back of the dash bulkhead is fitted over the marks. Corrections are made to ensure proper alignment, and the lap joint is cut out of the intersections.

However, if you are running low on marine ply, and have to use offcuts to construct the sole, you could alter the spacing of the floor timbers to suit the dimensions of your ply.

To help measure the right length for the longitudinal floors, overlapping boards are employed, left, and scrap is used to gauge the housing, right.

In places where the longitudinal floors reach the hull bottom, there has to be allowance for them to accommodate the fillets which are reinforcing the athwartships floors. It is easy here to kill two birds with one stone and place the limber holes next to the intersections, such that they straddle the fillets. If the longitudinal members were not touching the hull that would unnecessarily weaken the joint, but where they do not touch the hull there is, of course, no need for limber holes anyway!

It is back breaking work cutting out housings in the floors, and progress is slow. I wish now that I had had the confidence to cut these joints before the timbers were laid, but eventually the pieces come together. The picture on the left shows the longitudinals laid out on top of the athwarts, ready for housing.

After one day's work, during which I had to start again on account of the first longitudinal splitting while it was being pushed home, I managed to get one only in place. I discovered that making the joints a bit loose helps a lot, and is probably to be recommended for the epoxy to be effective, but it certainly rubs against the grain for a wood worker, and I imagine that would be the case for traditional boat builders too.

Anyway, eventually the first stick is in, with its rear half 3 mm. proud of the transverse floors, and its forward half flush.

The crossover at the D bulkhead remnant (left), and at the floor immediately behind (right),
showing the 3 mm. difference in height.

One longitudinal floor in place after day one.

I am not gluing the floors together at this stage, because I want to be able to get the sander into the bilge and tidy up the epoxy. I also have to put in the remaining keel bolts and reinforce the shaft tube, and the bilge runners are yet to be attached. They will be screwed on from within the hull. So I want to get all the floors bedded in, but removable. Let's hope the rest of them go more smoothly with the experience gained!

Day 2:

The outer longitudinal floor touches the hull at its forward end, so I placed the limber hole up there so that it did not run foul of the epoxy fillet (see picture left).

There is one floor left to fit at this stage, and then there needs to be some thought given to adequate ventilation of the bilge before the sole is permanently affixed.

The provision of limber holes assists in air circulation, but cannot guarantee a complete solution. A series of small vents will be put into the sole segments as well. While they will not be effective while the sole is carpeted, they will help when the carpet is rolled up, which it should be after each use of the boat in any case, just to ensure a dry sole.

At the end of day 3 the floors are finally in place. There are a few tidying up jobs to be done still before they are glued, but that can be progressing while I am waiting for things like the hardwood edging on the sole panels to dry, so I will go straight on to the sole construction.

The floor is cut and laid.

Top of Page

44. The Cockpit Sole

The sole is made of 9 mm. ply panels resting on the floors, some of which need to be removable for access to the bilge. That would be easy if there were a stringer on the hull to which they could be screwed, but there is not. And besides, there is going to be a batten fixed at sole level for the attachment of the bottom of the lagging boards, so the movable sole itself can only go as far as the inner edge of the batten at most. Instead, a fixed sole, with the batten riding on top of it, can be used lateral to the removable sole, leaving only the central portion as free floating.

A Photoshop mock-up of some of the proposed sole panels.

The lateral sole should be epoxy filleted to the hull sides, and screwed and glued to the floor timbers. The filleting greatly increases the strength of the floor/sole/hull complex and also serves to keep anything which drops out of the glove boxes in the lining cavity from reaching the bilge.

The glove boxes in the cockpit lining are useful spaces for boaty paraphernalia.

The central panel of the sole immediately above the shaft seal has to be able to be removed quickly for emergency access to the bilge, so screwing it down is not the best solution. It will be covered with a carpet anyway, so should not rattle around if it is left loose, but possibly some sort of Velcro type fastener could be added for extra stability. The most obvious plan then is to support the central sole segments on top of the inner longitudinal floors, and have the fixed segments covering the rest.

Whatever the design solution, as there will be moving ply parts I intend to edge them with hardwood, as I did with the rear seat parts, to protect them from chafing.

Ply sole panel being edged with Red Mahogany (Eucalyptus resinifera).

Finally, I should mention that an alternative to the ply sole is mentioned in the plan, namely slatted duck boards. My view here is that the extra weight of those structures would not be welcome, and that, even if I did make them, I would not want to cover their impressive looking joinery with carpet, and that omission would be contrary to the decadent feel of this boat. Too much boat and not enough boudoir!

The slatted duckboard sole in a Freebody launch.

There are two ways of preparing the sole for fit. One is to scribe a piece of ply to both hull side panels so that it fits up tightly against them all the way from the rear seat half-bulkhead to the dash bulkhead (bulkhead C). The other is to reproduce the dimensions of the floor timbers directly onto the ply which is to be used for the sole, and use a batten to trace the outline of the hull's inner surface onto it. This then becomes the sole in one piece prior to segmenting it into fixed and movable sections, or, in Ariadne's case, because the widest part of the sole is wider than a sheet of ply, it would need to be done in segments from the beginning.

The first piece of sole is fitted, then the second, leaving the bilge access plate.

Once the fixed segments are fitted, the line of their junction with the lagging should be drawn onto them. This can be determined simply by temporarily attaching straight boards to the existing carling, as was done in April '06. Where the boards meet the ply is their line of intersection with the sole. Since the real lagging boards will be set slightly outboard of the inner face of the coaming, allowance must be made for the setback. Alternatively, and somewhat more simply, scribe a line onto the sole a fixed distance from the hull, in this case the same as the distance between the hull and the coaming, minus the thickness of the lagging boards, or 160 mm. In practice this is not as easy as it sounds, because the fillets joining the bulkheads to the hull make it impossible to measure accurately the distance of any point from the hull surface, so the first method is better.

Next comes one of the more demanding jobs in the boat's construction, reminiscent of the skills that the traditional boat builders exercised. The batten which rides on the sole for the fixation of the lower end of the lagging boards has to be shaped. This means that its inner surface has to be angled to allow the boards to lie flat against it, and that angle changes for every board because their inclination parallels the hull side. Planing a constantly changing angle onto the batten is difficult to say the least.

Marking the position of the lagging batten.

Hardwood edging the sole panels.

As each sole panel is cut to fit it is edged with the hardwood. Above, you can see that the small panel which will give access to the shaft seal has been finished, while the neighbouring ones are drying. Even although the two panels on the left will be fixed, I have edged them where they lie on top of the bulkhead D remnant. It is probably pedantic, but it makes it look better to me. Only the shaft seal panel has finger holes so far, but they will all get them before they are glued down.

The joint line on top of the bulkhead D remnant is one of convenience. It would have been possible to cut a single panel to run from the rear seat to the dash bulkhead, but breaking it at the bulkhead D level makes it easier to fit. The line of the join will be covered by carpet anyway, so let's make for an easier job. The bulkhead was trimmed off at floor top level and the sole segments will meet over its cut end. There are two transverse floor timbers attached to that bulkhead, so screwing the sole down to them is no problem. At some of the other timbers, where two pieces of sole are being screwed to one timber only, it is expedient to add some ledger strips so that the screws are not too crowded and too close to the edge of the ply. In addition, as I have not used absolutely clear grade Oregon for the longitudinal timbers, there are one or two significant knots in them. I overcome the weakness of the knots by gluing a strip of ply over the surface of the floor, just underneath the ledger.

When in is finished, the sole is almost too attractive to cover with carpet, but you have to remind yourself that, when it comes to woodwork, nobody cares.

Even after all the joinery is done, and the hull is ready to be covered I will not glue the sole down. Certainly it can be screwed into place, but just in case of unforeseen necessities I want to be able to gain access to the bilge again for as long as possible. So, until the time comes to fix the lagging to its frames, by which time the filleting on the sole will have to be completed, there is no need to glue. In the meantime, some contingency arrangements, in the form of (possibly unnecessary) holes in floors and bulkheads will be employed. One obvious use for that will come if the batteries are too heavy to be placed in the motor compartment, and have to be moved into the cockpit to balance the boat in the water. If that is the case there will need to be cabling running from their location under the driver's seating to the motor compartment. That could be routed through the sole and forward underneath it, and then through the dash bulkhead, or the space behind the cockpit lagging could be employed.

The completed sole.

During the construction of the cockpit sole and the rear seat I have been using polyurethane glue. I have been less than completely satisfied with it for a number of reasons. Firstly, it leaves deep brown discolouration of the skin which takes a few days to come away, but more importantly it creates a line of foamy discharge out of the joint, which proves to be very adherent and difficult to remove without damaging the wood, especially ply.

The foamy discharge of polyurethane glue.

However, the final straw came with the arrival of the current edition of Fine Woodworking magazine, in which the strength of a number of different types of glue was tested. Polyurethane glue was the weakest of those tested, and especially in loose joints. Loose joints are not a problem in this boat, but the test results confirm my suspicion that that foam is simply not a substitute for tightly fitting timber. It has nothing to offer in the way of gap filling.

The best performer was PVA glue, and if you use one with a type 1 waterproof rating, such as Titebond 111, you get a product both easy to use and easy to clean up, which is suitable for boats. I'll be trying that for the next major item in the construction, the cockpit lagging.

Top of Page

An examination of the photos available to me seems to demonstrate that the lagging in Slipper Launches can be placed either on a diagonal, or perpendicular to the water line.

These boats appear to have almost perpendicular lagging, whereas the one below definitely shows a diagonal lie.

The diagonal arrangement gives the cockpit a more rakish appearance, and the joinery involved would not be much more difficult, but it does alter the geometry a little, as explained below. I note also that the glovebox openings are sometimes put into a contrasting wood. I presume that the reason for that is that the opening requires a very wide board which will be stable enough not to want to expand or contract with humidity. Just such a timber would be mahogany, especially quarter cut mahogany, but to save weight and/or expense the remaining lagging boards are of something lighter and cheaper. They are either shiplapped or tongue and grooved, so their expansion is catered for. At this stage I am thinking of Western Red Cedar for the lagging, which should be stable enough to allow wide boards to be glued up without too much trouble.

For the sake of lightness and of being able to bend the lining boards into the required shapes, I am going to use fairly thin and narrow boards. Because the carling, to which they are attached, is twisting to remain parallel to the hull sides, each board will be slightly angled to its partner, and therefore each board will have to twist a little itself. The twist is very slight, and can easily be accommodated by these light boards.

At their top ends I attach them to a cleat which is fixed to the underneath of the outer lamination of the carlings. The coaming will eventually be attached to the inside of the inner lamination of the carling, so I have the choice of running the coaming board down a little to cover the top ends of the lagging, or of using a kind of cornice to soften the transition from coaming to lagging. The traditional solution was to tuck the lining boards up under the coaming, and that seems perfectly satisfactory.

At the bottom end of the lagging there is another cleat screwed down onto the fixed part of the cockpit sole. But it is fanciful to imagine that those two points of fixation will be sufficient. There needs to be a third longitudinal member introduced at about the height of the rear seat platform running forward from the rear seat to at least as far as the dash bulkhead, and possibly further if any lagging is to be continued into the engine compartment. (That will only be necessary if there are footwells which make part of the engine compartment visible.) This third member will have to lie parallel to the hull side, as the carling does, so may also have to made up of laminated stock. It will need to be housed on the front of the rear seat and the back of bulkhead C, and will need to fit into a notch in the remnants of bulkhead D.

Clearly this is a challenge, and one which is far more enjoyable than rasping out keel scoops, for example.

Once the backbone for the lagging is in place allowance must be made for the conduit of such things as electrical wiring and steering cable to pass through the bulkheads to their destination, and the boundaries of the gloveboxes have to be fixed into position. Wiring has to accommodate the riding light(s) on the rear deck and any power outlets into the passenger compartment. A car fridge, for example, could be plugged into a cigarette lighter in the rear, and one needs to consider whether a music system is desirable. As the boat is to be mostly a day boat there is no absolute need for cabin lighting, but, depending on where the batteries end up, a conduit may be needed to the engine compartment for such items as horn, spotlight, navigation lights, etc. as well as the main wiring for the motor.

Finally, there is a choice as to whether the lagging should continue across the cockpit side of the dash bulkhead, or whether the bulkhead should remain unadorned. The cockpit below (left) shows a plain dash bulkhead blending in nicely with the rest of the work, whereas the example above shows a more finished look at the expense of extra weight. I presume that the dash boards in the traditional launches were not attached to ply bulkheads, but to a frame like the one below (right), which is an Andrews day launch. Nevertheless, it would be possible to fix a purely cosmetic thin board dash to the bulkhead to achieve the traditional look.

Traditional Slipper Launch (left) with a plain ply dash bulkhead, and Andrews day launch (right) with a board-on-frame dash.

Considering the thickness of the lagging boards, 9 mm., the shiplap is going to be far preferable to the tongue and groove, because it only bisects the thickness of the board, instead of trisecting it. The trouble with all of these types of joint is that the thin, rebated part tends to warp away from the surface, but, the thicker it is, the less likely that is to happen.

Because of that it is better to cut the cockpit side of the lap to be thicker than the hull side. It does not matter so much if the unseen hull side twists, but it is important to keep the cockpit side even. Therefore, a 5 mm./4 mm. distribution is preferable to a 4.5/4.5 split.

To break up the plain look of the shiplapped lagging, a bead can be cut onto the cockpit surface, but not on the thin rebated part, as that could weaken the joint. Rather, it should be situated over the other side of the board on its full thickness.

The shiplapped board with beading. It should not be beaded over the rebate, as shown above, but on the other side.

So, on to the mark out for the lagging frame: one of the costs of running the lagging parallel to the hull side, instead of perpendicular to the waterline (as is called for in the plan), is that you lose a good deal of beam at the sole level, especially near the driver's feet, where the hull is narrower than at the passengers' level. Measuring in 160 mm. from the hull onto the sole I came up with a line which appeared to come perilously close to constricting foot room too much.

The photo on the left has been enhanced to show the pencil line on the sole. It demonstrates how much leg room is being lost by the slope of the lagging.

Although I am not planning to use Lloyd Loom chairs in the boat, I dropped a similar one into the driver's position to check for fit. I seem to have only just enough room for a snug fit.

Next, I used the batten-on-the-carling procedure to check the newly marked line on the sole, and, to my surprise, I found that the extension of the inner surface of the carling to the sole did not coincide with the drawn line at all. In fact, the carling surface seems to be more steeply inclined than the hull at every point, except at the dash bulkhead level, with the result that the line of its intersection with the sole moves lateral to the drawn one.

The theoretical (solid) line of intersection of the lagging with the sole, and the actual
(broken) line as taken from the batten.

At the rear of the cockpit the actual line results in a gain of 40 mm. on each side, but as the dash is approached the theoretical line is correct again.

The broken line approaches the solid in the driver's area.

The outcome then is an increase in beam for the passengers, but not so much for the driver. But there is still enough room at the front of the chair for feet, and it is perfectly adequate for shoulders, as can be seen by the curve of the chair.

The front leg of the chair almost touches the line of the lagging.

So, the "actual" line on the starboard was reproduced on the port, and a trimming line was established for the remnants of the D bulkhead, which would allow the lagging boards to cross it.

Now, it should be borne in mind that the "actual" line under discussion is one plotted by holding the batten at a perpendicular angle to the waterline. If the lagging is to be put in on a diagonal, as I wish to do, the line will not be the same. As long as the diagonal slopes forward at sole level, (and, therefore, back at the carling), the line of intersection with the sole will move further laterally. This is because the bottom of the batten is at the same athwartships angle as its top, but at a more forward position on the sole. Hence, to widen the available leg room, you only have to increase the diagonal angle of the lagging. That, however, will come at the expense of the lagging and the coaming not lying in the same plane, unless the face of the carling is subsequently altered to parallel the lagging.

With the batten perpendicular the "actual" line (which is now solid) is touched, but when it is moved into a diagonal lie
the projected "actual" line moves outwards.

It also means that the lagging will not be truly parallel to the hull side, but that is not of major importance. The main consideration for now, is at what diagonal angle I really want to place the lagging, not only to plot a course for the cleat on the sole to which the lagging will be attached, but also to determine exactly how much of the bulkhead D remnant has to be removed. Currently is has been trimmed back to a point 8 mm. lateral to the existing "actual" line, but when the diagonal is introduced some more will need to go.

Also, I am keen to preserve a continuous plane from lagging to coaming. It seems likely that that will involve altering the inner face of the carling so that it slopes less inwards, so it will be a good idea at this stage to allow for a set-back of the lagging boards a bit more than the 9 mm. required for their thickness. That way, the face of the carlings can be planed off later without impinging on the lagging.

The first step in the actual construction of the lagging is to fix the cleats to the under surface of the carlings. Because they will be rather narrow I chose not to simply screw them up into the carling timbers, but, rather, to make them L shaped and fix them to both the under- and outer-surfaces of the carlings. That offered the advantage of giving somewhere to attach clamps to force the cleats into the twisting shape of the carling.

The cleat is glued onto the carling. Only the section behind bulkhead D is on at this stage.

And, while this method offers good clamping, it has the added benefit of providing a shelf on the outboard surface of the carling, which will be very useful later for the placement of backing blocks between the sheer clamp and the carling. These will be needed for the support of strain-bearing deck hardware, such as fairleads.
There is also a specific benefit for the section behind bulkhead D: in the original plan for a 20' launch, the arc from bulkhead E to D was only about 1300 mm. When the plan was altered to produce the 21'9" version which I am building the span was increased to 1600 mm. That section of the carling is quite springy, and the effect of an additional lamination from these cleats has been to stiffen it a bit. The area will eventually get further support from the decking and any blocks which go in there, but in the meantime it feels a lot better just from the attachment of these members.

A perfect spot to rest a backing block.

After the clamps come off there is a 20 mm. cleat hanging below the carlings, to which to attach the lagging boards. I took the opportunity to cut a piece of scrap for test fitting beside the rear seat, where the upper and lower cleats are now available.

Left, the cleat behind the remnant of bulkhead D, and right, the test fitting for the lagging.

It is gratifying to see that the scrap fits perfectly once an appropriate bevel is cut on its bottom end where it sits on the seat platform, inclined at 15°.

A good fit of the test scrap lagging board on the side of the rear seat platform.

Now, the rest of the carling cleats can be added, starting with the two sections in front of bulkhead D. Then the remaining section on the starboard side goes in, leaving only the third horizontal member yet to be positioned before the small lengths of lagging frame, which will complete the structure, can be placed on the rear seat front and on the dash bulkhead.

Port side behind the dash.

The complete port side cleat is shown (left), while the final starboard section is being glued on (right).

Both side cleats attached.

The attachment of the central ribband for the lagging involves fixing small blocks to the rear seat front, and to the dash bulkhead, and cutting a rebate in the remnant of bulkhead D.

Fixing blocks of the rear seat and the dash bulkhead, and the bulkhead D rebate.

The first of the laminations of the ribband is then inserted into position to check for fit, and angles are cut on the front and rear ends across its thickness to let it sit flush. A second lamination is cut to the same dimensions and then the whole lot is epoxy glued into the housings. The ribband is run parallel to the sole, which moves it away from the carling towards to front of its length, but this will allow deeper gloveboxes at the driver's level, and will be useful for the business end of the boat.

The first lamination of the central ribband.

The second lamination completes the port side ribband.

The lagging frame is continued over the dash bulkhead next, followed by the rear seat front.

The forward portion of the lagging frame is attached to the dash bulkhead in two sections.

The critical point in this ribband construction is the area of the rear seat, where the plane of the lagging must move in continuity from the side of the seat to the cockpit. It may be necessary, as it was in this case, to wedge the angle of the ribband a little while the glue is drying, to ensure that continuity.

Wedges hold the position of the ribband at the level of the rear seat.

The addition of the starboard ribband is a little more complex than the port because of the need to make it an exact mirror image, but it goes smoothly enough. The position of the cleat on the sole can now be checked again with scrap pieces cut to fit and angled at the desired degree of diagonality.

The first lamination of the starboard ribband (left), and the scrap lagging used to locate the sole cleat position (right).

Along with the second lamination of the ribband I decided to reinforce the cut end of the ply of the old bulkhead D. The ribband was not quite as tight a fit in the bulkhead D housing as it was on the port side, so needed to be held in position because of the weight of the clamps tending to twist it out. Two sprung battens do the job while the glue dries.

Incidentally, the cut end of D shown here is a glaring example of what should not occur in marine ply purporting to conform to BS1088. I has voids, and these are far from the first ones I have found. I gather it is a common problem; but it should not be.

These ones will naturally be able to be filled. Pity about the others!

With the addition of the starboard dash bulkhead frame the lagging structure is now complete, except for the actual lagging boards themselves, and the cleats on the rear seat half bulkhead and on the cockpit sole.

Before the lagging boards can be affixed to their frame the structures which sit behind them have to be put in. These include the conduits, discussed next, and the glove boxes. Then, the sole can be pulled up again, and the floors can be glued together. There will be a bit of fiddly work to level the floors, and some miscellaneous jobs, like limber holes and keel bolts to attend to, before the final sub-sole project, the bilge runners. They will be attended to next month, and then the sole can be screwed down permanently, allowing a shift in attention from the cockpit to the motor compartment.

Top of Page

46. Conduits for the Steering and Electrical Systems

For an inboard engine boat of the 7 metre mark the choice of steering systems seem to be between a hydraulic one and a rack and pinion type. The plans allows for a pulley and wire arrangement, or the Morse system if the extra expense is not a major factor.

I have decided to use a single cable push/pull system rather than pulley and wire, basically for simplicity of installation, so incorporated into the lining cavity of the cockpit are conduits for the passage of the cable, plus any electrical wiring which needs to reach the rear enclosed compartments or the passengers.