Wednesday, March 30, 2011

Three Bolger boats under construction at once

Here are a couple of photos showing three Phil Bolger-designed boats under construction in our old shed, all at the one time (there were three other boats being built as well!) To paraphrase somebody else, "I can remember doing it, but I can't remember how!"

The boats in the photo are two of Phil's Hope design (see last posting) and a Harbinger. Phil Bolger has often been associated with what some people call "Bolger Boxes". Well, the boxes are works of art in their own right, and demonstrate a rare understanding of hydrodynamics, allied with the application of common sense - but none of the three boats shown here could be called boxy!

Two examples of Hope (foreground and upper-left) and one example of Harbinger (upper right)
Here you can see the wooden pattern I made to allow for the casting of a bronze "Y" shaped propeller strut.
The shapely bow of one of the Hope designs in slings. The cuddy-cabin design was mine, as was the clinker construction plan - but Phil Bolger was consulted in writing and approved of the alterations.

Saturday, March 26, 2011

More Interesting Work from the Workshop circa 2004

Around 2004 we were building at least six boats at one time in the old workshop - with only three-and-a-half people on the job. It was a hectic time, and looking back on it, we did some creative stuff. At one point a customer decided that his outboard-powered boat (a Phil Bolger Hope which we were building with a lapstrake hull and a small cuddy cabin) would be better as a diesel in-board, even though she was nearly finished in the outboard configuration.. I quickly designed an inovative skeg arrangement, and we carried out substantial structural modifications to accept the Yanmar 1GM10 deisel.

The nature of the retro-fitted keel/skeg was such that I thought it needed extra support against side loads and so I made a pattern for a cast bronze propeller shaft strut which had an inverted "V" on the upper side of the bearing. Not an original thought, but my first attempt at pattern making, which had been made necessary by the extreme urgency of the situation. The foundryman was a marine specialist (Marine Castings and Manufacturing - still in business) and was (to my surprise) happy with the pattern. Not only did he cast the prop strut, but he cast and machined the propeller, and the water-injected dripless-seal. Everything worked!

Propeller and strut, loose fitted without the proper pads etc - but still looking good

12" x 10" propeller from memory.
Custom-made dripless-seal
Boat on launching night - secret launching before the official one to come the next day!

Me, looking very relieved after the owner had gone home from the secret launching. Now I could get back to business on the details (and have a quiet beer, I think)

Launching Day (Official) for the in-board boat

The outboard boat awaiting the owner's arrival for the final inspection.

Sunday, March 20, 2011

A Dory for Gazela Primeiro (and at the end, tips on glass application)

A while ago I received a nice email from Tony Souza regarding construction issues with a new dory he intended building for the old Portuguese barkentine Gazela Primeiro.

Hello Ross,
Thanks for your recent article "Stems and Trailers".  In that article you discuss natural timber vs plywood construction and show the merits of plywood glued lapstrake for boats that are dry stored.
I'm trying to choose a construction technique to use for a reproduction dory that will be "dry sailed".  The dory is a Grand Banks dory of 17 feet overall, 13 feet on the floor. The new dory has to resemble closely the last of the three dozen Portuguese grand banks dories that sat on the deck of Gazela Primeiro, the 100+ year old Portuguese barkentine on which I am one of the all volunteer crew.  Last winter we patched and repaired the last dory, and this year I'd like to build a dory to be used often for the teaching of boat handling, rowing, and dory sailing.  the boat handling includes launching and retrieving over the side using hand operated boatfalls. The new dory will spend most of its time stacked inside the old dory, the two as deck displays of what was used in long line fishing of the early 20th century.
As a matter of resemblance the new dory will have four frame sisters overlapped, removable thwarts, 5/8" cedar planking, solid gunwales, a mast step on the floor and two sets of oarlocks.
For ease of construction and in keeping with the dry storage idea I plan on using modern materials where I can.  The flat bottom will be Meranti plywood already on hand, no need to soak the floor seams closed before putting boat into the water.  The frames are already built from laminated ash. Tree crotch and buttress sections are too difficult to find.  I plan to epoxy the garboard plank to the bottom, glass tape and epoxy the inside seam.  After planking I plan to turn the boat over and glass the bottom and garboard with cloth.
After that long prologue we come to my uncertainty area, the clinker planking joints.  It's tempting to use lap-stitch aka ship lap joints, easily cut with a router and epoxy fastened.  On natural, i.e. not plywood, timber that may lead to splitting along the grain at the join.  Likewise dory lapped (rolling bevel) joints likewise glued might suffer splitting.  I should have said earlier that using modern glues and eliminating metal fastners is a goal.  The old dory is a pincushion of steel nails on the planking joints which has resulted in rusty streaks and rotten wood.  (Of course dories were not supposed to last more than a couple years in ocean service. Old dory is probably 40+ year's old.) 
Would it be better to caulk the planking seams with Boat Life, a polysulfide, rather than epoxy fasten?  The frames are there to give cross grain strength.  With your expeience you might see that I'm headed for problems not yet envisioned.
Any suggestions will be very welcome and gratefully accepted.
Gazela can be seen at
A picture of the old dory is attached. 

Here is part of the text of my reply: -

I agree totally with your concerns regarding "hard" gluing of either "lapstitch" or dory-lap joints when using natural timber rather than plywood - I believe that they would definitely crack at the point where the planking thickness returned to single-plank. Harry Bryan wrote about the matter when he did a couple of articles for WoodebBoat Magazine about the building of his Daisy design. She had either double-planked cedar (glued with epoxy) for the bottom, or a single sheet of plywood. The topsides were planked with lapstrake cedar planks in the normal manner (I think using copper clench nailing from memory). The important thing in our context is that he expected her to be dry-sailed, and after completing the planking in the normal way, he dragged a sharp, flat-blade screw driver along the underside of the lap on the outer side of the planking. This produced a sort of square-shaped groove of about 1/8" x 1/8" in the underside of the lap. He then filled the groove with a bead of polyurethane (I would use 3M 5200 or Sikaflex 292, using a polyurethane-specific primer)

Having said all that, I wonder whether you could just glue the entire lap with polyurethane (I don't know enough about polysulfide to say anything about its adhesive qualities where no fastenings are used). The polyurethane is strong enough, but I don't know whether a fully-glued lap would allow enough movement to overcome the cracking problem, even given the flexibility of the compound - but my guess is that it would be ok, especially if the glue-line was thick.

If I was doing it, I'd go the copper-fastened route, with the polyurethane (or polysulphide) run into the groove under the lap.

I think that the garboard arrangement you propose would be fine. The glass on the outer face should supply cross-grain reinforcement, and the tape on the inside will extend some distance beyond the vulnerable line of intersection between the inside faces of the bottom and the garboard.

On two boats I built (one of which ended up in the Jody Foster film, Nim's Island - see the recent comment made on Duckworks ), I made the bottom from Western Red Cedar glued strip, continuing around the turn of bilge, after which I continued the planking as plywood glued lapstrake. I cut the bevel on the bottom planking to accept the first lapstrake plank before glassing the bottom. When glassing the bottom, I continued the glass right around and onto the bevelled face. This meant that when the first plywood plank was glued on, the glass was sandwiched between the WRC strip planking and the inner face of the plywood - the aim being to prevent splitting in the WRC.  This explanation is a bit clumsy, but I haven't got time right now to do a sketch. In your case this will not be necessary, as you will have the glass on the outside of the cedar planking anyway.

Western Red Cedar bottom planking

Bottom covered with 400gsm (12oz) double-bias glass and the first two planks of plywood lapstrake attached - the first plank going over the glass, which itself had been laid over the bevelled edge of the cedar bottom planking.
Here is Tony's latest up-date: -

Hiya Ross,
Here are three pix of the dory project to date.  This boat is a 'copy' of the last original belonging to Portuguese fishing barkentine Gazela Primeiro.  You can see more pictures on and my facebook page. My version of the dory has a meranti plywood bottom and white cedar planking over laminated ash frames.  The gunwale and cap are white oak. The dory will essentially be dry sailed i.e. living on Gazela's deck most of the time and occasionally be used for crew training and exhibition at home and ports of call.
After corresponding with you some time ago I took your advice and epoxy joined only the garboard plank to the bottom.  The rest of the construction has followed traditional methods. The upper edge of the garboard and the remaining planks are dory lapped and copper riveted.  Planks to frames are joined with Si bronze screws.
I think we talked about covering the bottom and up to the upper edge of the garboard with synthetic cloth and epoxy. 
Now areas where I could use advice:
What cloth would be appropriate? I don't think the dory will ever see a beach, but it might. So a light cloth should do.
Is there an advantage to graphite additive to the epoxy?
Any tricks to applying the cloth?
Thanks for your help,
Tony Souza

And part of my reply: -

Dear Tony,

Thanks very much indeed for the up-date, and for the nice photos. The boat looks super to my eyes, and I'd love to take her out in the rough stuff - preferrebly with some weight in the bottom to represent the ballasting effect of a load of Cod!

My appologies for the delayed reply - we have been recovering from the devastating flash-flood which wiped us out on January 10, and priorities tend to be re-arranged!

For the cloth, the simplest would be 200gsm (6oz) woven glass. This is light and easy to use, and gives a good level of protection, and more importantly, provides a sort of screed to ensure an even thickness of epoxy. Dynel is also a good option (4oz, I think, but you would need to check with the supplier). Dynel is bulkier than glass for the same weight, and it has superior abrasion resistance - I like it on decks - but it doesn't have the tensile stiffness of glass so it will have less structural effect on the garboard-to-bottom joint. For a hefty boat like yours which may get handled roughly, I'd give serious thought to using 400gsm (12oz) double-bias glass cloth. It isnt woven, so it takes bends fairly well, and with the fibres aligned 45 degrees/45 degrees, every fibre crosses the longitudinal joints. It is cheaper than woven glass, but is heavier than the normal 6oz stuff.

I normally use the 'dry' application method (for weights up to 12oz). I lay the cloth over the dry, sanded, and vacuumed surface, and them smooth it into position with a dustpan brush or a wide, dry paintbrush. Tape any troublesome edges down with temporary bits of masking tape. Then start by mixing small quantities of epoxy and pour them onto the glass (or Dynel) and spread them with a squeegee - I use rectangles of 1/16" model aircraft balsa as they can be bent along the grain if required, and the corners don't snag on the glass.  Don't press too hard, as you will end up aerating the epoxy and making it go creamy with minute air bubbles (just like what happens when whipping cream). Just use a gentle figure-eight sweeping motion to get it out onto the surface. Don't fuss about getting it to wet-out - that will happen automatically. Keep on mixing, pouring and spreading until the entire surface is covered. Small batches are good, as they don't heat up so rapidly in the container.

When the surface is covered, use disposable brushes and/or disposable foam rollers with about a 1/8" nap to distribute the resin evenly. I use dry brushes and rollers - they pick-up from the excessivcely wet areas and put down in the dry areas. When all is even, use the squeegees again (held at about 45 degrees to the surface) to scrape off excess resin so you end up with just the wet cloth, but no visible pooling of liquid resin.

After the epoxy has gone off enough to be certain that the cloth won't float up off the surface - this depends on temperature and rate of cure - but when it gets to a "green"state of cure, lay on several more coats of epoxy to fill the weave so that when you finally sand the surface (after removing any amine blush with water and cloths, sponges or Scotchbrite pads) you are only sanding epoxy and not going through to the glass.

Have a look on my website under the button labelled "First Mate Photos 3" for a brief pictorial demonstration - the thumbnails enlarge if you click on them.

Friday, March 18, 2011

Using Peel-Ply

There are two significant problems when laying up a lamination of epoxy and a reinforcing fabric.

1. There are points where you transition from a single layer of cloth to two (or more layers), or where a single layer of cloth transitions to bare timber, (such as at the edge of a taped-seam), tend to be ragged and the edge, or selvege, of the cloth lifts up to form a ridge which is difficult to sand and fair;

2. The surface of the cloth is usually uneven and requires further filling with epoxy resin/hardener mix. The problem here, apart from time and labour, is that you need to ensure that subsequent layers are applied while the preceding coats are still chemically active. If not, you must remove amine blush and then sand the surface to provide a keyed surface suitable for a mechanical (rather than chemical) bond - commonly referred to as a secondary bond (rather than a primary bond).

In many circumstances, you can get around all of these issues by applying a layer of "Peel-Ply" over the epoxy/fabric layer while it is still wet. The peel-ply is simply a fine-weave fabric - a bit like shower curtain - which allows epoxy to wick through but does not adhere strongly to the wet epoxy. When the epoxy cures, the peel-ply is 'peeled' off to reveal a nicely filled fabric (due to the magic of capiliary attraction or surface tension, or something like that!). In the process, the amine blush is removed and the resulting surface is ready for the next coat - or it can be lightly sanded and then painted.

Here are a few pictures of a small job I did yesterday and the day before. It involved the application of three staggered layers of uni-directional carbon-fibre over the outer surface of a snotter attachment cleat on a carbon mast I'm setting up for a customer - it could have been anything, but this little job demonstrates the process quite well.

The bare timber (Celery-Top Pine) snotter cleat epoxied to the forward face of the mast. The carbon-fibre mast had been sanded on the mating surface to provide a good gluing bond.

Three overlapping layers of carbon-fibre laid over the outer face of the timber (done to strengthen the cleat against the tension forces imposed by the snotter). The carbon was laid in WEST System epoxy, and laid up so as to go from a single layer at the outer ends, to a double layer, and then a triple layer in the middle. Over this combination was laid a roughly-cut strip of peel-ply.  

Next morning I peeled off the peel-ply reveal a nice, well-filled surface

...which only needed to have the edges trimmed with a spokeshave or a sanding block.
 Try some peel-ply (or shower curtain) on your next job!

Wednesday, March 16, 2011

Going Professional

Back on 3 July 2000, I drove out of the staff carpark at the Brisbane Air Traffic Control Centre for what I believed was the very last time. I had been working as a controller continuously for twenty-five years, and things were about to change. My decision to get out of the job at age 45 years of age had been planned, as I wanted to follow my passion for the building and designing of small wooden craft.

What followed was an odyssey of business risk, independent learning, support from friends, seven-day work weeks, and formal qualification. The first five years were the most intense, but after an interlude caused by a serious heart illness and associated surgery, I've been back into fairly constant work. These days most of my workload is made up of design projects and answering questions from email enquirers. My good friend Paul Hernes (builder of the first Phoenix III) tells me to, "Keep Smiling", and most of the time that is what I'm doing.

Here are a few random photos from the early days in the workshops.


First Boat after going full-time - a Green Island 15 designed by Mike Roberts. I was building a number of other boats at the time, being helped by Allen Danvers

Here she is on an early sail - son David at the helm, and Steve on the rail.

This is a very serious and young me starting the layout of an Iain Oughtred Whilly Boat. My current workshop is not as tidy!

Not long after, we were building an Oughtred Tammie Norrie and a Phil Bolger Micro at the same time. In-between times we had built a 24ft Oxford Punt of my design, a Surf Ski also of my design, and had finished the Whilly Boat. The dapper-looking fellow in the tie is David Neumann, the Hempel Paint rep.

Here is the finished Whilly Boat

My first commercial design - the paddle ski

Customers were friendly, and brought in lunch!

Boat Cupboard designed and built for a local high-class restaurant. Can still be seen in Manly, Queensland

Five boats - four 18 footers and one 22 footer for the movie, "Diamond of Jeru" starring Billy Zane. We built these (fully functional with buoyancy tanks and provision for outboards) in 20 days of 5am to 12pm work. That is Allen and me feeling very exhausted. The boats lasted so well that they were sold after the filming was complete. We also built two Phil Bolger Payson Pirogues for the movie at the same time!

Jeru boat on the movie set - this is one of the small ones

Here I am at nearly fifty-seven with fifty-five boats under my belt, a tolerant wife, and about three glasses of red wine inside my belly. This is what full-time boat building does to you folks!

That is probably too many photos for one posting, but if you are interested I'll put up the odd interesting photo from time to time. We did some interesting stuff (and still do) in those days, like making patterns for complex bronze castings etc. Let me know if you want to see more.

Monday, March 14, 2011

Harbinger and the High-Peaked Gaff

It is with direct relevance to two of my recent postings (High-peaked Gaffs and Building Phil Bolger's Harbinger) that I show the picture below.

This shot is of Harbinger just moments after pushing-off from the launching. It may appear that the motor is dragging, but in fact it is just ticking over (Honda 2hp Air-cooled 4-stroke) so as to be ready for emergency use in getting out of the marina. That is me you can see standing at the tiller, concentrating intensely - both of the owners were on-board and I was unfamiliar with the boat's handling characteristics.

The point is that the photo shows two things - Harbinger is a beautiful design, and more importantly, you can see how the combination of a short hoist (luff) and a long, high-peaked head gives one a really versatile rig. Even in the light conditions, it is apparent that the gaff is bending nicely under the tension of the peak halyard. I guess that it could have been released a little to get more draft into the sail, but we didn't know what conditions would be like outside, and the sail and lacing were all brand-new and unstretched. For the record, we didn't need the engine, and the boat sailed superbly - just look at the curl of water coming off the forefoot!

Thursday, March 10, 2011

High-peaked Gaffs

Some of you may have noticed that all of my gaff sail plans have a short hoist (luff) and a long, high-peaked gaff. To many, this is a somewhat strange-looking shape compared with the more commonly seen longer luff, shorter and lower peaked gaff.

Pegasus 18.5

So, why do I do it that way?

The main reason is to keep the mast as short as possible. This may not be such an issue in a boat which sits on a mooring, but for a trailer boat, every inch you can remove from the length of the mast is important. Added length in the gaff means reduced length in the mast - although there are diminishing returns, as the peak halyard needs to intersect the gaff at as close to a right-angle as practical. This is so that the gaff stands well, with minimum sag, when under a weight of wind - in the process reducing twist in the sail. The result is that the mast needs to project quite a distance above the gaff jaws, and a quick glance through your boat books will show that many designers and builders make the mast too short above the jaws, and the halyard then works at a very inefficient angle. This is one of the reasons that gaff rigs have a bad name for windward work.

The second reason for my use of long, high-peaked gaffs is that the light, flexible gaff tends to bow around the halyard attachment point when the wind gusts, automatically flattening the sail at just the right time. Also, by tweaking the peak halyard, you can produce very effective sail shape adjustments when changing point-of-sail, or during varying wind conditions. The high-tech racing boats spend hundreds of thousands doing this with carbon-fibre and stainless steel - here we can do it (in a cruder, but effective way) using home-made wooden spars and a bit of line.

I frequently draw a tiny, short-luffed jib for these rigs. The short luff and small sail area means that the little sails stand well, even though the masts have no shrouds or backstays of any sort. However, the effect of the jib is remarkable, as it smooths the airflow around the section of the sail attached to the bulky , turbulence-inducing mast. The jibs are so small that the rig can be used with, or without the jib, and will still balance well.

Saturday, March 5, 2011

More on the Centreboard Shape Issue

I've been working on the possibility of producing an altered internal layout for both Phoenix III and First Mate, but as always seems to be the case with designs, it is better to go with an entirely different boat, with the desired element designed-in from the start.

Before the flood did its job on us, I had been working privately on a new design for my own use. She was to be the same size as Phoenix III and First Mate but was to be slightly higher in freeboard, have a rectangular centreboard to allow a forward rowing station in addition to the midships thwart, and to be of a shape which is suitable for both glued-lapstrake and strip-planking. The rigs from the original two boats would work, but I wanted to use an un-stayed gaff-headed cat with a small (optional) jib set flying.

This hull shape has less bearing aft than the other two, so it won't be necessary to pay lots of attention to sailing her flat - she should balance quite well at significant angles of heel. In fairness, Phoenix III is pretty good in this regard anyway.

Work was interupted by the floods, but here is a rough drawing for the sake of interest.

The boat is shown with a vertical transom in the drawing, and with five planks of lapstrake or batten-seam (or stitch-and-glue) planking. The final design (if ever finished) might well have a raked transon - say 10 degrees - and would also has a set of lines showing a completely round hull for those who may wish to build strip-planked, cold-moulded, or lapstrake with a larger number of planks. The cut-away forefoot is to make planking with wide strakes easier.

Dimensions to the inside of the planking are 4610mm/15'1-1/2" length x 1373mm/4'6" breadth and a total of 104 sq.ft of sail including the flying jib.

Tuesday, March 1, 2011

Centreboard Case Shape

I try to think carefully about every single element which goes into each of my designs, and a good example is the centreboard shape in Phoenix III and First Mate.

When I was first designing Phoenix III, a major consideration was proper rowing geometry. Rowing is a wonderful way of providing auxilliary power, but it is very important to get the rowing geometry correct - done well and the rowing will be a pleasure, but if it is wrong the rowing will be an unpleasant effort.
There are a number of considerations, but the two most important are: - 1. The horizontal distance from the aft edge of the rowing thwart to the oarlocks, and 2. The vertical distance of the oarlocks above the rowing thwart.

Because these are relatively small boats, there is not a lot of vertical distance available under the rowing thwart, which limits the width (or chord) of the centreboard. Not only does this mean that board area is is reduced, but it also means that the "bury"of the board in the case is very small when the board is lowered. This means that the board and the case are heavily stressed by the side loads when sailing.

My solution to the problem was to make the board and case wider at the forward end - something which used to be common in working boats such as Swampscott Dories. The end result is a board which has much more area than the rectangular one, and it has a lot more "bury" in the case when lowered. Also, the open part of the top of the case is higher in the boat, which helps keep things dry. 

The only real disadvantage with this system is that it makes it impossible to have a thwart for a forward rowing station, as the oars would strike the centreboard case. The forward rowing station is important if you want the boat to trim properly when rowing with a passenger on the aft seat (stern sheets), but I judged that Phoenix III and First Mate were primarily sailing boats, and adequate centreboard area is very important for good sailing performance to windward.

However, I've received quite a few requests from plans buyers who really want a forward thwart, and so I've started investigating the possibility of putting in a rectangular board. It will have to be wider, and therefore the rowing thwarts will have to be higher. This in turn means that the oarlocks will have to be mounted higher. But I think it is workable.
As you can see, the rectangular board is still a little smaller in area, but I might be able to fiddle around with the length to make up the difference.

Regarding the underwater shape of the original board, I ran a spline through the apexes of the angular outline, and you can see how it approximates the shape of a shark's pectoral fin. Probably a good structural/performance compromise.