Deck camber

My brain isn't mathmatical in the slightest, so I had to work it out on paper. And its pretty nice to see two sources agree perfectly on the camber. I dont know if Dutch vessels varied but it will be period, and as you say, look much better than flat.

By the way, I think you would enjoy seeing a superb model of Golden Hind on ebay at the present, item number 180159392678. Its in a live auction located in Germany, and appears to be taken directly from the same plans used for Revell's version - I think this was Hoeckel? It also appears to be old, and with its German location, perhaps one of the earlier models built to his plans. Clicking on the images will get larger versions.

Several details are worth noting - I recall you had a question about the poop cabin extending above the line of railing - in this one, it does not, but just comes up to the railing. Also, the interior bulwarks are planked up to just above the gunports, with the remainder open, perhaps to serve as a visual shield (I remember you had a question about this too, or at least were working on a fitting solution). Overall there are several refinements in proportion worth noting, and the execution of the model is outstanding - there are even catheads carved into the catheads. If anyone would appreciate this model and its design, I think you would. Its truly wonderful.

I'm working on deck hatches now, to get them out of the way so I can get to planking. Then there will be bulkheads to make and so on.

Thanks again for the help with the camber.

Hmmm....My decidedly non-mathematical brain doesn't function at even its ordinarily modest capacity this late at night, but I think 1/2" for every foot of half-breadth does indeed equal 1/4" for every foot of breadth.  As I said earlier, I'm pretty sure there are other rules for different nations at different times, but that ratio does seem to be remarkably consistent for ships of the western tradition.  Or perhaps I should say, models and drawings using that ratio seem to look right.

The Nautical Research Journal is a terrific publication, but members of the NRG do occasionally go a little bit overboard in their enthusiasm.

That's a great idea, thanks! I measured it out as suggested and the rise is subtle but definitely present.

I finally found the camber formula in Brian Lavery's book on Susan Constant, which confirms your suggestion: according to Thomas Harriot's ca. 1608 work, the camber is 1/2 inch for every foot of half breadth, or 5 3/4 inches for the lower deck of Lavery's reconstruction of Susan Constant. (The Susan Constant's lower deck scales out to just under 22 feet at widest, using an O scale ruler on the 1:48 hull section drawings.)

I think the next step is to make a test cross section of the deck to make sure the method I'm thinking of will fit the camber. I'm using thin scribed sheets of basswood, cut to shape using the template in the kit. The scribing runs lenthwise, allowing the sheet to curve into the camber and also providing a visual reference to the planking, so they will run straight. The sheets are narrow, so the center joint will be supported below by an additional sheet, which turns out to be close to the 5" scale height. A cross section a few mm in width, with planking, should show whether this will maintain the camber, and how high the deck will be with the various layers. I'll have to make the hatches, mast step, etc., before I cut the sheet for final installation, so it will fit over them when installed.

The quarterdeck will be where the beams come in. The 1909 replica, like Anderson's 1926 Mayflower, has a very short quarterdeck. It extends out from the cabin below, but not by much. A few old Dutch paintings do show this, mostly the work of Adam Willarts, whose paintings of the 1620s seem to have strongly influenced the 1909 reconstruction. However the majority of paintings, and more importantly, earlier ones, show extended quarterdecks, more like William Baker's reconstruction of Mayflower, and the modern replica of Duyfkin. The latter two have an open bulkhead here, contemporary artworks also show a solid one with two doors or openings. To support this, I'll have to build up all the stanchions (since the hull is reformed, the kit's pear shape stanchions are useless) so the arched deck beams will come into play.

The kit comes with very wide planking strips, annoyingly in fine walnut or something like that - just the thing you would not expect for a realistic deck. I bought some Mt. Albert scale lumber last night in 1x8" S scale (just now I find that Lavery recommends 12" planking, argh, so my deck planks are now slightly too narrow) and plan on using a thinned alcohol and india ink wash to get the greyed look of a deck. (From reading posts here, it seems that decks were not holystoned at this time.)

After reading a few issues of Ships in Scale, the camber story is very believable.

He demonstrated, using calculus (which is completely beyond my capacity) that the difference between a deck beam "cambered" in a circle arc and one "cambered" in a parabolic curve would amount to about .002" in the real ship.

   And, after a couple of licks with a holy stone even that difference vanished.

Bear in mind that none of this is chiseled in granite.  Some old artwork suggests that higher decks on old ships did indeed have steeper camber than the main decks did.  But 1/4" of rise per foot of maximum beam seems to be a pretty safe generalization - a good figure to fall back on if you don't have any evidence to the contrary.

A good, simple way to make a template for cambering deck beams is:  1.  Draw a straight line on a piece of paper.  2.  Make two marks on it, representing the maximum beam of the ship.  3.  Bisect that distance, and mark the center.  4.  Figure out the distance between the height of the deck at the ship's side and the height of the deck at the centerline (i.e., 1/4" for every foot of the ship's maximum beam).  Make a mark at the center of your line that distance away from it.  5.  Take a thin piece of reasonably flexible wood (plastic strip would work, too), lay it on the paper, and flex it so that it touches all three points.  6.  While holding both ends and the middle of the wood strip in position, trace along the edge of it with a pencil.  You now have a template for every beam in the ship.  Just be sure to keep the center point lined up.

There's another, clever way to do it if you have a band saw - and you're working on a fairly large scale.  This trick isn't original to me; it comes from an old issue of Model Shipwright.  The guy who wrote the article was more comfortable with trigonometry than I am; he established that, if the camber does indeed amount to 1/4" for every foot of the ship's maximum beam, the top of the deck beam is a circle arc with a radius six times the ship's maximum beam.  So he built a simple jig that would hold the timber he was using to make his deck beams exactly that distance away from the blade of his bandsaw, mounted in such a way that it could swing in a horizontal arc.  By swinging his deck beam timbers through the blade, he cut all of them to the proper camber in a few minutes.

Quite a few years ago there was an interesting debate in the pages of the Nautical Research Journal about how to plot the exact, authentic curvature of a deck beam's camber.  It seems that various sources from the seventeenth, eighteenth, and nineteenth centuries offer different formulas for doing it - some saying the camber should be an arc of a circle, some saying it should be a parabolic curve, etc.  The debate on the "letters to the editor" pages got rather emotional, until it got halted decisively by a ship modeler who was also a professor of mathematics.  He demonstrated, using calculus (which is completely beyond my capacity) that the difference between a deck beam "cambered" in a circle arc and one "cambered" in a parabolic curve would amount to about .002" in the real ship.  No more was heard from either side of the argument after that.

Thank you, I appreciate the help and guideline.

Using this formula, a deck width of about 18 scale feet, which would allow for a 4 1/2" rise, a deck widthe of 20 scale feet would allow for a 5" rise and so on. Its interesting that the camber diminished with narrower decks at the bow and stern - the poop cabin deck is about eleven scale feet at its widest, so a 2 3/4" camber would be in place. I'd just assumed the same rise would be in place, but now I see how this would make for a very steep narrow poop.

The main deck template is cut in two halves, and the stiff wood gives a peaked boxcar roof effect. I copied the pattern onto thin scribed wood and plan to use that as the base for the planking, so I can get a softer curve. The scribing provides a reference line for planking, so it will go on straight. Your suggestion for a template to determine the rise in bulkheads is extremely helpful, thanks again.

The Billing model is pretty bad, but it is getting me to scratch build just about everything, which is proving to be good. Thanks again, Jim

So far as I know, there is (as usual) no firm written evidence about deck camber from this period.  (There is, in fact, scarcely any firm written evidence about any details of shipbuilding practice prior to the mid-seventeenth century.)  The few extant sources (e.g., the Matthew Baker manuscript) do make it pretty clear, though, that deck camber was indeed a normal feature of ships of the period.

Actually deck camber seems - with plenty of exceptions - to have been a fairly consistent feature throughout the sailing ship period, and even later.  If you don't have any evidence to the contrary, it's fairly safe to assume the deck camber amounted to about 1/4" of rise for every foot of the ship's beam.  That ratio doesn't vary with the scale.  (Think about it and that will be obvious.)  So for a model with a beam of 4", the deck at the point of the greatest beam would be about 1/12" higher at the centerline than at the side.  The amount of camber obviously gets less as the hull tapers at the bow and stern.  The usual way to determine it is to make one template for the deck beam at the point of maximum beam (i.e., a circle arc with its center point 1/12" higher than its ends), and use that same template for all the deck beams. 

There's no guarantee that this system will produce an absolutely authentic result, but it should be pretty close.  And yes, though the eye may not detect that difference of 1/12", the eye probably would notice if the deck was perfectly flat.

Hope that helps a little.  Good luck.