Deck Crowns

[Sayla]

Question - is the deck crown always produced as a centreline part of a template set out to the widest beam, or is a ratio used to re-jig, as the beam changes - i.e. a parabola based on a ratio is a different curve to a portion of another parabola

The curve of a ten foot width of the centre (top section of the parabolic curve) of a template set for a crown of 1' in 20', is a different curve from a crown that is 1/2' in 10'.

So does everyone use the upper part of the same (max beam) template along a deck?

sayla

[Peerie Maa]

There may not be one right answer, as the appearance will be effected by the amount of crown, the amount of shear, and the fullness or fineness of the deck in plan. With a motorboat that has a flat shear and full deck line the centre line of the deck in profile can have a nasty hump, or if the cl profile is straightened, then the shear can start to kick up towards the stem. You need to sort this out on the loft floor and the result will depend on the type of boat.

[Ian McColgin]

There are some very complex methods for getting this right, ranging from the mathematically silly to extensive use of shims. I don't get the WB magazine's new search system, but there was a great article and a fairly straightforward way that involves using two crossed beams slid from one side to the other - it comes down to two cords with two points fixed and the third moving - to describe a true arc. It's easy enough that you can do it to mark the curve every few deck frames so they are all individually correct, each with its own arc.

[wizbang 13]

Peter Spronk (catamaran genius )taught me the system Ian mentioned, had to make a different sliding jig and a different shape for each deck beam for an old Island sloop I modified. This was 1976.
At the time I thought it was cool, any progress was good progress, and boat building was magic.
Then I built a boat that used the same mold for all the deck beams. Things went from slow and magical to practical, fast and cheap.
I still do not understand it,( all math turns to silly inside my head).

[Ian McColgin]

Wizban13 is right that some hull shapes - the combination of sheer and changing beam - work out that you can lay a fair or nearly fair deck across deck beams that have the same arc or perhaps a curve mathematically different from a true arc. Usually there's at least a little planing or shimming to get each deck plank to lay fairly on each beam but it's not bad. It also often happens that the placement of the trunk cabin eliminates that part of the deck camber that would have been too approxomate.

All this can be figured out during lofting.

[TerryLL]

Here's a pic of the method Ian mentioned. This set-up will give you a true arc without doing the math.



A complete explanation can be found in Boat Building In Your Own Back Yard by S.S. Rabl.

[Ian McColgin]

Thank you TerryLL.

[Gib Etheridge]

I used a changing radius here in order to have a straight king plank, but I suspect that a uniform radius, which would have been much less work, would have been just fine.

Here is the formula for determining the radius of an arc when you know the width and the height.

http://www.mathopenref.com/arcradius.html

[TerryLL]

I used a changing radius here in order to have a straight king plank...

Same here with the end decks on a Caledonia Yawl. A straight king plank required every deck beam to be at a different camber. The deck turns out to be a conic section rather than a cylindrical section, both of which are easily planked in plywood.

That's a good looking build you've go going there Gib.

[Jones]

Peter Spronk (catamaran genius )taught me the system Ian mentioned, had to make a different sliding jig and a different shape for each deck beam for an old Island sloop I modified. This was 1976.
At the time I thought it was cool, any progress was good progress, and boat building was magic.
Then I built a boat that used the same mold for all the deck beams. Things went from slow and magical to practical, fast and cheap.
I still do not understand it,( all math turns to silly inside my head).

Happily I found this thread, as I've been pondering the matter in preparation for building my first boat. So far, I only have Boatbuilding Manual by Steward to go by. Oddly, it glosses over the subject of all the rest of the beams except for the above illustrated layout for the widest beam. My carpenter sense took me to the same arc for all the beams; as the width decreases toward the ends of the boat, the arc height decreases and things should be fair.
But my plans are for a sailboat with some shear to the deckline, which complicates things, so I gravitated to the simple solution as making the most sense. I even contemplated putting beams in where the cabin will be, then cutting them out for the headers to make the header layout simpler (the carpenter again). Then use the cut out pieces for the fore and aft decks. It's a small boat, so the waste wouldn't be much.
I thought the book would cover it. I was surprised that it didn't, being that it goes into exhaustive detail on the hull.

But that's why I'm here--to ask the gurus. If anyone knows of a book that covers it, I'd be interested...then there's the cabin top...

[Sayla]

Thanks for the input guys - there's 2 dog houses in this deck, so the visuals are fairly broken up
I think I'll go with the consensus of H. Chapelle and R. Parker and use the top of the same parabolic curve set out for the widest beam, and moved vertically on centreline until top of template curve hits sheer / deck line, rather than have a consistently changing parabolic curve along the deck (even though the computer seems willing to do it).
Using the 'same template' method produces a flatter curve fore and aft (at bow and stern) than if the crown ratio were reworked, which is probably really good for walking on decks - R. Parker writes this is also good for side decks and hence the original selection of this curve form.

I believe this setout produces a basic parabolic curve type, and as does the diagram from TerryL (see below)

Chapelle (in Boatbuilding) gives the parabolic curve setout



And R. Parker the same, with commentary





A mathematical view of the two stick method

[wizbang 13]

The above illustration (#6) is missing the third stick , the anti sway brace , that it does not change shape while moving it.

[TerryLL]

The above illustration (#6) is missing the third stick , the anti sway brace , that it does not change shape while moving it.

The rigid connection between the two slats is enough to retain the shape. I've used this set up as drawn numerous times without any problem. If you want to add some extra bracing for peace of mind, then go right ahead.

[Ian McColgin]

I don't quite understand why a single arc is faster than scribing each one - no matter what you have to mark the arc and cut it out beam by beam.

I do understand that the single arc and other approxomation methods will often give a 'close enough' fit that the deck planks can be forced on with only occasional shims or even none since the bobbles are cured by fairing the deck after. It's most common in production boats as one can see if one looks closely and in detail. Marmalade, for example, has three places (two starboard and one port) where a little shim was needed to get the covering board to lay fairly.

[wizbang 13]

Fer me , deck beams have always been laminated, so it was more about one mold.

[Canoeyawl]

I have used this layout a number of times, a similar layout works for masts...

[mcdenny]

If you want to know what the deck will look like before building it, loft it first. You can then pick up the hgt and len of crown and make the correct radius beams to create the king plank profile you want.

[Jones]

I have used this layout a number of times, a similar layout works for masts...

I just got Boatbuilding (Howard I. Chapelle), out of the library (from which the above iluustration comes). It confirms the method I arrived at independently.

Place the pattern on the body plan at each station with the center aligned and the curve touching each deck line (make sure the bottom of the pattern is at right angle to the centerline). That gives the full size patern for each beam.

If the deck line is fair in the plan view, and the shear line is fair in the profile view, the deck should be fair as it progresses down the length of the boat.

Hovever, the sailboat plans I have show a steadily decreasing crown to the cabin top as it progresses forward, without an appreciable decrease in beam. I'm thinking a constant ratio of decrease and a pattern for each beam may be necessary here(?)

I haven't haven't been able to find that part in the book yet.

[Daniel Noyes]

If you have a large (especially full length) deck having a straight king plank will look ridiclious in most cases, it is necessaryu for the crown of the deck to echo the sheer line of the hull rather than being ruler straight. The best way to do this is cut a pattern for the camber you desire at the widest point, then use that pattern to cut all deck beams, any thing else is busy work.

Eg. camber follows curve of sheer (my own design)

[Jlaup]

If you have a large (especially full length) deck having a straight king plank will look ridiclious in most cases, it is necessaryu for the crown of the deck to echo the sheer line of the hull rather than being ruler straight. The best way to do this is cut a pattern for the camber you desire at the widest point, then use that pattern to cut all deck beams, any thing else is busy work.

I did the same on the Indian and a) like the look of the up swept arc of the deck, and b) had no difficulties convincing largish pieces of plywood to bend in two planes.

[Jim Ledger]

In some cases it's best to loft the crown of the deck...or more importantly, the cabin top.

Take, for instance, a catboat cabin, an oval shape when viewed from above. In section it's heavily crowned, very heavily crowned. The profile shows a strong sweeping curve along the top of the house sides. If you make all the roof beams to the same camber, as you approach the forward end, where the width quickly decreases to zero, the roof will take a turn downwards.

In this case, I would start with the desired roof line in profile and work backwards from there, adjusting the shape of each deck beam, and in this case the shape of the front edge of the roof. Doing it this way gives you control of the shape you're building...it will come out looking like you want it to, instead of relying on a rule-of-thumb method in a situation where ease of construction will result in a disappointing appearance.

That said, the deck of the same catboat will offer less problem, as there is only a foot in the stern and four feet in the bow that show in profile...not enough distance in either case for things to go too far wrong.

The example of an ideal shape hull to apply a single crown pattern would be a narrow decked-over garvey where there isn't an extreme change of beam and the bow ends in a transom and the crown is modest. For all shapes in between these two examples there will be an increasing effect as the hull widens and the crown gets higher.

[PeterSibley]

Same here with the end decks on a Caledonia Yawl. A straight king plank required every deck beam to be at a different camber. The deck turns out to be a conic section rather than a cylindrical section, both of which are easily planked in plywood.

That's a good looking build you've go going there Gib.

That might be the most important aspect for a lot of us, not too many plank decks are laid these days.