Gardner's Clamming Skiff

[Quick Karl]

PAR,

Early on you suggest I might could find the offsets and lines online but darned if I can find em.

For what it is worth I went and dug through my storage unit but I'll be dipped if I can find that old book. I guess I am going to have to just break down and buy a new one but I sure would like to do a 3D model of it in the mean time.

Here is a lil example of what I do 3D:

So I know I can make a boat look good!

 

[Quick Karl]

My better half says she thinks "we" "mighta put it in a box of books we took to the used book store"...



I ordered a new paper back version: http://www.amazon.com/gp/product/007142797X/ref=oh_aui_detailpage_o00_s00?ie=UTF8&psc=1

Last I recall some of the pages had been slipping out of the old hardcover anyways.

 

[gdavis]

qcarl, checkout www.spirainternational.com(i think that's right) he's got some sweet skiff designs that are easy to build and there's lots of photos of finished boats and some study plans to look at. Perhaps gardners skiff design is a bit out dated these days.(sorry john) Another real fine one is the San Jaun dory skiff, plans can be found at woodenboat mag. All are built of plywood with taped seams, pretty straight forward. Stay away from the skiffs with the real wide bottoms they'll pound your teeth out. Well then maybe this info will help. PAR, that's a super fine looking skiff in the photo, something like a san jaun but yours has a sweeter sheer. Remember texas dory boat plans? I still have a study plans package from them and still look at the plans and wonder.................................later...............g

 

[Quick Karl]

Thanks George! I've been to that website but I have my heart set on the Gardner Skiff - I just like it.

I am still operating from memory but I believe the Gardner Skiff had alteration information that would make for a better planning hull with an outboard, as opposed to the rowing version.

I am certainly no boat builder or naval architect but I always wanted to build the boat with the framing in Gardners specs, but screw and epoxy it together with a plywood bottom instead of his cross planked bottom.

 

[gdavis]

aye, it will be a good skiff. I like skiffs, always have and always will. Their fun to build and are a good economical way to get on the water. Basically for a skiff to perform well with an outboard the bottom should be a flat run for the last two thirds of the bottom running aft. If there is rocker aft of midship it will probably porpoise when you twist the throttle. Also the bottom should hold the same width from midships to the stern or pretty close to the same. This will keep it from squatting when you give it some throttle. And building it because you just like it sounds like the best reason there is!............have fun..............g...........

 

[Quick Karl]

Thanks again, George!

I am starting to recall what you said about the bottom rocker and width now that you mention it.

I have an off-the-wall question that maybe you or another reader might know the answer to -- where can I find out the mathematical formula for calculating the conic sections for the sides of a boat like the one we are talking about?

A flat piece of material like a sheet of plywood can only be bent in one direction, so the sides must be some section of a cone. Therefore, there must be a formula to determine the exact offsets at each station for a perfect (or near perfect) shape in terms of offsets from a centerline. This would also help develop a perfect stem as well.

I am sure there is a traditional Boatwright's way to determine these numbers but I am sure there is a mathematical way too.

Does anyone know?

 

[Quick Karl]

Thanks again, George!

I am starting to recall what you said about the bottom rocker and width now that you mention it.

I have an off-the-wall question that maybe you or another reader might know the answer to -- where can I find out the mathematical formula for calculating the conic sections for the sides of a boat like the one we are talking about?

A flat piece of material like a sheet of plywood can only be bent in one direction, so the sides must be some section of a cone. Therefore, there must be a formula to determine the exact offsets at each station for a perfect (or near perfect) shape in terms of offsets from a centerline. This would also help develop a perfect stem as well.

I am sure there is a traditional Boatwright's way to determine these numbers but I am sure there is a mathematical way too.

Does anyone know?

 

[DCockey]

......

I have an off-the-wall question that maybe you or another reader might know the answer to -- where can I find out the mathematical formula for calculating the conic sections for the sides of a boat like the one we are talking about?

A flat piece of material like a sheet of plywood can only be bent in one direction, so the sides must be some section of a cone. Therefore, there must be a formula to determine the exact offsets at each station for a perfect (or near perfect) shape in terms of offsets from a centerline. This would also help develop a perfect stem as well.

I am sure there is a traditional Boatwright's way to determine these numbers but I am sure there is a mathematical way too.

Does anyone know?

Developable surface is the technical name for a surface with curvature in only one direction, and which can be unrolled flat without distortion. Not all developable surfaces are sections of a single cone. They can be a composite of sections of multiple cones, or a "tangent developable surface" which is essentially a composite of an infinite number of infinitely small cones. There is no simple mathematical formula or the general case of a developable surface.

Developable surfaces can be modeled using software such as Rhino to objain the "exact" shapes of stations, stem, etc.

 

[Quick Karl]

I'm a pretty fair CAD guy with more than a casual amount of surface modeling experience. If someone developed shapes in a CAD program from published offsets it would be very difficult to know how theoretically perfect those surfaces actually are. All you would have are NURBS surfaces that might look pretty good, but if you use published offsets you will never really know how fair the surface really is (could you spot a 1/8" defect in an offset for a developed surface in a CAD program with a 24" monitor?).

Working backwards a person could model a pleasing shape and intersect it with planes that represent the stations to derive very close estimations of the offsets.

What I want to do is figure out how fair the developed surface, based on a set of published offsets, really is, just out of curiosity (it isn't going to make a boat do magic tricks ).

 

[DCockey]

I'm a pretty fair CAD guy with more than a casual amount of surface modeling experience. If someone developed shapes in a CAD program from published offsets it would be very difficult to know how theoretically perfect those surfaces actually are. All you would have are NURBS surfaces that might look pretty good, but if you use published offsets you will never really know how fair the surface really is (could you spot a 1/8" defect in an offset for a developed surface in a CAD program with a 24" monitor?).

Lots and lots of discussion on this topic in previous threads.

Can you spot an 1/8" defect by simply looking at an entire hull in a 24" monitor? Of course not.

Can you, in appropriate software such as Rhino, spot un-fairness in a surface which needs a 0.005 inch movement of a control point to correct by zooming in and using tools such as curvature combs, zebra patterns and "fluorescent tube" reflections? Absolutely, with practice.

By the way, what is your definition of a "theoretically perfect" surface?

Working backwards a person could model a pleasing shape and intersect it with planes that represent the stations to derive very close estimations of the offsets.

What I want to do is figure out how fair the developed surface based on a set of published offsets, really is, just out of curiosity (it isn't going to make a boat do magic tricks ).

Simplified version of my workflow in Rhino for a boat to be built using sheet materials:
- Import the published offsets. I use an Excel spreadsheet template and the process of data entry and importing into Rhino usually takes less than 30 minutes.
- Create curves for sheer, chine, rabbet, etc which go through the appropriate points. Check the curves and adjust/fair as needed. Frequently problem points are due to typos or similar. Fix those first, redo the curves, then fair the resulting curves if required.
- Create developable surfaces using DevSrf in Rhino which fit the faired curves. The curves almost always need to be extended past the ends of the boat to be able to create surfaces which cover the entire hull. Trim the resulting surfaces as needed. The curves may also need adjusting in order to create developable surfaces. Check the developable surfaces for fairness and agreement with the offsets. If there is a problem the defining curves are adjusted and new surfaces created. Iterate until the developable surfaces are satisfactory.
- Create inner or outer surfaces as offsets from the developable surfaces as needed.
- Create transom surfaces.
- Add keels/keelsons, stems, chines, rails, etc.
- Create planking by trimming the developable surfaces and transom surfaces using the keels/keelsons, stems, chines, rails, etc.
- Cut sections and create mold/frame shapes.
- Unroll planking surface to create shape of flattened shape of planking.

 

[DCockey]

John Gardner's initial clamming skiff in Chapter 13 of Building Classic Small Craft, Volume 2, 15' 8" in length, has rocker extending the full length of the boat and the aft portion of the bottom curves upward. An early builder reported the boat porpoised until he moved weight forward.

John then redesigned the boat without rocker aft which he labeled on the drawing as "LINES ALTERED FOR PLANING SPEEDS". He added a version lengthened to 18 feet.

 

[Quick Karl]

Could I scale the 18-ft boat down to 14-feet? I am not looking for a performance boat, just something nice to get me out in Matagorda Bay for some redfish and speckled trout fly fishing. I won't be out there during gales trying to catch a fish or trying to break any speed records. I like the workboat look if Gardner's skiff.

The reason for my inquiry into the math is because I am curious as to how Gardner and other boat builders derived their developed surfaces, pre-computer. How do we know that somewhere along the curve, the developed surface isn't deflexing a bit, instead of being the progressively curved, or de-curved surface, I presume it to be?

Nice workflow outline! I've done work in Unigraphics but most of what I do is SolidWorks and 3D Studio. Even with curvature combs and zebra patterns you're still only getting close with respect to the question I am curious about.

 

[DCockey]

I wouldn't try scaling down a plywood skiff from 18 feet or even 16 feet to 14 feet. I'd find a design of the size I wanted, or if I felt confident of my knowledge create a new design.

John Gardner described creating the first clamming skiff design as "the result of some preliminary sketching and the construction of a scale half-model in soft, dry pine". He said he used a "thin poster-board template" to obtain the shape of the sides.

"Even with curvature combs and zebra patterns you're still only getting close with respect to the question I am curious about." Only getting close to what???? You can get closer than you can build in plywood. In fact with sufficient skill you can create a surface suitable for milling dies for automotive exteriors. An important tool in addition to curvature combs and zebra patterns is surface reflections.

 

[PAR]

The Gardner skiff is developed, but he didn't intend or plan the boat this way. The boat's design is a traditional one for solid wood planking, not sheet goods (plywood).

Developability, pre-computer was a pain is the ***, but simple geometry. It's pretty complicated to describe quickly.

A 23% reduction of this skiff wouldn't look very good, nor would it be recommended, as it's just too much to ask. There are 15' skiffs available and you can place the Gardner sheer on it, to get the look. It's Gardner's sheer that makes the boat, plus it's general proportions, but there are also lots of skiffs with pretty sheers too.

I'm pretty sure offsets for both the 15' 8" and the 18' skiff are in his book. My 17' skiff can be made 15' simply by respacing the station molds. The same may be possible with the Gardner 18' skiff, but I'd have to look it up. Lastly, I'd recommend you go with a 15' or 15' 6" skiff, instead of the 14' you've requested. It's more economical of 8' panel lengths (plywood) and you'll like the stability and volume this length offers.

 

[DCockey]

Gardner specified plywood for the sides, cross-planking for the bottom

Offsets for both lengths of the revised design are in the book. The 18' version is just the 15' 8" version with a 2' 4" extension.