Development of intersection of two cones and two planes.

Rhino, 15 minutes, two cones, a bit of slicing and dicing and …........

 

Howdy PhilSweet,

How have you found the calculated errors to work out in practice.

My experience is when there is significant twist in the panels when the real boat is put together - noticeable by lots of tension in the copper wire ties - that the shapes start moving away from the ideal computer shape.

Panels always try to get rid of stress, so, often the ruling line angles turn out to be significantly curved in real life - showing that the ply shares out the stress by putting some of the curvature that the computer says should be all at 90 degrees to the ruling lines and throwing some of it in the direction of the ruling line too.

Effectively compounding the panel in high stress areas.

Building a pure stitch boat allows things to move away from the calculated shapes a bit to make sure panels meet up.

But I've done a few trials over accurate strongback and traditional moulds and found that if the original shape is forced then the calculated panels have significant deviations from the actual shape close to any areas of high stress.

But if you put it together as pure stitch with no constraints there's nothing that epoxy is not able to bridge ... but the shape has moved away from the design "a bit".

Usually where the bilge panel goes from near horizontal in the mid body to vertical to meet the stem.

Does anyone else have something like that experience building real boats from computer projected panels.

 

I did find, when planking my rowing boat, that there was noticeable deviation between the Delftship prediction of the garboard and the actual shape on the boat. It was pretty accurate aft, where it was strictly developable, but ran off for'd (by about an inch, IIRC) where it was slightly compounded.

This didn't matter, since I was only using the projection as an indication of how many sheets of ply I needed and the plank shapes were taken off the real frames on the strongback. I thought it was mildly interesting though, and worth noting for similar builds.

 

Ok, so really all you are saying is that you, personally, can't think of any more sophisticated way of drawing boats.

That's not exactly an awesome recommendation, and may not result in the best shape. Your use of "desired outcomes" is an extremely limited use of the term. What do you do if the desired outcome cannot be obtained by this method? This is not a hypothetical, since it's easy to think of situations where this method would be inadequate.

Sure, in principle all boats shapes can be described by suitable geometry, but it's rarely going to be as simple as you would like it to be.

 

Incidentally I did play around with this stuff years ago. This was back when everybody used pencil and paper to do lines, and offsets always needed to be lofted. I had the bright idea that if the boat shape could be generated from mathematically defined curves it would come out fair and the offsets would be fine.

It works too, but of course it depends on your choice of numbers and functions. It's really easy to make utterly horrible shapes. It's also possible to make some very good shapes, including a very close variant of the Gokstad faering* (swoopy lap lines and all) but it requires a lot of messing around with functions and variables and can get quite complex. Some of the combinations of functions can get quite elegant.

Don't have any of the old stuff left, since the calculator it was on crashed yonks ago (this was in pre-PC days, or before most people had them anyway). It was an interesting exercise at the time though, and I learned a bit more maths from it.


*Didn't build this, just drafted it. Since this was pre-PC days, I used to manually draft lines from the generated offsets.

 

There are some older threads on the forum about "mathematical hull design".

http://www.boatdesign.net/forums/boat-design/request-your-input-my-design-47644-8.html

Here's an interesting link:
http://developable-surface-boat-des...7/sample-developable-design-calculations.html

 

On the other hand, you can get a plywood and let it develop the shape it naturally does. Then just tweak it to look good and to take a shape that is within parameters known to perform well. Seems like there is a lot of effort put into mathematical formulas to do something that is quite simple: let panels bend into their natural curve. Experience allows a builder/designer to use those curves to advantage. This goes back over a century when Herreshoff made fun of designer talking about cycloid waves. He said they didn't look like a cycloid from the cockpit of a kayak running down his back. In essence, he says that experience in the water and the shipyard are pre-requisites to the understanding of boat design and behavior.

 

Okay, okay, mathematics and descriptive geometry are simple tools. What really used to design a boat are the navigation hours of the designer.
With all my respect for the professionals, would you ask a truck chofeur to design a trailer, or a taxi-driver to design a car?. Why does a sailor can know how to design the hull of a boat?

 

@boatmik,

That probably has more to do with the material than the software. Thin, three or five ply fir isn't going to behave the same way if you rotate the grain direction, it will minimize it's energy by stretching in a addition to just bending. This is yet another thing to consider when designing for ply, and why I think it is worth the money to custom order 9 ply panels that are A/A/A with no seams on even the internal plys. A cheaper alternative is to skin the plywood with a veil that has its fiber orientations at +/- 45 to the plywood. That helps stabilize the cheaper fir marine panels, but it adds stiffness, so it's a bit of a bargain with the devil.

So what you need to do with plywood is avoid those highstress areas on the design in the first place. Deformation is an energy minimization effort. The ruling lines in the model you posted show you exactly where to look. Keep the focal points or directrix a safe distance from any part of the panel. And Don't Suddenly Swap Sides with the focal points, doing that forms a hard point that will be impossible to fair in practice. (It creates a stress concentration at the edge of the panel. This is due to plywood not behaving like an isomorphic material.)

The ability to stretch plywood into shape can be very handy, but it only works to the extent that you aren't fighting the the panel's attempt to stretch itself out of the developed shape. In practice, I could get the chines on the hull I posted to practically vanish at the stern because I wouldn't be adding that much energy to the system. It wouldn't have any global affect on the shape because I can intercept that with a small frame. That's by design. I use a fair bit of light framing in stitch and glue. It helps control the shape during the build. The idea is not to stress the shell with the frames, just cradle it in place.

One last point is that you need to carefully plan the stitching so that the stitches don't load the panel. Keep them close to the edge, less than the thickness of the panel, and very precise. Like 1/100 inch precise.

It doesn't cure all the troubles, but it sure helps.

 

Just thinking about this raises a question. Why on earth would you want to remove subjective aesthetics from design? It's one of the most enjoyable bits, so getting rid of it doesn't seem like progress.

Note that I'm not saying functionality doesn't matter (far from it) but aesthetics are part of the good things in life.

 

Good point, I didn't notice that part in Pdmclean's reply.

I will go even further by saying that aesthetics (which is always subjective) is a part of the functionality of a boat.

It is partially so even in case of workboats, and definitely makes a fundamental part of functionality of pleasure boats.

 

Like Herreshoff said: A thing of beauty is a joy forever.

 

And Hereshoff did design some true little jewels of the seas.

 

Then “John Keats” should have designed boats, from his tombstone "HERE LIES ONE WHOSE NAME WAS WRIT IN WATER".

 

I hope this drawing helps define shell expansion.