Project Runaway Bunny

[CDBarry]

Project "Runaway Bunny"

A couple of years ago, I published a paper at the Chesapeake Sailing Yacht Symposium on a revival of the old technique of "composite construction"; wood planking over a metal frame. This generally died out prior to WW II and was subsequently largely supplanted by fiberglass mass production, but technology developed subsequently, notably arc welding and Computer Aided Drafting / Computer Aided Manufacturing / Computer Numeric Cutting offers the possibility of making it more affordable, especially for one-off or limited run construction. The paper was subsequently published in the 2022 volume of the Journal of Sailing Technology. At the symposium, one of the organizers suggested that I prove the method by designing and building such a yacht, then write a paper on that, having sailed to boat to the symposium venue as an exhibit.

I am accepting the challenge and will occasionally post updates as I go. The first is attached to this post for your interest and comments.

The paper is "CNC Enabled Wood/Metal Composite Construction of (Relatively) High Performance Sailing Yachts" and is open access at:

CNC Enabled Wood/Metal Composite Construction of (Relatively) High Performance Sailing Yachts | Journal of Sailing Technology | OnePetro

If anyone wants the slide deck, let me know.

 

[jehardiman]

And here I thought it was going to be an offshore capabile Wylie Wabbit....

 

[fallguy]

How do you avoid crevice corrosion?

(Also a concern on my build aluminum beams in frp sockets exposed to seawater)

 

[wet feet]

It looks like an interesting project and I doubt that many of us would prefer spending weeks on our knees on a lofting floor when sitting in front of a monitor in a warm room could achieve the same,or better results.I can't be the only one to have seen people's reaction when they learn that the whole boat can be developed on a computer and it isn't just an electronic drawing board.

The best summary of crevice corrosion I ever saw was in Professor J E Gordon's book about science and strong materials The New Science of Strong Materials https://www.penguin.co.uk/books/13533/the-new-science-of-strong-materials-by-je-gordon/9780140135978 .He explains in simple terms that if the protective oxide film on the surface of stainless steel is damaged maybe by stretching and there isn't any available oxygen to allow the formation of more of the oxide,corrosion will follow.I recommend the book.

I didn't see that the project under consideration makes specific reference to stainless steel in any case,so it might not be relevant.Back in the 1980's I knew an NA who did a study for his own amusement on the feasibility of building an IOR 1/4 tonner from 2mm steel.He decided that the project was technically feasible and what had led him to it was the popularity at the time of alloy construction for one off level rating boats.Epoxy and carbon was still quite exotic and alloy 12 metres were considered the norm in the class,which links in with the references in the summary above to classic designs such as Dragons,Tumlaren and 22 sq.metres.The unfavourable element at the time was the difficulty of arriving at a really fair hull using traditional plating techniques,but he felt that there was no other downside.With the explosion in computer capability since then,many things have become feasible and I will be watching this thread with interest.

 

[CDBarry]

How do you avoid crevice corrosion?

(Also a concern on my build aluminum beams in frp sockets exposed to seawater)

That is actually part of the experiment. In the past (before WW II) corrosion of the internals was not a problem if they were properly designed, coated, and maintained, but that's a bunch of ifs. So more ifs; if it's built and if I last long enough, we will see.

For this, the standard for steel would be NACE SP near white blast with two coats of Mare Island epoxy everywhere, polyurethane adhesive where it touches wood, with wood also epoxy treated everywhere. If I can get HSLA grades with copper, (i.e. CorTen or equal), I will. Copper doesn't eliminate the need for coating but reduces the tendency to undercut damaged coatings. I would like to spray aluminize, but I don't think I can afford that. Galvanizing is nice, but there is too much post coating welding required and welding galvanized is a hazard I don't want to deal with. Another possibility is high inorganic zinc primers. All of those are common for interior exposure. I have to do some research on all that; put my AMPP membership to work.

For aluminum, similar epoxy / urethane on contact surfaces, otherwise bead blast to Coast Guard bare aluminum standard. This seems to work, even on external exposure, but I don't expect to be able to use aluminum. Though I qualified to weld aluminum a long time ago, the equipment is expensive, and the power supply is an issue.

 

[CDBarry]

And here I thought it was going to be an offshore capabile Wylie Wabbit....

"Relatively high performance ...", not "High performance".

 

[CDBarry]

It looks like an interesting project and I doubt that many of us would prefer spending weeks on our knees on a lofting floor when sitting in front of a monitor in a warm room could achieve the same,or better results.I can't be the only one to have seen people's reaction when they learn that the whole boat can be developed on a computer and it isn't just an electronic drawing board.

The best summary of crevice corrosion I ever saw was in Professor J E Gordon's book about science and strong materials The New Science of Strong Materials https://www.penguin.co.uk/books/13533/the-new-science-of-strong-materials-by-je-gordon/9780140135978 .He explains in simple terms that if the protective oxide film on the surface of stainless steel is damaged maybe by stretching and there isn't any available oxygen to allow the formation of more of the oxide,corrosion will follow.I recommend the book.

I didn't see that the project under consideration makes specific reference to stainless steel in any case,so it might not be relevant.Back in the 1980's I knew an NA who did a study for his own amusement on the feasibility of building an IOR 1/4 tonner from 2mm steel.He decided that the project was technically feasible and what had led him to it was the popularity at the time of alloy construction for one off level rating boats.Epoxy and carbon was still quite exotic and alloy 12 metres were considered the norm in the class,which links in with the references in the summary above to classic designs such as Dragons,Tumlaren and 22 sq.metres.The unfavourable element at the time was the difficulty of arriving at a really fair hull using traditional plating techniques,but he felt that there was no other downside.With the explosion in computer capability since then,many things have become feasible and I will be watching this thread with interest.

There was a fairly successful Dick Carter 1 tonner built in steel during early IOR.

 

[CDBarry]

It looks like an interesting project and I doubt that many of us would prefer spending weeks on our knees on a lofting floor when sitting in front of a monitor in a warm room could achieve the same,or better results.I can't be the only one to have seen people's reaction when they learn that the whole boat can be developed on a computer and it isn't just an electronic drawing board.

The best summary of crevice corrosion I ever saw was in Professor J E Gordon's book about science and strong materials The New Science of Strong Materials https://www.penguin.co.uk/books/13533/the-new-science-of-strong-materials-by-je-gordon/9780140135978 .He explains in simple terms that if the protective oxide film on the surface of stainless steel is damaged maybe by stretching and there isn't any available oxygen to allow the formation of more of the oxide,corrosion will follow.I recommend the book.

I didn't see that the project under consideration makes specific reference to stainless steel in any case,so it might not be relevant.Back in the 1980's I knew an NA who did a study for his own amusement on the feasibility of building an IOR 1/4 tonner from 2mm steel.He decided that the project was technically feasible and what had led him to it was the popularity at the time of alloy construction for one off level rating boats.Epoxy and carbon was still quite exotic and alloy 12 metres were considered the norm in the class,which links in with the references in the summary above to classic designs such as Dragons,Tumlaren and 22 sq.metres.The unfavourable element at the time was the difficulty of arriving at a really fair hull using traditional plating techniques,but he felt that there was no other downside.With the explosion in computer capability since then,many things have become feasible and I will be watching this thread with interest.

A composite metal boat only has metal internal framing; the shell plating is wood. Stainless steel actually doesn't make a lot of sense for this because SS isn't as strong as even regular ASTM A131 steels, is subject to a lot of welding distortion and generally can't be welded well with internally cored wire feed or stick, which are the two low cost processes.

 

[jehardiman]

I think the real issue is going to be how light you can make the actual skin. IMHO, this type of modern "composite" construction should have more features like a skin-on-frame kayak or geodesic hull than "traditional" composite construction like the tea clippers, mid-Victorian era warships, or even the J's. Stressed skin construction anyone?

 

[wet feet]

I have heard one or two old timers mentioning bending slabs,dogs and furnaces for shaping the frames.the reference in an earlier post to profile cutting services would presumably only apply to the transverse faces of the frames,with the flanges being formed with suitable bevels before welding takes place.Might we see the use of intermediate steamed frames at perhaps two out of three frame locations?Metal deck beams could be integral with the frame assembly or would you use a metal beam shelf with wooden deck beams?I suppose due consideration of weight versus space would apply in a number of instances and will be interested in following the whole process.

 

[philSweet]

For very low quantity builds, you want to avoid having to consume material for molds and jigs which don't end up in the final boat (or it's trailer). CNC metal parts which stay in the boat, principally define shape, and also provide enough strength to facilitate the build process are a very efficient use of time, money, and materials. You only need to box out the center half of a sailboat with a grid and maybe four ring frames. And you need a structure for global bending to deal with forestay and backstay loads. These can be easily customizable as far as bow style and transom style is concerned.

For low quantity build, the weight that matters is the weight of materials you have to buy. The more your finished boat weighs for a given procurement, the better. I've done some small stich and glue craft where my scrap didn't fill a five gallon bucket. With CNC, you want to look at the scrap rates very carefully. Product weight is a driver for mass production, but it is not very important for a one-off. Or put another way, for a demo project, I think you want to look at the sort of boats where weight isn't very important . Carvel and lapstrake are both decently efficient with respect to material use. Glued carvel over metal frames with a glass skin is a very attractive combo. The CNC metal frames can be small so as not to eat up space and can integrate mounting lands and other reference flats to speed up the build. I usually try to integrate the trailer design right from the start. I want to be able to build the trailer out of the strongback. While you are looking at skinny boats in your paper, I think the proposed method is begging for something in the style of a Nonesuch 30 or Larry and Lin Pardey's Taleisin.

The last item is scale. Choosing a traditional design means it is designed for munchkins. Ever been in a wood-framed H28? Adult bunks need to be 7' minimum, and I try for 7'2". 6'6" is a child's bunk these days. Boats for couples cruising used to assume 275# was a good weight for a couple. I use 550#. So if you want to build a 34' boat, the interior will have the general arrangement based on classic 29'er.

 

[CDBarry]

Thank you for all of your comments, I really appreciate them.

A lot of further details are in the attached document and even more in the linked paper, which (like every paper in the Journal of Sailing Technology) is free open access.

Further points are that it is likely to be wood longitudinals on about 300 mm spacing or so and metal web frames on 1 m spacing (both TBD), and yes, probably with a metal sheer clamp. The comment about places to pick up point loads is correct and is an advantage of metal internals, and an important part of the detail design process.

As to internal accommodation sizing, it is a custom one off boat, mainly to demonstrate and get some practical tests and comparative numbers of the construction process at minimum cost and labor, and I'm only 5'10" and my wife is shorter. I will take the 5th regards weight, however.

The whole point of the project is to see if this process actually makes any sense as an alternative method for building one-off or limited run boats, not to build a boat that would be at all popular in any numbers. It is either a stunt or an experiment.

Perhaps it could re-establish a tradition of one-off custom boats or update open design racing classes like the square meter classes or some of the later "box rule" classes.

Fiberglass is clearly the best material for making more than six to twelve or more identical boats, and they have to have a broader appeal.

I intend to publish whatever I can open access, including the drawings, including all the metal piece parts for the record and if anyone is interested in various details, but I would be very surprised if anyone else built one. (I will be surprised if even I manage to build one, for that matter.)

The shell of the parent craft is about 230 ft sq, so in 3/4" Doug fir at about 32 lb/ft3, it's 470 lbm for the wood shell plate + 5% for epoxy and reinforcing/sheathing cloth, 10% for wood longs and 20% for metal framing, total about 647 lbs. The shell is thus about 22% out of an estimate of 3,000. This is pretty reasonable compared to other boats, and it has a lot less internal outfit weight. But ... the design spiral.

Thanks again for your thoughts.

 

[Rumars]

Sometimes a little bit of "waste" material brings great labor savings. Case in point, the metal clips needed to mary wood longitudinals to the frames. These are small fiddly pieces that have to be fabricated and welded in position individually, each at a different angle. It also requires at least one sacrificial wood stringer to hold them while welding.
The alternative way of doing them is to bend a U profile over the entire lenght of the boat, weld it in place as a longitudinal then cut the part between the frames that's not needed.
An even better way is to use cnc cut flat plate, cut the two flanges of the clip as continuous longitudinals, weld then to the frames then add the back strap to close the clip. To minimize waste you can cut the plates in alternating lengths covering only two frames, or cut only one continuous flange and small pieces for the other.
This way frame stringer alignment is done as for any metal boat, without fiddling with clips, spacers and small wedges.
A trick to aid frame alignment is to not cut out the stringer tabs, leave them attached (depending on cutting technology either partially cut trough or with interrupted cuts), and have the clip plates marked for the frame position (scratch cut, laser engraved, etc.). This way the clip plates can be aligned for welding without the use of spacers, and it's easier to get everything squared. The tabs can be easily broken out afterwards and the edge dressed with a suitable tool (finger, belt, die grinder) before welding on the back plate of the clips.

Next trick is related to corrosion prevention. In scandinavia composite yacht repair has a love affair with stainless steel, and it has worked well for their temperature and humidity ranges. For a more tropical use, hot dip galvanizing is the way to go, and for that the ticket is to incorporate a few mechanical connections into the framing. Where exactly to put these connections depends primarily on the galvanizing plants tank dimensions and the boatyards handling equipment. If we consider a complete ring frame we could have it split for example into three pieces, a floor section and two sides each incorporating half of the deck beam and the hanging knee. The floors could already be welded to the keel, and there could be several keel sections, as well as a separate stem and horn. Other schemes are possible like separate deck and floorbeams bolted to sideframes, with the knees beeing part of either the frame or the beams, there could be a bolted connection to a continuous keel, or each floor could have knees with long arms connected by bolted side plates, etc.
While this implies dismantling the frame after welding and errecting it again after galvanizing the labor hours could be less then applying epoxy primers and topcoats by hand. Some galvanizing plants also offer powder coating or painting over the galvanizing, or as they call it here "double coating". Mounting the connecting plates and bolts with PU adhesive or thickened epoxy takes care of the inevitable paint scratches from bolting, and all that's left is to paint the bolts and nuts.

For transverse framing the cnc can cut beveled wood futtocks that bolt alongside the steel frame. Wich brings us to the principal problem of composite construction today, if we can have cheap prebeveled futtocks we can assemble double or triple sawn frames glueing them together with epoxy. Plywood futtocks take care of the old grain runout problem where necessary. Such a frame is heavier then a normal laminated one, but much quicker to make. If additional strength is required it's easy to just laminate more wood on the inside of such a frame. Wich means steel framing is only required for really high load areas on big boats (as you noted in the article), or for spreading out point loads in otherwise thin shell structures (see the RM keel grids).

These are just my 2 cents on the subject, I would love to see your boat buildt, regardless how.

P.S.
A long time ago on this forum someone proposed and even buildt a crazy composite vessel. The framing was steel, made in the conventional manner, then a complete plywood ceiling / furniture was installed. The whole assembly was then sprayfoamed from the outside, the foam faired to the stringers and glassed over. Unfortunately the guy didn't stick around to tell us how the boat did over time. I believe that with a proper grade of foam and the addition of more carefully designed metal to glass attachment this could work well, and the added weight could be offset by having an unsinkable boat.

 

[philSweet]

One advantage of duplicating Taleisin, besides it being a wonderful Lyle Hess design that was designed for cast bronze floors and knees, is that the original build (a one-off) was very well documented. Full documents of materials, labor, wall clock time, and fitout exist to compare yours with.

 

[CDBarry]

I have heard one or two old timers mentioning bending slabs,dogs and furnaces for shaping the frames.the reference in an earlier post to profile cutting services would presumably only apply to the transverse faces of the frames,with the flanges being formed with suitable bevels before welding takes place.Might we see the use of intermediate steamed frames at perhaps two out of three frame locations?Metal deck beams could be integral with the frame assembly or would you use a metal beam shelf with wooden deck beams?I suppose due consideration of weight versus space would apply in a number of instances and will be interested in following the whole process.

When the flanges of the web frames are in the 3D model with the required bevel, they unroll into flat curved pieces because they are the surfaces of part of a generalized cone (called "j-ing"). They can then be numerically cut flat and when they are welded onto the flat web, they naturally take the required bevel (with a bit of encouragement with clamps to get them on the flanges) because the beveled condition is the lowest energy state.