Wood/Epoxy Keel and Keelson

[CanuckGuy]

Hi all,

Sorry for starting a new thread if this topic has been covered anywhere else. At least if people post links, there will be an easily searchable thread entry...

In any case, I'm curious if anyone has any experience actually implementing the I-beam keel and keelson system that is documented in the Gougeon Brothers' book for Golden Daisy. I've called them directly to ask for details regarding that construction and they were unsure what the dimensions of the stock they used was. They also did mention that it did snap on the maiden voyage due to grounding, and that they had to make a thicker version for it.

I plan on using laminated Hard Maple for both keel and keelson. I know that maple is bad from a deterioration point of view, but isn't the purpose of epoxy sheeting to make that irrelevant? Does anyone know from experience how wood covered in epoxy behaves over decades?

And does anyone have experience with this type of construction?

Anyways, any information on the above is much appreciated.

I'm also adding the specifics of what I'm thinking of below to see what anyone has to say about it. Suggestions/criticisms are welcome.

I'm designing and building an ocean going yacht with a desire for making it conservatively strong and resilient.

I'm planning on laminating the keel/keelson flange out of two 1/2" x 5" strips of scarfed wood running from stem to stern. For the web, I'm considering some sort of lamination totaling maybe 3" high by 2" wide. (prelim SM is ~300 cm^4)

Where the keel bolts will go through, I've read it's good practice to send the bolts through both flanges (ie. Keel and through to keelson). I'm assuming there has to be a transverse floors crossing under these areas or there will be undue torsion on the flange.

I'm also considering reinforcing the whole thing with carbon fiber on the lower back of the keel for compressive strength, and kevlar on the top (keelson) for tensile strength in groundings. This is just a consideration as I'm not sure the stress/strain properties of the fibers will be compatible with the maple.

Any ideas, suggestions, advice, relevant experiences are welcome.

[TeddyDiver]

Have you plans for the project?
You can protect the wood with monoetyleneglycol if wan't to be sure. It's good with epoxy..
Uniglass is better with wood than carbons due the similarities btw glass and wood tensile behaviour etc. Carbon is too stiff and it takes the whole load and you gotto use a lot more of it..

[alan white]

IT seems to me that an I beam would be totally inappropriate when a solid beam would be heavier to such a small degree that you could literall carry the extra weight (between flanges) on your shoulder. Maybe not but awful close.
Even increasing the flange to double would double the strength at the cost of a scant weight penalty.
Then all the expensive exotic stuff---- so expensive! Are you racing or fast cruising?
A tall I beam makes more sense, but a short one doesn't pay it's way.
4" tall is not much to begin with.

Alan

[TeddyDiver]

Originally Posted by alan white

IT seems to me that an I beam would be totally inappropriate when a solid beam would be heavier to such a small degree that you could literall carry the extra weight (between flanges) on your shoulder. Maybe not but awful close.
Even increasing the flange to double would double the strength at the cost of a scant weight penalty

The idea of I beam here is having a "real" keel (if the OP refers to the Gougeon bros and I recall it about right), the floors and the keelson in a way they are most easily done, not due weight issues allthough it's also an advantage.
Too late and too lazy to take the book and look ref what else was there..

[CanuckGuy]

TeddyDiver:

Thanks for the tips regarding Carbon over Wood. Anything regarding Kevlar? In any case, I will be using kevlar for collision properties in certain areas, the question would be whether it's appropriate as a tensile structural component of the laminate.

Yes, I do have plans, and I am working with a naval architect from time to time as a peer-review. However, I'm not sure he's worked with Wood/Epoxy before. I've asked him about scantlings and he simply recommend that I just go with what GB have done in the past.

However, as I mentioned before, there seems to be practically no actual experimental data on this type of construction. GB themselves (or rather the guy I talked to) was not sure about the dimensions. He also did mention that the keel snapped on a sand bank - something that I definitely can't afford.

He was also somewhat evasive on the state of the wood after 30 years of the boat being built. He mentioned that if bilge water remained was not pumped out etc, it might pose an issue. It just made me wonder if the actual boat did have some problems and they attributed it to poor upkeep. (I think that even if the upkeep is not great, it's a good data point to know what happens).

Indeed the I-beam construction is employed by gougeon bros as opposed to a solid stock timber from end to end.

[alan white]

Well, it's true that an I beam is more subject to catastrophic failure due to a vertical impact tending laterally. What prevents failure of a solid piece is the fact that the inner part (now being replaced by a web) can stay in column, while a web can easily distort. Were it to remain in column, the web would be nearly 100% as strong as a solid beam. There are shear stresses as well, not present to the same degree in the solid beam.
For a savings in labor, then, the I beam has little to offer in return, I think.
Go-fast designs, especially multis, have far less reserve strength built in.
A monohull is much deeper (shoal challenged) and it also has more mass. If the idea has failed when slightly less beefy, what is your safety margin going to be?

Alan

[CanuckGuy]

Originally Posted by alan white

Well, it's true that an I beam is more subject to catastrophic failure due to a vertical impact tending laterally. What prevents failure of a solid piece is the fact that the inner part (now being replaced by a web) can stay in column, while a web can easily distort. Were it to remain in column, the web would be nearly 100% as strong as a solid beam. There are shear stresses as well, not present to the same degree in the solid beam.
For a savings in labor, then, the I beam has little to offer in return, I think.
Go-fast designs, especially multis, have far less reserve strength built in.
A monohull is much deeper (shoal challenged) and it also has more mass. If the idea has failed when slightly less beefy, what is your safety margin going to be?

Alan

Well, to answer your previous questions as well, I can get 1 yard by 56" of 6oz. kevlar for 40 bucks. That is 14 square feet. As a reference: I can get the same amount of area in Maple veneer for 10 dollars. Given that the area of the keel is maybe a total of, say 5" by say 5 feet, that's 2 square feet. If I laminate, say 7 layers, that costs me 40 dollars.
I'm not going to go broke over 40 dollars when I paid 6000 for the white pine on the hull.
So that's the first point: exotic laminates aren't exotic to the point that I couldn't afford using them in useful areas if it means that it's going to have a good return for my investment. And I don't feel any qualms in doing this either. After all, I am using Epoxy which was considered exotic a few decades ago.

Now regarding the keel. The issue is not a weight saving from the i-beam. It's a cost saving from the keel. I will be laminating half inch stock that is 12 feet long. Finding that kind of stock in select quality is trivial. Finding 5 inch by 6 inch solid timber of any kind of tree is a difficult undertaking. It's difficult to transport, it's difficult to handle on the shop floor, it's difficult to work on, and it's also difficult to insure that it's dimensionally stable. It's also going to be very expensive (for example if I choose to use dimensionally stable, but ecologically unfriendly, and also expensive mahogany).
Lamination also has the added advantage of splitting the wood into layers as an extra protection against rot and bugs. Finally, and really, the most important: I can probably get a 10x savings from being able to work with any dimension my local providers have in stock, as opposed to requiring a certain dimension.

To answer your points about the flange buckling under the load, I do have the same concerns, however, the floor timbers which have the job of spreading the transverse load of the keel evenly on the hull skin go "through" the i-beam structure, effectively turning the whole thing into a solid piece at the keel bolt connection points. (see illustration).

[Tad]

There are a great number of issues here....

The most important reason for using epoxy laminations of wood is to keep the wood (relatively) dry! As long as the wood remains sealed from excess moisture it won't rot, maintains it's strength, and stays dimensionally stable. If the wood does not move (much) with changes in temperature and moisture content, the entire structure can be bonded in one piece. Thus loads are transfered through the wood (rather than at points with fastenings) and easily spread over the structure.

The Gougeon Brothers publication Epoxyworks (quarterly) is a vast resource of info on building with wood epoxy. Find all the back issues you can, plus read the ones available on their website.

See Epoxyworks #17 for info on the longitivity of boats built using wood epoxy techniques. The oldest examples are now approaching 40 years old. Built of good materials, using good workmanship, and with excellent maintenance (including ventilation) they are in good shape.

Epoxyworks #17 does discuss GoldenDazy, which used the I-beam keel structure. They had trouble with the ash used in part of it's makeup, but no problems with the mahogany caps. The problems are attributed to 30+ years of laying in bilge water, and poor ventilation.

At Bruce King Yacht Design we used the I-beam keel structure a number of times. The first was in the 90' WhiteFin, it was also used in 80' Liberty, and in the 154' Scheherazade. The builders hated it because of the numerous individual pieces that must be cut/fit and then sealed before bonding into place.

One of the tricky parts of the I-beam keel involves the method of building the boat. If it's built upside down on molds you need to remove the boat from the molds to laminate in the keelson/cap. Will the boat support itself while you do this? And are you adding the blocking while she's upside down? Whooo boy...no fun.

[alan white]

Originally Posted by CanuckGuy

Well, to answer your previous questions as well, I can get 1 yard by 56" of 6oz. kevlar for 40 bucks. That is 14 square feet. As a reference: I can get the same amount of area in Maple veneer for 10 dollars. Given that the area of the keel is maybe a total of, say 5" by say 5 feet, that's 2 square feet. If I laminate, say 7 layers, that costs me 40 dollars.
I'm not going to go broke over 40 dollars when I paid 6000 for the white pine on the hull.
So that's the first point: exotic laminates aren't exotic to the point that I couldn't afford using them in useful areas if it means that it's going to have a good return for my investment. And I don't feel any qualms in doing this either. After all, I am using Epoxy which was considered exotic a few decades ago.

Now regarding the keel. The issue is not a weight saving from the i-beam. It's a cost saving from the keel. I will be laminating half inch stock that is 12 feet long. Finding that kind of stock in select quality is trivial. Finding 5 inch by 6 inch solid timber of any kind of tree is a difficult undertaking. It's difficult to transport, it's difficult to handle on the shop floor, it's difficult to work on, and it's also difficult to insure that it's dimensionally stable. It's also going to be very expensive (for example if I choose to use dimensionally stable, but ecologically unfriendly, and also expensive mahogany).
Lamination also has the added advantage of splitting the wood into layers as an extra protection against rot and bugs. Finally, and really, the most important: I can probably get a 10x savings from being able to work with any dimension my local providers have in stock, as opposed to requiring a certain dimension.

To answer your points about the flange buckling under the load, I do have the same concerns, however, the floor timbers which have the job of spreading the transverse load of the keel evenly on the hull skin go "through" the i-beam structure, effectively turning the whole thing into a solid piece at the keel bolt connection points. (see illustration).

I didn't mean to imply that solid meant as-from-the-tree, but rather laminated to a full rectangular section as opposed to an I beam of the same outer dimensions.
I understand what you're doing. One simple variant would be to box in the open sides of the beam, which would add a large amount of stiffening for your efforts (like adding a shoe box lid to the shoe box).
BTW, I'm interested in your boat. You say pine. Strip-plank? Got any pictures?

Alan

[CanuckGuy]

Originally Posted by alan white

I didn't mean to imply that solid meant as-from-the-tree, but rather laminated to a full rectangular section as opposed to an I beam of the same outer dimensions.
I understand what you're doing. One simple variant would be to box in the open sides of the beam, which would add a large amount of stiffening for your efforts (like adding a shoe box lid to the shoe box).
BTW, I'm interested in your boat. You say pine. Strip-plank? Got any pictures?

Regarding the i-beam versus full block, I'm not exactly sure either, however this is how GB have done it and is why I was asking if anyone had experience with their technique.

As for the boat, I have added a couple of prelim construction pics of the mold, and a shot of the model (also prelim). I'm inspired by the Spirit class of boats built in England. Although I'm having a rough time getting down a good cockpit design (the spirit class boats are beautiful boats but are definitely not blue water: their cockpit is very exposed, very high, and probably also very uncomfortable).

My boat will be of same Lwl as spirit 52, but loa of only ~40.

Regarding construction, it will be 1/2" inch white pine strip plank with 3 layers of 1/16" Hard Maple veneer ±45˚ and a final layer of thin glass (it will be a painted hull, no wood exposed on the outside). This is roughly equivalent to the Golden Daisy scantlings btw. You can see the pictures of the scarfed white pine stock. It is so far the most expensive component of the boat. I will be reinforcing key areas (along keel, on the flare of the bow, the bow itself, and the fore foot) with an initial layer of kevlar cloth under the layers of veneer, and a final layer of kevlar on the exterior. GB have some tests on this sort of laminate from an impact resistance pov, and it looks quite good for the amont of extra effort/money.

Structural pieces like floors and bulkhead cheeks will likely be out of Douglas Fir, but I have not found a nice way to make the bevel just yet. (also included is a pic of lamination tests for the bulkhead cheeks out of white pine). I'm having difficulty making the bevels just perfect and so that is bugging me. Considering maybe doing the laminations *after* the hull is complete, however that's not my best case scenario as it would put in a pre-torsion of the laminate, and also why skip a chance to make the laminate shape exactly molded to the mold shape when waiting till after hull completion there is a good chance that the hull skin will sag and change ever so slightly in shape... Ideas on good beveling also welcome.

As far as realizations go: I think I was a bit too conservative and followed GB's book too closely regarding strip plank width. It's 1 1/8" wide strips. Which will come out to little over 100 strips per half of the boat, whereas looking at the spirit yachts, I can see that they clearly used wider strips (possibly even 2 1/2" wide). That would have saved me much time, money, and sawdust on the shop floor.

[CanuckGuy]

Tad: what do you guys do as an alternative to i-beam style keels?

One of the tricky parts of the I-beam keel involves the method of building the boat. If it's built upside down on molds you need to remove the boat from the molds to laminate in the keelson/cap.

I've thought about this at length. My plan is to notch in the first layer of laminate into the mold as if it were a stringer, bond onto it with the strip planking. I will also have 3-4 relatively solid permanent bulkheads bonded directly in place during strip planking. However, this poses several issues, one of them being the notches on the permanent frames, and also the lamination of cheeks as I mentioned previously.
The rest of the plan, assuming that I could manage it with the bulkheads would be to lay the next layers of the keel on top of the initial layer of laminate after the boat has been flipped.

What are you referring to when you say adding blocking while upside down?

[Butch .H]

Canuk Guy what are the dimensions of that thin layer of glass. I have a hull that is coverd with veneer and I would like to also finish with glass

Regards
Butch

[CanuckGuy]

Butch H: Sorry no idea yet. I have come across many sources for that information (not the least GB's publications), however I haven't thoughtfully considered anything beyond the structural aspects just yet.

[Butch .H]

Cool I wait in anticipation

Regards
Butch

[alan white]

Originally Posted by CanuckGuy

Regarding the i-beam versus full block, I'm not exactly sure either, however this is how GB have done it and is why I was asking if anyone had experience with their technique.

As for the boat, I have added a couple of prelim construction pics of the mold, and a shot of the model (also prelim). I'm inspired by the Spirit class of boats built in England. Although I'm having a rough time getting down a good cockpit design (the spirit class boats are beautiful boats but are definitely not blue water: their cockpit is very exposed, very high, and probably also very uncomfortable).

My boat will be of same Lwl as spirit 52, but loa of only ~40.

Regarding construction, it will be 1/2" inch white pine strip plank with 3 layers of 1/16" Hard Maple veneer ±45˚ and a final layer of thin glass (it will be a painted hull, no wood exposed on the outside). This is roughly equivalent to the Golden Daisy scantlings btw. You can see the pictures of the scarfed white pine stock. It is so far the most expensive component of the boat. I will be reinforcing key areas (along keel, on the flare of the bow, the bow itself, and the fore foot) with an initial layer of kevlar cloth under the layers of veneer, and a final layer of kevlar on the exterior. GB have some tests on this sort of laminate from an impact resistance pov, and it looks quite good for the amont of extra effort/money.

Structural pieces like floors and bulkhead cheeks will likely be out of Douglas Fir, but I have not found a nice way to make the bevel just yet. (also included is a pic of lamination tests for the bulkhead cheeks out of white pine). I'm having difficulty making the bevels just perfect and so that is bugging me. Considering maybe doing the laminations *after* the hull is complete, however that's not my best case scenario as it would put in a pre-torsion of the laminate, and also why skip a chance to make the laminate shape exactly molded to the mold shape when waiting till after hull completion there is a good chance that the hull skin will sag and change ever so slightly in shape... Ideas on good beveling also welcome.

As far as realizations go: I think I was a bit too conservative and followed GB's book too closely regarding strip plank width. It's 1 1/8" wide strips. Which will come out to little over 100 strips per half of the boat, whereas looking at the spirit yachts, I can see that they clearly used wider strips (possibly even 2 1/2" wide). That would have saved me much time, money, and sawdust on the shop floor.

It's an interesting technique. Moreso because it uses (and this is unusual), American species. I could literally source that wood in my back yard.
Was the choice of pine over cedar due to price? Eastern white cedar is very hard to get clear, and western red costs too much.
Given pine's reasonable qualities of strength/weight, I can understand the choice at maybe a couple of pounds per cubic foot penalty.
Good luck with the bevels. You'll find a way.

A.