BUT FOR DISCUSSION let me throw in some things that imco have to be taken into account for woodwork on a boat.
NEVER USE AN "INTERIOR GLUE" ON A BOAT.
White Oak in lamination cannot be glued with any glue used in the Fine Woodworking test.
Titebond 3 so far as I know has not been tested for white oak bent laminations glued up under tension. Nor in a wet/dry cyclic boat environment.
I have HEARD for years that boatbuilders ("conventional wisdom") will not use any yellow glue because it will eventually creep because it is hygroscopic in the damp. The FW test is primarily for wood joints that will live in a fine moisture controlled frufru situation.
A glaring omission from the FW test is water resistant plastic resin glue you catalyzed with water. It has a long open time, requires clamping, and has almost no glue line. Commercial tillers use it.
Titebond 3 (not yellow but grey) seems to be a different animal, but will need testing for veneer laminations and other wood-to-wood gluing in damp.
The conventional test for glues used for wood-to-wood bonding is a four hour boil test!
There is only one glue that survives this test 100% and that is Resorcinol. This glue requires milled surfaces and pressure clamping, no gap filling, THAT I have always taken to mean very light sanding to deglaze surfaces.
Gorilla Glue has no place on a boat. I have witnessed a number of failures used for furniture. While epoxy is a glue, polyurethane* is not!
I have had my own failure with T-88, and wrote extensively here about my experience with it. My conclusion is that the best off the shelf 2-part STRUCTURAL EPOXY ("tropical wood") is made by Smith & Co.
I have used this glue for white oak laminations (untested) and noticed some starved joints. Imco it is very difficult to squeeze this product out of a clamped joint, BUT it seems to be possible and therefor is not used for bent in form lams.
I have some suggestions which have been reiterated elsewhere. Imco no trustworthy lamination can be made with ANY glue on white oak. Except Resorcinol.
And it may be that the oak must be pre-treated/de-natured to remove tannins.
I epoxy-laminated a white oak beam for A338. I put an epoxy saturated layer of fiberglass between each piece of oak in an attempt to not starve the joints. I also attempted to remove tannin oils with solvent and roughly scoured the faces with 40grit. Years now later I see some tiny separations appearing and decided to bolt the whole damn thing together - which I've described in the Gallery.
I'll always worry about it. Wood will ALWAYS move in a cyclic environment. In my case I have isolated the wood pieces with a hard non-moving material. Doesn't really work. So you see what happens using Resorcinol - there is NO glue joint. Can only describe the phenomenon as a chemical bond - rather than a secondary mechanical bond provided by ALL the other stickums.
WITHOUT QUESTION, it would be much better to bandsaw a new strongback out of solid white oak timber!!!
Ultimately, the best bond/attachment is always made with bolts and screws.
Maybe we should see glue as only in a support role for metal fastenings and as a caulk to keep moisture out of seams.
For wood-to-wood furniture work on the boat my gut feeling is that Titebond 3 can be used.
It is easy to use, there's no mixing, it's water cleanup, and hardly a glue line. Mahogany and fir and plywood, conventional joints like the open joints used in the FW test. Only an assumption.
Would like to see anecdotal evidence on this. [EG, what did Larry Pardey use for the extensive interior woodwork in Tallesin?]
For attaching wood to fiberglass there is also no 'boil test' guarantee that you will get bonding. Therefor fillets and tabbing must be used. And that has to be done with epoxy. Could say this method is an attempt to OVERPOWER the wood.
__________________________________________________ _______________________________
*Polyurethane tube rubber is another discussion.
Reply With Quote