Gluing Scarfed Inwales

George F.

Curious about Wooden Canoes
I am restoring an 18' OT Guide. I have read numerous posts on scarf joints, but am still having difficulties with them. Following are the steps I have taken.

-I am replacing inwales with ash. They have been steam bent (still a bit of spring back) and dried.
-I have cut 10-12" scarf joints and sanded flat. The mating surfaces match well. If it was previously glued and failed, I sanded down to wood before gluing a second time.
-I have tried two types of adhesive: System Three 5 min. epoxy and Titebond III, both recently purchased. Both have failed (epoxy at least 2 times).
- applied well mixed epoxy to both sides, then clamped with 3-4 clamps for 3-6 hours. With Titebond, I coated both sides, let get tacky and clamped for 8 hours. Both cases have squeeze out once clamped.
-I glued the inwale on a flat table (as opposed to clamping and installing in the canoe)
-For the epoxy, I clamped the glued inwale to the canoe ribs from 4-8 hours after the initial glue up, figuring the joint had cured. For Titebond, I clamped them in place 24 hours after initial glue up.
-Both the epoxy and Titebond joints failed after 4-24 hours.
- I have another one in process with epoxy. Coated both surfaces even more liberally and will not unclamp or move them for 24 hours. I will then clamp them to the canoe and cross my fingers!
-The Stelmok and Thurlow book (pg. 171) shows a 9-10" scarf joint with two screws (from the out facing side of the inwale). I want to avoid screws, but will do it if necessary. If it is, anyone know what a good screw size is?
-I also have inwale ends to scarf glue on once I get them laid out and cut to join the stem and deck. They are about 3' long and also have well cut scarf joints from 10-12". I am now worried about these scarfs not gluing well as there seems to be more stress on this joint due to horizontal and vertical bending.

WHAT am I doing wrong? Should I be gluing while the inwale is clamped to the canoe (ribs)?
 
Hmmm. That is a puzzle. I’m sure others with more experience will chime in, but I would have had a lot of confidence that TIII would have worked, had I been doing that job, based on good results I’ve had with TIII.

What I would try in your shoes is dowelling the joint from the back side: that is, drill (say) 4 1/8th holes from the back side all the way through the inner piece, and almost all the way through the outer piece. Then glue up the surfaces as before but also glue dowels into the holes you’ve made for them. This would increase the strength of the joint and be better than using screws.
 
I have scarfed with both TB 3 and with epoxies, although not the 5 minute kind. I do remember having trouble with a piece of white oak one time, but not with ash. Some things to consider is whether the glue is old, the temperature during the dry time, clamping pressure, and joint prep. With TB 3, I put a layer on both pieces and immediately clamp tightly and let dry overnight. The only thing I do differently with epoxy, is I don't clamp the joint as tightly. All joints were glued prior to being installed on the hull. I do about an 8:1 or longer on my scarfs, and use a hand plane to prepare them, to minimize gaps. I suggest you set the inwales aside for now and make practice joints on some scrap until you are satisfied with your results. You should not be able to break the two pieces apart. The wood should fail before the glue. Please report back when you have it figured out. Good luck!
 
To start with, five minute epoxy is mostly junk and shouldn't be used on anything serious. Of that five minutes of hardening time, it probably only has a two or three minutes to penetrate and get a decent grip on the wood's surface before it has become too thick to do so - and that isn't enough time for a good bond on structural woodwork. The proper epoxy procedure would be to use a standard boatbuilder's epoxy, mix it well, give both sides of the joint a light coat of plain resin/hardener mixture and give it a few minutes to soak in. Epoxy doesn't penetrate all that far, and generally doesn't need to for a good bond, but you don't want the wood sucking the resin out of your joints. When scarphing, you are basically glueing to end grain, so it may actually take a couple of rapid-fire coats before the wood stops obviously sucking up resin. Once that happens (no need to wait for it to harden) it is a good idea to mix in a little bit of a structural filler (cotton or linen fibers or short-cut glass fibers, for exampe) for strength and to give you a little better gap-filling capability and coat at least one piece evenly. Assemble the parts and clamp them, but don't go crazy with the clamp pressure. If it's a clean, well-fit joint, excess clamp pressure can squeeze all the epoxy out of the joint. Let it harden and remember that a full epoxy cure can take as long as a week or better, so don't rush it with too much bending stress. For a structural scarph on a piece like that, your scarph should be 10-12 times as long as the stock is thick.

Most wood glue, on the other hand, requires higher clamping pressure and a cleaner fit. They will have varying amounts of gap-filling ability due to their thickness (or not) and also due to shrinkage as they lose their water content while drying. You still want to pre-coat the joint and give it enough time to soak in and stop, to keep glue in the joint, and on structural timber you want a similarly long scarph.
 
Thank you all for your input. Interesting comment regarding System 3 epoxy. I will see if I can find some "standard boatbuilder's" epoxy. Is there a brand you recommend and a place to purchase?
I have plenty of fresh Titebond III. Perhaps I will try that on one of the joints and leave it longer before stressing.
If all else fails, I will give Larry's dowel idea a try. Also sounds like a good option.
I will report back.
Thanks again!
 
System Three makes perfectly good epoxy - but nobody makes great five-minute epoxy. The concept itself is the problem, not the brand. For serious use, especially on porous materials like wood, you want more penetration/saturation time. I use the term penetration with a certain amount of care, since any sort of deep penetration for epoxy bonding doesn't happen much anywhere other than in B.S. advertising and isn't even needed, but still we don't want to rush the limited absorption that we can get.
 
Only 1 small thing to add to the comments received, how smooth are you sanding the joint to?
Be sure to leave some depth to the surface finish so the adhesive has something to grab.

Dan
 
I'd guess that your joint is at fault. My thinking is that if you felt you had to sand it then the 'fit' wasn't perfect to begin with. If that's the case sanding isn't likely to improve the fit. When you glue wood together the fit should be as perfect as possible as the glued bond is strongest where there's wood-to-wood contact.

What I've done is make a fixture I can c-clamp to my radial arm saw that helps me make accurate (and straight) scarf cuts. The fixture consists of a piece of plywood maybe 2' square. It has a slab of wood attached at some angle - say 40deg - to which I can clamp my work to. To use I set the blade to 45deg and position the fixture so the end of the work piece gets cut as I work the blade. Using the above 45deg & 40deg specs I'd get a 5deg cut on the part. With this I can get perfect scarf cuts. The requirement though is that the part being cut must have parallel edges.
 
George
I know it's frustrating, but to date I have had no trouble scarphing joints as Todd describes. I use ash and the same technic as Todd describes..

Any joint that has stress, I leave clamped at least two or three days.

Jim
 
As recommended, I purchased some System 3 T-88 structural epoxy. I then roughed up the joint with a rasp, glued and left clamped and unstressed for at least 24 hours and most often more. Every joint turned out great and I have not had a failure since. Thanks for the help!
 
Good information above, all I might add is using a utility knife to make a cross hatch pattern on the glue surfaces gives the glue a little more penetration. I prefer epoxy, System 3, with a little thickener for the bond, don't strive for perfection on the mating surfaces, no failures so far with white oak, mahogany and cedar.
 
Back
Top