Lane, I recall testing Devcon 5 years ago and it held up OK. I hadn't tested since. It may have changed formula but I'd suggest to any knife maker now that they avoid it.
Ray, you are right about seemingly testing for sheer only but we're trying more than that. Sheer stress - whacking the snot out of them - is the only way to accelerate failure in our meager testing shops. I've cycled through the heat furnace three times, the freezer at least as many and the dishwasher once. These are all different for heat, cold, metal expansion contraction, hot detergent soaking ultimately tested by sudden shear shock - ala the big whack. Steve, who's short movie on testing should be nominated for an Oscar (technical division) I'm still giggling about, testing by vibration, I am testing by impact.
My current test is under way, soaking for two hours in water and then straight to the freezer. This is stabilized wood that has not been sealed so it should have soaked up quite a bit of water. We'll see how that works.
Couple other tests to report. You might recall I was amazed at Loctite 324 Speedbonder load holding 100lbs. I glued up another test piece to verify the results and get a picture this time. The #### stuff did it again. It held 100lbs, the metal started to bend, held for 45 seconds (as I was trying to get the camera into position) and ultimately failed again but I still claim it held 100lbs. It lasted about as long as last time. No picture, again and this time the metal was really bent which is way cool. I did notice that the mechanical force on the test piece being bent changes from shear to peel force and that may account for the failure. In any case, the 324 is amazing stuff for how well it holds. I still think it is too expensive but if it comes out on top, it is what I will use.
I did the same load test with gorilla glue and it failed at loading 50lbs which is the weakest of the bunch so far. This is very disappointing but not neccesarily a deal breaker. I'm not sure two pieces of metal being glued together is fair for this adhesive that requires moisture to kick off. Even though I put plenty of water in the joint, it still didn't look quite right when examining the sheared joint.
One concern with gorilla glue was expansion and the need to tightly clamp to keep the scale from pushing away from the tang (or whatever). I glued some up and put three big clamps on a full 5" strip of metal to wood. When it cured, I ground down the edge along the entire length. There is no visible glue line. (I'll post a pic later) I had a concern that tightly clampling this stuff would some how change how well it held so I banged on it harder than any piece so far and with more thumps than any piece so far and it didn't budge. I continue to gain respect for the polyurethanes to use with metal/wood material handles. I would not use it for metal to metal or metal to micarta/g10/ivory.
I will be retesting the gorilla glue with metal to metal and then a sandwich with liner material soaked in water in between. That should provide the moisture it needs to kick off. I have some wood/metal curing for round 2 if I ever make past round 1. I also have some golf shafting epoxy curing for the load test.
If nothing else we have learned and confirmed that surface preparation is the single most important factor in a sucessful glue up. I understand this maybe a premature leap of logic at this point but I'm going to include my last 5 years of glue ups and aircraft composite studies prior to that for my opinion. I'll remain open minded if testing proves other wise, but I'm fairly confident it won't. I know some adhesives will outperform others but I will bet any one nearly any amount at this point, the best adhesive will not perform as well with poor or even mediocre surface preparation as any of the appropriate, 'average' adhesives would with excellent surface preparation.
Debate and investigation on the best surface preparation process may end up being the most important learning we get from this entire process.
Any one have any opinions?
(now that's funny and I don't care who you are...)