Home Articles Forums Gallery Marketplace Videos

WoodCentral Forums

Knowledge You Can Build On — Est. 1998

Nearly 100 Exotic Wood Species, Direct-Imported Since 1972. Shop Slabs & Turning Stock.

Panel joint failures

Posts

Panel joint failures

#1

Panel joint failures

Bill Tindall, E.Tn.

All the glues commonly used to assemble panels are stronger than the wood. (This statement is only true at when the glue is not in a hot environment or a severely damp environment, both of which weaken most glues substantially.) Hence, stress on a glued up panel should fracture it at some location other than the glue joint.

If the joint failed there was some defect in the glue, machining of the joint or its assembly. The Franklin trouble shooting guide to wood adhesives lists more than a dozen ways that the process and materials of making the joint can be defective. I will describe 3 in addition to excessively damp and/or hot ambient conditions.

Insufficient clamping pressure to squeeze excess glue from the joint- If the glue is stronger than the wood why does a wide glue joint fail? The glue is a mixture of water and polymer particles, as I recall about 50% solids. When the water leaves from a fat glue line voids are created from the vacant space where the water used to be. Microscopic examination of the cured glue would reveal that it looks like Styrofoam, or empty honeycomb in the worst cases. While the glue remains strong the air bubbles holding the joint together are not.

Too long to assemble the joint- The open time for our common glues can be less than 5 min under hot or dry conditions. If the glue begins to dry before the parts are mated several bad things can happen that will prevent the mating surfaces from being wet by the glue and/or the particles in the glue from coalescing to form a continuous polymer film between the mating parts.

Glue applied to cold lumber- The glue forms a bond by water leaving the glue line and the particles of polymer that remain coalescing into a continuous tough film. The mass of the lumber is vastly greater than the mass of the glue, so it is the lumber temperature that rules, not that of the glue. To coalesce, the particles must be mushy. If they are too cold they are not suffieiently mushy. 55 degrees Fahrenheit is the lowest temperature that ordinary Titeboand can be used, for example. (Latex paint and Titebond glue are essentially the same physical materials, what in polymer chemistry is called a "latex". Apply the paint when it is too cold and the surface will be grainy because the particles have failed to coalesce. All the problems and precautions of painting apply to gluing, for the materials are similar )

Re: Panel joint failures

#2

Does this suppose...

John in NM

That a 100% correctly made and assembled joint will stay together indefinitely - that glue failure over the course of decades is attributable to improper use alone?

I tend to think that this is a reasonable assumption, within certain limitations. However, I know of no way of proving it beyond a 50 year test. Anything less still requires a degree of simulation that is not necessarily valid when testing requires a time component.

Re: Panel joint failures

#3

Joint failures

Lee Schierer in McKean, PA

The joint that is most likely to fail is either the one that is most visible or the one that will cause something inordinately expensive to crash to the floor.

Lee

Re: Panel joint failures

#4

Re: Bad science and good science.

Tom in Toronto

Failure is not a single cause thing, even when everything about assembly was done correctly. Loads differ. Most woodworking journals take the smash it approach. But long term failure are often creep related, and will vary a lot in environments that cycle many small loads and a few large ones. An interesting example of the latter being when a spar falls over in a marina. The failure is often the result of all the 10% loads over a high cycles thing. So just what the wave motion on the dock does, and the wind. So creep is a big factor there.

http://epoxyworks.com/34/pdf/Adhesive%20Bonded%20Structures.pdf

http://epoxyworks.com/31/pdf/So%20Stands%20the%20Mighty%20Oak.pdf

http://epoxyworks.com/25/pdf/Understanding_Flexible.pdf

http://epoxyworks.com/21/importance.html

http://epoxyworks.com/19/Protective.html

http://epoxyworks.com/21/woods.html

I recently saw a youtube video that pointed out that yellow glue type bonds did not fail at well below normal levels when there was a gap. This was big news to me, though it was in a whack-a-mole testing environment. See here:

http://www.youtube.com/watch?v=MUWu-N85oXM