Glue Joints

[Ron Solfest]

View: https://youtube.com/watch?v=m7HxBa9WVis&feature=shares


Came across this video that does an excellent job of testing and describing glue joint failures. Well worth watching if you glue up wood joints. Short answer for those that just want the conclusions:

Lignin strength < Glue joint strength << Cellulose fiber strength

In layman’s terms, all else equal (glue joint length/geometry), roughly:

Side-side strength = side-end strength < end-end strength

Recognize that side-end joints will have long term expansion stresses that will defeat these over time, this was not discussed or tested in the video.

Still, some good testing and discussion that is not what I thought I knew…

 

[Phil Hamel]

Good info.

 

[Richard Coers]

How do you measure lignin strength? What is long term? I have furniture I made over 50 years ago and have had no joint or panel failures. I won't trust this kind of YouTube video unless it came from a certified scientific lab. A resident mythologist can pass on myths as easily as the next guy.

 

[Darryn Achall]

Newton, Edison, Aristotle, Einstein, Tesla, Steve Jobs, Les Paul and many, many more didn't work in a modern certified scientific lab. I don't take this video at face value without further research, but not because it is from a lab without known scientists and certified test equipment. Many of the most important discoveries in history are from home labs and accidental discoveries.

 

[Ron Solfest]

@Richard Coers - I am familiar with hundreds of technical experiments whose results impacted life/death issues, and $billions; along with statistical analysis techniques, 6sigma, etc.

With that said, I’ve also done several glue experiments (and othe woodworking experiments) on my own, that certainly weren’t done in “certified labs” or frankly even using NIST traceable equipment. I base many of my woodworking decisions on experience and “preponderance of evidence.” This particular video appears to use well controlled techniques and adds to my body of knowledge. If you want to dismiss it purely because of lack of certified lab status you’re welcome to. But as @Darryn Achall said, you’ll be dismissing most of learned knowledge that way…

Don’t get me wrong, I agree that “mythologists” have done tremendous harm to our society with conspiracy theories etc lately. But good critical thinking involves understanding and weighing information, not just accepting it because xyz said it without real data; nor categorically dismissing valid data.

 

[John Hebdon]

View: https://youtube.com/watch?v=m7HxBa9WVis&feature=shares


Came across this video that does an excellent job of testing and describing glue joint failures. Well worth watching if you glue up wood joints. Short answer for those that just want the conclusions:

Lignin strength < Glue joint strength << Cellulose fiber strength

In layman’s terms, all else equal (glue joint length/geometry), roughly:

Side-side strength = side-end strength < end-end strength

Recognize that side-end joints will have long term expansion stresses that will defeat these over time, this was not discussed or tested in the video.

Still, some good testing and discussion that is not what I thought I knew…

I watched this the other night. and many more like it. I was shocked about the end grain joints being stronger than edge clued joints. Very good video. Thanks for sharing.

 

[Roger Wiegand]

I do a lot of repair and rebuilding of moderately old wooden constructs (50-150 years). My non-rigorous observation is that most flat out failures of glue joint involve 1) end grain and 2) cross-grain joints, and that I rarely (water damage being the exception) have occasion to need to repair side grain joints. In rebuilding organs from that period I'd say that pretty much every time I've found an end grain glue joint (not often, those guys were pretty careful woodworkers) it has failed. So I don't depend on glue to hold end grain to anything.

 

[Richard Coers]

I guess I hit several nerves by criticizing Patrick Sullivan.

 

[Ron Solfest]

I guess I hit several nerves by criticizing Patrick Sullivan.

Sorry if I came across harsh.

I don’t know Patrick Sullivan ‘from Adam’ but the methodology he used seemed sound and credible. I was just reacting to dismissing data out of hand as a poor example of critical thinking…and poor critical thinking has become a hot button for me the last few years…

Again, sorry if I overreacted

 

[Brian Gustin]

I was just reacting to dismissing data out of hand as a poor example of critical thinking…and poor critical thinking has become a hot button for me the last few years…

You and me, both.

 

[Glenn C Roberts]

Critical thinking? Just the opposite.

 

[Kirk Yarina]

Patrick Sullivan has caused some controversy in the wood-not-turning worker community with his glue strength tests. This an his other videos are well worth watching, form your own opinions

 

[robo hippy]

I think I saw this a long time ago, but not positive. The testing process seems pretty complete. It does appear that when he glues the end grain pine together, he applies more glue to those surfaces than he does to the side grain glue ups. That may be a critical point since the glue needs to penetrate the end grain fibers a bit more than side grain. Main question this brings up for me is how many glue failures are due to bad glue up technique?

robo hippy

 

[George Kuipers]

Testing after 2 days is no problem, but what about after 2 years. Wood moves, dries out etc.
Test again after 2 years and a total different picture emerges.

 

[Russ Braun]

Us Segmenters typically stagger our joints for the added strength. My pieces have many glue joints; I have seen them bounce off the floor w/o structural damage at all.

I think I have seen this or a similar glue “test” before. There are variables that many shops encounter that may not enter in these test results. One of my favorites is, “how old is your glue”? I know glue has a shelf life; I have never looked for an exportation date, the glue has always stuck!

 

[Tom De Winter]

From Titebond's website:
Our literature states the shelf life of a majority of our wood glues as two years. Most of our yellow and white glues, including Titebond Original and Titebond II, remain usable beyond two years. Should Titebond Original become thick and stringy, or Titebond II turn into an orange-colored gel, these changes signify that the glue is no longer usable. The minimum shelf life of Titebond III is stated as one two years. When stored appropriately at room temperature, Titebond III is expected to last beyond its stated shelf life. If thickened, shake vigorously by firmly tapping bottle on a hard surface until product is restored to original form.

 

[Richard Morgan]

Main question this brings up for me is how many glue failures are due to bad glue up technique?

My guess, based on 40 years as carpenter, cabinet maker and furniture builder is more than 90%, including some of my own early pieces that I have rebuilt and still live with. Of the glue failures I've seen and worked on the most common problem would be glue starved joints, especially where the woodworker applied a thin line of glue and clamped up so that there was only a 1/4" wide line of glue trying to hold a joint in 3/4 stock. Poorly fitting tenons would also be a big part of that.