In Shop Blog

We may receive a commission when you use our affiliate links. However, this does not impact our recommendations.

Editor’s note: The following post is from James Wright of Wood By Wright on YouTube. He recently conducted a massive test of common wood glues and I asked him to share a few discoveries on the Popular Woodworking blog. Links at the bottom of the post are affiliate links that support James. Be sure to check out his YouTube channel that has his complete presentation of the results! – David Lyell

Which is the best wood glue? This was not the original question to start started this test off, but it soon became the goal of the whole test. Initially, I just wanted to know if there was any benefit to using dry hide glue over one of the commonly available liquid choices. I decided that if I was already doing a glue test, I might as well throw in a couple of the PVAs and see how those do, and in that case, I might as well test epoxy and the list began to grow from there. In the end, there were a total 32 types of glue pitted against each other to find out which was the best choice for one of four different situations: long grain to long grain, long grain to end grain, gap-filling, and exterior conditions. Each glue would be tested 10 times in each of the four scenarios. The data would be collected by using a calibrated load cell to test the actual breaking strength of every single block of wood. What was originally going to be a simple test in one afternoon with a set of lifting weights, ended up taking over 2 months and over $1,000 in materials and testing supplies. I was really interested to see if any of my assumptions stood up to the rigors of an actual test.

A full overview of the test can be found at Wood By Wright on Youtube:

1. Super glue is surprisingly resilient

In the long grain to end grain test, the gap filling test, and the exterior test, the super glues or cyanoacrylate (CA) glues did exceptionally well. In most cases, they beat out the standard epoxies and PVA glues. The tests demonstrated a strong correlation between the thickness of the CA glue to the strength of its bond. The thinner the glue the weaker the strength, but with the thick gel CA glues actually work very well as a gap-filling or end grain glue. The glues were also tested with the activator to see if the strength of the glue changed, most of the time it did not. With the thick gel CA, the activator did harm the strength of the glue. Though more tests are needed to assess for sure why, I believe it has something to do with not allowing the gel enough time to soak into the wood before it was activated. I am intrigued to see if CA glues can maintain their strength over time. Further testing needs to be completed, but that will probably be the next glue test.

2. Polyurethane glues are not for gap-filling

When I first got started in hand tool woodworking, I built a bench and the joints were not as clean as they should have been. It was suggested I should use gorilla glue or another polyurethane glue to fill the gaps in the joints. This solution seems to make sense and tends to be the common knowledge answer to gap filling in less than perfect joints. As a result, I fully expected the Gorilla Glue to hold up very well filling a large gap, but I was extremely surprised. Not only did the Gorilla Glue fail early, but it was consistently at the bottom of the list of glues used for gap-filling. You would be far better off using a thick or gel CA glue to fill a gap than polyurethane. The only thing worse at gap filling was the contact cement, and that’s not by much. Moral of the story, do not use Gorilla Glue to fill gaps.

3. There’s little difference between liquid hide glue and dry hide glue

One of the key things I wanted to find out in the test was whether there was a difference between liquid hide glue, such as old brown high glue and Titebond hide glue, and dry hide glues that you mix with water in a double boiler. Though there is much to be said for the fun of making your own high glue in a double boiler, the test showed that in most instances the liquid hide glues did just as well, if not better, than their dry counterparts. The data was enough to show me that, unless I really want the fun of making my own high glue, the liquid Titebond hide glue would do just as well as an average wood joint glue.

4. Titebond II and Titebond III did not stand up well in exterior conditions

The Titebond III label has “waterproof” and the Titebond II label has “water resistant”. In the past, if a project was going to get wet, I would use Titebond III to glue the joints together. All PVA glues, including the Titebond series, did very poorly in the exterior conditions. Several of the blocks from Titebond II and Titebond III did not even make it to the test as they fell off when being carried to the testing rig. The evidence was more than enough to convince me that from now on, in exterior conditions, I will be using epoxy or even CA Glue instead of the PVA glues.

5. PVA glue is a fantastic wood glue.

One expectation I had was that the PVA glues would be vastly blown away by the superiority of epoxies and other high tech glues. I was pleasantly surprised that for average uses in gap-filling and long grain to long grain, the PVA’s stood up very nicely to the far more expensive epoxies. The strength for most of the PVA glues was far more than would be needed for any joint. This reaffirmed my trust in PVA and I will continue to use Titebond II as my go-to wood glue.


This test was designed to find out the strength of the glues regardless of the strength of the wood. Almost all of the glues tested would cause the average wood to fail long before the glue would fail. While some of the numbers are very impressive the glues tend to be far stronger than they need to be. Many of the assumptions going into the test were destroyed, several were upheld and reaffirmed. For most wood glue joints, PVA is a fantastic glue that will serve the user very well. Hide glues are a very respectable glue with several benefits other than their strength. For exterior conditions, epoxies really are worth the expense and time needed to complete the project. Although in the past I have laughed at the idea of using a CA glue, I will now be adding them to my normal use as legitimate wood glues.

Spreadsheet of compiled test results

Testing equipment

Load Cell used:
Cell Reader:

A complete list of glues used

Titebond hide glue,
Old Brown,
315 hide glue,
251 hide glue,
192 hide glue,
Homemade hide glue,
Titebond Original
Titebond II,
Titebond III,
White wood glue,
White Elmer’s craft glue,
Super glue and wood Glue, (2P10 Thick and Titebond II)
Cheap Super Glue,
Fix All Super Glue
Gorilla Glue,
3M 5200,
West systems 205 fast,
West systems 206 slow,
east coast resin,
5-minute epoxy,
Dap Dynagrip foamboard,
Thinned barge,
Super 77,
Casein glue,

For detailed information on how the test was conducted, you can watch this video.

Product Recommendations

Here are some supplies and tools we find essential in our everyday work around the shop. We may receive a commission from sales referred by our links; however, we have carefully selected these products for their usefulness and quality.

Recent Posts

Start typing and press Enter to search