Why is a floating tenon weaker than a tenon?

[Michael Rutman]

I can watch YouTube videos of it being tested and see it’s true, so I don’t doubt it’s true. I fully believe it’s true, I just can’t grasp why.

In every single test I see the wood on the mortise side rips apart before the glue. If the loose tenon pulled out I’d understand it, but that isn’t happening. If the glue failed, again, I’d understand it. I understand that the tenon side is going to be extremely strong, but again, even in a loose tenon that’s not the failure point.

For what reason is a loose tenon like a Domino weaker if it’s the wood itself that fails?

[johnny means]

It's not. All those videos are flawed.

[Doug Garson]

Maybe post a link to one or two of the videos you refer to.

[Thomas McCurnin]

I don't believe its true, and can give you YouTube videos which allegedly prove that dowels as strong as mortise and tenons joints. I think it is all BS, and none of my furniture would be put to that kind of stress to worry about the strength of floating tenons, traditional mortise and tenons, or dowels joints.

https://www.youtube.com/watch?v=WoOEwEuB1ag

[Kevin Jenness]

From a theoretical standpoint a "true" tenon is rooted in the rail and fully integrated with the surrounding long grain fibers and possibly stronger. As a practical matter a properly sized, fitted and glued spline tenon is fine. In a rail with grain runout an inserted tenon may be stronger. There are situations like glass bars or chair rails with tenons close to one face where there is not enough cross section to use a spline tenon safely.

[Carl Beckett]

and none of my furniture would be put to that kind of stress to worry about the strength of floating tenons, traditional mortise and tenons, or dowels joints.

Except chairs. Chairs take tremendous stress in the joints. I read one person that would only use hide glue on chairs because he felt it was inevitable they would loosen with age, thus needed to be repairable.

[Michael Rutman]

https://www.youtube.com/watch?v=Nh4H9maVjFM

I have a vague memory of tests where they pulled straight out but I’m not finding those anywhere.

I am comfortable with my dominos and pocket joints as strong enough, just curious really.

[Scott Clausen]

Could be splitting hairs here but a floating tenon will not likely have a bond on either end and may even have a gap on one or both mortise bottoms. A normal tenon will have a firm solid connection on at least one end with no room to start movement as failure is induced. How this effects the real world I have no idea. I am only speculating that there could be a marginal increase of strength on one end. I say make the joint the way you want, it is too stressful to stress over stress points.

[David Kumm]

Any difference is only relevant if you are attaching wings to an airplane with them. Dave

[Prashun Patel]

This isn't a formal study, just my anecdotal experience: It's hard to make a perfectly fitting tenon in both width and height. Unless you get it perfect, the integral tenon seems to have a mechanical advantage. So, a one-sided tenon feels more stable than a floating one.

I use a ton of floating tenons (Dominos). Occasionally, well after glue up, I have needed to take the joint apart. I am usually able to shock the joint apart. Now that might be because I am striking it with a hammer and it's shocking the glue, but still. The integral tenons when struck, almost always dislodge from the mortise; they don't split on the integral side. That being said, floating tenons on tables I've made have never failed. Integral tenons - with wedges - on a windsor chair or two I've made have loosened . So, I strongly suspect it's a moot point. The strength of these two joints is more determined by usage and the maker than the type of tenon.

.

[Ronald Blue]

Except chairs. Chairs take tremendous stress in the joints. I read one person that would only use hide glue on chairs because he felt it was inevitable they would loosen with age, thus needed to be repairable.

If the joint has loosened wouldn't it be possible to disassemble the joint regardless of the glue used? What am I missing here?

[Edward Weber]

It's not. All those videos are flawed.

Agreed
While every situation is different just saying a floating tenon is weaker than a traditional tenon is a bold statement to make. There are so many variables that go into proper joinery, it's pretty arrogant of some of these videos to make such claims.

[Jim Dwight]

I've seen some tests too where domino joints came out "weaker". They all used pre-fabricated dominos, however, which were smaller than the integral tenons. It is not a secret that smaller mortise and tenon joints are weaker than bigger ones. Scott's point is valid. Festool pre-made tenons are about 2mm shorter than the labeled length to allow room for glue at the bottom. But I don't think the lack of good end grain to end grain connection matters much. Correctly made mortise and tenon joints get their strength from the side grain to side grain connections. I make my own loose tenons but I allow about the same amount of gap.

I have 10 chairs sitting in my dining room with domino made loose mortise and tenon joints. They have been used with no issues. I have four in my great room with integral tenons that are over 10 years old of basically the same design. They are all fine too. The "secret" is making a good fitting and correctly sized joint. One reason I make my own tenons is I don't want to increase the temptation to get lazy and use a smaller tenon than the project needs because I have those ready to go. I do make sticks about 3 feet long of tenon stock in advance so I still have some temptation but for something like a chair, I do not use a smaller joint than called for in the plans.

There is no real limitation from the domino machine except the maximum plunge depth of the 700 I use is 70mm (~2 3/4 inches). If you want a mortise wider than 14mm you can overlap the holes to get what you need. Normally that is not an issue for me but the width is often a bit small. It is even easier to overlap holes in this direction, just plunge about every 1/2 inch (depending a little on the cutter). Making the joint wider significantly increases side grain connection so it significantly increases strength.

You do have two glue joints to glue up per joint which also adds opportunity for error but I don't think that is often the problem. I put glue on both the mortise and the tenon - standard practice.

In short, the issue in the tests I've seen is size, not method of making the joint. If you make the loose tenon joint smaller it will be weaker. But why do that? And then pretend the issue is with the machine?

[Cliff Polubinsky]

If the joint has loosened wouldn't it be possible to disassemble the joint regardless of the glue used? What am I missing here?

The issue is in putting it back together. Hide glue will adhere to hide glue so you can just add new glue. PVA glue doesn't adhere to itself, it needs fresh wood so you have to remove all the old glue. You end up making the mortise wider and the tenon thinner.

Cliff

[Rich Engelhardt]

If the joint has loosened wouldn't it be possible to disassemble the joint regardless of the glue used? What am I missing here?

I can answer that from recent experience!
Sometimes you have to take apart a tightly glued part to get one of the loose ones repaired.
If you can't get the old glue soft enough, you can snap off a part - like I did. I had a loose stretcher & needed to get the legs a little bit apart so I could get the stretcher out.
I ended up snapping the leg off at the seat.