joinery strength tests

[Michelle Rich]

A month or two ago, I posted about my use for years and over 70 projects, of a beadlock jig and said how well it worked for me and that it was on sale at Rockler. I called it the "poor woman's Domino" . Last night I was reading old Fine Woodworking magazines and ran across a test they did on all methods of joinery. What I found VERY interesting was the tests done on the Dowelmax, the Beadlock and the Domino. I thought ya'll might be interested.
The list shows pounds of force to break the joint:
BEADLOCK 836
DOWELMAX 759
DOMINO 597

Interesting, YES? So if strength and $$$$ matters, the Beadlock is a heck of a choice.
FWW Jan 2009

[Lee Schierer]

Did they test traditional wood joints as well? If so how did they compare?

[Matt Day]

We're the sizes of the loose tenons the same or close? How many dowels, spacing of dominos, etc.
It's tough for us engineer types to my have all the information.

[Jim Dwight]

It's worth googling to find the pdf entitled joinery_failure.pdf. It's an interesting read. Wood did a similar test but I couldn't find that on-line, just a video. The Fine Woodworking test used 8x2.5 inch pieces of cherry 3/4 thick. The strongest joint was a halflap which tested at 1603 lbs. A bridle joint was 1560, a splined miter 1498, a 3/8 mortise and tenon 1444 and a floating mortise and tenon 1396. A 5/16 M&T was 988 and a 1/4 was 717 lbs. I found the M&T results interesting. They suggest we should make the tenon 1/2 and not 1/3 the thickness. The commentary in the article seems reasonably balanced and notes that the stress they tested is not the only consideration. A half lap wouldn't resist a twisting type load very well, I suspect. I think the low results of the domino relate to it's width. I think they used a 10mm but I also think it was 25mm wide. That looks like it is not much more than half the width of the M&T joints. The dowel joints came out well but they used 3 so the area is greater than the domino. The results correlate well with the amount of long grain gluing area - except for possibly the splined miter. The miter wasn't that big, 1/4 thick by maybe 3/4 wide but it ran the length of the miter so it was pretty long. They just made up right angle pieces and pushed on the ends until the joint failed.

Overall it is interesting and I think you can draw some conclusions but I wouldn't say that the domino was exactly dismissed by the results. It does indicate that the lack of multiple widths of the little loose tenons could affect the joint strength for some material sizes, perhaps.

In my googling I found another test where the author claimed to test tight, loose, and normal fitting M&T joints and found little difference in strength. I also found that of interest. That would suggest that the really nice mortises the Domino machine makes are not a big advantage in joint strength (just speed and perhaps accuracy).

Pocket screws tested at 698 and a single biscuit at 545lbs. I wouldn't rate a pocket screw joint stronger than a domino but that is the test result they got. But other applications of dominos are highly likely to test differently. They used one 10mmx50mm and might have squeezed another in likely making the joint far stronger.

[Bill Adamsen]

Jim, I'd seen that but it was a great refresher. Fascinating that the cope&stick is weaker than the butt joint!

[Art Mann]

I use the Dowelmax and so have been interested in strength tests using that jig versus other methods. Every reasonably scientific experiment I have ever seen showed joints created with the Dowelmax using the methodology from their website yielded joints that were the equal of most other methods and slightly superior to Domino floating tenons. I don't remember as much about Beadlock joints but I recall they were also very highly rated as well. In spite of this, I wish I had a Domino due to the speed and convenience. People need to remember that if a joint is strong enough to survive the application in which it is used, then it is, by definition, good enough. In applications where appearance is not of very great importance, like shop cabinets, I mostly use pocket screws and I have never had a joint fail.

[Steve Baumgartner]

I've seen many examples of this kind of tests over the years and found that they are rarely consistent. One says joint A is strongest, another joint B... From reading all these tests, I've come to some skeptical conclusions. People don't want an engineering analysis, they want a single number or an A,B,C ranking. So the tests are oversimplified. People who like a particular joinery technique tend to pounce on any report that favors it and reject ones that favor another.

- The tests rarely include enough samples to have any statistical significance and the reports don't talk about consistency from sample to sample. They only test a single species of wood and kind of glue, but interpret the outcome as applicable to all.

- The loads at failure are nearly always greater than what you can expect in real life, so the differences are somewhat academic unless you design very spindly stuff. Other factors such as cost of the tooling, precision (consistency), and time spent making each joint may actually matter more than raw strength of a new joint. This is part of the point the OP was making.

- The tests nearly always involve loading the joint in one plane parallel to its parts. Most often they make an "L" and squeeze the ends of the legs together or clamp one leg and push the end of the other one parallel to it, lever-fashion. That kind of test may be relevant for the seat frame to back leg of a chair when someone tilts back. But even for a chair, squirming around in the seat is more common and involves multi-directional loads including twisting.

- The tests are always of a single, isolated joint. Real furniture is an assembly in which multiple joints often reinforce each other. For example, a rectangular frame with a joint at each corner is much stronger than you would conclude by testing one corner without the others. The stretchers between chair legs are added for a reason!

- The greatest strength factor revealed by these tests seems to be the area of glued surface. With most joint systems you can increase the area by using more tenons, larger tenons, etc. But the tests assume that you are just using a particular system "out of the box" without analyzing the loads and adjusting the joinery. Comparing a 6mm domino with a 1" wide beadlock tilts the playing field!

- A large fraction of the joints fail by the wood breaking adjacent to the joint. The amount of "meat" left in critical stress areas is therefore nearly as important as glue area.

And after all that, my bottom line is that the tests don't tell you much about the way that real furniture joints fail. After years of aging, seasonal cycling of expansion and contraction, compression due to loads, etc., the joints become loose. Then they either just fall apart or break because the looseness puts all the stress in a small area.

[Prashun Patel]

I wish someone would do a test of 'how much pounds-of-force are required for an application'.

That is, for the average bench, or chair or table, how strong should the joint be? Then I could calculate how many dominos and dowels I need or whether a stronger joint is required.

These discussions tend to get into the apples-to-oranges flawed design of the test. But I'm much more interested in knowing what's good enough for what application.

I'm learning over a time, but it'd be neat to have a reference so I could be comfortable when building something new.

[Brian Holcombe]

IMO these tests are minimally useful in practical application. Yes, you need strong joinery, however in my opinion it is far more important to know what stays strong over time. What will stay tight over time.

[Bill Adamsen]

Or to Prashun's point ... even the physics of furniture building ... how to calculate the stresses (and strength required) would be great. There has got to be a manual out there for that!

Separately on testing, quite a number of years ago people came up with the idea of skinning their wooden racing sailboats with fabric and plastic to add strength and especially impact resistance. One of the boats subject to this had a large number of boats built, and a group of people in leadership position concerned about the impact to the class, especially if a chosen approach led to failure and loss of participation in racing. The question became ... what can we expect from the alternative solutions, and what's the best approach (combination of materials) leading to a durable durable. What is the solution everyone should follow! Some engineers at Union Carbide helped develop and run tests on the strengths of various combinations (cloth types and weights - dynel, glass, polyester) and adhesives (polyesters, epoxies). Replications of different woods representing the wood types used (mahogany, luan) in the boats were created, and the different combinations tested on a test stand. One could find fault in the approach (not enough replications, did it simulate actual use and aging) but the results were definitive. My recollection was that the bending oscillations before failure ranged from about 6 ... to 50,000. Hyperbole for sure. But testing of this ilk can provide directional (like focus groups for user experience) insights that in this case could likely have contributed to saving the class - which is racing and more popular than ever - today.

[Malcolm McLeod]

Any testing regime needs to be repeatable, both for the tester, and for others trying to validate the result. Otherwise, the results are questionable. That's why most of the tests you see in mags are rather basic - even one-dimensional - as cited above. If the tester removes many of the variables and focuses their test on a single feature, it is MUCH easier to get repeatable results.

As for the strength of a joint, I believe a great deal of any joint's strength comes from the width of joinery 'device' (2.5in. wide lap joint > 2in. beadlock > 25mm (1in.) Domino). You just have to figure out how to align the greatest width of the device with the greatest load to be applied to the joint. (...I'll let you know if I ever get close to this light-bulb moment.)

[Mike Henderson]

For furniture, most of the joints are not under much of a load, so it doesn't matter a whole lot what kind of joinery you use.

The one joint that is critical is on chairs (dining room type chairs to be specific) and the joint that joins the seat to the back. On that joint, the major factor in survival of the joint over time is the long-grain-to-long-grain surface area. And the reason is that the joinery fails by wood failure, and not by glue failure.

On that joint, I would expect that a beadlock, domino, shop made loose tenon, or regular M&T would all work pretty much the same, provided they all had the same long-grain-to-long-grain surface area.

The reason that dowels often fail in that joint is that you generally can't get the same surface area with dowels as with the above techniques.

Mike

[Art Mann]

Take a look at the report to which Jim Dwight referred and you might feel differently. I was an engineer for a major car company and then a defense contractor for most of my career and have done countless studies using this kind of methodology. While some of your points have some validity, they do not negate the value if this sort of study. See comments in red.


I've seen many examples of this kind of tests over the years and found that they are rarely consistent. One says joint A is strongest, another joint B... From reading all these tests, I've come to some skeptical conclusions. People don't want an engineering analysis, they want a single number or an A,B,C ranking. So the tests are oversimplified. People who like a particular joinery technique tend to pounce on any report that favors it and reject ones that favor another.

- The tests rarely include enough samples to have any statistical significance and the reports don't talk about consistency from sample to sample. They only test a single species of wood and kind of glue, but interpret the outcome as applicable to all.

The study referred to above used 5 samples of each type of joint, which is enough to be statistically valid. I don't know whether the test was done, but an "ANOVA" calculation would have predicted whether the variation was due to sample variation or method variation or the percentage of each. In view of the credentials of the researcher, I would guess he did that or something similar. Although I would be interested in such an analysis, I would estimate that 99.95% of all the readers of this study would not. It is understandable why such information would be left out.

It isn't reasonable to claim that if you only test one species of wood, one type of glue, etc. that you can't infer generally applicable information. In this case, the length, cross sectional area and spacing geometry of the connecting pieces, among other things, was what was being measured. Using multiple species of wood and multiple glue types would have dictated that several hundreds or thousands of samples would be needed and little additional information would have been gained. The important thing about this type of study is to define what it is you are measuring and eliminate all other variables.

- The loads at failure are nearly always greater than what you can expect in real life, so the differences are somewhat academic unless you design very spindly stuff. Other factors such as cost of the tooling, precision (consistency), and time spent making each joint may actually matter more than raw strength of a new joint. This is part of the point the OP was making.

Yes!! I often use pocket screws for non critical and non appearance applications.

- The tests nearly always involve loading the joint in one plane parallel to its parts. Most often they make an "L" and squeeze the ends of the legs together or clamp one leg and push the end of the other one parallel to it, lever-fashion. That kind of test may be relevant for the seat frame to back leg of a chair when someone tilts back. But even for a chair, squirming around in the seat is more common and involves multi-directional loads including twisting.

By definition, this particular study set out to measure the ability of various joints to withstand racking. That is the most important property you can measure. Racking resistance in one joint is what prevents the twisting forces you describe in other joints.

- The tests are always of a single, isolated joint. Real furniture is an assembly in which multiple joints often reinforce each other. For example, a rectangular frame with a joint at each corner is much stronger than you would conclude by testing one corner without the others. The stretchers between chair legs are added for a reason!

Refer to previous comment.

- The greatest strength factor revealed by these tests seems to be the area of glued surface. With most joint systems you can increase the area by using more tenons, larger tenons, etc. But the tests assume that you are just using a particular system "out of the box" without analyzing the loads and adjusting the joinery. Comparing a 6mm domino with a 1" wide beadlock tilts the playing field!

No. The greatest strength factor in these tests is the cross sectional area, the length and the spacing geometry of the connecting material. Surface area is a measure of glue strength and except in 2 or 3 predictable cases, (a ridiculous but joint being one) the glue didn't fail. The "playing field" as you call it, is defined by the joint the woodworker chooses. That is precisely what we want to know. In this particular study, the Beadlock joint was stronger than the Domino joint because the Domino limited the size and shape of the connecting material. That is valuable information.

- A large fraction of the joints fail by the wood breaking adjacent to the joint. The amount of "meat" left in critical stress areas is therefore nearly as important as glue area.

In nearly all cases, wood breakage near the joint is the root cause of the failure. The glue joint didn't fail and isn't relevant. At the expense of repeating myself, the greatest strength factor in these tests is the cross sectional area, the length and the spacing geometry of the connecting material. Obviously, that is related to the joinery method used. That is, in fact, the exact information we are looking for in this type of study.

And after all that, my bottom line is that the tests don't tell you much about the way that real furniture joints fail. After years of aging, seasonal cycling of expansion and contraction, compression due to loads, etc., the joints become loose. Then they either just fall apart or break because the looseness puts all the stress in a small area.

In the short term, studies like the one Jim mentioned will tell you a lot about whether a particular joint is adequate for a particular application and are, therefore, very useful in my opinion. In the long run (decades or centuries), the glue material becomes more and more important. There are tests which will artificially age assemblies to test for long term material failure. They are typically outside the scope of what a typical woodworking magazine can do.

[ian maybury]

I'd be much less comfortable about this sort of testing. (and i haven't read the linked piece, only others outside of the US) The devil is often in the detail. (or in the mind of those setting up the trial - and i'm not even sure that mags necessarily understand what's happening) I've seen other mags run tests which to my mind were either badly designed, or were purposely configured to favour one or other (well actually in the case i'm thinking of probably one) jointing system. (for a possible hint take a look and see which system has a large ad running in the same issue)

There's multiple variables involved, but (a) one scenario sees a joint width/dimension picked that maximises the number of joining elements for one system, and disadvantages the other by placing it in the opposite situation (depending on the recommended pitching or recommended size of the dowels/tenons/whatever for the width and thickness of wood); and (b) lots depends on the direction in which the force is applied/the mode of loading of the joint versus the direction in which the particular joint is strongest.

The classic T joint that gets tested by applying a hydraulic cylinder at right angles to the vertical leg of the T can as an example deliver a less than level playing field too. When the force is applied down close to the joint the jointing elements are placed in almost pure shear (simple sideways slip), if higher up the bar then it gets converted into mostly racking/a bending moment - which concentrates a heavy pull out force on to the first jointing element nearest to the side the load is applied to. The other end of the joint is subjected only to a simple compressive force as the bar tries to pivot around the furthest away corner. i.e. it bears on the horizontal, and the jointing element(s) at that end don't have to do very much at all).

It's always the same. A chair for example at first sight has to carry certain absolute loads arising from the weight of the sitter, but it's probably in the end the tendency of people to lean back in/push chairs back and the like (loading it in heaven knows what ways) that eventually kills the joint. Think it's always necessary to either use a joint layout proven in the application, or else to very carefully work through the likely ways in which the joint will be loaded in use and which jointing method best handles this before making a selection.

Simplistic tests of the sort normally used by and large load joints in one of many modes, and may not be representative of any useful reality at all. Joe Punter unfortunately has a tendency to respond at the level of stronger = better, and to hell with the details/precise format of the test - and most makers of jointing systems will probably be happy to grab anything (and to hell with the detail/qualifying the claim) that justifies them in making such a boast. There's to my mind in practice unlikely to be a single generic test using a single mode of loading that delivers better than a highly qualified suggestion of what a given joint is capable of in a given situation....

[Jim German]

As others have mentioend there are lots of flaws in this sort of testing, and claiming that one type is better than another because of the results. Pocket screws particularly seem to be hammered on because they aren't the strongest joint and alot of people look down on them as cheating.

Two things to consider, the first is that you don't usually need a ridiculously strong joint. Frequently the joint is under minimal stress, and in those cases there is no point in going nuts making a ridiculously strong joint. There are also many times where even the strongest joint isn't going to survive, so its not worth the effort. Leaning back in a chair may be one of those situations. Lastly, there is no point in making a joint that is stronger than the surrounding material, you just have to keep it as strong.

The next point is that there are many different types of strength. There is tensile strength, compressive strength, shear strength, bearing strength, fatigue strength, impact strength, and so on. While one joint may be strong in one method, it may not be in another.

Therefore its worth recognizing the strengths and weakness of all the different types of joints, and not looking down on certain types just because their ultimate strength isn't as great as a mortise and tenon.

[Malcolm McLeod]

Bill,
Ian & Art have let slip the dogs of Mechanical Engineering, so I'll add my $.02.

Its relatively easy to do a force diagram - simply a stick representation of the piece with distances and applied forces. (If its a chair, who will sit in it? A grandchild, or your favorite NFL tackle?) Generally, for any design the torque loads are more critical, so calculate force X distance and sum them around a given joint.

Figure out what joinery method you want to use & the magazine load ratings will get you close.

[Art Mann]

More than one person has stated or implied that test-to-failure joint testing is not valuable because it doesn't allow for every possible joint loading geometry. I say that is ridiculous. Simple racking tests will be a very good predictor of joint reliability in most furniture building and other applications. If an individual has some other primary type of load - like pure shear or perhaps pure tension - then the racking test is only moderately useful for comparison purposes - not definitive but possibly better than nothing. I think it is also pretty silly for people to imply that that testing is typically biased by the investigator toward one method or another. If you compare several of these studies, they all come to pretty much the same basic conclusion. This includes the ones done by the tool manufacturers.

[Pat Barry]

I wish someone would do a test of 'how much pounds-of-force are required for an application'.

That is, for the average bench, or chair or table, how strong should the joint be? Then I could calculate how many dominos and dowels I need or whether a stronger joint is required.

These discussions tend to get into the apples-to-oranges flawed design of the test. But I'm much more interested in knowing what's good enough for what application.

I'm learning over a time, but it'd be neat to have a reference so I could be comfortable when building something new.


There are too many variables to consider to make this sort of a rule of thumb thing that anybody can use for any situation. Take a chair for instance - who is going to use it. Are they wigglers or rockers? And by the way, strength tests are just initial point testing. Has anyone ever seen a life test done? Unlikely. With your joints, if possible, make them as large a glue surface as possible. This is a case of more is better so if in doubt look for ways to increase the area.

[John TenEyck]

I'm pretty sure Finite Element Analysis (FEA) has been done by several folks for a chair, table, etc. of interest. Those studies would show what the applied forces are under simulated load conditions. Short of that, the comments above about simple bending and torque calculations would get you in the ballpark of the forces. Once you have the force you can design the size of the joint required to withstand that force based on the shear strength of the glue and surface area of the joint. The required cross sectional area of the frame components would be similarly calculated based on the properties of the wood to be used. This is what engineers do all day long, and I was proud to do my part for several decades - in another discipline. There's a science and methodology to it. That no one (of whom I'm aware) has published an article on the whole process is rather astounding, but I bet it's been done somewhere, perhaps a furniture manufacturer who, rightly so, has kept the info. proprietary.

Without such information people follow what's been shown to work well over time. Some push the envelop to make very light and graceful designs - and have failures. Others are super conservative and over build - with ugly proportions the result. The only way you really make progress is to do the engineering analysis and testing required to understand what's going on. We didn't get to the moon by intuition.

John

[ian maybury]

Hi Art. Guess i'm not saying that a simple test can't be broadly indicative of something - the resulting information if correctly interpreted and applied with caution may well prove very useful. It likely beats the hell out of having none at all. One issue though is that the choice of the best jointing method is generally a multi-dimensional and quite highly nuanced call, and simple tests seem unlikely to fill in all the blanks.

That said my caution relates primarily to the human factors.

As implied in the original i've seen mag tests (none mentioned here) the configuration of which somehow ended up favouring one jointing system over some of the others.

Then there's the more than reasonable likelihood that the maker of a given system will if handed a positive result trumpet a one dimensional version of it from the rooftops - no matter how qualified it was in reality.

Finally there's the likelihood that many punters will resist looking past this, and as a result walk themselves into difficulty...

[Jim Dwight]

I'm a mechanical engineer too. So when I look at tests like this I consider them data but I also realize some of the limitations. So when I see ~50 lbs difference in a floating tenon versus a traditional tenon when both seem to be the same size, I don't get excited. I don't think the tests are good enough to determine a 50 lb difference. If the joints are made properly and there is no difference in the material (which will never be the case with wood) there should be no difference. But there was a little suggesting either some variation in the wood or possibly incomplete glue application.

The butt joint being as "strong" as it is also seems like a test anomaly. We all know they aren't real strong joints. The cope and stick has more area so it should have been stronger. And a stub tenon is even worse? Maybe these joints were not made very well. Or they didn't get enough glue.

I don't think dowel joints are poor but I've seen them fail in older furniture (chairs) and I think it may be the fact that you don't get a lot of long grain to long grain gluing area. That may be a bit different with beadlock, I'm not sure. It looks a bit like a wavy loose tenon which should work better. The other issue dowel joints could have is with the glue failing (perhaps because of seasonal movement) and then loading being applied to individual dowels. But still, this type of joint also works well in a lot of applications. Tage Freid (sp?) used it a lot on his fancy stuff. It is not a bad joint.

I mentioned the size of the parts partially because I am sure it affected the domino result. 2.5 inches width in an inconvenient in-between size if you use loose tenons 1 inch wide. 2 leaves little room outside of them and 1 isn't really enough. Do a test with 1.5 inch wide parts and with conventional M&T joints that also have 1 inch wide tenons and I am confident that the domino would be right there with a conventional M&T joint. If you were making something with 2.5 inch width parts and needed a strong joint, however, the results suggest you should cut down some dominos or make a custom loose tenon rather than relying on a single domino. An experienced woodworker would come to the same conclusion looking at the joint, however.

I think this is useful data but it can't be extended to cover every situation and needs to be reviewed in light of other data and experience. Where other experience and logic say the data is suspect, additional tests are indicated (like the butt joint versus stub tenon results). But generally, I think the data is interesting and somewhat useful.

[John TenEyck]

I would get a little more excited if the tests had used 25 or 30 specimens per sample set. With 5 specimens the variance would have to be almost zero for the there to any real difference between many of those results. I once set out to do a bunch of joint testing because I still had access to our universal test machines at my former employer, but I couldn't get enough people interested in making the specimens to actually move forward. Now I no longer have that access, and frankly, I've lost interest in doing the work. I've settled on my own joinery approaches, usually loose tenons custom sized to the application. After 30 years with no failures I'm fairly confident they work well. I've pushed the envelop on several occasions, too, fully expecting some of them to fail. To my surprise, none yet have. As an example, I built some chairs with loose tenons where the front leg to rail joints are open at the top. If the glue were to let loose there is nothing to prevent the leg from falling off. They are now 20 years old and a couple of them get daily abuse. Glue is amazing stuff.

John

[Phil Thien]

I wish someone would do a test of 'how much pounds-of-force are required for an application'.

That is, for the average bench, or chair or table, how strong should the joint be? Then I could calculate how many dominos and dowels I need or whether a stronger joint is required.

These discussions tend to get into the apples-to-oranges flawed design of the test. But I'm much more interested in knowing what's good enough for what application.

I'm learning over a time, but it'd be neat to have a reference so I could be comfortable when building something new.

I think Prashun is getting to the crux of it.

There are few applications for simple right-angle joints that will be subjected to tortuous loads. And there are lots of acceptable solutions (a simple diagonal brace, for instance).

And I don't deny that a conventionally-made M&T or half-lap is going to be about the strongest right-angle joint.

But if you're making (for instance) a table where each side has two legs connected at the top and bottom by an apron and a stretcher (respectively), now the very design of the structure is going to spread forces across multiple joints. An assembly made of multiple comparatively weaker joints becomes quite strong in this way. It becomes difficult to apply racking forces in such a way that they a single joint is isolated and compromised.

Two 2x4's nailed together at the ends (in a right-angle) makes for a pretty weak joint. And yet many of them together form the houses in which we live (and they're quite strong). And that again is because loads are spread across many joints.

I have tool stands made of two-by material held together with biscuits and glue. I can't apply enough racking force in order to break one, not in the way in which the things were intended to be used.

[Mike Henderson]

I'll just comment that you can make Dominoes wider. Just cut one side of the dominoes and glue two together to make whatever width you want. It's easy enough to make the mortise any size you want by doing multiple plunges. I've done that for loose tenons on chair repairs.

Mike

[Chris Friesen]

The other thing that nobody has mentioned is that these were brand-new well-made joints. Things are going to behave differently a few years down the road after humidity fluctuations and such. A drawbored M&T starts out weaker than a non-pinned one, but will continue to hold tight even if the glueline cracks.

[Brian Holcombe]

More than one person has stated or implied that test-to-failure joint testing is not valuable because it doesn't allow for every possible joint loading geometry. I say that is ridiculous. Simple racking tests will be a very good predictor of joint reliability in most furniture building and other applications. If an individual has some other primary type of load - like pure shear or perhaps pure tension - then the racking test is only moderately useful for comparison purposes - not definitive but possibly better than nothing. I think it is also pretty silly for people to imply that that testing is typically biased by the investigator toward one method or another. If you compare several of these studies, they all come to pretty much the same basic conclusion. This includes the ones done by the tool manufacturers.

Art, I think they should be tested in practical application. It's my understanding that Danish chair manufacturers will test their chairs by rocking them until failure....the tilting backward in your chair type thing that your mom would yell at you about....they do that until the chair fails.

the takeaway from that test is not how long those joints will be in service, but instead that they can hold up a lot of weight. I don't intend on building a table that will hold 2000lbs and the practical side of me knows that if a table were put under such load that it would break the moment it rocked.

so being that I have reasonable stress requirements I'm back looking at joinery that will stay together for many years and frankly I'm not convinced that it is a domino or a dowel or whatever else floating tenon.

If you look at chairs from fifty years ago, where did they fail? The glue died and then the chair came apart.... I expect Most often if they broke it was actually because they became loose and weren't taken out of service. What stays together is often the joinery that is allowed to move in unison with all of it's parts.

[Michelle Rich]

Good
golly!
The reason I posted this info on the 3 methods was because when I posted my experiences with the
Beadlock so many asked how it compared to the other 2. I posted this one to answer their questions and not to decide which method , whether M&T or pocket holes were superior..just these 3 jigs/methods.

[Mike Henderson]

If you look at chairs from fifty years ago, where did they fail? The glue died and then the chair came apart.... I expect Most often if they broke it was actually because they became loose and weren't taken out of service. What stays together is often the joinery that is allowed to move in unison with all of it's parts.
My experience with repairing chairs is not that the glue fails - especially with chairs that used dowels. When you take the joint apart, what you see is wood attached to the dowel (with the glue). The failure is that the wood fails - it comes apart from the main part of the wood. There's just too much stress for the amount of glued surface area.

Especially with dowels, another tip off is that the hole is now too big for a replacement dowel. If you were to use dowels to repair the chair, you'd have to drill out the dowel holes and put in the next larger size dowel. A better repair is to use a loose tenon.

Many chairs continue in use after the joint fails, held together by the corner blocks. The chair "rocks" but people keep using them for quite a while before they finally decide to get the chairs fixed.

Mike

[Roger Pozzi]

Good
golly!
The reason I posted this info on the 3 methods was because when I posted my experiences with the
Beadlock so many asked how it compared to the other 2. I posted this one to answer their questions and not to decide which method , whether M&T or pocket holes were superior..just these 3 jigs/methods.

I find the info very interesting. And, I have been a Beadlock fan, and user since I first saw it demonstrated on the TV show "Cool Tools" when it was hosted by David Theil.
And, I must add, as I've noticed on most of the posts about tools or techniques, everyone will jump on board to justify their beliefs and try their best to dismiss the OP's original posting.
So, Michelle, I will say that I understood what you posted, and was glad to see the refernce to FWW magazine, and your thoughtfulness to share that information. Thank you!!!

[Pat Barry]

Keep in mind that as you build your chairs you just don't copy the joinery designs of the past because they won't hold up for long under today's load conditions.

[Michelle Rich]

Keep in mind that as you build your chairs you just don't copy the joinery designs of the past because they won't hold up for long under today's load conditions. I TOTALLY do not get this ...I have been making shaker & windsor repros for 30 yrs and use those "out of date " methods..what are you referring to?

[Michelle Rich]

I find the info very interesting. And, I have been a Beadlock fan, and user since I first saw it demonstrated on the TV show "Cool Tools" when it was hosted by David Theil.
And, I must add, as I've noticed on most of the posts about tools or techniques, everyone will jump on board to justify their beliefs and try their best to dismiss the OP's original posting.
So, Michelle, I will say that I understood what you posted, and was glad to see the refernce to FWW magazine, and your thoughtfulness to share that information. Thank you!!!

You are welcome Roger..thanks for your comments, much appreciated

[Mike Henderson]

Keep in mind that as you build your chairs you just don't copy the joinery designs of the past because they won't hold up for long under today's load conditions.
Our ancestors did not have the glues that we have today so they learned, through trial and error, which joinery would work (hold up) in the various chair joints. (Joke: it's said that surgeons bury their mistakes. Woodworkers use their mistakes as firewood.) Although people have gotten heavier, those joints are still good designs. If you're worried about the additional weight they might have to carry, I'd just size them up a bit but I'd still use the same joinery.

Beyond those reasons, I'm not sure what alternate joinery would be appropriate. While glues have changed, the various joints we make have not changed that much. We have lock miter joints for drawer corners but I can't think of any alternate chair joinery off the top of my head.

Mike

[Brian Holcombe]

I'm sure pat is making a fat joke.

[Bill McNiel]

Mahalo to all for clarifying the various joint strength capacities and issues. I am no longer confused about the validity of the test results and now know exactly which technique works best.

Michelle -special mahalo for the attempt.

[Pat Barry]

I'm sure pat is making a fat joke.

Not exactly a joke, but we need to recognize that folks today, on average of course, are bigger and heavier than ever before.

[Pat Barry]

I TOTALLY do not get this ...I have been making shaker & windsor repros for 30 yrs and use those "out of date " methods..what are you referring to?

I don't recall saying those are "out of date" methods, just that you should consider rescaling them based on today's consumers.

[Pat Barry]

Our ancestors did not have the glues that we have today so they learned, through trial and error, which joinery would work (hold up) in the various chair joints. (Joke: it's said that surgeons bury their mistakes. Woodworkers use their mistakes as firewood.) Although people have gotten heavier, those joints are still good designs. If you're worried about the additional weight they might have to carry, I'd just size them up a bit but I'd still use the same joinery.

Beyond those reasons, I'm not sure what alternate joinery would be appropriate. While glues have changed, the various joints we make have not changed that much. We have lock miter joints for drawer corners but I can't think of any alternate chair joinery off the top of my head.

Mike
What I'm suggesting is that those joints of the past, for chairs in particular, maybe a bit undersized for people nowadays, Just because they worked 200 years ago doesn't mean they will still work today. Unfortunately, unless we consider this our furniture of today will not last 100 years

[Mel Fulks]

I think it is more how things are used. I don't think guests of old leaned back and put two chair legs in the air. Some very fine period side chairs would be reduced to kindling by a super bowl gathering. Some of the finest federal chairs had no strechers but have survived because they were used for formal occasions, not by people "hanging out".

[Larry Edgerton]

I kind of chuckle at the "This information is useless" crowd on these kind of tests.

[Roger Pozzi]

I kind of chuckle at the "This information is useless" crowd on these kind of tests.

http://www.stationwagonforums.com/forums/images/smilies/2_thumbs_up_-_animated.gif

[Brian Holcombe]

The information is useless. :D

[ian maybury]

Think information of that sort is at best highly specific, and that using it requires extrapolating what it may actually point to (as opposed to what various vested interests would say it does) in real world situations. If applicable.

While there may be stuff that maintains it's altitude assisted entirely by lots of hot air, it's not by accident that most of the various joint designs and jointing systems we are familiar with (new and old) have survived. As in they tend to work if properly used, and have a fair degree of tolerance built in. e.g. while some may be a bit weaker in a given situation than others, they have compensating advantages. The wrong joint may equally be used, or the right one misused.

It'd be a pity to get sucked into arguments about this and that being the best, because it's for these sorts of reasons futile to talk of joint choice in isolation.

Horses for courses. For sure there's something (some combination of joint type and implementation of it) that's probably best in a specific situation (as in representing the best possible balancing of all of the myriad considerations that match a joint to the situation - but only if a long list of parameters are precisely controlled/it's done right), but even that 'best' may not survive contact with a particular unskilled individual's take on reality. As in it can be thoroughly screwed up if used/done in a manner which fails to maximise its strengths and minimise its weaknesses.

:) Or vice versa...

[Larry Edgerton]

Keep in mind that as you build your chairs you just don't copy the joinery designs of the past because they won't hold up for long under today's load conditions.


Pat, you are one to something here.....

For years I fixed the chairs at a friends 500 seat restaurant. Nice, strong, well designed chairs with one exception. There was not enough room between the arms for the size butts that were going to flop down in them. The wide bodies were pushing the chairs apart.

They were doweled. I fixed a few with oversize dowels. They came back, more big butts. Then I switched to West System, hogged out the holes with a dremel, used threaded rod for dowels, one wet coat and a filler coat with microballoons. Never saw one of those back, and I fixed that set of chairs for a few more years. Traditional, not even close. But it worked....

Hang out by a Micky Dee's and you will see why steel furniture has its advantages.

[Andrew Joiner]

http://assets.rockler.com/media/catalog/product/cache/1/image/720x720/9df78eab33525d08d6e5fb8d27136e95/3/7/37801-03-1000.jpg

Of course the the beadlock is superior. Look at how perfectly it fits together :) The photo is from Rocklers site. One key to joint strength is intimate contact of the mating pieces. A poorly cut beadlock will not be as strong as a precisely cut beadlock. Given the beadlocks complexity it may be time consuming to get a good fit.

I like simple strong joints. I test them like Sam Maloof did. It's a complicated scientific test, but fun to do.