long term integrity of different glue joints

[Will Blick]

After reading the FWW article on joint strength, and the thread that followed on this forum, I am curious....

It makes sense the lap joints, M&T would win the test, they have the most surface to surface area, as well as the the obvious structural joining at critical stress areas.

However, since wood expands on its width, and glue is unforgiving... over time, why would the glue on these joints, specially the half lap not give to expansion / contraction. You have two woods joined at 90 degrees, one will expand on the X axis, the other on the Y axis. What prevents the glue line from weakening by natural expansion / contraction?

Also, this makes a good argument for loose tenon joinery with compressed woods, or woods that don't move in a direction opposed to the wood it is glued to. Dowels and Dominoes meet much of these objectives. Make sense?

On a side note.....I agree with Jim from Dowelmax...had the test used more dowels, I think the results would be similar to M&T...and have the benefit of not having the glue bond become vulnerable to natural wood movements. Of course, the same applies to the Dominoes.

[Peter Quadarella]

The problem in general with these tests, is that they don't tell us what really matters. Furniture built with any of these methods will be plenty strong immediately after completion.

What most people who use mortise and tenon, or any people who care about joint strength really want to know, is which joint will stay strong over time. Will my chair still be strong in 10 years, 20 years, 100 years?

[Eric Brown]

I've seen pictures of through tendons that had wedges.
I've seen pictures of tendons with draw pins.

Would using both make for the strongest joint of all?

Thanks - Eric

[Dan Barr]

The longevity of the joint is determined by many more factors as well.

In my short-lived experience:

Dowel joints subjected to racking give way the soonest.

draw-bored and through-keyed mortise and tenon joints with no glue have been know to last hundreds of years even in bad weather and total neglect.

I've attached a picture of a through-keyed mortice and tenon. The timber is about 10" x 30", the tenon is about 2 1/2" x 7" and the wedge is about 2 1/2" x 4" x 16". I'm taking my best guess at these measurements, don't go getting all nit-picky. :D

I took this picture while I was on a tour of the west tower of Ely Cathedral in Ely, Cambridgeshire, England about two or three months ago.

"The West Tower was built between 1174 and 1197" (Wikipedia)

1174!!!

how is that for longevity? :D

Overall I believe that mortice and tenons will be the best performers while avoiding using the potentially ugly pin of a draw bored M & T. If you absolutely have to have the strongest joint possible, I guess go with the draw-bored M & T.


ciao,

dan

[Mike Cutler]

However, since wood expands on its width, and glue is unforgiving... over time, why would the glue on these joints, specially the half lap not give to expansion / contraction. You have two woods joined at 90 degrees, one will expand on the X axis, the other on the Y axis. What prevents the glue line from weakening by natural expansion / contraction?

Also, this makes a good argument for loose tenon joinery with compressed woods, or woods that don't move in a direction opposed to the wood it is glued to. Dowels and Dominoes meet much of these objectives. Make sense?



Will

Nothing will stop the glue from weakening. All glued joints will eventually fail. It may take years, decades, or centuries to happen, but eventually the glue will break down. Furniture makers centuries ago learned this, and designed their joints to maintain mechanical integrity between the two pieces of wood.

A glued M&T joint is not as exposed to ambient environmental conditions as the rest of the piece that is exposed. Wood absorbs and releases moisture through the end grain, for the most part.. Having the end grain protected in an enclosed space, the mortise, and filled with glue, the rate of expansion/contraction will be diminished at the joint.

M&T joints designed for longevity will always utilize some form of a mechanical device in the joint. A tusk tenon, a wedge, a haunch, or locked shoulder, or pin to drawbore, etc. I've seen barns in New England that are 200+ years old and all of the framing is drawn M&T joints, Those barns are still as straight and tight as they day they were finished. No glue either and totally exposed to the elements.

Dowels, and floating tenons have been used in furniture making and construction in some form or fashion for centuries, as has the M&T joint. Nothing new there. But, each type of joint has it's place and usage. Used in the wrong application, the joint will fail early.

The "traditional" M&T joint is a labor intensive joint, even with a machine it still takes time. Individuals and furniture makers have tried for decades to find a joint that speeds up production and that is more forgiving than cutting M&T's. The problem comes from that they wanted to use one exclusive joinery technique for all joints, and that just can't happen. The joint needs to be designed for the application. They all have their respective place.

[Mike Cutler]

I've seen pictures of through tendons that had wedges.
I've seen pictures of tendons with draw pins.

Would using both make for the strongest joint of all?

Thanks - Eric

Eric

Strongest can have multiple meanings. It can be related to compressive and tensile strength. It can be applied to quantify shear load, or related to an individual materials overall modulus. It really kind of depends. Strength is a relative term.

The joint that you referred to above is used in a lot of the Mission Style, Arts and Crafts Style, and Art Nouveau Style furniture from the late 1800's to the early 1900's

In the Stickley and Morris furniture the through tenon was a massive joint that got it's mechanical from the shear size of the material involved. The through tenon was used to draw all of the members into place. The limitation on this joint is that the wedge acts against the long grain of the through tenon and can blow out the material behind it if hammered home too hard.

Greene and Greene,and Limbert also used the joint, but sometimes more so as a design/decorative element, and you can find examples of where the actual mechanical component was a draw pin or some other device.

[Jim Becker]

The thing that was most striking about the failures illustrated in the test article was how many of the "failed" right where the tenon/dowel, etc. ended in the opposing piece--notice how the failure was a splitting of the wood with the grain. What I learn from that is "length matters" ;) when it comes to how much load a joint can handle. Longer tenons/dowels/cheeks (the latter in the case of the bridle joint, etc) provide both more physical support and more glue area. Both are essential to handle higher stressing.

Since we always have a choice on the type of joinery we employ, we can use this knowledge to better plan our projects.

[Will Blick]

> Nothing will stop the glue from weakening.


Yes, but my point was, if you subject a glue joint to - one board expanding in the X axis, and the other board in the Y axis, you have created the perfect "natural" scenario for a glue line to break. The most obvious of the joints tested that met this criteria was the Half Lap.

So I would think, that although the half lap performed the best in the testing, over time, it could become the worst performer, as the integrity of the glue line is highly compromised? I am putting aside all the other variables of loads, end use, etc.

Whether all glue will fail in time, is not too relevant here....as I am referring to reasonable time periods, not hundreds of years, where we have no test data of modern glues.

[Mike Henderson]

The expansion and contraction of wood is why glues like Titebond remain somewhat flexible even when cured - to allow for wood movement without glue failure.

My belief is that the joint with the greatest face-grain-to-face grain contact will last the longest.

I reason as follows: The glue does not fail all at once but probably as a percentage. That is, given a certain age, maybe 50% of the bond will fail. If you have a lot of surface area, the remaining 50% is still strong enough to hold the joint together if you start with a lot of surface to begin with.

But if you have a small surface area, such as two fairly short dowels, the 50% glue surface remaining is not enough to hold the joint and it fails.

Mike C. also made an excellent point that joints that bury the end grain will last longer because exposed end grain sucks up moisture, causing more wood movement.

Mike

[Will Blick]

I think its safe to assume, that end grain is not the ONLY place wood absorbs moisture. Although it would be interesting to learn more about how significant of a role the end grain plays... my guess is, end grain absorbs "more" moisture then the rest of the board, but covering the end grain with another board as part of the design, by itself, is not a solution to expansion / contraction. (but definitely a smart design consideration)


I understand that TiteBond type glues do have a bit of flexibility when dry.....again, it would be interesting to understand "how much". It must be a very small amount, otherwise glue would not do its job.


As Mike suggests, gluing the face grain together, such as a butcher block glue up, will provide max. glue life IMO, as the boards expand together in the same direction. Conversely, a lap joint has each board moving in directions opposed to each other.... i.e. when each board expands on its width, the X movement on one board, and the Y axis move on the other board, create the most possible stress on the glue line.

Also, if you use a lot of dowels, such as a dowelmax application, I think the total glue surface area can match the glue area of a M&T.

[Mike Cutler]

Will

I see your point a little better now. Not having seen, or read the article, limited my understanding.

I "personally" would not rely on glue alone in any joint, but that's me, and the way I was taught. Not that I don't ever do it, but I really try to limit it.
I also agree with your assessment of the half lap, using glue alone as a joint. I wouldn't do it. All the half laps I've done have used some form of a pin, or part of the design element as the mechanical component.
The frame that I made in the following thread is half laps at each corner. The tops are pinned, and the bottom uses the relief of the stretcher, and the shape of the end of the stile to keep it together. All joints are epoxied.

http://www.sawmillcreek.org/showthread.php?t=60965

The beadboard in the following thread uses a classic breadboard end, with a through M&T joint in the middle. The glue area of the joint is the M&T section, the ends of breadboard end allow the center section to expand and contract.

http://www.sawmillcreek.org/showthread.php?t=61157

This particular breadboard expands and contract about an 1/8"-1/4" across the center section seasonally. As you rightfully pointed out, glue alone would not have held up in that joint, (your x and Y axis illustration). I doubt it would have lasted a single season.

Good discussion you generated.

[Mike Henderson]

Also, if you use a lot of dowels, such as a dowelmax application, I think the total glue surface area can match the glue area of a M&T.
I agree with you but just want to point out that when you're doing the area calculation, you only get half the surface area of the dowel as face grain - the other half is facing end grain in the hole. So, a 3/8" dowel 4" long (2" in each piece of wood), only has about 0.59 sq inches of face grain surface area in each piece of wood. (the formula for the surface area of the sides of a cylinder is 2*PI*r*h - remember the radius is 3/16")

Compare to a tenon about 2" by 2", which has 8 sq inches of face grain to face grain contact. You'd have to put in 14 dowels 2" deep (on each piece of wood) to equal the same square inches of contact as the M&T.

It might be tough to put that many in the space of a 2" high tenon.

Mike

[Neal Clayton]

The thing that was most striking about the failures illustrated in the test article was how many of the "failed" right where the tenon/dowel, etc. ended in the opposing piece--notice how the failure was a splitting of the wood with the grain. What I learn from that is "length matters" ;) when it comes to how much load a joint can handle. Longer tenons/dowels/cheeks (the latter in the case of the bridle joint, etc) provide both more physical support and more glue area. Both are essential to handle higher stressing.

Since we always have a choice on the type of joinery we employ, we can use this knowledge to better plan our projects.

there's also a consideration of repair. which is why you see those lap joints on older windows and short mortises on muntins in divided light doors and such. should a piece be broken it would be preferable to be able to reproduce and replace the broken piece without trashing the rest. so while those lap joints and 1/8 or 1/4 deep muntin mortises might not be initially stronger than a hidden mortise or deeper mortise in the case of the muntins, they are serviceable.

[Will Blick]

Mike, fully agreed on the dowel contacting "some" end grain.... but the dowel is compressed so tight, the overall strength at the dowel to end grain will probably be at least 50% as strong. Remember, the dowel itself is not end grain, so its not end grain to end grain. But good point....


> You'd have to put in 14 dowels 2" deep (on each piece of wood) to equal the same square inches of contact as the M&T.


Agreed, there will be certain sizes where M&T will win on surface area for sure. But remember, I was referring to "loose tenons" in general, not just dowels.... (although I did lean on the dowels in my explanation) in some cases Dominoes or a loose M&T made from compressed wood could be used to avoid the X Y expansion issue. Anyway, not too loose focus of the original "issue"...which is, how to best manage X Y forces on the same glue line.


Mike C, I think your approach is excellent.... the added mechanical fasteners have to help the XY stresses. In addition, it adds a nice decorative touch to your work...


I think there is some great metal fastening strategies that don't get much mention, of course mainly when the joint is not visible. The popular one is "figure 8's" for table tops, a brilliant fastening technique to account for movement. When I have a leg joining an apron, I often use a lag bolts, with the hole in the leg drilled a ~ 1/8" wider than the lag diam., then use large washers.... rock solid, but will allow for wood movement in both the X & Y direction. This is not applicable to all pieces, but as mentioned above, you pick n chooze wisely. The more options you are aware of, the better choices.

[Andrew Joiner]

What I love about woodworking is improving what has been considered best in the past.
The very fact we all care about a joint that will make our wood product last for centuries is incredible.
There's a good chance that the things we build will be around for many lifetimes and even go up in value. Not many people get to say that about the work they do.

[Mike Henderson]

Mike, fully agreed on the dowel contacting "some" end grain.... but the dowel is compressed so tight, the overall strength at the dowel to end grain will probably be at least 50% as strong. Remember, the dowel itself is not end grain, so its not end grain to end grain. But good point....
It's a bit more difficult to visualize the wood-to-wood contact in a round hole than in a square hole, but I believe I'm correct in saying that 50% of the surface area is end grain and 50% is face grain. Let me explain:

Let's visualize a hole in a piece of wood that has the grain running vertically. We can easily agree that the top and bottom of the hole is fully end grain and the sides are fully face grain. Think of the hole as the face of a clock. At 12 and 6 we have end grain and at 3 and 9 we have face grain. If we start at the top and go clockwise, we find that at 1 we have the grain at an angle to the the side of the dowel - we can say that the joint at that point is part end grain and part face grain.

If we look at the side, and move from 3 to 4, we find that the grain is at an angle to the dowel - we can say that the joint at that point it part end grain and part face grain.

Summing over the full circle, we find that 50% of the joint surface is end grain and 50% is face grain. Of course, most of the joint area is a combination of end grain and face grain, and the more end grain, the weaker the joint contact.

This might be easier to visualize if we consider the "dowel" to be square but of equal surface area. In this case, it's easy to see that 50% of the surface area is face grain to face grain, just like the sides of a M&T, and 50% is face grain to end grain, just like the top and bottom of a M&T. Making the dowel round does not change that - it just makes it a bit harder to visualize. If you could get more face grain to face grain contact (and a stronger joint) by making the joint round, our ancestors would have discovered that long ago and all of our joinery would be round instead of square.

The dowel is not end grain, except on the ends, just like the tenon is end grain only on the end. The top and the bottom of the tenon are face grain to end grain (12 and 6 on our clock) and we don't assign much strength to that portion of the joint. Likewise, we don't assign much strength to the end grain to face grain contact of the dowel - only to the face grain to face grain component.

I do not see much advantage to compression of the dowel. If compression was an advantage, the builder could build the tenon so that it had to be driven into the mortise to gain extra strength. According to an article in one of the early issues of FWW, compression in a joint doesn't provide any advantage because the compressed wood takes a "set" and just remains smaller. The only advantage I can see for compressed joint components, such as a biscuit, is that they will expand to fill a sloppy joint, while a standard tenon requires that it be made to a good fit. The disadvantage of an expanding joint component is that it must be assembled quickly or you won't be able to put it together (once it gets wet from the glue and expands) and you can't adjust it after you get it together and the component swells. You only get one shot.

There really is no free lunch that I can see in the use of dowels compared to M&T. I do believe that given equal face grain-to face grain contact they will be of approximately equal strength.

Mike

[Will Blick]

Mike, in general, I agree with your position... but its hard to fully assess the angled end grain touching the dowel.... at some point, highly angled end grain will match the hold of side grain, but your point is well taken, and actually demonstrates why non round loose tenons might have an edge over round tenons....... kudos to Dominos and Beadlock system in this regard. :-)

More importantly, in most cases, the glue to glue area in the dowel joints we are discussing, will not be the weakness of the joint. The reason is, the force applied to a dowel joint is NOT trying to pull the dowel out from its hole, which would be a test of the glues holding power. Instead, it's the shear strength of the dowel and its surrounding wood that will be tested, not the glue line.

I appreciate your thoroughness here..... however, the original issue I had was with joints which fail the X Y expansion / contraction scenario... and while non round loose tenons may have greater glue surface area (in the side grain component), it doesn't address the X Y issue.

An example.... a half lap joint may fail due to excessive X Y movement over time. It has no end grain, and maybe 4x the glue line area vs. dowels. The dowels / Dominos experience only X movement, their glue lines continue to hold, even under stress, as the stress on the joint is only shear. That is the basis of my original post. Its OK we drifted into this interesting tangent, but I was hoping to keep the train on track, as I am interested in how others feel about this...


> There really is no free lunch that I can see in the use of dowels compared to M&T. I do believe that given equal face grain-to face grain contact they will be of approximately equal strength.


After this thread has now matured....do you feel there is a potential benefit for dowels / Dominos, since they both face only X pressure, vs. the M&T and Half Lap that both are confronted with constant X Y pressure? If the joint is well designed, lets assume all types of joints will pass the sheer test over time. What I am curious about is, will the glue life be longer in the joinery that only contended with X movement? (vs. X Y movement) My guess is, yes, it will be... just my opinion...


A good indication of this would be those who restore antique furniture that may encounter these situations. Of course, a glue line failing in a M&T will not make the piece fall apart in most cases, but it would help answer our questions. I have seen dowel joints on chairs 100 years old still strong, which is remarkable. I never come across many old M&T antique pieces though, so I am can't compare...

[Ron Dunn]

Will, why are you shouting at us? Please write at normal size.

If you're having trouble reading normal size, make your browser text bigger.

[Will Blick]

Is that better Ron?

[Ron Dunn]

You're a kind-hearted gentleman, Will ;)

[John Keeton]

On the dowel issue (percentage of face grain contact), it would seem that a dowel only has face to face contact at the extreme tangents on two sides - a very minimal area. The entire remaining area is compromised by end grain to a varying extent. But, again, looking at the test results, nearly all of the joints held to the extent of glue contact - the wood failed. Which would lead me to conclude, at least for cherry, that it doesn't take a lot of face grain glue contact to exceed the strength of the cherry. The point is to create a joint that will provide the strength needed in the particular application.

As to long term integrity, probably 100% of joints in furniture that has survived the ages are hide glue, which is subject to moisture and heat. Perhaps it is that quality that provides movement and flexibility - and thus, long term integrity of the joint. Similar to building into a skyscraper the ability to bend so as to not break. In humid, hot times, the glue "releases" its adhesive quality slightly, allowing for movement, etc.

It may be that our modern adhesives have all been formulated to resist those forces (moisture/heat) in order to retain joint integrity, and this becomes an actual detriment of the long term stability of the joint.

[Larry Edgerton]

I favor mortise and tenon in any work that matters.

One point I have not seen discussed is that no matter what kind of loose joinery you are using you have weakened one half of the connection for no gain other than expediancy. When you drill/rout holes on the end of a rail you have weakened that piece expotentially. No matter how deep you run a dowel/domino in the hole you have drilled, it will not be as strong as it would if you had cut a tenon on the end of the rail. You can not get a better glue bond than no joint at all. When you dowel/domino you have introduced one more joint to the work that needs not be there. Further you have introduced a piece that has a different expansion/contraction rate than the original wood, again promoting eventual failure.


I use mortise and tenons with West System on anything that will see a lot of stress, the epoxy strengthens the wood surrounding the joint, something other adhesives do not do. For all others I use regular wood glue. I do not like Titebond II.

For large work such as this post and beam house I use blind wedged tenons so they become tighter as they settle in. Code departments hate no fasteners so I glue cut off heads on with hot glue until they are gone, easier than arguing. :)

[Dennis Peacock]

Is that better Ron?

Oh yea....much better Will. Thank you kindly. :)

[Will Blick]

> You can not get a better glue bond than no joint at all.


Larry, certainly no one can argue this point :-).

But when viewing the test results, the glue line is rarely the breaking point. This suggests the glued loose tenon is not the weak link, on either end, even if its weaker than the true tenon. Of course, this mainly refers to small work. With the architectural work you do (very nice), I too would NEVER gamble on loose tenons.


Anyway, the test only represent a fresh glued joint.... my guess is, with the joints that experience the most X Y stress over the years, we would see different failures than what the tests demonstrated.

[Larry Edgerton]

It's a bit more difficult to visualize the wood-to-wood contact in a round hole than in a square hole, but I believe I'm correct in saying that 50% of the surface area is end grain and 50% is face grain. Let me explain:

Let's visualize a hole in a piece of wood that has the grain running vertically. We can easily agree that the top and bottom of the hole is fully end grain and the sides are fully face grain. Think of the hole as the face of a clock. At 12 and 6 we have end grain and at 3 and 9 we have face grain. If we start at the top and go clockwise, we find that at 1 we have the grain at an angle to the the side of the dowel - we can say that the joint at that point is part end grain and part face grain.

If we look at the side, and move from 3 to 4, we find that the grain is at an angle to the dowel - we can say that the joint at that point it part end grain and part face grain.

Summing over the full circle, we find that 50% of the joint surface is end grain and 50% is face grain. Of course, most of the joint area is a combination of end grain and face grain, and the more end grain, the weaker the joint contact.

This might be easier to visualize if we consider the "dowel" to be square but of equal surface area. In this case, it's easy to see that 50% of the surface area is face grain to face grain, just like the sides of a M&T, and 50% is face grain to end grain, just like the top and bottom of a M&T. Making the dowel round does not change that - it just makes it a bit harder to visualize. If you could get more face grain to face grain contact (and a stronger joint) by making the joint round, our ancestors would have discovered that long ago and all of our joinery would be round instead of square.

The dowel is not end grain, except on the ends, just like the tenon is end grain only on the end. The top and the bottom of the tenon are face grain to end grain (12 and 6 on our clock) and we don't assign much strength to that portion of the joint. Likewise, we don't assign much strength to the end grain to face grain contact of the dowel - only to the face grain to face grain component.

I do not see much advantage to compression of the dowel. If compression was an advantage, the builder could build the tenon so that it had to be driven into the mortise to gain extra strength. According to an article in one of the early issues of FWW, compression in a joint doesn't provide any advantage because the compressed wood takes a "set" and just remains smaller. The only advantage I can see for compressed joint components, such as a biscuit, is that they will expand to fill a sloppy joint, while a standard tenon requires that it be made to a good fit. The disadvantage of an expanding joint component is that it must be assembled quickly or you won't be able to put it together (once it gets wet from the glue and expands) and you can't adjust it after you get it together and the component swells. You only get one shot.

There really is no free lunch that I can see in the use of dowels compared to M&T. I do believe that given equal face grain-to face grain contact they will be of approximately equal strength.

Mike

Mike I think you are being too generous to the dowel. I would sum it up as 5% good gluing surface, and 95% that is of compromised quality because of the reasons that you have sited.

I have fixed a lot of old furniture over the years, and far and away the biggest failure cause according to what I see is the use of dowel. The rail breaking because the dowel holes have stolen its integrity is just as common as dowels failing.

Dowels are great for making plate racks.........:)

[Howard Acheson]

>> Yes, but my point was, if you subject a glue joint to - one board expanding in the X axis, and the other board in the Y axis, you have created the perfect "natural" scenario for a glue line to break. The most obvious of the joints tested that met this criteria was the Half Lap.

Let me make this point. Some adhesives, even when fully cured, have some "give" to them. One of these adhesives is the very popular PVA or yellow adhesive. These adhesives will allow a small amount of movement without loosing their adhesive integrity. That's one reason to use these adhesives for what are essentially cross grain joints. Most epoxy adhesives are also somewhat flexible so they too, are good for cross grain joints. These two adhesives are also the best choices for joints that are subject to constant changing stress. Chairs are a good example of the type of item where the strength of epoxy and it's flexibility are a real advantage.

On the other hand, adhesives that cure hard are also quite brittle. This is one of the primary disadvantates of hide glue and one of the primary reasons that old furniture joints fail over time. Because they have no flexibility, the frequent "working" of the joint cracks the adhesive and the joint fails.

[Dewey Torres]

Wow... what a topic:

Unless any of us are trying to top Dan Barr's post (creator also dead), lets just use any of the time tested joints with the modern glues and find that they will outlive every Creeker alive today if properly fit and assembled.

[Dan Barr]

i forgot to include in my post that that big joint is holding the cathedral's west tower together; literally. It holds the tower to the main part of the building.

cheers,

dan

[Tom Adger]

I guess that some years down the road I am destined to have some of my glued mortise/tenon joints fail. I read one poster say in 50-100 years. I am now 66, so at age 116, I will nervously watch my workbench joints. If they are OK, I will relax until the 100 year mark, when I am 166.

[Will Blick]

Please do report back Tom....

[Dan Barr]

Quote: John Keeton

"On the dowel issue (percentage of face grain contact), it would seem that a dowel only has face to face contact at the extreme tangents on two sides - a very minimal area. The entire remaining area is compromised by end grain to a varying extent. But, again, looking at the test results, nearly all of the joints held to the extent of glue contact - the wood failed. Which would lead me to conclude, at least for cherry, that it doesn't take a lot of face grain glue contact to exceed the strength of the cherry. The point is to create a joint that will provide the strength needed in the particular application.

As to long term integrity, probably 100% of joints in furniture that has survived the ages are hide glue, which is subject to moisture and heat. Perhaps it is that quality that provides movement and flexibility - and thus, long term integrity of the joint. Similar to building into a skyscraper the ability to bend so as to not break. In humid, hot times, the glue "releases" its adhesive quality slightly, allowing for movement, etc.

It may be that our modern adhesives have all been formulated to resist those forces (moisture/heat) in order to retain joint integrity, and this becomes an actual detriment of the long term stability of the joint."


BINGO!!!


This may be the best observation of the lot on long-term integrity.

cheers,

dan

[Darius Ferlas]

As to long term integrity, probably 100% of joints in furniture that has survived the ages are hide glue, which is subject to moisture and heat. Perhaps it is that quality that provides movement and flexibility - and thus, long term integrity of the joint. Similar to building into a skyscraper the ability to bend so as to not break. In humid, hot times, the glue "releases" its adhesive quality slightly, allowing for movement, etc.

It may be that our modern adhesives have all been formulated to resist those forces (moisture/heat) in order to retain joint integrity, and this becomes an actual detriment of the long term stability of the joint.

This is my first on this great forum. I have been lurking around for a while and I decided to join.

Hi all!

Anyway, another BINGO from me.

I am amazed by the longevity and quality of some of the furniture I came across while visiting various historic places here and there. No power tools, no laser guide lines and yet the results are simply stunning.

I have seen a table top made of 19 varieties of wood. No inlays, but solid pieces of wood fitting nicely into the surrounding pieces and creating various interesting patterns. The table was about 10'by10'by 4". It ain't new, pushing its 200th birthday, and yet it looks as if it was made yesterday. From what I have been told, other than the types of glues that were used the critical part has been maintenance - wax, wax and then wax some more.

The place where I saw the table has beautiful floors. Immaculate. Most made made into patterns of 5 to 20 wood species. Not a single squeak throughout the castle, and the surface looks as if it has been just polished. As a matter of fact it has, as visitors are required to wear thick felt shoes over whatever footwear they came in. The result is that every step in the castle helps maintain the floors.

The attached picture is that of a table (from the same castle). It's made of 20,000 (that twenty thousand) knots. The same knots we all frown upon and most of the time tend to throw out. The structure and edging is made, again, of a bunch of wood varieties. All hide glue, all waxed, and immaculate. The table is about 200 years old.