joint strength

[daver828]

Hello, I have some questions about joint strength, and specifically options about glued joints, vs. mechanical means, etc.

Just got several videos, one of which is "Forgotten Hand Tools" by Christopher Schwarz. In this DVD he discusses the drawbored mortice and tenon and how modern woodworkers tend to avoid this, and nails, etc. in favor of "piston fitting" joints (dovetails, mortice and tenon glued, etc).

I can readily see the advantage of a drawbored mortice and tenon. I can even go along with using more nails in carcase joinery. What I am not following is if what he says is true, that all glues wear out over time, then how does that affect things like butt edge jointed table tops? I posted another question some weeks ago about biscuit joiners asking if they were still used. I got varied responses. But I understood the majority to say that edge joining butt to butt was plenty strong enough. But again, if "all" glues wear out, what happens to table tops, wide carcase joinery, etc.? Will they all separate over time?

Please someone teach me.

David

 

[DaveO]

I wonder about the statement that "all" glues will wear out over time. I haven't been at this woodworking thing for a long time, but I have never had a glue failure due to time passing. I have pieces that are glued by my wife's grandfather and are easily 40 years old, and the glue hasn't failed. Hide glue is the only glue that I have heard to have a failure rate and that is after many, many decades.
I don't recall which woodworking mag ran this review but they found that pinning/pegging and drawboring M&T joints actually weakened the joint compared to a well fitting glued M&T joint.
I don't think that the weakening was significant enough to cause furniture with drawbored M&T to just fall apart, but in the "scientific" test they faired less than the glued joints.
There is a role for mechanical fasteners (screws/nails) physical reinforcements (dowels/pegs/wedges) and plain ol' glue joints. I think that they all should be used, the key is knowing when and where to utilize them.
MTCW,
Dave

 

[Bas]

In the old days, all they had was hide glue. Hide glue isn't terribly strong. By comparison, pretty much all the modern glues (PVA, plastic resin, epoxy, polyurethane etc.) are stronger than the wood itself. Mortise-and-tenon and dovetails offer great mechanical strength, something vital with weak glues. Today, with new glues, rabbets, dado's, biscuits joints and even butt joints are almost impossible to break apart. Also, nails used to be expensive. Today, you can blast them in using a brad nailer for pennies.

I think that for almost any project that doesn't receive a lot of stress, a simple glued joint will last for decades. But, box joints, dovetails etc. do look much nicer than simple butt-joints. Cosmetics matter in this game! And traditions are hard to break.

 

[Travis Porter]

I don't really buy glue "wearing out". Glue can fail, but if it wore out, there wouldn't be any antiques around. Hide glue works, and it still has its place, but it isn't something I have used yet.

I have done some of the joints you speak of, and will do them again in the right scenarioes. If I can do what I consider a better or stronger joint I will. For example, a dovetail only needs the glue to keep the joint from coming apart, not for strength.

As for the biscuit jointer, I am like most others. I use it occassionally, now adays. I no longer use it at all for edge gluing boards for a few reasons. One being I have screwed up more than once putting the biscuits too close to the edge and have biscuits showing, and two, I have never really gotten it to work that well to help align the edges.

 

[nelsone]

I think you have to consider the joints he is talking about. MT and pegged joints are usually connecting the legs and aprons of chairs and tables and such. They endure a significant amount of stresses through racking over time, much more so than an edge joined table top or even an apron skirt. I would find the likely hood of glued butt joint failing if done properly. I also think a pegged MT joint with no glue, ore hide glue, would also suffer due to racking and wood compression, but not to the extent of failure. This is just my opinion and I have no scientific study to back up that opinion other than the projects I have done.

The base of my workbench was assembled with draw bored MT and glue and shows no sign of racking after 3 years of use. I have also glued up cabinet panels that show no sign of failure after about 6 years.

 

[daver828]

Thank you all for the input. I will follow up the first question with a "what do you use" question. Say, you are making a dining table, or a nice wide coffee table. How do you join the boards to make the top? Also, if you were making something like a Philadelphia Highboy, how do you join the boards for the sides? Thanks,

David

 

[McRabbet]

I would use the same method for either of your examples -- after selecting the set of boards to glue up by matching the best grain pattern, I'd select a pair of adjacent boards, fold them with the top side in and edge joint them at once against the fence of my jointer (this assures any slight angle off perpendicular is eliminated). Continue joining all internal edges. Apply Titebond II or III to each glue edge and clamp with good parallel faced clamps (I use Bessey K-Body clamps) every 12", alternating clamps above and below boards. Be careful to wipe off any squeeze out with a damp, clean rag to avoid glue closing pores of the wood. I also use waxed paper over my clamp bars under the glue-up to keep the bars cleaner.

 

[Travis Porter]

I would use the same method for either of your examples -- after selecting the set of boards to glue up by matching the best grain pattern, I'd select a pair of adjacent boards, fold them with the top side in and edge joint them at once against the fence of my jointer (this assures any slight angle off perpendicular is eliminated). Continue joining all internal edges. Apply Titebond II or III to each glue edge and clamp with good parallel faced clamps (I use Bessey K-Body clamps) every 12", alternating clamps above and below boards. Be careful to wipe off any squeeze out with a damp, clean rag to avoid glue closing pores of the wood. I also use waxed paper over my clamp bars under the glue-up to keep the bars cleaner.

I use pretty much the same method. No biscuits, dowels, etc. I used to do dowels and biscuits back when I was fearful that the glue wasnt strong enough and I was paranoid and didn't believe glue could hold boards.

 

[NCPete]

I had always figured, the bigger the joint, the stronger the buzz.... I MEAN JOINT! i mean... oh it's way too late!

 

[jeff...]

Depends on what wood I'm working with - most all hard wood I just edge glue with no dowels- but if it's pine I always use dowels - why? because pine contains low amounts of turpentine which is a solvent.

 

[DaveO]

I find it better to wait until the glue squeeze-out has skinned over before trying to remove it. Unless you are extremely diligent in rising your damp wiping rag, you run the risk of smearing diluted glue across the wood.
I wait until it has skinned over, about 30 minutes and remove it with a utility knife blade used as a scraper.
MTCW,
Dave

 

[JimD]

I also just edge joint without dowels or biscuits most of the time. I too have placed biscuits where they would show. I have also read about biscuits telegraphing to the surface as a depressed area after the glue dires. They are just not necessary for edge to edge glueing.

I do not own parallel jaw clamps, I use pipe clamps. 3/4 are nicer than 1/2 but I have a bunch of 1/2 inch pipe clamps I still use (in 3 and 4 foot lengths). If the wood is straight, I may glue 3 or 4 together at a time but if the boards are not as flat, I will glue two at a time so I can align the boards as I go. I start at one end and work my way down the boards spreading the other end up and down to get the area the next clamp will go flat. I ususally just try and keep the pipe off the board but a nice trick I haven't tried yet is to put pieces of shower curtain cover on your pipe clamps at the glue area to avoid black marks.

Jim

 

[Howard Acheson]

The issue of "stronger" is different from the issue of longevity.

It's clear that old time hide glue was quite strong but it does deteriorate over time. It is a brittle glue and does not tolerate stressed joints well. With stress or just the periodic expansion/contraction of some joints, it's brittleness works against it and it becomes powdery with time and fails. Witness all the old furniture that needs repair.

Modern adhesives like PVA, Urea formaldehydes, casein and epoxies have the potential to last forever. While all are less than 100 years old, there is very little evidence that they deteriorate with time in normal environmental variation.

All the above adhesives are "stronger than the wood" itself for all but a number of exotic woods.

PVA (yellow or white) and epoxy are both slightly elastic. This is goodness for joints that are periodically stressed. PVA and epoxy are particularly good for cross grain joints like half laps and mortise and tenon. They will tolerate the slight seasonal movement and still maintain their strength. Epoxy is particularly good for chair construction or repair. There is probable no piece of furniture subjected to more variable stress than chairs and there is probably no piece of furniture that is dangerous than a chair that fails.

Unless we were repairing a true, valued antique, all chair repairs were made with slow set, two part adhesive. New chairs were all made with epoxy.

The bottom line is that I know of no modern adhesive that has any life limiting characteristic in normal use. They should last forever.

 

[Joe Scharle]

I would use the same method for either of your examples -- after selecting the set of boards to glue up by matching the best grain pattern, I'd select a pair of adjacent boards, fold them with the top side in and edge joint them at once against the fence of my jointer (this assures any slight angle off perpendicular is eliminated). Continue joining all internal edges. Apply Titebond II or III to each glue edge and clamp with good parallel faced clamps (I use Bessey K-Body clamps) every 12", alternating clamps above and below boards. Be careful to wipe off any squeeze out with a damp, clean rag to avoid glue closing pores of the wood. I also use waxed paper over my clamp bars under the glue-up to keep the bars cleaner.

When I use pipe clamps, I put small pieces of pipe insulation around the pipe to keep the pipe off the glue line.
Joe

 

[daver828]

thank you for that Howard, and all others. You are all helping me greatly.

 

[SkintKnuckle]

Howard is right on!

Some corrections to some of the other posts, in the days of old, not only did they have hide glue, but they also used blood (protein) and casein adhesives. A high casein content adhesive will give a stronger than wood joint in pretty much all domestic species, but costs about 10 times more than a PVA to produce, and is now where as convienient to use. In addition dried blood produces better glueline wetting than any synthetic in use today... some high end applications still utilize a small amount in their gluing, but it is some nasty stuff.

I suspect that the longivity of a glue joint depends more on the stresses imposed on it during it's lifetime.

 

[DavidF]

I think there is a place for long splines in a table top glue up when the pieces are not perfectly flat, which unless you have a massive jpinter is quite likely. If you rely on clamp pressure alone to hold the edges of slightly bent boards together there is a good chance that you are exerting too much pressure and may starve the glue joint and even then they might move slightly before the glue dries.

 

[JackLeg]

I like a glue joint bit and Titebond III. Works for me.

 

[BumoutBob]

My wife brought home an old Maple dining room table that had glue joint failure. It was a large table and appeared well made but some of the glue joints on the top had opened up and the legs showed either wood shrinkage or glue line creep. When trying to repair the top, I found what I thought was starved joints. The glue was clear to white and I thought it might be 1950's white glue.

Can any of you more experienced woodworkers talk about how much glue to use and how much clamp pressure?

 

[JimD]

There is an article in a FWW within the last year that provides optimum clamp pressure. My take-away from reading it is that I can essentially never get up to the optimum pressure - I do not have enough clamps. My suggestion is that you do not worry about starving the joint. The optimum pressure is a function of the wood, however, so maybe with softwoods or balsa or something you could have an issue. With maple, I do not think you will get there. I typically clamp large surfaces with pipe clamps about a foot apart.

Best glueing technique is to spread out a thin layer on both surfaces and clamp. I usually just do one surface and do not always spread it out and I have not had problems. I like to touch the second surface on the first, however, to put some glue on it (and reduce the thickness on the first piece) prior to glueup. Flux brushes work well to spread it out. I think best technique for removal of excess is to wait an hour, remove the clamps, and remove the excess (I use a cheap old chisel). Titebond says you can remove clamps in half an hour so my hour wait is probably conservative. Glue excess is much easier to remove if it is not fully hard. I like to see a little bit of squeezeout all along the joint. Then I know I got enough on. A lot of squeezeout is OK for strength but makes a mess.

Jim