My latest bottle is gorilla and it works well enough. But exactly like you said, I don’t think I could pick it out from every other bottle I’ve used in the last 20 years.
White Elmer’s glue is pretty much the same formula as their “washable” school glue. It bonds wood quite strongly but it tends to be slimier than wood glue so when you go to clamp the boards together they tend to slip around out of orientation. It’s not as fun to work with as yellow carpenter’s glues which tend to be tackier so the boards don’t slip around as much.
Titebond 3. It’s a pretty easy choice; it has one of, if not the highest strengths of wood glues on the market, and it’s water resistant. If you want the wood to break before the glue does, that’s the stuff you want.
That is usually what I go with, because I normally only keep one bottle of wood glue around and it covers pretty much any use case I could ever have for wood glue being waterproof, safe for indirect food contact, etc.
But honestly, for general gluing furniture together and such, even the cheapest no-name brands of wood glue have always done just fine. Pretty much any wood glue out there is stronger than any wood you’re likely getting the be gluing (inb4 some carpentry nerd chimes in with some rare wood that only grows in New Zealand or something that is stronger than steel or something)
I’ve seen plenty of bonds on furniture fail, rather than the wood. It seems most typical on things that are a dowelled construction rather than a mortise and tenon joint. I’ve seen it most often with chairs, since they’re under a lot of stresses. Maybe I’m in a uniquely bad environment that’s harsh on wood glue; I don’t know.
Chairs, like wooden dining room chairs, are some of the most dynamically stressed woodworking projects. A bookcase may carry hundreds of pounds of books but you put the books on the shelf and they mostly just stay there. A dining room chair has people sitting down, scooting forward, shifting around, leaning back, standing up etc. so there’s a lot of force moving around trying to bend the frame members and shift the tenons around in their mortises. This often causes the glue, or the wood immediately around it, to fracture under repetitive stress and causes loose joints.
Some woodworkers prefer to use hide glue (or its modern synthetic equivalent) rather than PVA glue specifically because it isn’t as strong, and because the bond can be released with heat. That allows the glue to fail while the wood itself remains intact, and then a chair with a failed joint can be disassembled and repaired. A chair assembled with PVA is likely to break in the middle of a board or dowel and is impossible to disassemble in any intentional way.
Yeah I think the way a lot of chairs are constructed is just a bad use case for glue. Like you said, chairs are under a lot of stress (tension, compression, shear, cleavage, peel- glue can handle some of these well and others not,) there’s a lot of weird ways you can put leverage on the joints, people don’t tend to sit perfectly still so those loads are dynamic and always shifting a bit, and to make it worse they kind of have to be designed to be somewhat lightweight, easy to move around, small enough to fit under a table, etc so there’s always some compromises made and they’re never as overbuilt as they probably should be.
Different kind of construction, but I work in a 911 dispatch center, we have some ridiculously overbuilt chairs that are supposed to be rated for someone to be occupying them 24/7. They cost a ridiculous amount of money and they’re still breaking in new and spectacular ways almost constantly. It’s tough to build a good chair.
There’s also of course issues that can arise from bad surface prep, poor fitment, improper clamping, too little glue, not letting it dry long enough too high/low temperature/humidity/moisture, the wood shrinking/expanding, poorly thought-out joints that don’t have enough surface area or are putting the glue under the wrong kind of stresses, and of course sometimes you’re asking the glue to do something it doesn’t do well, it’s good at gluing wood to wood, but not nearly as good at gluing paint to paint or varnish to varnish.
Unfortunately, wooden chairs that don’t suck tend to cost a fuckton. The styles that people tend to like are usually on the fragile side by their nature.
Wood glues like Titebond are PVA-based glues. So is Elmer’s white school glue, which is also very good at bonding wood. Wood glues are yellow either because of added resins that make it tackier when wet so clamped boards don’t slide around as easy before the glue dries, so that the glue dries to a harder, more rock-like consistency rather than staying slightly flexible, or because wood is kind of yellow so they wanted it to look like wood, I’ve heard all three and I’m not sure which is true. Titebond does sell a brown wood glue so that it blends into darker woods like walnut and ebony though.
A face grain to face grain joint will usually fail at least partially along the growth rings in the board(s) rather than at the glue joint, yes. But an end grain to end grain joint (which are rarely made for practical reasons) will typically fail at the glue joint.
Wood is kind of a composite material; it’s cellulose fibers bound together with a polymer called lignin. PVA wood glue is stronger than lignin but not as strong as the cellulose fibers, so a broken face-to-face joint will break along the weakest, the lignin.
If you edge glue a panel together for say a table top, gluing boards edge-to-edge, that board will be at least as strong as if it was one wide board; it will take at least as much force as a single board to break.
But, if you glue two long boards end to end, it won’t be as strong as a single continuous board of the same overall length. It will fail at and along the glue joint, maybe pulling a couple splinters out of one board. Which is why we basically never do that; if a board has to be spliced it’s common to add a doubler so there are fibers crossing the joint line.
But yes PVA glue like Titebond is amazingly good at bonding wood.
Wood glue, no particular brand recommendations, is one of the pew products I trust to do exactly what it claims to - glue wood.
I wonder if there is any bad wood glue out there. I use it quite a bit and I don’t think i ever used the same brand twice.
My latest bottle is gorilla and it works well enough. But exactly like you said, I don’t think I could pick it out from every other bottle I’ve used in the last 20 years.
For some reason I have a thought in my head that I don’t like Elmer’s wood glue. I don’t know why, I don’t remember it ever letting me down.
White Elmer’s glue is pretty much the same formula as their “washable” school glue. It bonds wood quite strongly but it tends to be slimier than wood glue so when you go to clamp the boards together they tend to slip around out of orientation. It’s not as fun to work with as yellow carpenter’s glues which tend to be tackier so the boards don’t slip around as much.
Titebond 3. It’s a pretty easy choice; it has one of, if not the highest strengths of wood glues on the market, and it’s water resistant. If you want the wood to break before the glue does, that’s the stuff you want.
That is usually what I go with, because I normally only keep one bottle of wood glue around and it covers pretty much any use case I could ever have for wood glue being waterproof, safe for indirect food contact, etc.
But honestly, for general gluing furniture together and such, even the cheapest no-name brands of wood glue have always done just fine. Pretty much any wood glue out there is stronger than any wood you’re likely getting the be gluing (inb4 some carpentry nerd chimes in with some rare wood that only grows in New Zealand or something that is stronger than steel or something)
I’ve seen plenty of bonds on furniture fail, rather than the wood. It seems most typical on things that are a dowelled construction rather than a mortise and tenon joint. I’ve seen it most often with chairs, since they’re under a lot of stresses. Maybe I’m in a uniquely bad environment that’s harsh on wood glue; I don’t know.
Chairs, like wooden dining room chairs, are some of the most dynamically stressed woodworking projects. A bookcase may carry hundreds of pounds of books but you put the books on the shelf and they mostly just stay there. A dining room chair has people sitting down, scooting forward, shifting around, leaning back, standing up etc. so there’s a lot of force moving around trying to bend the frame members and shift the tenons around in their mortises. This often causes the glue, or the wood immediately around it, to fracture under repetitive stress and causes loose joints.
Some woodworkers prefer to use hide glue (or its modern synthetic equivalent) rather than PVA glue specifically because it isn’t as strong, and because the bond can be released with heat. That allows the glue to fail while the wood itself remains intact, and then a chair with a failed joint can be disassembled and repaired. A chair assembled with PVA is likely to break in the middle of a board or dowel and is impossible to disassemble in any intentional way.
Yeah I think the way a lot of chairs are constructed is just a bad use case for glue. Like you said, chairs are under a lot of stress (tension, compression, shear, cleavage, peel- glue can handle some of these well and others not,) there’s a lot of weird ways you can put leverage on the joints, people don’t tend to sit perfectly still so those loads are dynamic and always shifting a bit, and to make it worse they kind of have to be designed to be somewhat lightweight, easy to move around, small enough to fit under a table, etc so there’s always some compromises made and they’re never as overbuilt as they probably should be.
Different kind of construction, but I work in a 911 dispatch center, we have some ridiculously overbuilt chairs that are supposed to be rated for someone to be occupying them 24/7. They cost a ridiculous amount of money and they’re still breaking in new and spectacular ways almost constantly. It’s tough to build a good chair.
There’s also of course issues that can arise from bad surface prep, poor fitment, improper clamping, too little glue, not letting it dry long enough too high/low temperature/humidity/moisture, the wood shrinking/expanding, poorly thought-out joints that don’t have enough surface area or are putting the glue under the wrong kind of stresses, and of course sometimes you’re asking the glue to do something it doesn’t do well, it’s good at gluing wood to wood, but not nearly as good at gluing paint to paint or varnish to varnish.
Totally agree.
Unfortunately, wooden chairs that don’t suck tend to cost a fuckton. The styles that people tend to like are usually on the fragile side by their nature.
Wood glues like Titebond are PVA-based glues. So is Elmer’s white school glue, which is also very good at bonding wood. Wood glues are yellow either because of added resins that make it tackier when wet so clamped boards don’t slide around as easy before the glue dries, so that the glue dries to a harder, more rock-like consistency rather than staying slightly flexible, or because wood is kind of yellow so they wanted it to look like wood, I’ve heard all three and I’m not sure which is true. Titebond does sell a brown wood glue so that it blends into darker woods like walnut and ebony though.
Yeah. If it fails it’s the wood around the glued joint, not the joint iself.
A face grain to face grain joint will usually fail at least partially along the growth rings in the board(s) rather than at the glue joint, yes. But an end grain to end grain joint (which are rarely made for practical reasons) will typically fail at the glue joint.
Wood is kind of a composite material; it’s cellulose fibers bound together with a polymer called lignin. PVA wood glue is stronger than lignin but not as strong as the cellulose fibers, so a broken face-to-face joint will break along the weakest, the lignin.
If you edge glue a panel together for say a table top, gluing boards edge-to-edge, that board will be at least as strong as if it was one wide board; it will take at least as much force as a single board to break.
But, if you glue two long boards end to end, it won’t be as strong as a single continuous board of the same overall length. It will fail at and along the glue joint, maybe pulling a couple splinters out of one board. Which is why we basically never do that; if a board has to be spliced it’s common to add a doubler so there are fibers crossing the joint line.
But yes PVA glue like Titebond is amazingly good at bonding wood.