What type of glue for a workbench?

UKworkshop.co.uk

Help Support UKworkshop.co.uk:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
I have a little French's mustard bottle for use at the bench for PVA and a HP one for PU. Much better than lugging a big bottle about
 
I have a little French's mustard bottle for use at the bench for PVA and a HP one for PU. Much better than lugging a big bottle about
Now you mention it I am reminded that I use a Heinz sauce bottle, the sort with a silicone valve that reseals itself for shellac, works surprisingly well and doesn’t gum up. Ian
 
Jam jar, stiff paint brush (trim the bristles on a normal one), palette knives, and a squeezy bottle with a nozzle.
Main thing is to spread glue over whole of both surfaces to be glued. Too much is better than too little.
 
Last edited:
I used Titebond 3 on mine:

20210327_180344.jpg


The important thing, I believe, is to have co-planar surfaces coated all over with glue, joined and then held in place properly until everything is dry. I did the central laminates one day and then put the end sections on a few days later.
 
PVA is rumoured to creep under load, but it's just a rumour.
Surely that's a throwback to the sixties or seventies, Jacob? (urbann myth?) Try some Everbuild D4 PVA if you get the chance. Really good for exterior work where a bit of flexibility is required
 
I've not heard one specify, though I doubt it's D4 - but D4 technically isn't a PVA (though I (and everyone else?) treat it and use it as one). I suspect most are D3 now.
I though that D4 PVA was a PVA where the cross-linking capacity had been enhanced, and that the lack of complete cross linking might be why lower grades of PVA might be subject to an amount of "slip"
 
Agree 100%. I've never heard the reason why manufacturers give the instruction "apply to one side only", though.
Believe it or not but the chemistry shows that glue actually works best if you can get a joint tight enough for the glue to be 3 molecules thick. One on each piece and one in the middle for them to bond to. The phrase "never too much glue" is in fact complete rubbish, it weakens the bond and therefore the joint and means a crapload of unnecessary clean up.
 
Believe it or not but the chemistry shows that glue actually works best if you can get a joint tight enough for the glue to be 3 molecules thick. One on each piece and one in the middle for them to bond to. The phrase "never too much glue" is in fact complete rubbish, it weakens the bond and therefore the joint and means a crapload of unnecessary clean up.
Absolutely agree, I only ever put it on one side, never seen the point of wasting glue, or time. Ian
 
I put in on both sides but very thinly, works for me, I get very little squeese out, usually just a little bit, I'd love to try hot hide glue because you can re-use the squeeze out and put it back in the pot.
 
Hi all

Thanks to some help from various sources, I am now rapidly approaching the time to start milling the timber for my workbench.

I recently re-watched Matt Estlea's split top roubo build (that's what I'm going to be building) and he recommended Cascamite over normal PVA. So I had ordered a tub from Axminster as part of another order to qualify for free delivery.

Of course after buying it I just went on the search to discover that people have had bad experiences with it recently and have moved to a few other options for "stressed" joints.

Considering I'm building a workbench, I'm fairly certain that thing is going to see some stress. So didn't want to make the wrong call before the biggest glue up of my life.

Thoughts, recommendations, welcome.

Thanks in advance.
Bp

I used some kind of titebond - don't even remember which.

I have beaten my workbench to the point of senselessness sometimes (which is why I built it). The legs are laminated, the top is laminated. I didn't joint the boards because I was too lazy to (don't have a power jointer and it seemed they'd be fine if just run through a thickness planer and forced together.

I don't know how old it is now 7 or 8 years? I wouldn't overthink this.

Nothing hs come apart on mine or even shown a tendency to. I'm beginning to think that once in a great while, internet tips are helpful in building things, but 95% of the time, they prevent people from getting on with things (I don't know what cascamite is as it's not something I've seen here, but even the tip that the cascamite is bad is kind of a hold-up that just causes you to hold up and question things. One test board overnight before starting the bench will remove questions about the quality of the glue.

I mention not jointing mine because there would obviously be parts of the top that don't have Franklin's suggested joint thickness - it doesn't matter. There is no creep, etc, at any of the joints, either.
 
Believe it or not but the chemistry shows that glue actually works best if you can get a joint tight enough for the glue to be 3 molecules thick. One on each piece and one in the middle for them to bond to. The phrase "never too much glue" is in fact complete rubbish, it weakens the bond and therefore the joint and means a crapload of unnecessary clean up.

I'm not sure about the molecules, but there's a chemist on one of the american forums, and a brilliant practical chemist. He's not a glue guy, but went down the rabbit hole years ago and dragged all kinds of information out of Franklin about glues.

Not totally related to that, but his term is wetting or whetting - if a glue surface is whetted, the glue will have good adhesion.

the only glue joints I've ever seen fail in person are not too much or too little glue, though - they are when someone plays with something like a panel joint too long, lets it set up for a short period of time and then tries to adjust it laterally minutes later when they suddenly see a step on one side.

I don't want to say whether or not I have also been that someone!

Turners won't use glue due to creep, but one of the interesting things about creep and glue is that for decades, PVA glue was used in guitars (and polyurethane was used on the outside of guitars all the way back in the 1960s by Fender, and maybe earlier by others). Once the internet became big, there was an instant "common knowledge" that PVA makes terrible guitars "because the joint is too soft and tone is lost" and that "no good guitars are made with polyurethane finishes"

I have only made a few guitars, and some have been hide glue joints and some have been PVA. I've glued on fingerboards with tite bond and glued them on with fish protein glue, and the loudest unplugged electric guitar that I have is a rosewood necked telecaster style guitar with a PVA glued fingerboard (no glue needed on the body, it's one piece)

We are makers and most of us are keeping most of what we make. If the fingerboard on the rosewood guitar ever moves, I guess i"ll have to make a video. If I ever have to fix it (doubtful), I guess it'll take longer to steam off and I'll have to scrape it.

Some days, I follow the glue advice "well, if this is a nice guitar, i want to be able to say it's got only animal glue in it" and other days, I wonder why we torture ourselves.
 
I'm not sure about the molecules, but there's a chemist on one of the american forums, and a brilliant practical chemist. He's not a glue guy, but went down the rabbit hole years ago and dragged all kinds of information out of Franklin about glues.

Not totally related to that, but his term is wetting or whetting - if a glue surface is whetted, the glue will have good adhesion.

the only glue joints I've ever seen fail in person are not too much or too little glue, though - they are when someone plays with something like a panel joint too long, lets it set up for a short period of time and then tries to adjust it laterally minutes later when they suddenly see a step on one side.

I don't want to say whether or not I have also been that someone!

Turners won't use glue due to creep, but one of the interesting things about creep and glue is that for decades, PVA glue was used in guitars (and polyurethane was used on the outside of guitars all the way back in the 1960s by Fender, and maybe earlier by others). Once the internet became big, there was an instant "common knowledge" that PVA makes terrible guitars "because the joint is too soft and tone is lost" and that "no good guitars are made with polyurethane finishes"

I have only made a few guitars, and some have been hide glue joints and some have been PVA. I've glued on fingerboards with tite bond and glued them on with fish protein glue, and the loudest unplugged electric guitar that I have is a rosewood necked telecaster style guitar with a PVA glued fingerboard (no glue needed on the body, it's one piece)

We are makers and most of us are keeping most of what we make. If the fingerboard on the rosewood guitar ever moves, I guess i"ll have to make a video. If I ever have to fix it (doubtful), I guess it'll take longer to steam off and I'll have to scrape it.

Some days, I follow the glue advice "well, if this is a nice guitar, i want to be able to say it's got only animal glue in it" and other days, I wonder why we torture ourselves.
Try gluing a bridge on an acoustic guitar with white pva and see how long it holds out in a car on a sunny day! I've used titebond original for over 40 years for making guitars and never had a problem. I've also still got my first ever classical guitar that I made and the joints were glued with bone/hide Glue and the joints are all fine. My workbenches were glued up with Titebond original (these are made from beech 50-75 mm thick) and they have held up with no problems. There is quite a bit of movement sideways between winter 25% rh summer 80% rh on the bench top which I factored in when making the benches by putting an endpiece with a tongue and groove fixed with a coach bolt in the middle that allows for movement. I've used recently titebond extend (I think this is the same stuff as lmii sell a luthier glue) this gives a bit more glue up time and does also come apart with heat water alcohol and dries a bit harder than titebond original (more like hide glue hardness) only problem with extend is the shelf life, which is not as long as titebond original. As for the tone of different glues with instruments, I think a lot of nonsense is talked about by the cork sniffers about the tone of hide glue against aliphatic glues or whatever. I'd be interested if anyone could come up with any proper scientific way of measuring tonal effects of different glues, which I think would be minimal as on a good glue joint there should be virtually no glue left in the joint. A good musician and fresh strings make a helluva a lot more difference. Any way back to the original question. I would use in the following order for gluing a bench together 1: titebond extend (gives extra clamp up time) 2:Titebond original (it works and is more easily available than extend) 3: pva wood glue (should work ok if joints are good and wood is properly seasoned and is easily avaialble) 4: Cascamite (should work fine and is waterproof if you decide to work outside in the rain. 5: bone/hide glue (if you can afford the heating bills to get the workshop hot enough to glue up the panels and have enough people to get the glue and clamps on before it gels) altough this must have been the way they made the benches in the old days.
Cheers
Andrew
 
Try gluing a bridge on an acoustic guitar with white pva and see how long it holds out in a car on a sunny day! I've used titebond original for over 40 years for making guitars and never had a problem. I've also still got my first ever classical guitar that I made and the joints were glued with bone/hide Glue and the joints are all fine. My workbenches were glued up with Titebond original (these are made from beech 50-75 mm thick) and they have held up with no problems. There is quite a bit of movement sideways between winter 25% rh summer 80% rh on the bench top which I factored in when making the benches by putting an endpiece with a tongue and groove fixed with a coach bolt in the middle that allows for movement. I've used recently titebond extend (I think this is the same stuff as lmii sell a luthier glue) this gives a bit more glue up time and does also come apart with heat water alcohol and dries a bit harder than titebond original (more like hide glue hardness) only problem with extend is the shelf life, which is not as long as titebond original. As for the tone of different glues with instruments, I think a lot of nonsense is talked about by the cork sniffers about the tone of hide glue against aliphatic glues or whatever. I'd be interested if anyone could come up with any proper scientific way of measuring tonal effects of different glues, which I think would be minimal as on a good glue joint there should be virtually no glue left in the joint. A good musician and fresh strings make a helluva a lot more difference. Any way back to the original question. I would use in the following order for gluing a bench together 1: titebond extend (gives extra clamp up time) 2:Titebond original (it works and is more easily available than extend) 3: pva wood glue (should work ok if joints are good and wood is properly seasoned and is easily avaialble) 4: Cascamite (should work fine and is waterproof if you decide to work outside in the rain. 5: bone/hide glue (if you can afford the heating bills to get the workshop hot enough to glue up the panels and have enough people to get the glue and clamps on before it gels) altough this must have been the way they made the benches in the old days.
Cheers
Andrew

that doesn't make sense, there's no reason why white glue PVA is inferior to a huge greedy corporation like titebond, they haven't found some magic formula, it's all marketing BS and they need to be called out over it, I don't think titebond is better in any way at all from my own tests, there is no real evidence that it is. Also, you would never leave an acoustic guitar in a hot car on a sunny day, I've been playing over 20 years, mine stays in its hardcase all the time. 😂
 
Last edited:
Absolutely agree, I only ever put it on one side, never seen the point of wasting glue, or time. Ian
Believe it or not but the chemistry shows that glue actually works best if you can get a joint tight enough for the glue to be 3 molecules thick. One on each piece and one in the middle for them to bond to. The phrase "never too much glue" is in fact complete rubbish, it weakens the bond and therefore the joint and means a crapload of unnecessary clean up.
I've had a one or two glue failures and it's always been due to not enough glue and/or glue on one side only. When you look at the separated pieces you see the glue spread evenly all over one side has somehow not entirely made it to the other and there are glue free patches, in spite of cramping things tight.
I tried it as an experiment - glue one side, cramp up, quickly pull apart and see if the other side has got glue all over it. Quite surprising how it often does not spread and fill the gap.
So for me it's glue both sides, too much better than too little, if the glue is not being squeezed out into a little bead along the whole of the joint then there is not enough in there.
 
that doesn't make sense, there's no reason why white glue PVA is inferior to a huge greedy corporation like titebond, they haven't found some magic formula, it's all marketing BS and they need to be called out over it, I don't think titebond is better in any way at all from my own tests, there is no real evidence that it is.
Years back I had a chat with a chemist from a local glue manufacturer and asked him about Titebond. He said that "aliphatic resin" is more or less the same, chemically, as a D2 or D3 PVA and that it was to a great extent a marketing ploy because all these glues are fundamentally derived from starch which lends itself to being altered to improve cross-linking, change the viscosity and colour (by adding dyestuffs). To solve a problem we were having with dowel glueing he formulated a less viscous D3 PVA for us which sort of proved his point
 
Last edited:
that doesn't make sense, there's no reason why white glue PVA is inferior to a huge greedy corporation like titebond, they haven't found some magic formula, it's all marketing BS and they need to be called out over it, I don't think titebond is better in any way at all from my own tests, there is no real evidence that it is. Also, you would never leave an acoustic guitar in a hot car on a sunny day, I've been playing over 20 years, mine stays in its hardcase all the time. 😂
Well i've been making and repairing guitars since 1976 and i've had enough repairs to do with bridges lifting with white glue and people do leave their guitars in cars or in dry rooms (I doubt that's a problem in the UK where I was born and lived for 30 years) titebond will also give up in high temperatures although it's tendency for creep is less. I'm posting here from my own experiences and have found that titebond original and extend works fine and prefer it over most white pva's that tend to dry a little soft. All the stuff about big corporations is another topic. By the way which tests have you conducted on the different glue capabilities ? i'm not too old or stuck in my ways (excuse the pun) and ready willing to learn something new.
Cheers
Andrew
 
Last edited:
Back
Top