Official Luthiers Forum!

I guess empirically, glue must be flexing, or else the top braces would pop off when the bridge twists. But, there is often reference to hide glue and fish glue drying absolutely hard and will not creep. So, how much can glue joints on the top flex before they are compromised due to the glue cracking?

The ideal glue joint is one molecule thick. So the glue provides no resistance to flexing. That all comes from the wood.

If we could get wood surfaces so well-matched that a "one molecule" glue line was possible, we probably wouldn't need glue at all -- we could just rely on a wood-to-wood bond. In the real world, wood surfaces are not so smooth.

Studies (e.g., http://www.franklinadhesivesandpolymers ... f?sfvrsn=0) suggest that the most effective spacing for wood-wood gluing os 0.002" to 0.006". That's quite demanding and is why significant clamping is necessary to distort the wood surfaces into such intimacy. (That's on the order of 5,000 to 15,000 glue molecules thick -- for molecules of a typical glue.)

All that said, Barry is absolutely right, the flexing comes from the wood.

Still though it's best to think of two pieces of wood are held together by a molecular rather then a mechanical process.

glues is usually about a .007 in layer between the 2 pieces. If the glue flexed it would dampen the movement. That is why the animal protein glues are so good also they are not prone to cold creep like tite bond.
so a little glue 101
A there is drying glue such as the hide and fish glue
this means that they can reamalgamate and become glue again when steamed , so the glue will stick to itself
B Curing Glue
tite bond is a prime example of curing glue , once cured it never will become glue again
This means that once the curing process is over your glue will not stick to itself and you will need to remove the old
glue when regluing a joint
C chemical glues
epoxy is a prime example
Mix 2 parts the material goes through a chemical reaction and usually heat is involved

_________________
John Hall
blues creek guitars
Authorized CF Martin Repair
Co President of ASIA
You Don't know what you don't know until you know it

'Modern' cold glues, such as Elmer's and Titebond, tend to suffer from 'cold creep' which you don't get with traditional glues such as hide glue. I have seen thick glue lines with the modern glues, particularly Elmer's, that have shifted noticeably in shear. In one case a bridge on an import guitar had shifted toward the neck by about a mm, with a line of bare wood along the back edge and a ridge of finish in front. I saw a neck on an American made guitar that had shifted inward by about the same amount. The glue line between the UTB and popsickle stick braces had not broken, but the wood had moved. The inside of the top looked like it had been smoothed with very coarse paper, which left a thick glue line: they might as well have used a garden rake.

Hans Mattes wrote:
"That's quite demanding and is why significant clamping is necessary to distort the wood surfaces into such intimacy."

You should not need to 'distort' the surfaces to get good glue lines. Clamping is the usual way of squeezing out excess glue, but if the surfaces fit properly other methods work well. Violin makers traditionally make 'rub joints' for the center joins of tops and backs. Once the surfaces have been planed level one side is clamped in a vice to hold it, with the glue side up. Hide glue is spread on both surfaces, and the two are rubbed together with firm pressure to force out excess glue. When the glue starts to 'grab', but before it gels, you slide the two pieces into alignment and walk away. If they have been properly planed the glue line will be virtually invisible, and will hold up well. I'd say that proper work calls for well fitted surfaces and careful clamping that avoids distortion.

Does anyone know if 'cold creep' is a feature of the adhesive or a byproduct of poor gluing technique? And if so, which techniques may allow cold creep?

Thanks!

_________________
Peter Havriluk

Creep is a property of plastic, therefore it's inherent to plastic based glues. I'm sure different formulas have differing amounts of creep. A good quality glue joint probably helps minimize it though. There's been millions of guitars built with Titebond and by far the majority of them don't seem to be suffering from creep issues.
If you want to eliminate creep you probably need to move to hide glue.

_________________
Jim Watts
http://jameswattsguitars.com

"If you want to eliminate creep you probably need to move to hide glue."

That's been my experience.

It's plausible that the mechanisms that cause cold creep contribute to damping, but it would be very hard to isolate that well enough to be sure. Whatever sort of glue you use, good design and construction practices that produce minimum glue line thicknesses help.

Most cold creep is the glue itself . The better your joint stronger the glue bond. Like Al Hot hide and fish are the best for not having clod creep.

_________________
John Hall
blues creek guitars
Authorized CF Martin Repair
Co President of ASIA
You Don't know what you don't know until you know it

I've been using HHG and Lee Valley 2002 glue. The 2002 glue has a high solids content (45%). It appears to have some of the same qualities as protein glues. When I didn't get to the squeeze-out until after it dried, it chipped off clean and felt 'glass hard'. Only real downside I see is the brownish color when it dries, so I only use it on darker woods or unseen joints. I can't see that there would be any cold creep with it.

Brent