New WoodChat!
Home Articles Forums Gallery Marketplace Videos

WoodCentral Forums

Knowledge You Can Build On — Est. 1998

Exotic Hardwood, Personally Sourced Worldwide. Inspected at Origin. Shop Rare Stock.

Essential glue attributes

Posts

Essential glue attributes

#1

Essential glue attributes

Bill Tindall, E.TN.

An effective glue must have two essential properties. Focusing just on wood gluing, first it must wet the wood. In the context of adhesion wet means the electrons of the cured glue must be mingled with the electrons on the surface of the substrate. This "mingling" is not enough to say that a chemical bond has formed like what holds the glue molecules together, but it is similar. We can assume that anything sold as wood glue meets this criteria and we can not worry further about it unless we spilled linseed oil, or some such, on the wood before gluing and thereby changed the surface properties of the wood.

Second, the cured glue must be, ideally, as strong as the wood. When a joint is stressed the wood should be the weak link and we get wood failure. This failure is the goal of gluing. Or, the bond between wood and cured glue may fail. If the wood surface is not contaminated and the glue properly applied this failure won't happen using a glue designed for wood. A common failure here is to let the surface of the glue dry too much before assembling the joint. Or, in the case of protein glue, to apply at too low a temperature.

In a practical joint there will be a layer of cured glue between the two wood surfaces. To get wood failure the mechanical properties of this bulk cured glue must be better than the wood. Any glue recommended for wood will be stronger than wood unless the glue has been "damaged".

"Undamaged" wood glue of any kind will cure to a tough plastic. A thin layer of cured glue will be flexible and very strong like a piece of Mylar or Nylon.

In the case of the protein glues the protein molecules can be chopped up by bacteria, mold or other chemical action, enhanced by high temperature or retarded by additives. These processes render the glue brittle instead of tough. The particles in the yellow glues must coalesce to become tough. They will not coalesce below a certain temperature, nor if the moisture content of the wood is too high. The result will be a powdery or rubbery "cured" glue.

Because all wood glues are equivalent in terms of wetting and toughness the choice of what glue to use is made on the basis of secondary considerations- price, availability, use temperature, shelf life, working time, convenience, Period Correct, etc. "Better" only has meaning if the relative importance of these secondary considerations are defined.

Re: Essential glue attributes

#2

Re: Essential glue attributes

David Weaver

Don't forget repairability. I will use titebond III in my kitchen cabinets because of temperature constraints, but everything else I build (tools, furniture) gets hide glue now.

Unless I need gap filling, that goes to full time length/cure epoxy (as opposed to 5 minute types, etc).

Re: Essential glue attributes

#3

Re: Essential glue attributes

Warren in Lancaster, PA

When a joint is stressed the wood should be the weak link and we get wood failure. This failure is the goal of gluing.

I wonder what the rationale for this is. Certainly if a butt joint between two boards fails, it is easier to repair if the glue failed and not the wood. The worst is partial wood failure and partial glue failure where you have yellow glue residue.

Have you ever repaired a mortise and tenon joint where the wood failed and not the glue? Or a dovetail joint with pieces of the tails still stuck in sockets? How about chair tenons that pulled part of the seat along instead of coming out cleanly. These are not pretty sights. Frankly glue failure ahead of wood failure is a good thing.

It could be that Mack Truck Glue Co. likes to tell people that their glue is stronger than wood itself. Makes a nice story, but in the real world it is not all that helpful.

Re: Essential glue attributes

#4

Re: Essential glue attributes

wilbur

Also, don't forget the ability of the glue to interfere with later finishing steps. From that standpoint, hide glue has it all over the PVA glues in terms of how fussy you have to be with cleaning up the glue so that it doesn't interfere with the finish.

Re: Essential glue attributes

#5

Good point


Re: Essential glue attributes

#6

David Barnett

(Message Deleted by Poster)


Re: Essential glue attributes

#7

Glue as a "fuse" ?

Bill Tindall, E.TN.

I was inclined to not respond but then Ron revealed that there may be some that could be taking your recommendation seriously. If there is anyone that builds stuff and arranges to have the glue be the "fuse that bows" when the furniture is stressed I would like to hear how you arrange this situation, even if it was desirable, which it is not under any situation I can imagine. Provide examples of stuff built incorporating this design philosophy.

Let us examine your M&T example. If a leg is stressed and there is no glue(or pins that accomplish a similar purpose) the wood will fail unless the stress is simply pulling the tenon out. If the glue joint is stronger than the wood the stress will be born by both the tenon and mortise socket instead of only the mortise socket, which is why we glue these joints unless we are building in the style of 1220 AD.

Re: Essential glue attributes

#8

David Barnett

(Message Deleted by Poster)


Re: Essential glue attributes

#9

Re: Warren's position is shared by conservators...

David Weaver

Fantastic last line David. I agree with warren's views, too. I'm no conservator, but I do like to be able to fix things without too much effort.

To date, I haven't had to fix anything except tools where glue joints have let loose, but I envision with furniture bits that I build for the kids, it's not out of the question that some part of them could get bashed and benefit from easy replacement.

I'd rather never find out in practice, but.....you made one other good point, it's nice to assemble a piece and not have to be quite so anal retentive looking for glue. My very first piece suffered from that. It LOOKED to my untrained eye like I had sanded away all of the glue, but I didn't and an application of shellac showed it.

Re: Essential glue attributes

#10

The trouble is this...

John in NM

Conservators choose adhesives to make repairs, not to make furniture. As they are only ever working on historically valuable pieces, their adhesive choices are pushed far to one side of the spectrum.

Bill's point is that it is far more important to most of us, as builders, to produce work that will not need repair more than to produce work that will be easy to repair. Kind of like not running red lights so as to avoid accidents rather than running them very slowly to minimize the effects of the inevitable accident.

As I have no expectation that any of my stuff will ever be historically important, I am rooted firmly in the avoid breakage camp. I'm afraid I simply find this "engineered weakness" idea to be astoundingly foolhardy.

Fair warning as well, my sister is a conservator, so that may be knocking my opinion of their good judgement down a couple pegs :D

Re: Essential glue attributes

#11

The advantage is obvious to me...

John in NM

If the glue is stronger than the wood, thus the wood fails first, one avoids having to make repairs whenever possible. If I build and sell a chair, I will obviously want the potential failures to occur long after I'm dead and buried, because otherwise the chair's owner will come to me to fix it. Thus I use a glue stronger than the wood.

But then I'm like one of those evil capitalistic types you hear so much about in the news (except that I'm not so good at it as to have made enough money to actually be one of those in the news :D ). I fix the stuff I make for free, or replace it for free - why wouldn't I use the strongest glue I can possibly find?

It is absolutely no different than avoiding other silly mistakes that lead to callbacks - non-floating panels comes instantly to mind.

Re: Essential glue attributes

#12

Re: Engineered weakness, danger in relative terms

David Weaver

..I would consider engineered weakness to be a relative term danger.

If a force of 3x is needed to make the joint fail and a force of 10x is needed to make the wood fail, there is a lot of latitude for furniture to be built plenty well for general use and still be easily repairable.

I still have not had any joint fail in furniture that I've made (I suppose if I made chairs, it would be more likely), and not all of the glue joints have been made in ideal conditions. I'd be inclined to find the failure point before I concluded the equivalent of a 3 pass weld was needed when the item in question never sees the force to break a weld in a single pass.

Re: Essential glue attributes

#13

David Barnett

(Message Deleted by Poster)


Re: Essential glue attributes

#14

David Barnett

(Message Deleted by Poster)


Re: Essential glue attributes

#15

David Barnett

(Message Deleted by Poster)


Re: Essential glue attributes

#16

But we're not talking about glue creep...

John in NM

Though clearly that is one of the big disadvantages of PVA's.

The question we are discussing is whether one should look for a glue that forms a bond (ideally) which is stronger than your average chunk of wood, or whether one should select a glue that will fail before damage is done to wood parts by whatever toddler is so grossly abusing our fine furniture.

Personally, I opt for stronger - if I've managed to do everything else right, I simply don't want a callback due to using weak glue so I don't have to make a new chair rail or something (these days I'm selling gunstocks and grips, not chairs and tables, but its the same principle and the same warranty). I'd much rather make a new part for one customer than service failing glue for one hundred customers.

Re: Essential glue attributes

#17

I find use for several glues...

John in NM

Hot hide glue, PVA, plastic resin glue, epoxy, and even, horror of horrors, contact cement :D

I'm not arguing for one over the another at all. As far as I know, a properly made joint, glued with hide glue tends to be stronger than wood.

I'm arguing against the notion endorsed by a few posters here, that Bill so aptly summed up with "glue as a fuse".... I'm afraid I find the idea to be so silly that it hardly bears consideration. I don't mind a tough repair job now and then if it means there are fewer repairs to be made on the whole, end of story.

One could even argue that having the glue fail under stress before the wood does is just as bad as having the wood fail first. How many times have we come across loose jointed furniture that the owner simply hasn't gotten around to dealing with? Experience tells me that furniture with failing glue joints will be kept in service until the failure of a wood part removes it from service and it gets taken to the wood butcher to get fixed. Either way we end up making new parts and attempting to salvage any that are not too badly damaged.

Re: Essential glue attributes

#18

But isn't that what we are discussing?

John in NM

Go read Warren's post again. He's talking about the inconvenience of stronger than wood glues when repairing furniture. Consciously choosing a weaker glue to minimize this inconvenience seems to have been proposed and eminently fits what I would define as "engineered weakness"

Just like those cast iron axle shafts that Land Rover used to use :D

Re: Essential glue attributes

#19

David Barnett

(Message Deleted by Poster)


Re: Essential glue attributes

#20

David Barnett

(Message Deleted by Poster)


Re: Essential glue attributes

#21

Looks like I'm missing something...

John in NM

It seems this is really a PVA vs. hide glue discussion and I didn't get the memo. Oh well, nothing new in either of those :D

Re: Essential glue attributes

#22

David Barnett

(Message Deleted by Poster)


Re: Essential glue attributes

#23

David Barnett

(Message Deleted by Poster)


Re: Essential glue attributes

#24

When it comes right down to it...

John in NM

The ideal joint probably won't ever fail regardless of the glue used, making the entire discussion moot :D

Maybe we should switch to honing methods, this horse is looking a bit ratty. You can have the rump skin for a strop now that we've worn all the hair off :D

Re: Essential glue attributes

#25

Re: When it comes right down to it...

TomD

The ideal joint may not allow the ideal structure. Most serious work like say plane spars, or mosquito bombers, is done with good glues and loads spread over huge areas. Most ideal joints are poor structures. And this raises the further point that given what glues are actually used for (plywood say), are commercial glues being designed to do any of the stuff we actually do. The labs are impressive, but I doubt they have ever considered a gluing problem that comes up in my shop, or most of yours. One of the reasons WEST epoxy got off the ground was that they had boardroom level bro access, at I think Dupont, to pervert the coarse of the lab. Great things often take the personal touch.