Home Articles Forums Gallery Marketplace Videos

WoodCentral Forums

Knowledge You Can Build On — Est. 1998

Build it. Rock it. Rockler has your woodworking tools.

Joint life ( how long will glue last)

Posts

Joint life ( how long will glue last)

#1

Joint life ( how long will glue last)

Bill Tindall

There have been some discussions about how long various glues will last till the glue degrades and the joint fails. No doubt this fretting has left some wondering if the stuff they are assembling with modern glue will last as well as stuff assembled with hide glue. It will take another few hundred years to answer this question with certainty but apparently there is not yet any cause for concern(see below) if you are using glue of similar chemical composition to Titebond. If you are using cyanoacrylate adhesive there is ample cause for worry upon which I can elaborate if there is interest.

I met a retired chemist from Franklin glue company and he has been receptive to answering questions. He also sent me a DVD with loads of technical data on glue and gluing, the content of which will no doubt be dismissed by those unwilling to believe anything useful in woodworking has been invented in the last 100 years. However, I have found it informative reading. I asked him about what is know about the service life of polyvinyl acetate based glue (Titebond and the like). I have copied his answer below:

The product sold as Titebond Original Wood Glue has been marketed since the early 1950's, and we are not aware of any tendency for bonds made with that product to change or degrade over time. While some types of glues can physically or chemically degrade over time, particularly under exposure to moisture or heat, the most common cause of change in most bonds is a result of the dynamic nature of wood, perhaps coupled with extremes of heat or moisture.

As you know, all wood changes dimension as it changes in moisture content, and that dimension change can create stress in an assembly. Small changes, and those that occur slowly, like seasonal changes in moisture, cause little stress, and rarely affect a well made bond, but much larger stresses can occur when a glued project is exposed to large or rapid changes in moisture. It is those large stress levels that are prone to eventually affect a joint, with that change more likely to occur if the original quality of the joint is somewhat suspect. In the case of glues like our Liquid Hide Glue, the moisture itself also softens or weakens the glue, so a joint is likely to fail if it is exposed to extended periods of high humidity or if it ever actually gets wet. For the rest of the Titebond family, changes in a bond are more likely to occur due to exposure to excessive heat than due to high humidity or a little dampness. In that regard, the glue is

somewhat softened by the heat, with a product like Titebond Original only about half as strong at 150�F as it is at room temperature. While the heat does not directly damage the glue (so the glue would be just as strong after it returned to room temperature), the heat does tend to rapidly dry the wood, and the shrinkage associated with that rapid drying can sometimes create enough stress to distort the glue and damage the bond while the glue is warmer and weaker.

Ultimately, the ability to avoid excessive heat, and to finish a piece in such a way as to reasonably control the swings in moisture to which it will be exposed is the best way to assure that the integrity of the bonds will not be compromised by their age.

Re: Joint life ( how long will glue last)

#2

Re: Joint life ( how long will glue last)

Thomas Skaggs, Foothills of Mount Level

The best way to prevent joint failure is to build good joints in the first place and to construct to allow for expansion/contraction. Good construction will minimize the importance of glue. IMHO

T.

Re: Joint life ( how long will glue last)

#3

Re: Joint life ( how long will glue last)

Dave Bair, Eaton Rapids, MI

I think that glues like TiteBond are flexible to some extent (the dreaded "glue creep). While this may be undesirable in some applications, it may also make the glue less subject to changes in the joint's dimensions. All glue failures I've repaired in old furniture are from glue that has become so brittle it's crystalized and powdery. Wood movement seems to have caused the glue to actually fracture. My point is that modern glues may never give way during the life of the furniture because they stay flexible enough to keep up with wood's dimensional changes.

Re: Joint life ( how long will glue last)

#4

Good post Bill.

Keith Newton----Little Rock, Arkansas

I would like to have more information on the CA, if you don't mind sharing it, and if it isn't too much trouble.

Thanks,

Re: Joint life ( how long will glue last)

#5

Re: Joint life ( how long will glue last)

ThomD

That is probably true to a certain extent. In most structural applications creep is your big enemy. people seem to want to believe that flexibility is structures is as good as it can be in personalities, but the reality is that for most uses flexibility is a bad thing in glue, at least at the range we get. Possibly there is a glass like glue that would be too hard, but since it has never been sold, I haven't seen that as a problem. Titebond ORG seems to be low creep thus it's success holding bridges on against heavy string load. LMII makes a better lutherie glue, and guess what, it is noticeably harder. I have a hard time wrapping my way around the idea a glue is simultaneously more and less creep resistant at the same time.

I have seen failure in cabinetmaking due to wood creep. The most interesting example in my experience is the joints on my heavy trestle table. The top is nearly 1.5" thick, and about 5 pieces. These have a different expansion ratios in the minor axis. What happened was that the top was glued up and smoothed out perfectly, yet after a while one or two of the joints became noticeable after a year or two. If one runs ones hand over the surface, one feels sharp corners, just as if the surface had not been flattened. At first I was stunned thinking that I had missed finishing the surface, but then it became obvious that I could not have made such a mistake running my hands over the surface while finishing it. So the question is what would have happened with harder glue less likely to shift. Would it have split, or would it have controlled the wood surface. Don't know. The glue I used wasn't Titebond, by the way, at that point it wasn't sold locally, but I don't know whether it would have had the strength to hold up either.

One of the problems with modern glues is they are often combined in commercial work with modern joints, so you get joints that are a nightmare to fix, like finger joints or dowel joints. In my experience these have a 100 percent failure rate over 40 years. That isn't to say every joint will fail, but every piece I have seen with heavy use will have one joint that fails. I had to choose, I would do these new joints with old glue, and use the new glues on old joints.

While we are going through a renaissance of interest in hide glue, in the past, the few big name makers I ran into said they used new glues not because they were better to repair or over the long term, but because they were better in a single cycle. If you sell someone a 3K end table, you never want it back in the shop, and they felt they were more likely to get that service from modern materials, possibly both in the glues and finishes. Of course these guys were using the old joints.

My observation is that the hard glues are normally more expensive, if that tells you anything.

Re: Joint life ( how long will glue last)

#6

Re: Joint life ( how long will glue last)

ThomD

"the content of which will no doubt be dismissed by those unwilling to believe anything useful in woodworking has been invented in the last 100 years."

Well as someone who buys epoxy by the drum on occasion, and probably has 20 varieties of glue on the go at any one time, I have learned that not all glues are equal, or equal to the task. The problem with science is that even when it works 100%, it does so in response to specific design criteria. These criteria may include deliverables like: competitive pricing, shelf life, toxicity, ease of use, etc... that have nothing to do with bond. Also, it is normally unusual for the design criteria to match all the projects we have on hand. So to me it is unscientific to forget how well matched a particular product is to an intended use. Sorta like taking the scientifically designed puppy formula and feeding it to your cows.

Re: Joint life ( how long will glue last)

#7

It really is an interesting subject...

John in NM

To tell you the truth though, I tend to think that a properly made joint will last centuries under normal use, regardless of the glue (assuming its not CA of course). When the joint is less than ideal is where the modern glues seem to really outshine hide glue, or where conditions are outside "normal" - too hot, too wet, and really most of the modern glues just give you a little more tolerance of these conditions before they too fail.

Re: Joint life ( how long will glue last)

#8

Re: Coments on CA, please

AZ in Colorado Springs

I'd like to see your comments on CA, as well. I have no experience with it and would like to know the pitfalls.

AZ

Re: Joint life ( how long will glue last)

#9

stay tuned.....

Bill Tindall

My curiosity is kindled. What I thought I knew about the details of degradation of CA in gluing applications may not be true. There is no doubt that it is not considered a "permanent" adhesive in many applications but I am still looking into the details of how it breaks down . I am sure lots is known but I am still in the process of finding recent information. Water, and maybe just moisture, is not good for it. Lots is know about how it breaks down when it is used to glue body parts together, but the wood information is less abundant.

Re: Joint life ( how long will glue last)

#10

Dale Lenz

Worked on my dog's

Dale Lenz

"L" shaped skin tear.......for three days. After that, took the pooch to the vet and paid him 150$, ouch. Got 300$ in that drop off dog now!

Re: Joint life ( how long will glue last)

#11

Always a favorite for cuts...

John in NM

But for me, that's where the shop utility of CA begins and ends.

Re: Joint life ( how long will glue last)

#12

question of how

Bill Tindall

Early on we knew the military experimented with this stuff and used it for wound closure in Vietnam. I did more prosaic stuff like gluing glass or rubber horse shoes to horse feet (both were failures) or devious stuff like gluing unsuspecting persons fingers around a coffee cup handle or what ever. The person that first made it glued his arm to a lab bench before it was known that it glued flesh.

I knew it was used for wounds but never knew how. I would think that it would sting like crazy if applied to a cut. So how is it applied/used to seal, bring together ??? cuts. The formulation used for medical is not the stuff in the hardware store, but its related.

I tried it for sealing tail stumps after lamb tail docking but failed to stem bleeding probably because I didn't know how to use it. Burned the crap out of wife's leg when I spilled some on it and it rapidly polymerized. Boggling amounts of heat are given off when it goes. Not a problem on a joint but don't ever get a bottle of it going or you will be leaving the shop.

Re: Joint life ( how long will glue last)

#13

Doesn't sting for some reason...

John in NM

Maybe because only a little is needed, and it tends to be spread out on the cut and not generating a lot of concentrated heat. I've gotten accelerator in cuts (not trying to accelerate glue, just coincidental), and that does sting like crazy. Given how CA fumes burn the eyes, it is surprising that it doesn't seem to bother cuts.

The sheep tails might have worked if the bleeding had stopped before gluing. Also cuts that are producing lymph (clear liquid) don't seem to take well either. Then again, if there was moderate tension on the flaps, you'd probably have to re-glue every couple days.

Re: Joint life ( how long will glue last)

#14

Re: Doesn't sting for some reason...

FredD

For those who use CA and I have been known to close small cuts quickly should know that the edges of the cut are held together before the glue is applied ,so that no glue penetrates the cut, a spritz of accelerator will speed up the process and sometimes you even see and smell a little curl for smoke for a second or two. ;-))

As far as glue life is concerned , I have lost track of how many Picture frames and Wainscot panels I held together with this stuff in a Gel form, haven't any failures yet and we are going on thirty years for some the stuff.

My Brother for years was into heavy duty RC airplanes and boats and he used it by the case to beef up Balsa frames for 4 cycle V-eight model engines on the planes.

Somebody remarked that the secret to long life glue joints depended on how well the fit of the joint was and that is exactly right in my estimation.

FredD

Re: Joint life ( how long will glue last)

#15

Jason Roehl in Lafayette, IN

One of CSI's early mistakes.

Jason Roehl in Lafayette, IN

I've used SuperGlue to close small cuts for a number or years now. In one of the episodes of CSI, possibly even the first season, there was a clue that involved someone dripping SuperGlue because they tried to use it on a cut and it stung (or something to that effect). I knew immediately the exact opposite was true--there's no stinging or burning sensation.

What others have said is true, too--SuperGlue (and other CA glues) do not work well on "wet" cuts--you have to get the bleeding/oozing stopped first. Leaking bodily fluids just push the uncured CA glue out of the cut. Since CA glue is an anaerobic-curing glue (though there is solvent that evaporates also), you definitely need a way to keep the cut tightly closed until it cures--usually just holding it closed for 15-30 seconds is adequate. For that reason, it also works best on very clean cuts, such as from a razor. Ragged, torn cuts are more difficult to close with CA glues.

Jason

Re: Joint life ( how long will glue last)

#16

Re: question of how

Jim Dillon

Bill and others,

Interestingly, the 2010 Presidential Medal of Technology and Innovation was just given to the inventor of cyanoacrylate glue, Harry Coover, who was at Eastman Chemical at the time.

Anyway, CA is a generic name for a class of monomers that polymerize on contact with electron-rich chemical groups like amines, water, hydroxys etc. Depending on the type of CA, the polymerization can be very, very rapid, or much, much slower. Unfortunately, the faster the rate of polymerization, the more exothermic the reaction, and the faster the breakdown of the polymer. This rapid breakdown of some types of CA leads to an inflammatory response if used for tissue repair. This response is caused by the degradation byproducts of formaldehyde and a cyanoacrylate homologue based on the starting material. It is this response that led, wisely or unwisely, to the holdup on FDA approval for the tissue repair products.

However, CA can be synthesized in many flavors. Generally speaking, the longer the hydrocarbon chain (i.e. ethyl, butyl, octyl etc.), the slower the rate of degradation and the more likely the polymer will be stable over the "long" term.

The accelerators are simply dilute solutions of an electron-rich ingredient such as an amine that promotes rapid polymerization.

Regarding the use of CA in wood repair, I haven't really kept up on recent changes (if any) in CA formulations, but the chemical principles should remain the same and I would guess that water/moisture would be the Achilles heel of CA use in wood repair, but we could be talking years and years here depending on the type of cyanoacrylate used. I have no information on creep of CA glue, at least I don't recall anything, but all my textbooks are in storage. I would again guess that the longer alkyl CA would display better creep properties and may give better life expectancy.

I will be away from my computer for most of the day but will check back later to try to answer any questions.

Jim

Re: Joint life ( how long will glue last)

#17

not correct

Bill Tindall

There is no solvent in CA glue. The stuff in the bottle is pure monomer, plus some stabilizer and in the case of thickened glue a thickening polymer like styrene. Curing involves hooking all the small monomer molecules together to make very long polymer molecules and everything in the bottle become part of the solid....no evaporation involved. Water(nano amounts), and especially amines such as would be found on protein is what gets the reaction started.

Well, it may be that unscrupulous supplies dilute the stuff but that owuld not be the case with a reputable supplier.

While on the topic of CSI....If finger prints are exposed to CA fumes the stuff in finger prints initiates polymerization and in time a CA pattern of the finger print will develop and become visible.

Re: Joint life ( how long will glue last)

#18

doesn't creep

Bill Tindall

CA is hard and brittle when cured, not soft and creepy. Its chemistry and properties are similar to its close relative, Plexiglas. This property may be a problem in wood joints. They won't have any "give". CA is not very tough so any stress on the joint may make it fail. Complicated segmented assemblies may have lots of cross grain stress which could lead to failure. The moistrue issue is still under investigation.

Re: Joint life ( how long will glue last)

#19

Jason Roehl in Lafayette, IN

Oh.

Jason Roehl in Lafayette, IN

I just figured there was some sort of solvent evaporation based on observation. For one, the eyes can burn when using it, and two, a drop placed on a surface will appear to dry due to evaporation (albeit slowly), and will shrink in size. I wasn't fully versed on the chemistry of the product or it's reactions, just making an educated guess.

Jason

Re: Joint life ( how long will glue last)

#20

explanation.

Bill Tindall

The stuff in the bottle is a small volatile molecule. It has a large vapor pressure, like acetone, so it evaporates readily. It may indeed evaporate before it has a chance to polymerize and make a big non-volatile glue molecule.

Re: Joint life ( how long will glue last)

#21

Re: stay tuned.....

Jim Dillon

Bill,

Another thought I had was that the tannic acid content of the wood might influence the long-term creep/strength properties of the CA. The tannic acid content, combined with moisture/humidity etc., might catalyze the breakdown of the pendant ester groups - the weak link in the CA polymer. Degradation of that functional group would likely alter the properties of the CA, especially in a load-bearing (static or dynamic) situation. So, CA may give "acceptable" long-term performance with some woods, but poor performance with others. Personally, while CA has many useful applications in woodworking, I would never use the stuff for building quality furniture that I wanted to last a "long" time.

As an aside, I wonder if there is any open-literature on the long-term properties of CA in woodworking? I can't recall seeing its use in comparative testing of woodworking adhesives in woodworking mags, but then the CA suppliers likely don't recommend it for use in woodworking joints intended for long-term use, so it is not tested.

Jim