Glue joints and pre-cleaning - created 10-22-2011

DaveTays - 10/22/2011.03:43:43

I am on my first complete build and i was wondering if it is possible for a guitar body to have a five piece wood centre and have a bolt on neck. curious if the glued piece would with stand the tension of the string over time.

The glue line is stronger than the grain line, so go for it.

The only thing you should consider is the closeness of the dry fit. Hold your joins up against a bright light and plane until there is no light penetrating the join. I always clean the sufaces to be joined with acetone before gluing. This is especially relevant with oily timbers.

One more thing, use titebond original or hide glue. You can use titebond in a laminated body because the joints won't be under tension (pulling pressure). However, if you want to eliminate glue creep on glue joints under tension (flat top bridge), use hot hide glue. A hide glue joint eliminates the possibility of glue creep, giving you a join that is "brittle" rather than "plastic".

Wiping with solvents does not increase the strength of glue joints.

I always clean the sufaces to be joined with acetone before gluing.

This comes up here every so often, and our experience and testing show that cleaning the surfaces with any solvent can actually weaken the joint. It at the least has been shown not to help- all you need is a clean and freshly scraped surface. It is important to have a surface that is fresh, planed or scraped is the best.

According to my experience Titebond original does not creep. I have also successfully used body blanks laminated from several pieces. I see no problem here.

Titebond Original does indeed creep, but you're less likely to see it in joints of the type Dave is describing.

I beg to differ on Acetone weakening a gluejoint. The manufacturer of titebond says this about tannic acids and oily woods,

(from the Titebond website) "Get better results gluing woods that are oily or high in tannic acid. When working with woods that are high in tannic acid or are considered oily, wiping the joints with acetone before gluing them up ensures a good bond. Acetone clears the contaminants from the wood's pores on the bonding surface and dries quickly without leaving any residue."

We have, collectively, found that to be untrue, with many more joint failures reported after such treatment than after a plain glue-up.

Hmmmm, so who do the plebs like me believe? Has Titebond been told of the MIMF findings? Is there science to back up or refute each argument?

I don't have anything to back up my assertions except the manufacturers recommendation, which I have been faithfully following for years, and none of my instruments have come back with failed glue joins. Poor fool me?????

Alan Carruth is our resident expert on the need for fresh, clean joints. Yes, there is science to support it. Franklin is simply reporting the "conventional wisdom," but conventional wisdom is not always correct, as we on this forum have found by our collective experience several times. Otherwise we would not have spearheaded the resurgence of hot hide glue and French polish in amateur instrument building, both of which "conventional wisdom" said were inferior to more modern products.

Fine Woodworking did a test as well. Not scientific perhaps, but still well done enough to provide consistent results indicating that solvent wiped joints are not as strong as a freshly scraped, planed or lightly sanded joint.

I did a google search, and could not specifically find "general" recommendations.

The forest products publications have a on line book " Handbook of wood chemistry and wood composites" from 2005.

The only comment about washing glue line was about it is beneficial in dealing with Teak. No other recommendations for or against.

I don't wash East India Rosewood. This is the main wood I deal with which has a lot of oil.

"both of which "conventional wisdom" said were inferior to more modern products."

The marketing gurus have a lot to answer for. Maybe Titebond have a solvent manufacturing arm.

Allow me to cast a vote for urea-formaldehyde glue for lamination. It's a dry powder with infinite shelf life, mixes with water to the amount and consistency you want, dries hard, does not creep, and soaks into the fibers for a very strong bond. It's waterproof and stable at temperatures in excess of 200 degrees, so it's not coming apart.

DAP Weldwood Plastic Glue is the easiest to find.

Bill, is that a good glue to use where there may be some gaps? How does it rank on the scale between epoxy which is very good at it and hide glue which is the worst at gap filling.

It's really not intended for gap filling. It behaves more like hide glue, but with a long open time and lots of "slip" available if you need it. I'd choose a filled epoxy for rough, gappy surfaces.

Or make them non-gappy. :)

If you assemble something with UF glue, unlike hide glue, it's not coming apart.

UF glue also has the problem of off-gassing formaldehyde, a poisonous gas. I use my exhast fan during the cure stage.

Franklin has a good tech support line where you can discuss with one of their specialist on the phone if someone here wants to take the solvent wash recommendation up with them directly. I spoke to Joe Goodwin, a fine woodworker himself, a couple of years ago about the polyurethane glues and he was quite well informed and seemed interested in my feedback.

(800) 877-4583

I've done some relatively sciencey tests with tiny dots of oil and acetone on maple (density similar to rosewood) to see the exact effects of a solvent wipe. My finding was that it's effectively impossible to clean any porous surface, so you really are just spreading it around and gobbling up the clean areas the glue could have made a solid bond to.

The real science bit is that there is no surface prep superior to a clean and freshly scraped or abraded surface.You want to have some nice cleaved chemical bonds ready without giving them time to react with the air, and you don't want any contamination (dust is a HUGE one) competing with the surface for the glue's attention. I only consider a surface dust free if it's been thoroughly cleaned with compressed air from close range. The term to look up is 'surface energy'- that leads to all the research on this stuff, and it'll provide you with a lifetime of reading material if you're interested.

I'm about to be granted a masters of applied science for my work in this area, so I can finally claim some sort of credentials besides having made a bunch of stuff :)

That's a heavy bit of science you've got going on there.

My guess is that tightbond recommends an acetone wipe because the most important factor is a clean joint. Most users (particularly most who are going to read the instructions) are not as familiar with fine woodworking as many here are. Certainly most will not be making joints which will be asked to keep an airplane together. Wiping a joint with a solvent is a good way to ensure that Joe Public is getting the crud and dust off of joints that are not freshly prepared. I doubt the average user is gluing oily tropicals; the directions seem to indicate the wipe for all woods. When this pops up here, it is (almost) always in the context of oily woods. I think tightbond has much more to gain by recommending the wipe than it stands to lose. Over at the OLF there is a poster who works for Franklin (I think in product development) I bet he could weigh in on this issue.

Bryan, I think you are exactly right. Another indicator of this is that Franklin recommends a clamping pressure of 200 to 300 psi for hardwoods. This is way overkill for well prepared and well fitting joints, but it is proably appropriate for some klutz that is clamping a couple of warped two by fours together on his garage floor. As with finish manufacturers recommendations, they can often be dispensed with entirely if you have sufficient experience in the field.

Yes, we luthiers surely stand above other lowly glue users in our understanding and execution of joints... (excuse me while I remove my tongue from my cheek... there, that's better!), but I think Franklin would do well to update their directions in light of available knowledge and instruct that klutz to clean up his warped 2X4s with some kind of blade or at least sandpaper rather than rub them with some solvent before crushing them with big ol' C-clamps.

Tell me what group of users would have a better understanding of glue and execution of joints than luthiers.

I surely can't speculate about what groups execute better glue joints, but it can't be denied that there are instrument builders who jump ahead to the "art" part of the craft without first getting a thorough background in wood working, including joinery, and there are joiners in other fields who know more about joints and glues that I probably ever will.

I'm no even considering myself a luthier, let alone implying that luthiers are somehow above other woodworkers. I hope I did not come off sounding like that. My point is that many people decide to glue something together, and grab a bottle of glue off the hardware store shelf then look at the recommendations on the bottle. People who have been taught or who have taken it upon themselves to learn a bit about joinery, probably have a good enough handle on glue and its use related to joint preparation. My larger point was that extra insurance could be built into the instructions to ensure that most people can get decent results. I think the finishing product example is a good one. If I were going to try a new product (say waterborne) I would seek the advice of people I trust on the various fora rather than just read the can and go for it. In that case, both my resources of information and interest in getting good results are robust. A few weeks ago, I wanted to quickly and easily strip some paint/finish off of something that was of little importance, in that case I went to Home Depot and read the cans. . .

From the Franklin Titebond help page:

"Get better results gluing woods that are oily or high in tannic acid.

When working with woods that are high in tannic acid or are considered oily, wiping the joints with acetone before gluing them up ensures a good bond. Acetone clears the contaminants from the wood's pores on the bonding surface and dries quickly without leaving any residue.

A good bonding surface can also be achieved by sanding or planing the wood just before gluing the joints."

At least they mention the fresh surface prep option...

John makes a lot of good and common sense. .....and BTW, the inference I take from this is that I may be a "Klutz" in even suggesting to use the manufacturers recommendation. I have been a woodworker for 45 years, and yes, qualified, schooled,skilled and passionate. But it doesn't make me a chemist and in fact, why would I even question a manufacturer's rec. To have the audacity to question another expert in another field of endeavor assumes that "luthiers" are expert in all fields when nothing could be further from the truth.

And John, I am totally with you on the point you make about "Luthiers" jumping to the "art" of lutherie before gaining a sound knowledge of woodworking. A case in point is that some "luthiers" in my area of violinmaking are using "sharkskin" to smooth their instruments because Stradivari et al used it. That was 300 years ago and technology moves on. Stradivari used sharkskin because sandpaper was not invented yet. Until the manufacturer of sandpaper comes clean and says sharkskin is better, I will wipe my glue joins with acetone. Sorry, stubborn is as stubborn does. Until I see the science instead of innuendo, rumor, supposition and sharkskin soothsayers, I am unmoved by the righteous indignation.

It seems to me that just using common sense of how any solvent works argues against wiping an oily wood with a solvent to clean it. If the solvent is able to dissolve the surface oil to "clean" the surface, it will also dissolve oils below the surface because it is going to penetrate into the wood to some degree. As the solvent that has penetrated the wood evaporates, it is going to draw the deeper disolved oils in the wood to the surface with it. As the solvent then evaporates at the surface, it leaves these oils behind on the surface because they are not volatile, and you are left with oil on the surface you are gluing.

I've had nothing but lousy bonding with Cocobolo and BRW trying to clean with a solvent, and have no failures (hide glue) as long as the gluing surface was lightly scraped with a sharp scraper just before applying the glue. Granted I'm just an amateur builder, but many stupid requirement classes of inorganic and organic chemistry in college at least left me with some grasp of how organic molecules and solvents want to behave.

I'm 100% with Bob and "surface Energy". An easy way to visualize it is to pretend the wood surface starts "rusting" from the oxygen in the air as soon as that surface is exposed to the air, and the longer you let the oxygen rust it, the more you are just adhering your glue to flaky rust instead of good solid wood. A light pass with a really sharp scraper removes this rust (and surface oils for that matter) so you are gluing to wood instead of rust and oils.

I don't see it as arrogance on our part to assume that the directions the manufacturers give are aimed at the lowest common denominator of those that might use their product. But just because they say to do this does not make it really the right thing to be doing. I doubt very much that Franklin was picturing a gluing surface 9 or 10 one hundredths of an inch wide under continuous loads over 100 pounds all the while continuously flexing the joint while they were writing up their directions.

Almost forgot, way to go Bob! Boy, with that kind of backround, I think a bunch of us would love to have you expand and expound on your field of study. Any chance of you offering a class here (or at least a tutorial thread) on glues and gluing surfaces?

Barry D....So when all else fails, throw the instructions out? Granted that experience is invaluable, but experience without education (instruction... or whatever you like to call it) simply perpetuates doing things the way you think is the correct way, when it fact it may well be the wrong way .......and doing it so beautifully for so long...and so wrong!

Franklin might be right. Wiping the surfaces of oily woods to clear contaminants might be a good thing if you have no other surface preparation. Some woodworkers will glue up boards as long as the edges can be squeezed together with clamps.

Has anyone done research on how preping the surfaces with detergents to remove oils would work out?

Barry G., I in no way implied that YOU are a klutz, but if you want to take that mantle on then have at it.

Randy: My own mental picture of surface energy is sort of the same in reverse. I see the freshly scraped surface as a bunch of chopped up chemical bonds trying to grab onto something, things in the air if there's nothing else around for long enough, and the old surface as a smooth bit with nothing for the adhesive to react with.

We've known that mechanical bonding hasn't been a contributing factor compared to chemical for many years, ever since the first time someone isolated the mechanical bit by scratching up a piece of Teflon and trying to glue it to something. That said, there's definitely a strong level of irony in the fact that at the molecular level our goal is exactly that- a 'rough' surface.

I don't know about a tutorial or that sort of thing, but I have been planning to write an article for awhile about this sort of stuff. Since I'll be writing all about it for my thesis anyways, I might as well make it a bit more 'pop' so that it'll work for a general audience. I'll make sure to post a thread or something when it's done.

Clay:

The idea of using a detergent is an interesting one, since it's pretty much their job to bond to oils and greases. For anyone listening in, a detergent basically bonds to things that don't dissolve in water and turns them into things that do, and all of a sudden washing dishes or taking a shower gets a lot easier :).

The question, of course, is whether it's possible to get all of the detergent out of / off of the surface afterwards. Intuitively, I'd assume glue really doesn't like soap, though maybe the non-water-based glues can 'ignore' it once it's dried up in a way they can't with oils.

Barry D. .....A Klutz I might be for pursuing this nebulous thread. So far, nobody has referred to a definitive or conclusive paper that refutes the solvent wash question. Lots of blah but no substance. Is their a chemist out there who can put me out of my misery?

I've seen links to, and read papers on glues and adhesives and bonding to variously prepared wood surfaces. There is scientific evidence, but I don't have it at my fingertips. It's there if you search diligently.

Personally, I find it hard to doubt the opinion and/or motivation of someone who is nearing completion of a thesis on the subject and freely offers that knowledge.

Bob,

I was just using the rust thing as something that might make sense to a lot of people.

I know you mean at the molecular level, but your comment of the goal being a "rough' surface left me uncomfortable about someone thinking a truly rough wood surface is a good thing - as in the old advice to scratch up the bottom of the bridge before gluing, which at least to me seemed patently absurd.

If you do an article I'd love to get a copy when you do. (Heck I'd even like to take a crack at trying to wade through your thesis)

One thing I've wondered a lot about is what sort of timespan is involved for the oxidation of a significant number of available bonds. You run into anything shedding light on that?

A rather crude experiment might shed some light on the time period:

Get a spritz bottle of distilled or deionized water, plane the edge of a board, and spritz water on a short length of the freshly planed edge. The water should flow out evenly on the surface. Wait 10 minutes and do the next section, repeat until the water starts to bead up indicating that the surface is no longer a good gluing surface. Whatever time it took for the water to just start to bead up would be somewhere near the maximum time for a good glue joint. I bet it would be much longer than the traditional wisdom of 15 minutes for oily woods, but sooner glued is surely better regardless.

John,

Thank you. How elegantly simple.

One example I keep going back to on the timing of surface energy is an apple. It is a plant material and so may have some relevance. A cut apple starts to turn brown from oxidation within 10 minutes or so. Oxidation is basically oxygen making molecular bonds to the surface of the apple. Of course, people don't glue apples, but I think this still applies.

The apple (fruit) is not a good example. They have an enzyme which is triggered with injury. If you disable the enzyme by putting a little acid on the surface, it will not brown nearly as quickly.

This is probably unrelated to the oxidation that occurs on the surface of wood.

Oxidation is oxidation. The presence of an enzyme in apples and blockage when acid is added doesn't change the fact that the exposed surfaces adsorbs oxygen on both apples and wood. And my point is that they likely happen at about the same rate. Perhaps Bob investigated the speed of this during his research.

The Forest Product Lab did the timing experiment all the way back during WW II. At that time a lot of aircraft still used wood structures (the control surfaces on the P-47 were wood and fabric at first, for example), and sometimes they had problems with glue bonds failing (which is a drag in combat!). The FPL found that wood should be glued within 15 minutes of being worked for the best strength.

BTW, the water spray test works because water is a polar molecule that's attracted to those open bond sites, in case anybody was wondering. It's possible to determine the surface energy level exactly by measuring the angle where the edge of a water drop meets the wood, so I'm told.

I've brought up the 'surface energy' thing several times on newsgroups and such, and frequently get people saying that it's bogus. The usual line is to the effect that "I haven't bothered with that for the X years i've been building, and never had a glue line fail". Good for you. It reminds me of an old pilot's saying: "There are two kinds of pilots; the ones who have landed with the gear up, and the ones who have not done so yet". Just you wait. Personally, I think a lot of the resistance has to do with the name: 'surface energy' sounds so 'new age'.

Alan Carruth / Luthier

I almost hate to bump this one back up, but I just heard from Hugh Evans at Franklin. With his permission, I am pasting his response:

"Wow, sorry for my late response... I clearly haven't been around much lately. For future reference, if you have an urgent adhesives question 1-800-347-4583 will ring directly to the Franklin International (Titebond) Tech Service Line. I have several colleagues who are very capable of answering questions, and if the conversation turns technical enough about guitars you'll probably end up talking to me sooner rather than later.

Surface preparation is important for oily woods (ebony, rosewood, cocobolo, etc.) and can be accomplished both mechanically and chemically. Some wood species accumulate surface oils over time, and can impair bond formation with water based adhesives. Essentially, the oils repel water and prevent interaction with cellulose at the wood interface. Freshly machined pieces (generally within a few hours) can be glued up successfully without further treatment. Wiping the surface with acetone will never harm bond strength, and has only proved advantageous in our tests regardless of wood species. As a general rule, oily woods require additional clamping time of up to 24 hours only because the drying process can be slowed.

The question of solvents that can be used for this purpose comes up from time to time... and there is a reason we recommend acetone over others: it flashes off quickly and doesn't leave much behind. At the same time it is relatively aggressive towards oils. If anyone wishes to see this I can wipe a piece of cocobolo and post a picture of the orange oils that result.

When in doubt: wipe it with acetone and a clean rag."

I asked him about the tests they did and his response was:

Both formal and non-formal tests are used. Non-formal methods can be as simple as assembling rub joints and breaking them with a chisel and hammer. Formal methods are for the most part derived from ASTM specifications, most commonly the D905 shear strength test when working with wood glues. ASTM methods are copyrighted, so I can't share any of them directly. That doesn't prevent me from explaining them in depth if necessary. In most cases lack of surface preparation leading to failure will create an adhesive failure situation, where it literally fails to bond and minimal wood failure is observed. If I can track down any test data on this I will share it.

Feel free to re-post, most of the time it happens without my knowledge anyways. A PM'd link to the thread would also be appreciated.

There are a lot of myths and commonly held misconceptions in the world of adhesives, and few resources for answers. I enjoy public speaking/Q&A sessions, and could probably pull them off with some regional groups of luthiers... Which could be a *lot* of fun. In the interest of disseminating information I wonder if publishing monthly (at least) in depth articles over topics in adhesives as they relate to woodworking might be a good solution? Would anyone be interested in something along those lines?

"Wiping the surface with acetone will never harm bond strength, and has only proved advantageous in our tests regardless of wood species"

Thanks Bryan. It's good to know that what I have been doing for years is supported by experts.

I haven't done a real time test on wood yet to see exactly how quickly the surface energy degrades, but I can say that on aluminum it's on the order of a few minutes.

I do finally have a setup where I can do proper shear tests now, just finished it up last weekend, so I might put it to use on this when I get some spare time. Terrible waste of cocobolo, but I can make an awful lot of test samples out of half a board foot. Based on the results of my contamination spread tests, I'm still dubious about the wipe, but research is a lot more interesting if it's surprising.

Al: That's more or less the case. Often more than one liquid is used and the results are interpolated. It can be hard to get a really accurate measurement of surface contact angle, it can change over time between the same liquid and solid. In the best research I've read the 'receding' contact angle, when the liquid pool is shrinking, has a strong correlation to adhesive strength.

Alan Carruth stated:

>The Forest Product Lab did the timing experiment all the way back during WW II. At that time a lot of aircraft still used wood structures (the control surfaces on the P-47 were wood and fabric at first, for example), and sometimes they had problems with glue bonds failing (which is a drag in combat!)

IMHO, such failure would be a drag in any situation where safe landing is a desired outcome! ;-)