Official Luthiers Forum!

I've just had two PU glue failures in two days, luckily I caught them before proceeding with the next step....more on that later.

I'm gluing a maple ply pinblock to the rim for a couple of kiddie size banduras. The rim is maple compwood that's been thoroughly dried and as I mentioned, the pin block itself is an all hard maple plywood. I routed the curve for the pinblock using a template that was made directly from the bending form (I use an inside mold for bending).

Two days ago, I used some older "clear setting" gorilla glue. After a day of curing, I gave the frame a little push and the pinblock cracked right off - on both instruments with nearly 100% glue line failure. I went out and bought new glue, this time the full strength GG (it's all they have at the store near my house) and did the same thing - same result with the exception of one little spot of wood failure. Before re-gluing, I scraped both the rim and the pinblock until I was getting wood shavings.

So, what am I doing wrong? Unfortunately, I've got 4 other instruments which use PU glue to fix the pinblock and now I'm a little worried. Am I expecting too much for an "end grain" plywood glue up?

I'm also thinking that perhaps my joint isn't tight enough for PU glue and epoxy might be a better alternative.

I've attached a photo of the clamp up and the failed joint that shows what little wood failure I got.

Thoughts?

p.s. - this is a 100% never ever want it to fail ever even in a hot car type of joint. I've used PU glue successfully on face grain (non-plywood) in high stress areas with very good results in the past and even on end grain to end grain joints in non-stress areas. Heck, none of the prior 3 instruments are failing in that joint as far as I know either.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Did you moisten at least one surface with water before you applied the glue? (I always wet both surfaces)
I haven't used PU glue much, & not for instrument work at all, but have found it to give a very strong joint. I would have thought the old glue could be a problem, but fresh glue would have worked perfectly for you. Especially with so much surface area.
Bummer!
Time to reach for the epoxy, I guess...

How about titebond?
I've had really good luck with that glue.

I think it would be good to not think in terms of glues that react favorably or disfavorably to water as can some urethanes....as long as there are options. You make this sould like a classic glue failure....not a bond failure. With epoxy, you mix it and it's going to crosslink and cure....no way to screw it up if you carefully weigh and mix the components....and amine based epoxy catalyst is very aggressive on gluing surfaces. Though it is important to accurately process epoxy, when you do control it......it's predictable. I'm using epoxy to glue necks to bodies and fretboards and headstocks to necks. I don't like surprises.

But....there is a urea glue that works consistently for me....Pro-Glue. I use it for laminating body panels and neck laminates.

http://www.pro-glue.com/product_order.html

You mix it 100(powder):50 (water) to activate it. It has a nice long open time (about 15 minutes spread thin) because it was designed as veneer glue. I think the reason it seems so consistent is that the powder stays completely dry in the container until you are ready to use it. Not sure how good of a structural glue this would be....but that would be easy to test.

I hate mysteries like this. It would drive me nuts to think there are possible bond failures out there. Good luck figuring it out.

_________________
I read Emerson on the can. A foolish consistency is the hobgoblin of little minds...true...but a consistent reading of Emerson has its uses nevertheless.

StuMusic

I did not. I've had a problem in the past with too much water (glued together wet laminates) failing miserably as well. Judging from the amount of foaming from the joint, there was enough moisture. RH in my shop is right around 50% right now. I suppose it's a possibility though.

I'm thinking I'm going to try epoxy. I've got some West epoxy that comes in pre-measured packets that are pretty hard to screw up. I can afford to lose a bit of thickness on my pin block as well so I'll re-attach the template and rout off a 1/32nd to get a nice fresh surface on the pinblock as well. I did scrape before the second glue up but scraping itself can cause "waves". I'm hoping that the gap filling ability of the epoxy will make up for that. Honestly though, the biggest gap I could envision couldn't be more than .002" and even that's probably a stretch.

Would still be nice to know what's going on and if I should lie awake at night worrying about the other 4 instruments with this joint.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

I use titebond on everything, except my inlays( epoxy there). And have never had a problem. It's also an easy glue to get locally. Any lowes will have it.

I use titebond on a lot of other joints as well. I guess I just get nervous because of where the joint is. I have some scraps that I tested with the PU glue (which also failed) - I think I'll run a test with that as well. I intentionally didn't make the mating surfaces on the scrap test perfectly flat to test a non-perfect joint which I think may be the case with a curved surface like this.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

I don't use any Titebond!

_________________
Waddy

Photobucket Build Album Library

Sound Clips of most of my guitars

PU glue has absolutely no gap filling strength. Unless the joint is very tight it is not good.

And then there's the squeeze out.

Actually, the squeeze out tells you just how bad it is at gap filling: run your fingernail into some of that foam
once it's dry.

That's how strong your joint is.

50% RH should be good.

Some magazine did a test several years back, and found that about half the urethane glues bought new at a hardware store had already spoiled. Iirc, GG was among the worst. Any moisture in the bottle and the glue is doomed.

_________________
Howard Klepper
http://www.klepperguitars.com

When all else fails, clean the shop.

I'm still curious as to the cause of this failure... The polyurethane clearly reacted and you have sufficient clamping. If it failed to kick over at least we'd have a finger to point.

Titebond Original has superior bond strength compared to Polyurethane liquids with a notable exception being endgrain. While it is true that Epoxy and UF have superior thermal and creep resistance, we're not looking at a true structural load. In terms of convenience both are more of a hassel to mix and use. They are generally fool-resistant, however. Epoxies have wonderful gap filling ability and their chemistry allows handling characteristics and mechanical properties to be adjusted over a remarkable spectrum. The biggest drawbacks are ease of use and disassembly for repair.

Good luck.

Fine Woodworking issue #192 (sometime in 2007) tested glues and polyurethane glue,( specifically Gorilla was used in the test) was the poorest performer when compared to hide glue, PVA, and epoxy. That's all it took for me to stop using it for headstock scarf joints. PVA was rated the best value and in the case of waterproof PVA..the best adhesion (but...not as reversable as normal PVA (such as original yellow Titebond) which many of us use.

Andy
Whenever I have a joint that isn't very tight fitting (ie close to perfect), or that involves much end grain, I consider epoxy as the #1 choice.
Use brand-name (WEST or SystemThree) epoxy, coat both surfaces with straight epoxy and then apply epoxy with filler. Don't clamp too tightly- epoxy will fill gaps. If you have an 'inside corner' as part of the joint, make an epoxy fillet with the squeezeout.
Though there are differing opinions on this, I rough up the gluing surfaces a bit with sandpaper when using epoxy, especially if any of the wood has been put through the planer or jointer. If the machine blades aren't really sharp,the wood can get a bit 'glazed' looking. (Same with router-shaped surfaces...)

I use PU from time to time, but less-and-less nowadays. It's expensive and messy, and I don't have a lot of faith in the bond strength any more.

Cheers
John

John,

I'm totally with you on the faith in PU glue. I found an old discard piece of laminate that I glued many years ago and forced a chisel down the joint - it was fine. However, destructive testing of every glue joint is certainly not what I'm interested in.

Sigh....When will I learn: I tried epoxy two days ago - it was loctite extended open time. Don't have filler and I've also read in spots that it's not really necessary for tight-ish joints. Coated both sides with a thick-ish film. Clamped lightly, got a little squeeze out.

The next morning it pried apart even more easily than the PU

Now I've got the added hassle of removing the old epoxy before my next attempt. The worst part is that I have an un-opened syringe of System 3 .

So now, do I re-try epoxy or do I try titebond with a form fitting caul and mega-clamp pressure?

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Where did it fail? At one surface or within the epoxy itself? (Very strange....). BTW, I generally don't use any epoxy that comes in a tube for woodworking, so I don't know about the Loctite product.

I'd say that you need to get things cleaned up really well before proceeding any further. Definitely, Titebond (PVA) doesn't stick well to old glue of any kind, as far as I know.. Good quality epoxy will bond to well-sanded and clean epoxy - it can get complicated with amine blush if present.

Make sure you have good, clean wood surfaces before proceeding. Coarse grit drum sander can be handy cleaning up the inside of curves.

Cheers
John

I've got the same question as John: where is the joint failure? Is the epoxy sticking to the "real wood" or to the "end grain" plywood? If the glue remained mainly on the real wood I suspect that it might be that the "end grain" plywood is impregnated with some stuff that inhibits any gluing. I would go on by making a new pin block out of another (completely new, fresh) piece of plywood.

The failure was 100% epoxy. No wood failure what so ever.

I think it was bad epoxy in this case. I pulled the puck out of the cup today and put it in a vice - was still kind of sticky inside. Cracked with no problem what so ever. Seemed very soft when chiseling the slop off the top of the rim. Everything is all sticky and oily like too (blush?).

I did a test piece tonight using system 3 from a syringe. Only rub is that I can't weight it out of a syringe (not that I have an accurate enough scale for such small quantities).

I do have a few packets of pre measured West systems epoxy that would probably be pretty hard to screw up.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Andy-
It's quite handy to have epoxy on hand, and it doesn't 'go bad', so my advice would be to budget for some WEST and pumps rather than messing around with hardware store stuff. Probably cheaper in the long run as well......and easier on the stress level!
There's a lot of good user info on the WEST/Gougeon site. Jamestown Distributors or local marine place would be 2 sources for epoxy.


Cheers
John

I think you can mix both System3 and WEST by volume and get good results. Check the websites.

Fine Woodworking test on six different glues.

http://www.titebond.com/Download/pdf/HowStrongisYourGlue_FWW.pdf

_________________
.

Audiences and dispensations on Thursdays ~ by appointment only.


.

So here's some pics from the glue test I ran last night. Test subjects are:

PU glue with a moistened surface.

Titebond.

System 3 from a syringe.

Surfaces are freshly planed (power planer) hardwood and freshly jointed plywood on a router, using the same bit I used to shape the pinblock.

In these pics, I only have the PU and Titebond because the System 3 was still a little soft this morning. It had about 15 hours to cure at around 60 degrees so I'm going to give it another day and moved the piece upstairs where it's warmer.

To test, I put the plywood in a vice and pulled up on the solid wood.

PU glue held much better than before and I got a lot of wood failure.

Titebond did even better: the solid wood broke before the glue joint failed. I then used a crows foot to break the joint. In both cases, the glue held pretty good on the face grain and not at all on the end grain. Depending on how the epoxy works out, I may end up using titebond.

To be continued...

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

I'm glad to see you tested different glues Andy! Now you can see why i have only use titebond for the last 30 some years, with no failures.

It's new to me that epoxies are good for creep resistance. Not in the same class as a UF glue at all in my experience.

There are so many epoxies that just saying 'epoxy' leaves a lot to chance.

Andy, I happen to like that long set Loctite. I discovered it by accident when the local big box didn't have my old brand. I've only used it a few times, but no problems.

One other thing, while we are talking glue. I get that epoxy has material strength that makes it superior for loose fitting joints. I don't get why there should be any advantage to purposely having a loose fitting epoxy joint. Yet the packages and various people on the net keep saying to leave surfaces rough and not to clamp hard. I could see the point of rough surfaces if the glue has material strength but poor chemical adhesion, however I didn't think there was a problem with epoxy's adhesion to wood (with the possible exception of very oily woods). I can't see any advantage to not clamping tightly in any case. If the stuff isn't going to adhere well, how could it be beneficial to have a looser fitting joint?

In other words, if it has decent adhesion and good material strength, it may not be much or at all weaker when loose fitting. But how could it be better?

_________________
Howard Klepper
http://www.klepperguitars.com

When all else fails, clean the shop.

The first good user info on epoxy that I found was from WEST/Gougeon. Most of their early experience was in boatbuilding, and a lot of that was in using epoxy to laminate sawn cedar veneers and similar types of bonding. Thus there was a lot of emphasis in their guides on avoiding 'glue starvation' in the joints. Cedar and other softwoods really soak up the epoxy resin. If you have thickening additives like milled cotton fibers added to the mix, the resin can be pulled out of the joint leaving behind a weakened matrix of thickener and resin...that's the theory anyway.
Thus the emphasis on avoiding over-clamping.
If the joint isn't well-fitted dry, it's best to let the epoxy fill the gaps rather than 'squeeze' the joint together with clamping, I think. With a well-fitted joint, it's immaterial (and you probably don't need thickeners with the resin/hardener mix, either..)

And, of course, some tests always help to clear things up. (though the FWW test was not very 'scientific).

Cheers
John

Thanks for the link the Padma.

Hmmm.....
Both types of PVA glue and epoxy were rated higher than hide glue...and the liquid hide glue rated higher than HHG