Official Luthiers Forum!

Yesterday I strung up a dreadnought that's the first of three concurrent builds to reach the finish line. Played it for a couple of hours last night, and was floored by balanced it sounds right out of the gate. I think I even dreamed about it last night in anticipation of getting the last few small steps done today and officially celebrating it's completion. As soon as I give it a good morning strum this morning I knew something was wrong. I looked down to see the bridge had lifted and separated from the top of the guitar. Crappity crap! After masking off the rest of the guitar, I was shocked how little effort it took to pop the bridge right off. Just a gentle heating with the heat gun and fifteen seconds later the bridge was off. I'm hoping some of you more experienced builders can help me identify what went wrong based on the photos.

Actually I think I know. Did I:

a) starve the joint when I heated the area with a heat gun after clamping down the bridge (got lots of squeezeout)
b) break down the hide glue with excessive heat, severely diminishing its holding power
c) not hold my mouth right
d) all of the above
e) something different all together

Thanks in advance for any insight you might have.

Ken

_________________
Mountain Song Guitars www.mountainsongguitars.com

I'm seeing what appears to be a thick layer of glue bonded to the bridge and nothing on the soundboard. There also appears to be a thin line of around the bridge perimeter that I suspect rested on the lacquer. It was, at least, a total failure so it was easy to remove!

I don't like to leave a lacquer ledge under the bridge edges (if that's what you did). I know that some people do this, that it doesn't have the potential for an ugly edge that scoring around the bridge does, and that it's easier. But unless you rabbet the edge of the bridge a la Collings, there will be a gap of several thousandths under the bridge. Hide glue isn't the best glue for filling ill-fitting joints. That could be at least part of the problem.

Another possibility is that the glue gelled before the bridge was clamped. That's also consistent with the thick layer of glue and the lack of bonding to the soundboard. Were the top and bridge warm when you glued and clamped?

This is all speculation but hopefully it's of some help.

I'll give you my procedure FWIW. Assuming you have very good mating surfaces. I apply a thin layer of HHG to the underside of the bridge, place in position with light finger pressure and leave for 20 seconds or so. Remove. There should be a thin layer of glue on the soundboard bridge position. Apply fresh glue and position again.
I hold it there for a few minutes. Done, apart from cleaning a little squeeze out.
You may not want to risk finger pressure alone but it's worked every time for moi.

Thanks much for your quick responses. I did warm both the bridge and top with the heat gun before applying the hide glue. I suspect you're right Todd that it gelled before getting it clamped sufficiently. However, based on the amount of squeezout I got from heating it after clamping, I would've thought it reliquified. You're right though, based on how clean the top is, it just didn't adhere. I knew I should've applied glue to both surfaces...

Rick, I actually did rout a .005" rabbet into the bridge, and it seemed to key it into the corresponding ledge in the finish real nicely. I thought I was being slick...
And yes, when I fail, I like to fail utterly!

_________________
Mountain Song Guitars www.mountainsongguitars.com

That's a great tip, Michael, on working the glue into both surfaces. My mentor James Condino is a big proponent of scrubbing the surfaces together with hand pressure. That's all he uses to join mando plates.

Flippo -- Great input . Thank you. It could very well be that the surfaces weren't sufficiently warmed up. I'll definitely be making a fresh batch before trying it again -- the stuff I used has been reheated several times, though I try to never let it get over 140* F. It also may be a bit thicker than it should be due to water evaporation.

_________________
Mountain Song Guitars www.mountainsongguitars.com

Well, of course I had an alignment issue, but all told, still less than a minute from the time the brush came out of the bottle to the first clamp going on. I used the dainty little stewmac mini cam clamps for the wings (seemed sufficient for this, maybe not), then two of the skinny bridge C-clamps SM sells for the center mass. I think in the past I've used a big cam clamp in the middle with the C's flanking it, then the mini's on the wings.

_________________
Mountain Song Guitars www.mountainsongguitars.com

I've been using hide for braces, linings and other small stuff for a couple years now, and still can't find the courage to try it on a bridge . This thread is another setback for me. But when I'll try it, Michael's hand method sounds best to me. You get some perfectly controlled and applied pressure real fast. Once it gels in a minute or so you can apply the clamps too, making sure they do not try to skew the bridge. I'd first put a cam clamp through the hole for the middle followed by the wings. When I glue the bridge I have props inside and for the wings I simply clamp from the outside, with the back protected by a 3.5mm spruce test soundboard.

_________________
Build log

Here's one more thought.... all of the other suggestions make sense.

Are those sawmarks on the underside of the bridge? If so, it's likely that the fit between the bridge and the top isn't close enough - apologies if I missed something on this point in the other posts. In addition to heating the bridge, making sure the top is warm, and moving fast enough so that the HHG doesn't gel before it's tightly in place, it's critical that the fit between bridge and top be very, very good; e.g. no gaps. HHG doesn't fill gaps at all well (um, well,that's not entirely true - it'll fill em, it just wont have any strength at all).

Hi Ken,

On my last build I ran into a problem that looked just like on your pictures. Glued the warm bridge on twice (with HHG). Bridge came loose when the first strings were starting to pull on the bridge.

The solution in my case: The rabbet I had dremeled into the bridge wasn't deep enough. Therefore the Bridge was sitting on the edge of the finish and the bridge couldn't make sufficient contact with the top. After making the rabbet deeper everything held as expected.

Terrific suggestions everyone. A few things -- while there appears to be a lot of glue on the bridge, it's actually a very thin layer. I sanded the top radius into the bridge, so it's a near perfect fit. There was clearly excellent contact with the top. I think I know one other culprit -- I forgot to scuff up the spruce after buffing out the finish. I'd masked off for the bridge, but I removed it for sanding and buffing out, so it was super smooth and probably absorbed the buffing compound, so it took on a "wax paper" effect. There was nothing to bite into. Rest assured I've taken care of that.

I have a fresh batch of glue in the pot, a clean bridge, and a textured glue surface on the top. My next attempt will definitely include many of your suggestions, and I feel confident I'll be successful. I'll post an update with pics when finished. Thanks again, keep the suggestions coming, too!

Ken

_________________
Mountain Song Guitars www.mountainsongguitars.com

Ouch, that was surely the culprit. One way to clean the spruce when having the lacquer on is to scrape with a very sharp chisel or something of this sort. "textured" is not needed, just "as fresh and clean as possible".

_________________
Build log

The contaminated surface was certianly the major factor in the failure. I'm actually a little surprised the joint even held string tension!

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

Very interesting post. I've been having a hell of a time with finishing around my bridges and hadn't yet heard of that ledge thing with lacquer going under. Will have to try that, newbie that I am. Just wondering for us tyros, why use the hot hide glue? Is it better for sound? Seems like all that heat would be problematic and could interfere with braces and bridge plate adhesion. LMI sells its white glue I believe with a statement lauding it for bridge use.

Thanks.

_________________
"Preoccupation with an effect gives it power and enhances the error"
from "Your Owner's Manual" by Burt Hotchkiss.

Well all I can add is that the problem is unlikely to be the batch of glue...it stuck to the bridge OK didn't it. If the glue had somehow been ruined by bacteria or some other issue, it would liquefy in the bottle even when cold and smell putrid. I agree that the problem was most likely to be surface contamination of the spruce because it looks so clean of 'any' glue whatsoever in the image.

Here is an experiment I did with old HHG in early November 2010.

I mixed this batch in January 2008. The same glue had been heated used and stored in a fridge through many cycles. It has even been purposely heated well over 140f on a few occasions for use in broken headstock repairs. It has been thinned, sometimes a little too much, and then had a few dry crystals added to thicken it back up again. 17 months after it was first mixed, I used this same batch to glue on a bridge with absolutely no problem at all. In November of last year, that is 33 month after this glue was first mixed, I recorded the following basic strength test.

The glue sample:



This next image shows inside the bottle as it appears straight out of the fridge. You can see in both images that the is still quite clear..You will have to take my word for it that it is also solid, and does not smell.



I took two lengths of radiata pine, old dry wood salvaged from a single bed that lay in pieces on a kerb side dump day. It is destined to be ripped into stickers for wood storage but was diverted for use in this experiment first. Each bit of pine was 700mm (28") long x 90mm (3 1/2") wide x 35mm (1 3/8") thick and each board was run through the thickness sander to be cleaned up with 80 grit abrasive ready for gluing.

One end of each board was then marked at 100mm (4") forming an indicator for the glue spread area to create a simple face to face glue joint. The 33 month old hide glue was then heated and applied liberally to just one of the marked surfaces. The joint was then bought together, slipped a little to even the spread, and clamped securely to cure.



This next image shows the squeeze out indicating that adequate glue was applied to the joint.



The joint was left clamped for app 8 hours and then the joined pine was placed vertically in a bench vice so that one piece sat low and was well supported by the jaws, and the other presented as a considerable lever with which to apply great force to the glue joint. This was done which resulted in the length of pine which was the lever then snapping cleanly above the glue joint.




I decided that the above result was inconclusive as it was possible to view the image and question whether or not the breakage had been caused by an unseen fault with in the wood. What I was looking for was clear visual evidence of 'fibre tear' of the wood at the glue joint itself. To complete my test I took a piece of hardwood (jarrah) and placing it at that point which offered the most substantial support being just above the unbroken piece of pine. I then took to the jarrah with a gimpy hammer wielded with unbridled vigorous aggression



As you can see I again failed to get any real 'fibre tear' of the glue joint itself. But none the less I think the results are quite conclusive, and I would not hesitate to use this same batch of glue to fit a bridge to a guitar again even though it is 33 month old.

Cheers

Kim

Thanks so much for doing some real testing, Kim! There's a lot of speculation about glues, but relatively little testing of assumptions. I still haven't seen anything but hearsay to support the position that the glue rapidly loses strength if heated over 145, even for extended periods of time. If anything, my own experience has been that it's pretty hard to ruin hide glue (ie: abuse it all you want and it'll still work)

The old guys used to just let it dry out in the pot and add some water to reliquefy it later. That's furniture builders around here. Stuff's still together 250 years later.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

One more thought here. I can understand the poor adhesion due to the buffing compound but I would think that 1.)if the bridge made good contact 2.) the glue didn't gel (you had good squeezout) leads me to 3.)wouldn't the glue on the bridge look very smooth?
I have to think that the bridge was not in good contact with the top.

_________________
Beautiful and unusual tone woods at a reasonable price.
http://www.rctonewoods.com/RCT_Store
The Zootman
1109 Military Rd.
Kenmore, NY 14217
(716) 874-1498

I did that one time, with the same result. After that, I vowed to never allow the buffing compound to come in contact with the bare wood gluing surface.

_________________
John

Were you a detective before you were a wood pimp? That's some seriously astute observation!

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

I had something similar happen to a fingerboard glue-up before. At that time I almost thought hide glue wasn't good for guitar building! (the fingerboard came off under hand pressure, actually it was the truss rod that separated it initially. When it separated it was almost as though the glue wasn't sticking to the wood at all!)

The problem I had was the wood (on the neck) was heavily oxidized and I did not scuff it out before gluing, furthermore the glue was thicker than it needs to be, so it wouldn't be able to flow at all. After that correction I never had hide glue not working again. In fact later on I had to un-glue something and I used a chisel to remove the joint (as I did not have any heating blankets to do any proper heat removals) and I had a lot of fiber tears as a result.

That guitar is in Afghanistan now, so I am sure any rough-looking repairs I did on it will probably blend well with whatever war scars it will likely receive.

_________________
Cat-gut strings are made from kitten guts, stretched out to near breaking point and then hardened with grue saliva. As a result these give a feeling of Pain and anguish whenever played, and often end up playing themselves backwards as part of satanic rituals.

Typhoon Guitars
http://www.typhoon-guitars.com

This is a very common problem and puts many off HHG at first. I guess because it is easy to fool yourself into believing that 'more is better', people who are new to HHG quite often make the mistake of mixing it 'far' too thick. In my experience 'thinner' is better and for general use, the glue when warmed and ready to use should have a viscosity somewhat like that of 'warmed' honey. Too thick is a disaster every time.

Cheers

Kim

Too funny Bob. I guess it's from over 40 years of building with wood.

_________________
Beautiful and unusual tone woods at a reasonable price.
http://www.rctonewoods.com/RCT_Store
The Zootman
1109 Military Rd.
Kenmore, NY 14217
(716) 874-1498

Ken, I hate to tell you this, but that beatiful balanced sound you heard was the interface that the buffing compound presented. Don't remove all the buffing compound.