Official Luthiers Forum!

Meanwhile, on bridge stresses, here's what I originally wrote:


...which has been misrepresented as


Some ( ) bat for both sides...

It's clear from the posts currently coming in that there is plenty of room for discussion about the stresses on bridges. A common view with people who have looked at this closely is that the stress (load per unit area) concentrates on the back and front edge of the bridge. That is normally how stress distributes in these sorts of loading cases and has been well know (at least by engineers) for years. The late David Malicky produced some good finite element modeling of a bridge under load, including this stress diagram...



The areas with high tensile stress are the "hot" spots, concentrated at the back edge of the bridge. The difficultly is in determining how much the stress is concentrated. In theory, it is easy to find stresses that exceed the tensile stress limit for the materials. In practice, plastic deformation occurs, dissipating the stress and keeping it below catastrophic failure levels. This plastic deformation is important as it prevents cracks from forming and avoiding cracks is ultimately what keeps a bridge on. It can inform glue choice etc..

The degree of stress concentration depends on many things which I won't go into in detail, but includes type of material and the particular geometry. Many will be aware of stress concentrators like cracks and notches. To avoid one of the most common failure modes, a peel failure from the back of the bridge (the top, being flexible, peels off the more rigid bridge, similar to how sticky tape can be peeled off a bench) the design has to accommodate a transition in stiffness from the flexible top to the much more rigid bridge. A smooth transition tends to reduce the level of stress concentration. This can be achieved in a steel string guitar by transitioning from just the top behind the bridge, to the top and bridge plate and then the top, bridge plate and bridge. Tapering and curving the edges of the bridge plate and bridge further assists in obtaining a smooth transition in stiffness from the top to the bulk of the bridge.

On the other hand, a classical guitar transitions from a thin, flexible top directly to a bridge which traditionally has an 8 or so millimeter tie block. Pretty much an instant 8mm step with the resulting step transition in stiffness and a correspondingly high concentration in the local stress which can deliver a higher local stress (and corresponding propensity to fail) than a steel string bridge where the average stress may be higher. But, generally, it is not the average stress that causes failure.

As the actual stress is dependent on a lot of local geometrical and material properties, it is case case dependent, but it is readily arguable that the peak stress on a classical bridge can be higher than on a steel string bridge.

The lesson is to just do whatever you can to more smoothly transition the bending stiffness from top to bridge and that will keep the peak stress down.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

An interesting discussion, although altogether too heated at times.

_________________
A constellation only takes shape when one maps the whole.
- Beth Brower

Whoops! Sincere apologies Woodie. I'll try to correct that. Thanks for pointing that out Terrence.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

Done, Woodie! I hit the bottom quote button instead of the top one. Apologies again and I agree with you sentiments entirely.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

Thanks for the post, Woodie.

Last time I was involved in (a very constructive, BTW) discussion on bridge loading it ran to many, many pages. Cutting a very long story short, it's not a problem that is statically determinate without a lot of simplifying assumptions. Essentially, where we got to was the late David Malicky's diagram of my earlier post with the associated learnings. What matters are the stress concentrators which are difficult to put a generalized number on as they are so case dependent. So best just to work to keep them low.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

Hi Jeff,
My "divergence of opinion" was never about the tensile forces on the various bridge types, but rather on the statement:

"that includes all the classical bridges which are arguably under even more stress" .

As I saw it "stress" was used in an all inclusive way - to include all types of stress.

Woody, I do not want to be in any way condescending to you but explaining that to you would be a big job and it is too long since I have practiced as an engineer. I did an analysis of this quite a few years ago on MIMF and really don't want to revisit.
What you are attempting is a free body analysis where you treat the bridge as an isolated unit, look at the forces on it whether applied or reactions they are both needed and both forces and moments must add up to zero note that forces are directional and moments are clockwise or anticlockwise
Not a bad way to look at it but you are missing a lot in your diagrams

Although unscientific and I have no data to back it up I can't remember the last time that I had to reglue a bridge on a classical guitar. We do "fleet" servicing for a number of music programs from the universities in the area and from neighboring states too AND our community is well populated with classical players.

The school instruments often crack, have necks shrink, etc. from no RH management but we don't see the bridges lifting. We also work on high end classical instruments frequently and again no lifting classical bridges.

OTOH I'll see 3 - 5 lifting steel string bridges this week and it will increase until spring and then we see the lifting bridges from the folks too busy to hit us during business hours.

Sure there are far more steel string instruments out here but since we may.... have some agreement on the concept that bridge lifts are often a combination of factors which in my region often included RH abuse you would think that we would see more lifting classical bridges. I don't think this because it's not what we actually see.

It's not an absolute that classical guitars have less of a gluing footprint either. Many producers, particularly producers doing vintage instruments have small bridge foot prints on steel string instruments. Ever see one of Mario P's bridges? I'll also add that we all know that regardless of the physical size of a bridge on any sort of acoustic guitar there is no guarantee as to how much of the actual gluing area was utilized until it lifts and we get a look. I'm speaking of how much finish is cut back. I've seen 40% of available gluing surface not cleared of finish and this is not uncommon.

It's also worth noting that for bridge installs that are not rabbeted spanning the finish ledge creates an area at the finish ledge where the wood of the bridge is not in direct contact with the top. It's not uncommon to see film thickness of .003" or greater with steel string finishes but with FP shellac commonly found on higher end classicals that finish ledge is far lower because of less film thickness.

Analysis aside viewing a classical bridge as having a less ideal situation by any measure does not square with what we see that fails. Not trying to be argumentative but how do you square the fact that we are seeing far less, dramatically fewer classical guitar bridge failures than steel string bridge failures. That's a rhetorical question since my data is admittedly non existent but other folks here with more than hobbyist credentials when it comes to repair work may also wish to comment on the ratios of steel string bridge failures vs. classical bridge failures that they see as well.

I'll also add that the nature of how a bridge is removed, we use heat from a heat lamp, amplifies how seldom we have to reglue a classical bridge because they tend to be finished making the heat lamp method not advisable or prudent.

I see a lot of student classical guitars which frequently need bridges re-glued. This has less to do with tension and more to do with bridges being glued over a thick top finish without a chamfer routed on the bridge. This basically causes an air gap at the mating surface, which is filled with glue, which has low sheer strength. Tension is really not the issue. I see this with many classical guitars, up to mid-level instruments ($500 to $2,000). Some are Chinese, some are Spanish, et cetera. I'd be surprised if I've repaired less than 30 classical guitar bridges in the last few years. Classical guitars are frequent visitors, with tuner failures being second to bridge repairs.

Andy

Wow that's a lot of bridge failures.

Our experience with the lower end student guitars has been different especially since they are often from China and the mystery glues used tend to be pretty tenacious and unyielding and seemingly even gap filling since they tend to need be. As such we see classical guitar failures of mostly Luthier built instruments with a bigger price tag than the imports.

What's your climate like so we can see if RH issues may contribute to such a high failure rate? We know or strongly believe that since the classical bridge failures that we see much less frequently than steel string bridge failures are often accompanied by new top cracks that RH is often a contributing cause, likely.

Just worked on a 1920 Hauser that at some point lost it's bridge in the past and has a replacement. That was a tuner failure too but 100 years or nearly 100 years is a decent amount of value to get from a tuner.

I am left to wonder if the others here found this discussion to be as unsatisfying as I did in terms of the degree of exposition and illumination, and more to the point, the occasional coarseness of language and ill will displayed.

Obviously, the topic is a difficult one to discuss, where even the acknowledged experts seem to have a very hard time making themselves understood or, indeed, understandable. Even more troubling is the the lack of civility in what is usually a very civil place. While I recognize that this sort of low discourse is a tepid tea compared to the acrimonious exchanges routinely seen elsewhere in this digital wilderness, I still find it out of character for this community and a distinctly unpleasant experience.

I would hope that, especially with the holiday season upon us, apologies might be tendered and accepted, and our status as gentlemen and ladies affirmed by our actions.

_________________
A constellation only takes shape when one maps the whole.
- Beth Brower

Woodie what a great idea and I'll go first. Thanks for bringing this up.

I apologize to Trevor and anyone else that I offended.

Happy Holidays everyone!

I have no idea of percentages, but some classical builders do glue the bridge on before french polishing.
That will of course help to avoid finish under bridge problem altogether. I am not aware of any steel string
builders doing the same, but maybe there are those, too.

Gee whiz, some of us even masked off the spruce till the bridge is ready to HHG on...

I am primarily build SS with an occasional classical. I have thought about gluing the bridge on first but have been afraid I couldn't get a good level sand. (You know how a sprayed finish tends to build up in a corner?). You have me thinking if I do a french polish on the top only, like others (Trevor Gore for one, I seem to recall), it might be a great way to go.

_________________
Joe Beaver
Maker of Sawdust

I haven't had the opportunity to read the whole thread yet so forgive me. . .

I had a couple bridges come up right after I started using more bridge clamps. I suspect that the added weight of the bridge clamps may have caused the bridges to roll forward and distorted the tops (not unlike the belly from string tension over time) resulting in inadequate clamping at the back edge of the bridge (right where it is most important).

It could be coincidence though. One was a pinless mandolin so there was a fair amount of pull on the bridge and it lasted a couple years. The other was an ukulele, obviously string tension was not high but I suspect it may have gotten the hot car treatment (the owner denies it though).

EDIT:
Ha! I knew I shouldn't bother posting until I read the whole thread. By the time I posted, the discussion had moved way past my little observation.

_________________
Bryan Bear PMoMC

Take care of your feet, and your feet will take care of you.

Just thought I'd share this little tidbit.

There's some question as to whether or not we encounter slab sawn Ebony IRL.

This here is a premium grade blank from LMI. If it had been black, there'd be no way to know that it's completely slab sawn...

Money down the drain...

Impossible to photograph, but, gave it a quick wipe on the bottom with hot water and sure enough it warped...

Sure looks like slab sawn. With the way rosewood and ebony are disappearing, quarter sawn is getting harder and harder to fine. I don't use ebony for bridges anymore, mostly rosewood and african blackwood once in awhile. When I see how thin the wings are on that bridge, it no wonder it pretzels.

_________________
Joe Beaver
Maker of Sawdust

Optical illusion, wings are 1/8"...

Accepted, reciprocated and extended to all.

Hopefully, at least a little bit of knowledge was shared.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

Sure do, and french polishing must be quite bit harder with bridge on, too.

Personally, I mask the bridge area with frisket film, but leave it 2mm short from the edges of the bridge. Therefore I get maybe 80% of the area protected from the lacquer and I scrape/sand the rest all the way to the bridge edge. No ledges. No lifted bridges yet with this method but the scraping part is p.i.t.a. Reason not to mask all the way is that I find it hard to get completely clear edge when taking of the film. But I am tempted to try micro ledges Hesh and others have mentioned here.

That's how I do it now. I've used masking tape and frisket, not sure which one I like the better.

It just seems like a lot of extra work but it does come out nice. I would like to hear from those that put the bridge on, mask it, then french polish with one of the hard shellacs, if anybody does that. When I get closer to finishing a build I just might post a thread.

_________________
Joe Beaver
Maker of Sawdust

With it's high tendency to both warp across and split, flat cut wood for bridges ought to be outlawed. I tend to prefer skew cut, as that has the highest splitting resistance, and doesn't show much, if any, more tendency to warp. So far. Knock wood...

I too try to skew cut my bridge stock. I also like to do that on my necks since they are laminated. I reverse on side so the skew on both sides curves toward the center. Looks good and I believe (notice I didn't say know) it makes it stronger and more warp resistant.

It also gives me a lot more wood to chose from at my local lumber store.

_________________
Joe Beaver
Maker of Sawdust