Fixing the break angle on an Alvarez bi-level bridge

[dhalbert]

I have an Alvarez AP70 that could almost use a neck reset. The saddle is fairly low, and the strings are resting on the back of the upper part of the bridge, with a pretty low break angle.

Because of the shape of this bridge, regular slotting at the bridge pins won't work. Do you think the break angles are too shallow, and if so, what do you think would be most aesthetic fix? I am thinking I could slot the bridge, scallop it with a fairly wide radius at each string, or sand it the whole back side down a bit.

[B. Howard]

I see quite a few of these guitars and most of them have the strings rubbing on the backside of the bridge like in your picture. I have not had to modify any of those bridges to get them set up properly. As long as the string rings true without any fuzzy overtones when played break angle is sufficient. There is still far more break angle present on your bridge than on a conventional belly bridge with a saddle of similar height. It really only takes about 7 degrees of break angle to make a bridge functional.

[John Arnold]

The only string that looks too shallow is the first string. I would just cut a narrow ramp (slot) for that string.

[Alan Carruth]

I spent a fair amount of time several years ago experimenting with break angle, to see if it really made any difference. The short answer is that once you have 'enough' it doesn't, and it doesn't take much to be enough.

In my experiment I set up the same guitar with two different break angles: 25 degrees (less than you have), and 6 degrees (much less). I used a mechanical plucker to pluck the strings and recorded the sound. I strung together recordings of the different setups into 'synthetic strums' and played them back for people through headphones to see if they could tell the difference. They could not hear any when the break angle was changed. I also analysed the sounds to see if there was any difference in power, harmonic content, or rise or fall time, There was not.

As part of the experiment, I also put in a tall saddle, such that the six degree break angle was converted to 25 degrees. When that was paired up against sounds from either of the low-saddle setups people clearly heard the difference. There were also differences in the spectrum of the sound from the strings that are easy to account for when you understand how strings work.

So: if you change the break angle by putting in a taller saddle there will be a real difference in the sound, and you may hear it. If you change it by 'ramping' the strings there probably won't be any significant difference in the sound, but you might hear something anyway. People tend to hear what they think they'll hear....

[LouieAtienza]

Originally Posted by Alan Carruth

I spent a fair amount of time several years ago experimenting with break angle, to see if it really made any difference. The short answer is that once you have 'enough' it doesn't, and it doesn't take much to be enough.

In my experiment I set up the same guitar with two different break angles: 25 degrees (less than you have), and 6 degrees (much less). I used a mechanical plucker to pluck the strings and recorded the sound. I strung together recordings of the different setups into 'synthetic strums' and played them back for people through headphones to see if they could tell the difference. They could not hear any when the break angle was changed. I also analysed the sounds to see if there was any difference in power, harmonic content, or rise or fall time, There was not.

As part of the experiment, I also put in a tall saddle, such that the six degree break angle was converted to 25 degrees. When that was paired up against sounds from either of the low-saddle setups people clearly heard the difference. There were also differences in the spectrum of the sound from the strings that are easy to account for when you understand how strings work.

So: if you change the break angle by putting in a taller saddle there will be a real difference in the sound, and you may hear it. If you change it by 'ramping' the strings there probably won't be any significant difference in the sound, but you might hear something anyway. People tend to hear what they think they'll hear....

Alan, I remember you discussed this experiment at length on the MIMF. I think this is a very reasonable outcome. My conclusion is that the main factor concerning bridge torque influence is the "triangle" made by the leading and tail edge of the bridge where it meets the top, and the top of the saddle. If the bridge were shaped just like such a triangle with the string fixed on the top point it would probably affect the top tone similarly as a bridge with the same fixed points. Or in other words regardless of break angle (provided it is sufficient) if those three points are the same the bridge will function the same way mechanically against the top regardless of whatever extraneous shape is.

Thus the way to change the mechanical relationship is to change the "shape" of that triangle, by raising or lowering the peak, or moving it more toward the front or back. My thinking is getting that peak as close to the front as possible, which gives good bridge torque without necessitating a larger break angle (for lack of better scientific explanation!)

[dhalbert]

Thanks for the insights and suggestions. The photos I took show the situation at its worst. Right now the saddle has a .020" shim underneath for the winter. (I do humidify, but not up to summer humidity 60%-ish levels.) Adding the shim improved the sound of the high e: it is still touching the back of the bridge, but the angle is better. When the humidity rises again and I need to take the shim out I will listen again, check the angle, and do a little slotting if necessary.

[John Arnold]

As I look more at the photos, I think simple beveling of the corner will do the job.
My experience is that the low break angle I see on the first string does have a negative effect on the sound. Basically, any time I can pull up on the string and lift it off the saddle, I am ready to try a remedy.

[Alan Carruth]

At least one person has said that I didn't try a high enough break angle in my experiment; they hear a difference when the break is greater than 45 degrees or some such. It's possible that that's so, but I doubt it.

The forces that the vibrating string exerts on the saddle top are well defined, and much smaller than the tension of the string. So long as the string remains in contact with the saddle top throughout the whole cycle of the vibration, all of the sound will be transmitted to the saddle. If it doesn't stay in contact, you'll know it. Archtop guitars often use a break angle as little as 6 degrees, and my data indicates that this is (maybe barely) 'enough'. Beyond that a change in the break angle should not affect sound transmission from the string to the top.

There is, however, an effect on the way the top distorts that relates to the break angle. When you tighten up a string the bridge rotates toward the neck, and the top is pushed down. You can find the point behind the saddle that is the center of rotation and just moves down: I call it the 'centroid' for lack of a better term. As the break angle increases the centroid moves closer to the saddle, and the distance seems to depend on the angle, rather than where the string attaches to the bridge. Distortion of the top has small, but measurable, effects on the way it vibrates, and that could change the sound. In my experiment people could not reliably pick out any difference due to this. Maybe, if I'd looked at a higher break angle, they would have, but I doubt it. The changes are 'small', and certainly much less than you see from changing the height of the string off the top.

As usual, I find there are two things I learn from every experiment I do. One is that I need to re-do it better, and get more data next time. The other is that there's always something surprising happening that I never would have expected. That, of course, is why we need to do experiments.

[John Arnold]

My only concern with that particular guitar is the first string. Near as I can measure, it looks like about a 6 or 7 degree break angle.

[Alan Carruth]

Now that you mention it, John - I agree.

[dhalbert]

FOLLOWUP: This evening I sanded the back edge off the bridge and now all but the high e are not touching. Here are some photos. I had made some test slots before sanding to calibrate how much to sand. I didn't sand quite enough and you can see that the first string is still in a very shallow slot. It looks odd in the photo but it's not very noticeable in practice. The first string is at the best angle possible now without actually slotting at the bridge pin. I cannot say it clearly sounds better but it certainly sounds no worse. And if I need to take the saddle shim out later it will definitely have a better break angle. Thanks for all your help.