Saddle Break Angle Experiment

[gitnoob]

This is likely to be even more controversial than the bridge pin experiments.

Hypothesis:
Break angle affects tone. Or does it?

Procedure:
Tune two "treble E" strings to a common frequency. Plain steel strings chosen to reduce the variability seen in wound strings, and to minimize the impact of the bend near the saddle.

Place the strings at the "D" and "G" positions to minimize the differences in saddle height and saddle compensation.

As a control, use identical brass bridge pins for both strings and compare the plucked spectra.

To minimize mic placement and acoustical effects, plug the UST pickup directly into the computer and sample via audacity. This should give us "pure" string effects.

After the control, change one of the pins to a JLD pin. Similar mass, lower break angle. Repeat procedure.





brass pins: 4.3g/ea
brass JLD pins (with washer): 4.5g/ea

[gitnoob]

Others have suggested different ways to approach this experiment, including changing the depth of the bridge pin holes (mass change) and lowering the saddle. Lowering saddle height changes two factors known to impact tone -- mass and torque.

The idea here is to isolate just break angle as a factor. The JLD pins are made of brass, so the mass is similar to the "normal" brass pins shown in the picture above.

The break angle, however, is radically different.

[gitnoob]

Here's the control. Tuned both of them up to E. Plucked each 3 times at the 12th fret. FFT on the 3rd pluck of each.

Position "D":


Position "G":


A couple of interesting things to note here. You can see the 60Hz hum from my cheap piezo setup, which I now actually find useful as a level reference. You can also see that I mostly got even harmonics by playing at the 12th fret.

[gitnoob]

The JLD pin is in place. Most people would consider this break angle to be unacceptable.

[gitnoob]

Something happened to my computer that caused the input levels to be much lower than before (even the 60Hz hum). Not sure what it was, but it doesn't seem to impact the results.

Anyway, here are the results:

JLD brass pin with a "poor" break angle:


Normal brass pin with a "good" break angle:


Subjectively, I couldn't tell any difference in tone. More objectively, the FFT's don't show any significant difference.

[Bax Burgess]

Is the prevailing opinion that a higher action produces greater volume. And is that supposedly the result of a greater break angle? I have a couple of guitars with very shallow break angles, and they sound strong and rich.

[gitnoob]

Originally Posted by Bax Burgess

Is the prevailing opinion that a higher action produces greater volume. And is that supposedly the result of a greater break angle? I have a couple of guitars with very shallow break angles, and they sound strong and rich.

Action height (or string height from the top) is a different factor than break angle, and I'm only looking at the latter in this experiment.

A lot of people seem to believe that you need a fairly steep angle to ensure "good tone." That never made much sense to me, so I decided to do this experiment.

I believe Alan Carruth may be publishing something soon that covers both break angle and string height effects.

[murrmac123]

Originally Posted by gitnoob

A lot of people seem to believe that you need a fairly steep angle to ensure "good tone." That never made much sense to me, so I decided to do this experiment.

You can count me among the people who believe that breakover angle is important.

To take the matter to the "reductio ad absurdum" level, imagine, if you will, a bridge/ saddle set-up where the break over angle is 90°. ie the string sits on a straight line across the saddle top. What is the result ? Buzzing and no tone.

Now reduce the angle to 89°. What is the result ? Very slightly less buzz, and a very slight increase in tone/volume.

So continue the experiment down through 88°, 87°, 86°.... all the way to 45°(or even less).

At some point, it is possible that any further decrease in angle ceases to have any effect.

What I would like to know is, where exactly is that point?

I am quite sure that between you and Alan you will come up with the answer ...

[gitnoob]

Originally Posted by murrmac123

At some point, it is possible that any further decrease in angle ceases to have any effect.

What I would like to know is, where exactly is that point?

I am quite sure that between you and Alan you will come up with the answer ...

I won't give away the answer, but I believe it is something Alan has already determined.

The reason this should interest most people is that some guitars may seem like they need a neck reset because of the "low saddle." Maybe. Maybe not. If the only problem is a low break angle over the saddle, probably not.

[Matt Mustapick]

Of course in the most complicated hypothetical analysis (which nobody has yet undertaken...not by a longshot) *every* minuscule detail should matter, but only to the most vanishingly minute degree. While that's true, I don't agree that what we refer to as break angle, in and of itself, contributes to tone quality in a way that is amenable to any kind of generalization. It takes *very* little down-pressure and break angle at the saddle contact point to fully terminate vibration. No matter what height of the the saddle, the wooden bridge, or the distance between the saddle and the bridge pins, etc, the string terminates at the bridge plate. In any sensible arrangement, the strings' vibration is completely terminated at the saddle, the strings' tension is completely anchored either at bottom of the bridge plate, or at the back of the bridge (pinless bridge), and the vibrational energy is sent into the soundboard with nowhere else for it to go...end of story. The angle of deflection between as the string bends across the saddle to the segment between the saddle and the pin holes is a trivial detail that does not even affect, for instance, the amount of torque that the bridge footprint imparts on the top. That is determined mainly by the height of the strings over the soundboard.

That's my own personal take, and obviously I acknowledge that it runs counter to conventional wisdom and the considered opinions of many guitar makers I call friends. I'm prepared to change my view right quick.

[wrench68]

As Matt Mustapick said, it takes little downforce to terminate the vibration - if the mating surfaces are flat and smooth. My guess is the value of a sharper break angle and resulting increase in downforce appears when those surfaces are not optimal. At some point though, as the angle approaches vertical, the resultant force on the saddle is going to turn away from vertical and toward the front of the bridge.

I can also see the advantage of a shallow angle by decreasing the area of contact with the saddle and by keeping the angle of the force more vertical.


Gitnoob - you're a step closer to Stockholm.

[Matt Mustapick]

A greater break angle creates more downforce at the saddle, but it creates a corresponding increment in up-force at the pinhole! The net effect is zero, or at least it's *much* more subtle than familiar conversation suggests. No doubt others have much to add...I've said most of my piece.

[gitnoob]

Originally Posted by Matt Mustapick

A greater break angle creates more downforce at the saddle, but it creates a corresponding increment in up-force at the pinhole! The net effect is zero, or at least it's *much* more subtle than familiar conversation suggests. No doubt others have much to add...I've said most of my piece.

I don't disagree, but the net effect should be that break angle has a slight effect on the static deformation of the top, right?

That's why I thought I might detect some change -- in theory the top might vibrate in a slightly different manner. But this particular experiment wasn't sensitive enough to pick up any change.

It was even hard for me to formulate a working hypothesis. What should I look for? What should change? What would be the effect of a slightly different load distribution near the bridge?

[wrench68]

Originally Posted by Matt Mustapick

A greater break angle creates more downforce at the saddle, but it creates a corresponding increment in up-force at the pinhole! The net effect is zero, or at least it's *much* more subtle than familiar conversation suggests.

I see the net zero between the pin hole and the saddle forces, and perhaps I made a bad assumption of the experiment. Without thinking broadly objective, my brain plugged in the paradigm that increased downforce on the saddle improves tone. In only considering the pin and saddle, the higher downforce is present under the saddle, maximizing that joint even though the net is zero of the bridge to the top. You do make me wonder, however, about the effect of the string orientation to torquing of the bridge. I realize that may be known information, but I never looked at that before, as I am a tech, not a builder.

[gitnoob]

In theory, the string height from the top should be the factor affecting torque. How that affects tone is less clear.

Another experiment for another day. Or perhaps I'll just wait for Alan's paper.