From the July/August 2021 issue of Acoustic Guitar | By Martin Keith
What is the correct formula for placing a bridge/saddle on an acoustic guitar? —Marc Lucas, via email
This is a good and very timely question—I was just involved in a discussion with a few other builder/repair people about a brand-new guitar whose saddle was placed in a puzzling spot. In order for the guitar to play in tune, the luthier had to fill the saddle slot and reroute it in a different placement—pretty unexpected work for a brand-new instrument that sells for almost $2,000.
All of this serves to illustrate that saddle position is not always approached as a strict formula in the industry. A survey of several very popular makes (Martin, Gibson, Taylor, Takamine) shows differing saddle angles, split saddles, and compensated or ramped saddle tops on some guitars and straight saddle tops on others. If tuning is something that can be so easily measured, and if there’s a clear right and wrong, then how can there be so many differences in saddle position and treatment?
I’ve been asked this question before by clients and can’t always give them an easy answer. In my repair work, I prefer to consider each guitar on an individual basis—I measure the tuning accuracy with a strobe tuner, and determine string-per-string what adjustments are needed. This practical approach also gives me a convenient end-run around answering the larger question, but the question lingers nonetheless.
The simple math of fret scales suggests that the saddle should be placed exactly twice as far from the nut as the 12th fret. However, because strings are not perfectly flexible, and because that imperfection varies from string to string, the saddle needs to be moved away from that theoretical point. The strings’ stiffness causes the notes to play sharper than they should, so the saddle is moved to lengthen the scale, causing the frets to play a tiny bit flatter. When this is done precisely, the two effects cancel each other out perfectly, resulting in well-tuned notes up and down the fretboard.
The nut is also implicated here, and this is a great time to bring up the many ways this can impact intonation. A high nut (i.e., one whose slots are not cut deep enough) will force the player to bend the strings sharp just to get them in contact with the frets. This effect will be most prominent in the lower positions and is very common on new factory instruments. However, the effects of string rigidity do increase as you approach the nut, and modern luthiers have developed various systems to address those effects. Stepped or compensated nuts, first seen in the work of custom makers, are now starting to appear on mass-production instruments. Session guitarist Buzz Feiten developed a tuning system meant to sweeten the intonation of guitars, which includes a calculated change to the nut position. There are even aftermarket compensated nuts from companies such as Earvana, which allow the player to experiment with this idea in a non-destructive way on their favorite guitar. Finally, the most adventurous builders and players have jumped into the deep end by building instruments where the frets themselves are not straight, but instead zigzag across the fretboard in a slightly unsettling way.
Unfortunately, this brings us back to the same big question: Shouldn’t it be possible to measure each of these systems with an honest tuner and determine which (if any) gives a meaningful advantage over the traditional straight nut and fret? For readers who wish to engage in a well-documented technical analysis of the compensated nut, I would suggest the work of Trevor Gore and Gerard Gilet, two Australian luthiers who have done considerable research on the topic, including some fairly advanced math and physics calculations.
In my own instruments, I build-in a zero fret. This is an additional fret, located where the nut would normally be. The strings contact this fret when played open, so the guitar behaves as if it were capoed against a fret at all times. For this reason, I have not seen the advantage to pursuing any nut-end compensaion, and I’ll be honest that I have yet to get a straight answer about how nut compensation can be meaningful in any way once a capo is installed on the guitar.
However, for those that really wish to have a simple rule: Among the repair people I respect most, the consensus is that for a standard 25.5-inch scale, you should add 1/16-inch to the scale length for the first string, and 3/16-inch to
the length for the sixth string. This slope of 1/8-inch difference from strings 1 to 6 is likely to yield good results for the average setup, strings, and playing style. It will certainly get you close enough that any further adjustments can be made by carefully shaping the saddle top.
It can feel counterintuitive, even frustrating, to think that something as simply measurable as tuning can cause such differences of opinion among professionals. No guitar will ever be perfectly in tune. And in fact, no fretted instrument can be: The whole idea of 12-tone equal temperament is that every note shares the same small amount of error, which hopefully makes it too small to notice. Professional piano tuners often play with this by stretching the tuning to adjust the timbre and character of the piano, giving it additional warmth or brightness in specific ranges as needed.
I know of a session bassist (one of the industry’s best and most renowned) who intentionally sets up his low E to trend a bit sharp, to help it cut through in big mixes. And, most importantly, small dips and rises in pitch from playing technique are an essential component of what gives each guitarist their own sonic personality. It is with all this in mind that I usually tell my clients to trust their ears above their calculators. Do your best to get the guitar as close as possible, check your work with a tuner, and then get back to playing.
This article originally appeared in the July/August 2021 issue of Acoustic Guitar magazine.