Maybe I'll Call This "Microtuning"

older_wrench
Posts: 18
Joined: Wed Nov 13, 2024 5:28 pm
Status: Offline

Tue Jun 03, 2025 4:23 pm

On returning to guitar, I really want to play, and not ponder the physics so much. Then Neil did this! :roll:

https://totallylessons.com/lessons/2957

Neil discussed tuning, harmonics, intonation, and the role of equal temperament in complicating all of those things. He also talked about body resonances further affecting tuning, and suggested muting the strings not being tuned at that time. I agree with everything he said, and I have a development building on his instructions.

I won't re-litigate here the math and physics of equal temperament, because after studying it ad infinitum (and ad nauseum), guitars are still equal temperament in construction, and nobody is going to change that. I think the best one can do is tune the instrument in a manner so as to minimize the tonal errors caused by the non-integer pitch ratios of equal temperament. There are also physics-related variables (such as individual string height, mass, and stiffness), but all other things being equal, the instrument will never achieve perfect tune. This is the compromise of equal temperament tuning.

One of the million-dollar questions is, "can you hear it"? Another million-dollar question is, "can your tuner hear it"? And yet another important question is, "is your guitar lying to your tuner"?

I think most of us by habit tune our guitars before we start playing. If you were to start playing without tuning, would you know the instrument is out of tune? What determines the guitar is out of tune; your hearing, or your tuner? I similarly question intonation. Does your hearing tell you pitches go sharp up the neck, or do you rely on your tuner to tell you that?

Okay, enough of the fifty or more questions. Here are my observations from deep, deep, deep into the rabbit hole, and a solution that pleases my ear on all of the dozen guitars I play.

A most important point to make here is the guitars in my collection have a tremendously wide range of sound profiles (resonances), scale lengths, and string types. While I have guitars (Yamahas) that sound fine when tuned open with a Snark, I have others that never achieved their optimal tuning or intonation until I applied this technique.

So, among the discoveries I made are:
1) your ears may have better resolution than your tuner
2) tuners display the pitch of the highest amplitude signal it hears - and that might not be the note you plucked
3) the nut might not be the best place from which to tune a string.

Okay, back to taming equal temperament tuning. Before electronic tuners, the only manual tuning technique I knew of took place at the 5th fret. So one day as I'm staring at my most problematic guitar by far, it occurred to me that it was an old design - probably late 1930's or early 1940's, so, perhaps this guitar was designed to be tuned at the 5th fret. I tried a number of methods involving the 5th fret, and lo and behold, this worked really well for my ears:
1) lightly mute all strings with the index finger at the 4th fret (similar to Neil's concept of muting the open strings)
2) then use the middle finger to press strings one at a time at the 5th fret and use an electronic tuner to set the pitch of that string.

Things that shocked me doing this:
1) my Snark didn't make it - I used a Korg TMR-50, and it nailed it.
2) this guitar examined the usual way showed I needed to move the saddle over the bridge pins to intonate it. Clearly not true. After tuning with this method, the worst intonated string was within 4 or 5 cents.
3) This guitar sounded beautiful!

The guitar I first tried this on is an extreme example. It is highly resonant, and the amplitudes of body resonances exceed the plucked pitches. When plucking a fretted note with other strings open, the tuner showed something nowhere near what I plucked. And as those notes decayed, the displayed notes became wildly erratic. But the solution is simple; mute all strings at the 4th while tuning the target string at the 5th. Easy-Peezy.

Most guitars are not this finicky, but I will say every one of them sounds better to my ear tuned in this fashion. And every one of them tuned open (even muted) have a couple strings off a few cents at the 5th.

Now, I don't know everything, and if anyone can explain what I am about to disclose, I'd love to hear your ideas. I have four Yamaha guitars. Three are FG dreadnoughts, and one is an LJ jumbo. Three have factory 650 mm scales. I increased the vibrating string length of one of the FG guitars by 3 mm (without altering the fret spacing). Now, 6 strings x 20 frets + 6 strings open = 126 notes. All four guitars (including the altered string length one) show zero intonation error at all 126 positions. I have some hunches how Yamaha achieves this, but they are all exactly that - hunches.

Mute all strings at 4 while tuning one string at a time at 5. Thoughts?


User avatar
TGNeil
Posts: 1164
Joined: Thu Feb 18, 2010 1:09 pm
Status: Offline

Fri Jun 13, 2025 11:27 pm

Hey Wrench,

This is one of the most interesting posts I have ever seen here and I want to start with thanking you for the deep geek research. I am sorry it has taken so long to reply and plan to add more later. The quick thought is we have adapted to everything being close to in tune is good enough, the bar is set a bit low.

More thoughts later-

Neil


older_wrench
Posts: 18
Joined: Wed Nov 13, 2024 5:28 pm
Status: Offline

Tue Jun 17, 2025 2:30 pm

I spent a few hours today making some follow-up observations.

Is Equal Temperament really as bad as we think?

Math suggests it is not. Just temperament intervals are integer ratios of 4/3 and 5/4. That's 1.333333 and 1.250000. Equal temperament intervals are based on the 12th root of 2, and are 1.334840 and 1.259921. 1.334840-1.333333=0.998871 and 1.259921-1.250000=0.992126. That yields pitch errors off just temperament of 0.0012 cents and 0.0071 cents, respectively. I don't have the best hearing in the world, but I am certain I can't hear such small pitch errors.

So why does an E chord sound good, but a D chord does not?

I think it may have to do with B string compensation. Over the years I note many complaints about B string pinging and B string intonation problems. I observe today on two different guitars, both having 25.5" scales and standard compensated saddles, that strumming an E chord shows no pitch errors, but strumming a D chord shows the D note 15 cents sharp. So what is the difference between the E and D chords? The B string is open on the E chord, but fretted on the D chord. I then repeated on two other guitars, a Yamaha with 25.59" scale, and a Takamine with a 25.5" scale, but it has split saddles, providing more B string compensation than one could achieve on a single saddle. The vibrating B string length of the Yamaha and the split-saddle Takamine are equal. I would need to dig deeper into string compensation physics, but I'm thinking a B string with errant length or insufficient compensation sets up a standing wave vibration of higher amplitude than the D note.

I may look deeper into B string compensation when my fretting fingers start bleeding, but for now, 5th fret tuning falls within my hearing range.

But I really that split saddle concept........

OK, back to practice now. Ferry Cross The Mersey wants to beat me up some more. :ugeek:


Post Reply Previous topicNext topic