Octave errors, and why tuners misread low strings

A tuner that tells you a low E is an octave flat will have you tightening toward a broken string. Here is the physics behind the mistake and how a detector avoids it.

July 29, 2026

If a tuner has ever insisted your low E or your bass B was wildly flat, and following it felt like you were winding the string toward destruction, you were not imagining it. You had hit an octave error, and it is the most dangerous thing a bad tuner does.

The physics, briefly

A vibrating string does not produce one frequency. It produces a fundamental, which is the note you name, plus a series of harmonics at two, three, four times that frequency and onward.

The important part is that the fundamental is not always the loudest one. On a thick, low-tuned string the fundamental is a large, slow oscillation that a small instrument body radiates poorly. The second harmonic, an octave up, is often louder in the air than the note itself. On a bass low B at 30.87 Hz, the fundamental may be almost inaudible while its octave at 61.74 Hz is clearly there.

So a detector that answers the question “what is the strongest frequency here?” gets the wrong answer, confidently. It reports 61.74 Hz, concludes you are twelve semitones flat, and tells you to tighten.

Twelve semitones is a lot of tension. On a wound bass string you may well reach the breaking point before you reach the fake target.

Where it bites hardest

• Bass, both four and five string, and especially that low B
• Drop C and 7- or 8-string guitars, where the sixth string is down at 65 Hz or below
• Cello and double bass, where the C and E strings are low and the attack is slow
• Any low string played near the bridge, which emphasises the harmonics over the fundamental

How a detector avoids it

The naive approach is to take the spectrum and pick the biggest peak. That is what fails.

Better approaches ask a different question: not “what frequency is loudest” but “what fundamental best explains everything I am hearing”. Autocorrelation-style methods compare the waveform against delayed copies of itself and look for the period that fits the whole signal, which is much harder to fool with a loud harmonic, though it has its own failure mode of latching onto the wrong multiple.

The remaining work is explicitly evaluating the octave candidates. If a signal fits both 61.74 Hz and 30.87 Hz, something has to decide which is real, using the relative strength of the harmonic series rather than raw loudness. TuneHaven does that filtering, and where the evidence genuinely does not separate the two, it declines to state an answer instead of picking one and looking certain.

Protecting yourself in the moment

Regardless of what tuner you use, one habit prevents almost all the damage: know roughly where the string should be before you start turning.

If a tuner tells you a string that sounded approximately correct is an entire octave out, that is not a plausible thing to have happened by itself. Strings do not drop twelve semitones overnight. Treat that reading as a fault in the tuner rather than a fact about your instrument, re-pluck, and see if it settles.

Also useful:

• Pluck over the neck, away from the bridge, which favours the fundamental
• Pluck firmly and read after the attack settles, not during the first moment
• Bow it if the instrument takes a bow, since a sustained tone gives a far cleaner signal than a decaying plucked one
• Use the reference tone and compare by ear when in doubt. Your ear does not make octave errors in the way a naive algorithm does