Piano World Home Page Forums Our Most Popular Forums Piano Tuner-Technicians Forum Inharmonicity quotient survey

Hello all,

I would like to collect data on iH quotients for a variety of pianos. I realize that this has, on a very small scale, been done decades prior to now. We have here at PW, however, an opportunity to amass a significant database of iH quotients from professional tuners, technicians, and amateurs from all over the world. If you would like to participate, please send me the following data to the this email link - tannis.sprott@gmail.com

Using TuneLab Pro or an equivalent device (I suggest TuneLab Pro, since it is free for trial use.), provide me with the iH for A4, C1, C2, C3, C4, C5, C6 and C7, make , model, year, and modifications, if done.

If you have suggestions for additional information that would be relevant, please tell me.

I will collect the data, put it into a spreadsheet, and post the results here on a continuing basis.

Thanks everyone for your interest.

Here is my data -
A4-0.854,C1-0.179,C2-0.122,C3-0.162,C4-0.431,C5-1.040,C6-2.353,C7-5.835,M&H BB, 2009, no mods yet!

Note the high iH. I wonder if the "typically lush, American sound" is in part due to the iH?

Why do you want to do this?
What do you think these numbers signify?
How are they measured or computed?

I am curious to do a bit of data mining, using various statistical models to look for patterns. If two pianos of the same model only a few serial numbers apart and using the same materials have statistically different iHs, where is the variable that causes it?

Since you do not have a fixed method for acquiring the data, comparison is meaningless.

Regarding your second question - I have read a number of papers on inharmonicity and its perceptual significance to the listener. We tend to think of iH as a bad thing, yet there is evidence that it is more significant to the enjoyment of the sound of a piano than the relative amplitude of the partials themselves.

Provide me with a fixed, long distance method for acquiring the data.

Actually, your statement is quite interesting. It implies that everyone who uses an ETD for tuning, has no assurance that the data they have collected has any meaning.

Not my problem.

I'm sure someone will reach and correct me if they think I'm wrong, but it seems to me you're going to end up trying to quantify the inconsistencies between pianos due to their organic nature. Good luck with that.

There are those who claim that inharmonicity is a matter of design. Maybe, maybe not, but you need an agreed-upon method of measuring to find out.

This is not a problem as there is no such thing as an "inharmonicity quotient".

Kees

You might be right that standards of measurement are not standard. Here's what I have from my 1888 Steinway A1:

IHCon A1 0.223
IHCon G2 0.193
IHCon D3 0.187
IHCon A3 0.220
IHCon A4 1.114
IHCon A5 0.249
IHCon A6 6.330
IHCon C7 6.811

A4 looks suspicious I'd say.

This was always my understanding from instructors at university course. Simply explained at that time was a certain amount of mathematical error is built into the scale design, further any scale under six foot in length the error becomes more prevalent from what I can recall.

Long-time back to the 70’s though and the thought process on those things may have changed over time.

The age of the strings is to be taken in account, I believe iH raise in time on pianos.

The most significant number is at A 49, to me.

here is a small vertical 35_years old French brand Rameau (height 113cml)

A 49 0.81 c3 0.226 c4 0.435 c5 0.96 cts.

I will provide data from 2 recently stringed instruments I have in the shop.
Ih is not an absolute evil, it have bern categorized by the German scientist Klaus Fenner, but I have no access to this writing. It may mostly relate to the lenght at a49

How old are the strings ? (a5 also is more than that usually) a4 is in the 0.6 0.8 range usually on a grand. (less on some old pianos with soft wire)
Sure, for A4 , did you try more than once ? One string only (not the full unison)

You know, every time you guys use an ETD like it were a spectrum analyzer, Joseph Fourier up in heaven cries a little.

Please just stop.

Chris S.

I eagerly await your peer-reviewed research which substantiates this claim.

Chris S.

No, BDB's statement doesn't imply that at all.

It's not that the IH values don't have any meaning. It's that the IH values cannot be useful (or claimed to be useful) for anything other than what the ETD algorithm uses them for.

The IH values the ETD produces may be useful for tuning the piano at hand at any given time - they have the accuracy and precision needed to do so. That is the claim of the inventors. As far as I know, that's the only reliable claim that can be placed upon the IH data. To my knowledge, there is no assurance that the IH data can be used for anything beyond that task. There's no assurance that two different ETDs of the same algorithm will produce the same IH value from the same note on the same instrument. There's no assurance that the IH result of one ETD algorithm can be compared to the IH result of a different ETD algorithm for the same note on one piano.

Just measure IH a few times on the same note using your ETD and you'll see what I mean. I certainly don't find my ETD producing repeatable measurements. Nor does does mine produce anywhere near enough precision to do anything useful with the IH value outside of its contribution to the ETD algorithm.

If you want IH values that are useful for the database you're suggesting, and of sufficient accuracy/precision, you're going to have to do what BDB is suggesting - set up a reliable, repeatable, stardardized protocol for measuring it. (Preferably with a better microphone than is in a pocket computer.)

So can we agree to stop making Joseph Fourier cry up in heaven?

Chris S.

Should we conclude from your comments, Chris, that ETDs relying on iH calculations, or a least yours, are incapable of producing repeatable tunings, and that well implemented ETDs based on a full spectrum analysis of every note are a better proposition?