Piano World Home Page Forums Our Most Popular Forums Piano Tuner-Technicians Forum So in tune that it sounds terrible

Whatever the numbers may be, in reality isn't the main effect of inharmonicity that there is no such thing as the OP's "dead on" interval?

Actually, I see it as good news for Verituner users - it appears to show stable results through situations that confound other platforms... At least at the levels that apply to the realities of tuning pianos.

So it would seem.

Hello Ian,

It's nice to find you here.

I need to make sure that I get your point correctly: are you wondering.. due to iH, there isn't such a thing as "dead on" intervals..., meaning that because of iH it is no possible to think/talk about "perfectly in tune" tunings?

Please believe me, it is my poor English... if I'm not on your point, would you please re-word it?

The difficulty here is to define "perfectly in tune". Using a rank of organ pipes for example, which have no iH, it is theoretically possible to tune a 12ET with great precision, if the wind pressure is perfectly constant, the temperature and humidity remain unchanged for the duration of the tuning, and probably some other factors I have not considered. Or one could tune it to any UT with equal precision, if the math for the UT is well described. But on a piano, my guess is that there is a best compromise tuning that is the least grating on the ears. I can't produce it yet, by I am trying!

The reason the numbers don't look correct to you is that you're confusing two separate ways of representing inharmonicity.

http://www.pianoworld.com/forum/ubbthreads.php/topics/1493652/Help%20needed%20-%20inharmonicity

If it was noticed, and computations are good, no problem to me, but the workable numbers are for instance 0,7 ct at second partial level, it is way easier to read.

Ron you may have been looking at something else than iH in that case.

A small change in iH is supposed to change the computation, done by the VT100 but it may be so minimal you did not notice it. (it may not interfere on A3 A4 for instance) The amount of partial match used for for the first octave may have limits , tend to react more slowly than the top treble or bass for instance

For once the measured results correspond well to the theory you should have see a change in iH it is nothing mysterious, but the way the wire react to stress by being more resilient with more tension.

Hi Alfredo,

Yes, because the partials are not harmonic no one can say exactly what "perfectly in tune" means. You have to listen to the instrument and find what sounds best, as you said in your recent post:

I was too busy tuning pianos today to post in this thread...

Thanks for pointing that, in that case it is not "divided by 1000" indeed

Seems there's a lot of argument over whether iH changes enough to affect tuning alignment - in any case it's not really the topic. The question was why perfect tunings sometimes sound lifeless.

Nonetheless, all that you good tuners have to do is sit down at a piano, turn off your machines, and really hear the changes. I think you'll hear the temperament audibly change in behavior as you tune it.

I used the word inharmonicity in my description of this loosely because it describes the position of partials over a string. This is not equation behavior: No model will tell you the answer! You have to sit in front of a piano and just listen.

My feeling on it is that alignment of energy in the spectrum causes specific entrainment of frequencies - whether through the soundboard or just between strings themselves. In addition to tension changes and movement of the bearing points into speaking lengths, these are all inharmonic effects.

No formula or model is published on this in a way that is comprehensive and calculable.

But anyway, you guys do it, and share what you hear.

Hello Alfredo,

Nice to read you , I suggest (just to find something to detract from you that it is very possible that, due to the variation of the iH peception threshold with age, the "justness" perceived will depend of the age of the listener, younger people being more sensitive to higher frequencies, and then hear more the Ih of the bass than adults, for instance.

100% agreed about tone circulation within the tuning as a mean to enhance the spectra.

That was the discovering in the 70's that a pure frequency based tuning device can send us nowhere, and that the iH must be taken in account in a way or another.

Then derivations of "justness" have been tempted, that can be a help for the tuner, anyway a tool for him to produce a result.

I suggest that the pattern that is the result of those tuning schemes differ so ever slightly from what a tuner is actually doing, in the sense they push the tuner to listen less to consonance in the instrument, at worst they make him listen to a partial match, a worst they make him tune octaves he would not have tuned in absence of the ETD.


I may add that the tuner's hearing is also stamping the tuning, so it is important to have some points to refer on, as resonance, and clarity.

It is pushed a little far in regard of UT but I have envisaged that the preferred temperament sequence and the inherent imbalance obtained that I noticed could favor a better transition betwenn treble and basses (where iH jump is more noticeable, even on a concert piano there is some)

IH is well known since only a little time

Piano manufacturers have some choices in the design but it depends first from the size of the instrument

They use an iH limit, (they should, anyway) not to be overpassed, and some have tried to lower the iH so the tone would be purer (or they tried to gain a maximum energy output, hence raised the tension and stretched more the wire, lowering the iH )

If all known parameters are taken in account, in the end there seem to exist one scale allowed for an instrument size, and that is clearly noticed in many today designs

Hi Ian,

There I meant to say that for the instrument to sound at its best, a perfect tuning is not going to be enough, meaning that a (1) "perfect tuning" is only one of three principal factors: the other two factors are: (2) "timbre/color", related to the tone partials and the spectral envelope; (3) the amount of spendable energy and the way energy is released and maintained.

Those three factors may be seen as the effect of tuning, voicing and regulation, but what I mean to say is that if the target is "what sounds best", we must consider those factors as one.

Can a mediocre tuning make a piano sound dull?

Yes: easy to think that intervals can be all over the place, that beats can be annoying, or that dead-on unisons can suffocate the sound's sustain.

Will a "perfect tuning" be enough for a piano sound at its best?

Nope, because what is more important is that the piano can... sing, that the piano be sensitive to touch-dynamics, that it has a rich and generous and univocal sound/tone quality all across the keyboard, and keeps its whole shape reasonably.

..."...no one can say exactly what "perfectly in tune" means."...

If you like, I will be happy to expand on this... ...not here though. But let me say that, for what I can see, the problem arising here is different (no offence intended): it is how to move from an amateur to a professional attitude.

Best wishes,

Alfredo
.

Hi Alfredo,

I thought you might raise this point when I wrote no one can say exactly what "perfectly in tune" means.

Tunewerk has now expressed what I had in mind:



I'd say the result of listening and tuning with the exactitude you mentioned is as close to "perfectly in tune" as you can get. Others may hear things differently, as Isaac suggests, so their conceptions of "perfectly in tune" will not be quite the same.

Buongiorno Ian,

I shall expand on what you wrote in a different thread.

Let's keep in touch,

Alfredo

I haven't run the iH numbers by themselves in comparison but i can attest to the fact that RCT will render a slightly different tuning on a piano that is sampled before a pitch raise. There are numerous mentions about string tension being a factor in iH on strings. I hate using Wiki for citing but even there it states under iH that...

The inharmonicity of a string depends on its physical characteristics, such as tension, stiffness, and length. For instance, a stiff string under low tension (such as those found in the bass notes of small upright pianos) exhibits a high degree of inharmonicity, while a thinner string under higher tension (such as a treble string in a piano) or a more flexible string (such as a gut or nylon string used on a guitar or harp) will exhibit less inharmonicity

Since the topic is about very fine tuning, and some posters have declared that they can hear the IH change while tuning a fine unison, the numbers submitted show that is not reality. No one can hear .0001 cent much less measure it accurately in a real world piano.

I will repeat my first point: If the piano sounds lifeless when tuned perfectly; that is WHAT the piano sounds like. Voicing is more likely to improve this condition if the customer wants it. But I too find some pianos to have a too clinical sound-but I notice it even when the piano is out of tune.

Interesting opinion, Ed. I agree that .0001 cent is clearly not audible - even .1 cent is hard to hear correctly and consistently! Above A6, where approx. 1 CPS = 1 CEN, I'd argue 1/10 CPS becomes clearly audible. However, what the equation is declaring may not be the only factor in change. This is an important point I believe.

One more thing I wanted to add, is that there are tipping points in the audible spectrum where a smaller fraction of a cent may be audible. It's not linear. For example, when you get very close to a strong partial alignment, there might be a powerful difference in as small as .1 cent (guessing here). In other areas, you might be able to move a whole cent and not much will be done to improve or worsen the tuning.

I don't agree with your second point.. but that's hard to know over the forums because often what one person has in their head is different than another.



Thought I'd just add this here and save a post. Yes, one cent near F1 equals approximately 1/40 of a cycle per second, but if you look at the 6th partial at about 262Hz (or C4), that same deviation causes movement 6 times amplified. In the range of C4, each cent is about 1/6Hz.

Good bass tuning is defined by the precise placement of these partials, but the quality of the partials is different. They allow a little more movement because the very high frequencies are weak. Also keep in mind that at C4, we may be tuning to the 4th partial, or where each cent is about 1/2Hz in real value.

Here's to a good weekend! Cheers.

In the low bass, say around F1, one cent is just a tiny fraction of one Hz, yet many partials higher where, that same one cent is several Hz, causing beats. Is it possible that a tuner, in trying to tune both wound strings on F1 so they sound good together, may end up with the strings far enough out that each string's iH is different enough to make a good partial alignment difficult?