Piano World Home Page Forums Our Most Popular Forums Piano Tuner-Technicians Forum Unison detuning?

I'm reasonably pleased with my progress in using TuneLab to tune my own Knight K10 upright. I have always tuned the unisons as perfect as possible. Initially by tuning each string separately and recently by tuning the centre string with Tunelab then aurally tuning the outsides to the centre. I'm not sure that I can perceive any difference.

My next interest is concerning deliberate detuning of the unisons. I learned about this whilst reading my Pianoteq 5 manual. This is discussed and how to change it (within Pianoteq) on several pages the first of which says:

(quote)
4.3 Unison tuning
As very few people know, the three strings of each piano unison (the strings hit by each hammer) are not tuned at exactly the same frequency. To change the timbre or colour
of the sound, a skilled piano tuner introduces small tuning differences between these three strings.

Experiment yourself by gently changing:

- the unison width, that is, the difference between the lowest and the highest frequency produced by the three strings of a single note,

- The unison balance, which allows you to adjust the intermediate frequency of a 3 strings choir from the lowest (balance = -1) to the upper frequency (balance = +1)of the choir, balance = 0 being the factory setting value.
(unquote)

I searched here in this PW group for a discussion on "unison detuning" but found nothing. Is there a discussion somewhere about the pros and cons of this "technique"?

Ian

Unisons are tuned differently--not "detuned."

Search for DOA, smiley, colour, decay, sustain, Viennese, etc. The topic has been discussed on PW for about a decade now.

In my own humble experience, I've come to the conclusion that there is absolutely no place for dead on unisons in any piano tuning.

I don't think you will find any experienced piano technician who advocates them. The experience I am talking about is where one has spent years trying to tune clean unisons and has achieved such a high skill in it that they start to get complaints along the lines of "What have you done to the tone of my piano? There isn't any!".

The scientific explanation is that when you have equal frequencies of two or more oscillators, they can be in phase or out of phase. As tuners we have no control over that. So, a piano tuned with all dead on unisons will have some notes in phase and others out of phase. In phase notes will be louder. Out of phase notes sound like the tone has been "sucked into the piano". Besides being subjectively undesirable, it produces an unevenness of tone which is more objectively undesirable, IMHO.

I remember years ago describing that sound as the goal in a tuning class. One person asked "Wouldn't that sound bad?". I was too inexperienced to realize he was right.

But as long as you are not good enough to actually achieve dead on unisons, it is a good goal. And of course, if your stability is not great, that can detune a dead on unison as well.

And there is scientific proof from Professor Gabriel Weinriech that slightly out if tune unisons have a different tone and can project farther. See https://www.speech.kth.se/music/5_lectures/weinreic/mistuned.html

But to answer your question, search the posts of Isaac Oleg and Viennese Unisons.

My thanks to you both for pointing me to further reading.

Ian

There is a fine line, however, between moving a unison in effort of achieving a particular color and just not clean tuning. The technique necessary for such small changes in pitch demands a good deal of experience and a responsive instrument. In any case the piano still must sound clean and in tune.

The definition of what constitutes a tuned or deliberately detuned unison is not established. If the strings of a tuned unison are measured with very fine resolution-some differences may be noted. BUT if the differences are less than the coupling that occurs between unison strings-no beat will be produced. If there is no unison beat-the unison is not detuned in my opinion.

Weinreich's article in JASA assumed that the phase differences between unison strings caused by the slight mis-fit of hammer to unison strings was a normal part of the piano tone. Tone regulators know unison strings must be in phase at hammer strike. This changes the conclusions of his research.

The really great thing about pianos is that they have auto-detuning unisons.

I had to "de-tune" unisons on one occasion. For an elderly person with two hearing aids clean unisons just didn't sound right. Had to go back and put a slight wave in each treble unison. It was both lengthy and maddening. Glad it only happened once.

Unisons are never perfect. It's a balancing act. In the end, however, these must sound clean and in tune as stated above - that is, neither beats, waves nor twangs. "Dirty unisons" (as once humorously put) may be acceptable elsewhere. Here such will make a tuner of low reputation, if indeed he or she survives long enough to earn one.

Right on the money! ...and from this relative novice - - practice, practice, practice some more, then keep practicing. Temperament means nothing without consistent, clean unisons and octaves, regardless of their type.

Beemer, since you are using an ETD, you'll notice that the range for clean unisons is quite large. Near the tenor break, one could get c.-0.4 (L) 0 (C) +0.4 (R), up around the mid capo it's usually down to c.-0.1 (L) 0 (C) +0.1 (R); for the bass, you aiming on either side to centre the target c. -0.2 (L) +0.2 (R). However, it all depends on the piano...

As is everything with piano, there are pros/cons to every approach. One of the benefits of wider unions (aka colour) is that the temperament calms down considerably; if one tunes DOA, the beating in ET is MUCH more intense.

The more false beats and drifting strings that a piano has, the less significant the "type" of unison becomes. When the strings are unstable, one can use the ETD to observe the drift--in order to counterbalance the observed movement with the other strings.

I guess you are talking about top quality pianos. The pianos I tune usually, do not allow so much accuracy. I am happy with just clean unissons. And sometimes I have a bad time with meaowing unisons that can not be tuned to sound clean.

In rare occasions I've found notes in which if I tuned a clean unison the note seemed to be swallowed by the piano and then I had to work on the tone when tuning the second string to avoid this sort of dampening. I guess this is what you call a dead on and out of phase unison.

Here is a nice video of Isaac Oleg building tone in a vertical Seiler

[video:youtube]http://youtube.com/watch? v=PH7DzIytK7w[/video]

Maybe he can explain in detail what he is doing in this video.

Interesting discussion.

In the recording world (or electric guitar world), when you add
a chorus effect on a sound, you take a copy of the original and
pitch shift it up and down slightly, and add it to the original.

You can adjust the depth of shift, and the speed of the modulation.

But the point is, to get the thick, tonal illusion of many voices,
the voices cannot be exactly the same.

Has anyone else noticed a far too common occurance of, what seems to me anyway, deliberate unmating of hammers to strings on Yamaha upright pianos. This definitely produces a certain tone. It was my voicing out of these mating problems(?) coupled with the dead on unisons that got me the "You ruined my piano!" comment.

One important observation that must be noted regarding ETDs and unisons. Using Tunelab to measure the three strings of a unison can yield some pretty shocking results where the higher partials of the three unisons do not match up at all. Most ETDs only listen to one partial. This can produce less than ideal unison tone if using it to dictate where to leave each string.

Sure a chorus or phase shifting or out of tuneness can thicken a sound, pipe organs have used the principle for centuries...But we aren't talking about guitars or organs. A piano unison that goes too far out of phase is simply out of tune.

In my opinion the notion altering the tuning or coloring of a unison for sustain is merely compensation for a deficiency in voicing, scaling, or some other flaw in the inherent tone of an instrument.

Excellent analogy!!! When the unisons are DOA and the hammer fitting PERFECT, the sound of the piano is NOT. As strange as that seems: perfection is not the key--perfectly controlled chaos is what matters most!

Mark Cerisano, RPT, I had a similar experience. For my own edification, I wanted to know what a new fully-prepped Yamaha upright would sound like. Surprisingly, when it came to fitting the hammers, the sound did not dramatically improve like it "should." That is when I first realised there was a relationship between hammer mass and the fitting effects of the hammers.

Lighter hammers allow for a larger tolerance in hammer fitting (i.e., this doesn't work with hard&heavy hammers). Just like there is room in the tuning to adjust the "tonal thickness," so too can small variations in the hammer fitting be used to a musical advantage.


NOTE: I am not talking about variations that would produce an open string when blocked by the hammer; I am talking about variations in the length of sustain that each string produces when plucked.

I believe this to be a purely subjective view.

Why thank you, A443! Yes, absolute perfection is not the goal.

This would also explain the appeal of the "honky-tonk," or out-of-tune piano sound.

The chorusing effect of the out of tune strings gives the impression of many instruments.

Useful on certain types of songs.....

I was under the impression that the tuner "control" the phase when tuning, even if we only can hear the effect,

Seem to me that perfect phase is not attainable with 3 strings, only at the moment of impact and under ideal conditions the 3 strings with hae the same motion up/down.
I think this is disrupted immediately and tend to stabilise if possible in a more stable transitional "shape" ( 2strings up one down)

The only way we have to manage that is to "detune" .

One explanation (that need to be proved) is that a slightly higher pitched string , the wave will travel faster hence impact the bridge sooner.
Sound coherent that if that is the case, the phase is managed then, as for 2 strings for instance, we can have them immediately in the more stable phase opposal shape.

Easy to see in basses.
Hopefully for us the iH of strings allow the "detuning" to take place within the natural spectra of the strings.

If we are in 0.4 cts range this is a low level of iH , the amount of iH in the less inharmonic strings as find in low mediums for instance.
That amount of variation vs fundamental is totally unsuspected as "detuning" when listening.

That is how I imagine that anyway

PS Pianotech used a simple method and a simple terminology, but I dont know if the phase effects between the 3 strings are taken in account in the model.

You think so because you believe that the 3 strings are of equal lenght - it happens, but less often than one believe, be it voluntarily by design or because of the usual imprecisions (for instance curved capo shape is a classic there)

SO looking for an eventual absolute in unison is missing the point, which is more managing the available power provided by the hammer.
Of course unison may not hide bad voicing.
Some voicing types, without a lot of "meat" may also oblige to a certain type of unison may be to compensate, may be to find some crispness where it is not there naturally.