perfect unison does not sustain as long

[noambenhamou]

sorry for another topic but it's very interesting.
I was complaining to my tech today about a note that didn't sound right. He got a bit flustered because he heard it as perfect but I told him it's decaying too quick. I kept saying that the attack was way popping out, then a quick decay into a short and very soft sustain.

All the sudden a lightbulb went off in his head and he said - "oh, that's because that G is perfect in unison. You must like them a little imperfect". So he made all the 3 strings just a bit off. In his ear, it was terrible, but for any non piano tuner, it seemed perfectly in tune. And voila, the sustain was MUCH MUCH MUCH longer. We actually did this for the entire right side of the piano starting 1 octave above the middle C. The middle and left side of the piano has such long strings, that sustain is not an issue.

My tech is VERY good and is big on getting those unisons "perfect" which he says most tuners won't get to this level but he's very picky and very talented in my opinion.

I tried searching on google before I posted this topic but couldn't find anything regarding unison tuning and sustain. But my tech said that this is not just my piano, but any piano that has perfect unisons, the sustain is shorter than if they weren't 100% perfect.

Can anyone explain this scientifically? or direct me to a web page?

[richard black]

I've heard that said, though never witnessed a demonstration of it, and I can't for the life of me think why it should be true. I suspect it's more to do with psychoacoustics (i.e. how we perceive the sound) than physics (how the sound actually decays) though that should be easy enough to prove by making a recording and looking at it on screen.

[john90]

This is my guess:

If you think about the absolute amount of energy lost, when a single string is resonating loudly, more energy is lost per second than when it resonates softly, because it is making less noise when resonating softly.

Now if a pair of perfectly tuned strings resonate together in perfect tune, then they reinforce each other and synchronise wave forms. Now they reinforce each other when they are both at maximum amplitude, so more energy is as sound is output at the start. This is the most efficient path to silence. If they are slightly out of tune, then the strings transfer energy at less efficient points in terms of making maximum volume, and the total energy in the system is less efficiently released (as sound, and heat in the string), as the amplitude is less, but for longer.

[keyboardclass]

Your tuner doesn't know what he's doing.  No tuner in their right mind is going to tune to perfect unison.  As you seem to have discovered for yourself the art is in detuning them.

[pianolive]

"Your tuner doesn't know what he's doing.  No tuner in their right mind is going to tune to perfect unison.  As you seem to have discovered for yourself the art is in detuning them."

Nonsense, perfect unisons are exactly what should try to obtain.
The reason why you find some nonbeating unisons to have a shorter sustain is probably because one or all strings in the unison have got false beats, and when we (sometimes) manage to get the unison beatless, it might sound like the sustain is shorter. The real problem though is the strings, hammerfelt and voicing and also the steel strings going into the material in the capotaster.
To correct those problems by detuning the unisons is certantly not an option.

[richard black]

Of course there's no such thing as a truly perfect unison - it's just a question of getting beats so slow they're imperceptible. Keyboardclass may be thinking of octaves rather than unisons, which are normally tuned very slightly wide rather than perfect.

capotaster

The what? Do you mean capo d'astro?

Whether the shorter sustain is a physical or perceptual effect is really a question of the degree of coupling between the strings (in the strict sense of coupling resonators), which isn't actually very high. If there is a beat, the sound develops something akin to vibrato, which is well known to make sound appear more prominent as the brain somehow 'locks in' to a sound with vibrato, and I can't help suspecting that that's what makes a detuned note appear to resonate longer.

[keyboardclass]

Nonsense, perfect unisons are exactly what should try to obtain.

Then we'll have to agree to differ.  Though I'm not a tuner I know 'true' tone comes from how the three strings are tuned out of perfect unison.

Here's someone from another forum.  I couldn't trace the article.

One article in Science magazine some years back was a discussion of how the slight phase differences among three strings (which amount to slight differences in tuning) cause them to transfer energy among themselves. It is a commonplace that out-of-tune unisons actually sound louder the perfectly clean ones.

[noambenhamou]

I found this website to be very informative:
https://www.speech.kth.se/music/5_lectures/weinreic/weinreic.html


If I understand correctly, lets imagine two strings that are Dead On Unison.
Lets say the vibrate vertically - up and down. As soon as one goes up, and the other one goes down at the exact same time, they cancel eachother out - hense the shorter sustain.

[pianolive]

Yes, that is what happens when strings in a unison have false beats. Sometimes we actually manage to tune such a unison beatless, and it might sound more dull or short. Every tuner knows about this, but detuning the unison to get a longer sustain...
In the concert halls in Steinway D grands, played several times each week, it is quite often we after three years have to take off the strings in the two upper trebble sections, file and grind the capo dŽastro and put in new strings to solve this issue.
(Sorry for my english, not my language).   

[noambenhamou]

pianolive, are you saying that dead on unisons should actually have a good sustain but it's because of bad contact between the strings and the capo dŽastro?

[Bob]

I remember hearing something similar about synthetic pitches, that they had to be detuned a little to sound normal.  I'm blanking on exactly what that was though.  That was also working with "perfect" situations inside a computer though.  I didn't think it was actually possible in reality.

[noambenhamou]

oh it's possible. Whether anyone believes it or not, my tech tunes perfect unisons. I complained about the sustain after he hits a note, then he detunes it just a bit, and the sustain is clearly longer. I'm talking an extra 2 seconds on a note such as 2 octaves above the middle C.

When the unisons are perfect, I also complained about a loud "pop" just as you hit the note, then an "out of breath" sustain. Then when he detunes, the initial "pop" is gone, and the sustain is present for a much longer time. My tech was a little offended. He said if I didn't like perfect unisons, just have some other tuner tune my piano as most tuners can't do real perfect unisons.

[john90]

In the concert halls in Steinway D grands, played several times each week, it is quite often we after three years have to take off the strings in the two upper trebble sections, file and grind the capo dŽastro and put in new strings to solve this issue.
(Sorry for my english, not my language).   

On the D is it possible to push the strings slightly do one side on the capo d'astro bar to avoid filing and string replacement?
On pianos with a removable capo d'astro bar, (eg, Boesendorfer) do the strings still need to come off?

[keys60]

I believe that "pop" is like the clapper hitting the bell. Perfect unisons are 3 identical frequencies as opposed to differing frequencies. Could be, the way you hear, all like frequencies sound like a pop instead of varying frequencies having a slower effect on your ears.

[richard black]

Brainwave! It's obvious, actually. As I said above, you can't get truly perfect unisons (you can't get truly perfect _anything_!). But you can get the beats down to one per several seconds. When the beats are down in the region of one per 20 seconds of so, the sustain becomes a function of the beat itself and the note does indeed decay faster - though in principle it would to some extent recover its amplitude after 10 or more seconds.

This will probably only make sense to people who understand the maths of sum-and-difference frequencies. Sorry. When it's not late at night and I've a bit more thinking capacity left, I'll try to elucidate a little!

[iumonito]

I had net heard about this, and I must say it sounds a little suspect.

Could it be that in the absence of false beats you cannot hear the note as well?  Two notes slightly off (like the Celeste in a French romantic organ) will give you a pleasant vibrato, whereas without it you get a "clean" pitch plus harmonics or partials (whatever you want to call them), but no beats.

Your tech, by the way, sounds very knowledgeable to me from here, and I though pianokeyboard class was kidding.  If not, I am totally with pianolive. 

(and the original capo d'astro design of your D is definitely an area of improvement, if you ever cared to remanufacture your instrument; poor design).

[keys60]

Sorry Richard. I reread the posts and you did communicate my last sentiments. And I do agree, there are no perfect unisons, just optimum. Also, although most techs will strive for an optimally tuned (beatless) unison, it is not unheard of to tune slightly detuned yet still beatless unisons. A slow pulse is practically undetectable yet achievable and will create a different effect.

[pianolive]

pianolive, are you saying that dead on unisons should actually have a good sustain but it's because of bad contact between the strings and the capo dŽastro?

No, that is not what I mean. I shall try to explain, and when I talk about unison I mean beatless unison. Someone talked about beats over 10 sec or more, but that is of no interest for practical work.
If you play a unison string by string you might hear that one or all of them have beats. One string may beat rapidly, the next slower and the third maybe beatless. A tuner will try to tune the three strings together, so there are no beats. If he/she succeed, the unison very often will sound what you call "dead" and have a shorter sustain. The reason for this is that some partials "take out" each other, so to speak. That means that some higher partials simply disappear.
Other reasons why the sustain may be short: The striking point can be wrong. The hammer mass may not be right for the string mass. The top of the hammer may be too flat. The hammerfelt may be too soft or over voiced.
The unwanted beats in a string can have different reasons. The string may have been twisted, it may have been bent, maybe it is rusty. (Dont ever touch the strings with your bare fingers).
In Steinway grands the capo dŽis a part of the frame. Over time the steel strings make burrs (is burr the right word?) in the iron. Some strings then start getting these false beats and a terrible noice appears from the Duplex scale. By filing and grinding the Capo so it gets clean and smooth and put in new strings, the problems disappear.
In a repair situation if you dont care for the capo, the new strings will jump or slide inte the old burrs and the problems are back.
Hope I explained myself better now..

[iumonito]

The Capo d'astro can be hardened, by the way.  When it is time to re-string you should DEFINITELY make that part of the repair.  If I have my facts right, Steinway has modified the angle of the bar too - an improvement.  If by chance your piano is new to you but was manufactured a while ago, the problem would manifest itself in a tendency to break strings in the soprano area of the keyboard.  I don't think you will find that particular problem in a recently-manufactured Steinway, which is what I understand you got.  But if you turn to break strings with surprising frequency, that is likely the reason.

[john90]

The Capo d'astro can be hardened, by the way.  When it is time to re-string you should DEFINITELY make that part of the repair.

How is this done? Does it require removal of the frame, sending it off to be induction hardened or similar before re-guilding?

[iumonito]

You will have to get the details from the master rebuilder!  This is from Ron Overs website:

"In 1994 Overs Pianos adapted a recognised industrial process, for hardening the capo' and duplex bars. This process enables the capo' and duplex bars to be hardened subsequent to reshaping to a small radius. Furthermore, the string approach angle to the capo' bar can be reduced where necessary, and subsequently hardened (Some concert grands produced during the 1970s had a string approach angle to the capo' bar of up to 22 degrees. Premature string breakage has been a major problem with some of these pianos). Reshaped capo' and duplex bars allow for the production of 'clean' tone, while hardening the bars enables the tonal improvement to be retained for a longer service period. Recently, further work has been done on reducing sympathetic resonance of the front duplex system. Results of this investigative work will be incorporated into future rebuilds and the new Overs-Steinbach 225 grand pianos (from piano no. 2)"

https://www.overspianos.com.au/rebuild.html