Drifting Notes

[Jonathan the 2nd]

Easy question time for tuners. In a piano when the unisons have a string drifting out of tune , is there a certain point where it sound a lot worse . Above or below it won`t sound in tune but is there a "nails down the blackboard " feeling at one particular stage .

[Barrie Heaton]

All depend on you hearing simple as that

Depending on the time of year in summer when the pitch goes up its manly the left hand string that goes first, in winter its the right hand one.

The above rule is also effected by how hard the piano is played

Barrie

[Jonathan the 2nd]

Is that a wind up answer? What`s the science behind the left and right string then?

[Johnkie]

Wind up ? No ....it's all a direct result of differing string lengths. While the speaking lengths of each unison are the same. the overall lengths are completely different, resulting in each string being affected in a non equal way.

[Barrie Heaton]

And .. the interesting one is Becky uprights as they have outside return and inside eye So you can get the middle sting in with the octave and the 2 out side strings in unison with themselves but out with the octave.

I will now hang up my arrack

Barrie

[Jonathan the 2nd]

Well fancy that. It reminds me to ask about a similar subject. A tuner once said that he did not want each string in a unison group to be pitched identically the same as the other two. He said it sounded warmer and more musical .

[Barrie Heaton]

Well fancy that. It reminds me to ask about a similar subject. A tuner once said that he did not want each string in a unison group to be pitched identically the same as the other two. He said it sounded warmer and more musical .

Hmm was he tuning for Winifred Atwell then. Sorry you need pure unisons for a pianos to sound sweet OK some older and cheep pianos because they are false stops you getting as good a unison as you would like

Barrie

[Jonathan the 2nd]

Is it possible for a tuner to hear high notes differently to me when he tunes up a piano. The piano has an traditional unequal temperament and he uses a meter . The meter has 3 disjointed looking graphs on it. I thought the middle notes had to be tuned mostly by checking a meter and then octaves checked by ear. The phrase "stretched octaves" gives the impression that we could or should hear them as sharper .That seems a bit wrong to me. Surely they should sound just right to a player regardless of a meter . And more importantly , what if they sound sharp now, in the higher octaves ,just after a tuning? I mentioned them before too. What to do ? Have I got to sit and check all the octaves before I let him out again ? How accurate should a tuning be ? If Roman Abramovitch was having the job done would he ,should he , expect better ?

[Withindale]

Jonathan,

Due to something called inharmonicity, all piano tuners (human and electronic) have to stretch the octaves for the piano to sound in tune. If the octaves were not stretched it would sound out of tune. This is because the harmonics, which tuners call partials, produced by steel strings are at higher frequencies than you might expect. The first harmonic (second partial) is more than twice the frequency of the fundamental (first partial) and so on for the second, third and higher harmonics. At its simplest the octave (e.g. A5) has to coincide with the first harmonic of the base note (e.g. A4). A5 should be more than 880 Hz. It is not as simple as that in practice because tuners also allow for higher harmonics as well; this results in even higher frequencies and great debates among tuners. A German professor has recently shown that applying Shannon's communications theory to the sounds each piano string makes results in tunings very similar to the equal temperaments produced by aural tuners. See http://en.wikipedia.org/wiki/Piano_acoustics for details of the Railsback curve which shows how piano tunings are stretched.

Ian

[Jonathan the 2nd]

The Railsback curve describes the technical numerical frequencies but the sounds, subjectively to a player , should be in tune and not sounding ,subjectively sharp if all c notes are played in sequence from the bass to the treble of the piano. Leave aside for a moment the ghastly out of tune effects of equal temperament . (A case of truth is stranger than fiction .)

[Withindale]

You could certainly ask your tuner to tune your piano so that the c notes played in sequence sound the same as as pure c notes on say a cello. Then see if your left hand sounds in tune with your right.

Have any tuners ever tried tuning a piano in strict equal temperament or just intonation frequencies without allowing for inharmonicity at all?

[joseph]

correct me if I'm wrong, I think there's also the issue that each instrument has to be interpreted. So, you couldn't use exactly the same tuning on a Steinway D as a Steinway A, or a Steinway A and a Bluthner 6. Perhaps even two pianos of the same make and model will require a different treatment. I think it comes from the tuner learning to listen to the piano as he or she tunes, and then adjusting according to what is heard.

So, there is a basic tuning, if you like, based on a theory, and then the theoretical knowledge is adjusted accordingly. I guess it's the same as playing the piano. Different pianos bring out different things in the music.

Incidentally, many string players (talking of Cellos) will play in equal temperament since our western musical ears are so adjusted to the piano. Orchestral players sometimes, but rarely, complain about having to make adjustments to their tuning when they do a piano concerto or when a piano is in the orchestra. I must admit, I hadn't thought about that before it was pointed out to me. Of course, piano concertos are so commonly performed that I don't think it really affects them as much as they say it might...

[Gill the Piano]

I was once asked to tune a piano with the scale in perfect intervals (with a horrible wolf on the last) and the octaves pure from that octave. Horrid. And flat as wotsits the higher/lower you went. The lady was a copmposer who liked experimenting with different effects. Hmm...

[Jonathan the 2nd]

This offending high note ,a Bb , occurs in the Chopin Mazurka in G minor Op 67 no 2 . Bar 33 , beginning with Bb middle of the treble clef ,then the octave and then an octave higher. The last note is so sharp it pokes you in the eye with a bent stick ( on this piano 2 days after a tuning ).It needs the semitone key lower to sound closer to the right ( musical ) note. Why does it have to sound so gawkily , plain wrong ? Is that something we just have to put up with ? Notes that high are not played so often as chords . As a final note in the section you notice it far more . The tuner`s meter has about 4 different stretch levels available or should that machine not be used up that high ?

[Colin Nicholson]

It sounds very much as though you are asking the forum to explain what a "bodge job" your tuner has done??!! ...... why complain to us? .... just get on the phone and ask him to come back because you are not happy with the tuning. Also.... who knows how he's tuned it using a meter to your "whatever" unequal temperament scale? .... maybe its meant to sound like that.

Cut your losses.... ring a decent tuner, and tune it to Equal Temperament - then your Mazurkas will sound as they ought to sound...... IN TUNE!!!

[Withindale]

Tuners will have to comment but, for what it is worth, my take is the amount of stretch should come from the piano rather an electronic tuner, unless the piano's parameters have been measured and fed into the device. Dirk's Piano Tuner is one with software which appears to do this properly but some tuners think it is too time consuming to record all the notes before tuning begins. Other electronic tuners may achieve equivalent or better results with their algorithms. Even so fine tuning is probably still best done by ear because no electronic tuner will be able to predict and allow for all the musical resonances of each instrument. Maybe, as this topic suggests, the Bb drifted after tuning. What does your tuner say?

PS Originally posted before Colin's post above but edited afterwards.

[Jonathan the 2nd]

There was a reference earlier to string instruments matching the notes of a piano which has stretched octaves. It might be on the other forum . A violinist will usually adapt the intonation to match the piano . That`s what they do . Violinists also change the exact intonation for expressive purposes. If you mention that on a violin forum you could be attacked as if you are slanging off a very fine player. They sometimes refuse to accept that their favourite is anything but perfect ,as in "set in glossy unchanging marble ". They adjust to suit. So I would expect the notes on a piano to be blended for inharmonicity and also not to stick out like a sore thumb. It`s a valid topic for discussion. Maybe it`s a grumble as well .Maybe tuners are like plumbers. Do it ,get the cash and get out .
Basically the question (grumble or not ) is "Should you notice the stretch tuning or should it be there but not so it will frighten the horses ?"

[Jonathan the 2nd]

The Railsback Curve only gets one mention ,which is on this topic, so I shall ask a nice juicy question about that .
What is the effective relation between the Railsback Curve and the tuning system ? The equal temperament exists so ubiquitously for a technical tuning reason . On the unequal tuning videos we are told the piano resonates better than with equal tuning. The curves on the Railsback graph seem unlikely , to comply with a straight line mathematical system such as equal temperament. Should we worry about that or shove the problem down the back of the sofa and hope it goes away ? A Christmas Puzzle for you all.

[Withindale]

There are better puzzles in Christmas crackers.

Pianos are tuned to equal temperament because it is the best one for all the music all pianists want to play. You might think of it as the lowest common denominator. Equal temperament on a piano is the best that the tuner can do to make the intervals seem equally spaced in geometrical progression. The Railsback curve shows the departures of a tuning from mathematically pure equal temperament. It is a measurement rather than a prescription.

[Gill the Piano]

...and your piano was designed to be tuned in equal temperament. It's not surprising if it's having a bit of a wobble having been tweaked suddenly to what my friend calls 'stoneage' temperaments!

[Withindale]

...and your piano was designed to be tuned in equal temperament....

Really?

The maximum deviation from just is about 17.5 cents which equates to about a 2% change in tension in the string. Overall tension for a just tuning would be slightly less than for equal temperament. Most unequal temperaments will be somewhere in between.

Would a piano be able to tell the difference?

Oh, yes, of course, I forgot, it would start howling like a wolf!

[Withindale]

A Christmas Puzzle for you all.

Jonathan, Kirnberger is a good Christmas game. You play pieces in different keys. People guess the keys and describe their feelings about each key. It's a shame it's not played more often in these days.

[Gill the Piano]

Surely piano designers took into account the optimum sound for ET? Still, what do I know - I'm too far down a large bottle of Rebellion beer to be sure of anything....have a happy one, everybody!

[Chris Leslie]

The Railsback Curve only gets one mention ,which is on this topic, so I shall ask a nice juicy question about that .
What is the effective relation between the Railsback Curve and the tuning system ? The equal temperament exists so ubiquitously for a technical tuning reason . On the unequal tuning videos we are told the piano resonates better than with equal tuning. The curves on the Railsback graph seem unlikely , to comply with a straight line mathematical system such as equal temperament. Should we worry about that or shove the problem down the back of the sofa and hope it goes away ? A Christmas Puzzle for you all.

A Railsback curve plots deviation for all notes of a piano from theoretical, non-inharmonicity, equal temperament. For real pianos the curve will be essentially a smooth line with a gentle positive slope and curving up more steeply at the high treble and down at the low bass. Real imperfect pianos however, combined with tuning imperfections will mean that the curve will typically have peaks and troughs along the way but hopefully no more than a cent or two for most of the range.

An unequal temperament will ideally have a designed pattern of peaks and troughs, repeating systematically for every octave, with a shape that could characterise the temperament. In practice, I doubt though that a systematic repetition of curve shape could be recognisable in a measured unequally tempered piano.

If an unequal temperament is tuned using a machine, the pattern should in theory repeat for each octave when the temperament octave is expanded outwards. For aural tuning, the pattern will be blurred if the tuner uses techniques that involve balancing octaves and non-octaves.

[Withindale]

The curves on the Railsback graph seem unlikely , to comply with a straight line mathematical system such as equal temperament.

A Railsback curve plots deviation for all notes of a piano from theoretical, non-inharmonicity, equal temperament. For real pianos the curve will be essentially a smooth line with a gentle positive slope and curving up more steeply at the high treble and down at the low bass.

Hallo Chris

Very well put.

As you well know, your smooth Railsback curve for a perfect piano with a perfect equal temperament is a direct consequence of allowing for inharmonicity by increasing the interval from each note to the next ever so slightly. This works like compound interest.

The Railsback curve and equal temperament go hand in hand.

[rxd]

Johnathan,

To answer your first two questions in reverse order, it Is a phenomenon, depending on the acoustics of the hall, the distance of the listener from the piano, the tone quality of the piano and the player, that a piano sound can appear bigger if the unisons are spread slightly. The threshold at which this happens is approx 1 beat every 3-4 seconds at the first partial although I have never measured it accurately in those terms it is expressed in the sound of the attack.

There are tuners who make an art form of this phenomenon but in all the concert tuning teams I have been a member of, the policy has always bee to tune unisons as still as possible " so that if anything does drift slightly, it's in your favour" as was told me by one head tuner when I asked their policy on this. there are far to many variables to do it on purpose.

There are a couple of recital pianos that I see almost daily and I know I can let them go a little longer without tuning because I sometimes pop in at the back during a concert and listen for a few minutes. I get favourable comments when I am beginning to think of tuning the piano so the threshold does vary with the listener, as Barrie says.

This is not to be confused with the typical Mrs Mills and Miss Atwell sound. We must remember that perhaps 98% of people have never heard a live piano anywhere near in tune and are so used to an out of tune sound that they prefer it.

In most unequal temperaments, the Bb appears sharper than some notes but if the top a below it sounds more in tune to you, there may be excessive sharpening in the treble or there may be a picnik going on. Since you have found a tuner willing to bend over backwards and tune a neglected piano unequally I would assume you want to keep them on as your tuner. I would mention your perception of sharpness on their next tuning visit. They probably feel they've already gone the extra mile for you.

I was last involved in tuning two pianos unequally for a recording of a new work about 7 years ago. The composer was a big enough name for the recording company to shell out enormously to trundle both pianos all over town to the various rehearsal halls, we used 2 designated pianos for this job, dividing the work between two tuners involving a double tuning on each of them to retune in the new temperament, (it was a form of just intonation which is as far as you can get from eq.), two tunings before each rehearsal, 3 days of 10 hour attendance for the recording session at Abbey Road, and the usual tunings for the Albert Hall proms premier. Then 3 tunings on each piano to re-stabilise them in equal for their ensueing normal use I only mention this to put what your tuner did for you into some sort of perspective.

In answer to your other question, I have tuned dead to a strobe for Bob Ralston, organ soloist and featured organist and pianist with the Lawrence Welk orchestra where both pianos used on that long running television show were tuned dead to a strobe. Bob toured for Thomas organs and only used the piano reaching over from the organ console. I only heard it played that way so it always matched the organ. Even so, it only sounded decidedly flat to me in the top octave under those circumstances. Again, the charge was for four tunings total for each concert, two to change it and two to return to stretched. That was nearly forty years ago. The two pianos for the TV show sounded rather quaint when played solo but, again, only decidedly flat to me in the last half of the last octave. Lawrence's music was a bit to commercial for the taste of many, particularly amateur jazzers, but his musicians were meticulous in their playing, (some of them fine jazz players, too.) I don't know the history of how they decided to tune that way, but it worked well in those circumstances.

[Jonathan the 2nd]

I happened to drop into the music shop today and tried the sounds of a lot of electric pianos. I was surprised how much warmer a small "real" upright sounded . It was a midget compared to mine . The electronic ones were in tune but so very sterile . The upright had a sold sign on it and a real old twangy out of tune note low down that made me smile. The electronic ones did not make me smile though. One simple test is to try the top six notes. They are pretty useless on the electric ones . It`s hard to hear them .