Piano World Home Page Forums Our Most Popular Forums Piano Tuner-Technicians Forum The right wire for late 19th century grands

Original Chickering scales from that period were determined by tradition and reckoning, and perhaps a bit of experimentation, just like other pianos of the time. We can do better these days.

Wire changed a lot back then. There was a chart in some book I read, probably Edwin Good's, giving the breaking strengths of wire from various periods in the 1800s, and it increased quite a bit as time went on.

Send me 100 USD by Paypal and I tell you


Stress rate, on the typogram

I'd have to say that I'm a bit skeptical about all this talk about the exact wire type being important. There are really 3 inherent parameters of the wire that are important. I say inherent, because, for the moment, I want to defer thoughts about how well the wire was manufactured.

The three parameters are density (mass per unit volume), Young's modulus (sometimes called the modulus of elasticity, or the tensile modulus), and the yield strength. The modulus of elasticity is a measure of how much force is required to stretch the material a certain amount--one could think of it as stiffness. The yield point is the amount of strain the material can take before it starts to permanently deform.

The density of all steels is so close that for our purposes we can consider them to be the same. The modulus of all steels is essentially the same. The modulus for the stronger grades of stainless steel (400 series) is about 3% less than steel.

The yield point varies not only for the particular grade of steel, but also its temper. Many people don't realize that the tempering process only affects the yield point, but not the modulus. So, basically, all wire that would be suitable for piano strings has essentially the same stiffness, and the same density--it is really only the yield point that varies. Does variation in yield strength really affect sound. Well, I'm skeptical. Remember that all piano wire is tensioned well below the yield point.

Of course, we still have to consider the quality of the wire. The best wire would be extremely consistent both within any given length and between different batches. It's physical dimensions should be accurate and repeatable. It should have a smooth, shiny surface to help inhibit rust.

Well,now,
let's not think of our predecessors as cavemen knapping pianos out of flint. If, as you say, there was no calculation involved in the creation of these scales, then that leaves only tonal results as a guide. And the tonal results are what was being sold. My personal quest is to hear these pianos for what they were. My fear is that through modification all the personality will be wrung out of the piano. Back in the day decisions were made and fortunes bet on them. I want to hear what they were so confident about.




I have two Chickering scale77 9'4s in my space at the moment. One is from 1867, the other from 1897. The two are dramaticly different for all their similarities. The 67 has a bent rim, the 97 is a sectional. Both are all original, so the evolution of the wire and the scale should be apparent. Someday, when I have a free month, I plan on doing a comparative anatomy study.

The problem is that the tonal results of the stringing scale were modified by the voicing process. Poor stringing scales required more voicing, just to begin to have what they considered adequate tonal results. If you get a copy of Piano Tone Building, which is probably available from Del, you can see that this was a concern.

If you restring a few pianos, and rescale a few pianos, you begin to understand what differences rescaling makes. The results are subtile, but probably closer to what the early makers were aiming for. I have rescaled a Chickering from about 1890, and it has required very little voicing, and stays in tune very well. The main difference is an improvement in the transition from the bass to the treble, which is a problem with many old pianos.

The yeld point have been studied, I was also sceptikal until in an exchange with the director of a very reputed German factory, he confirmed to me it was an important parameter, assuring to me they begin a 50% (lowered yeld point in that computation) then going up to 80% in the medium range (I hear you US tech having the heair dressed on head !)

Hence 80% from there up to the high treble.
Since then, when I tune or listen to those pianos, I notice how the high mediums and the treble are crisp, clear and the strings react so fast to the hammer impact.

I measured the iH lowering also, with a few Hz raise in pitch even Tunelab can show you a lowering of iH.

Ther have been studies tending to prove that the more the string is near its breaking poin, the more it makes a well build spectra.

The other parameters, as you say, does not change much (mass of steel) , remains iH and BS% , which are linked.

AN old piano mounted with modern wire, the yeld point will be so low that the tone is too inharmonic and also nasal and hard.

"Soft" wire of old times was BTW way richer and warmer than the ones we have today.

PS I finally understand that the more you tense a steel wire, the more it is resilient, which by evidence is good for the tone, the more tense wire is more elastic than the less tense .. less flexible, more elastic

Percentage of breaking strength depends almost entirely on the length of the string, no matter what the gauge. Inharmonicity varies according to the gauge of the string. The percentage of breaking strength therefore has almost no effect on inharmonicity.

You may want to test that on a real scaling spreadsheet BDB (or I dont use the good term) Thicker gauge does not change much the BS, but not all gauges have the same.

it depends of the level of annealing of the wire, and the softness/hardness of the steel.

Softer steel have a lower level of BS

the idea of dealing with it induce choosing different wire quality, or choosing that when the plate is designed.

Tension have an effect on harmonicity

I have checked a spreadsheet. You are wrong.

You are correct, however given that a scale designer might be looking for a given tension, then for a longer string, the diameter must be reduced to maintain a given tension. The net result of the longer string and smaller diameter would be less inharmonicity. I suspect this is why people think that a string tensioned to a greater % of its yield point produces a clearer tone--it's not really the % of yield strength that's the issue, it's rather the longer, thinner string.

it is the mechanical behaviour of the wire that change depending of the level of stress it is subjected to. This was examined by acousticians and physicians before piano technicians begin to think of "rescaling"

this is something about how the elastic energy stocked within the wire behave

I remain unconvinced. Show me the research.

I tuned for a couple of physicians the other day. Neither have ever looked into the mechanical behavior of wire. They spend their time seeing patients.

That is because they dont speak French

Isaac,

La personne est un physicien n'est pas un médecin.

In English it is physicist.

Thank you Dan !

Valette et C. Cuesta, « Mécanique de la corde vibrante », Hermès, 1993.

original works on violin strings by Coulon

I wish i can find the abstracts, it seems to be difficult.

BTW the piano string tone is characterized by a "double decreasing" , it seem to be common term in the literature.
Coupling on a moving bridge make polarization on both planes, hence beats or double decreasing of the partials.

WHile this paraameter is evident to anyone listening well I cannot understand why it is never/rarely explained in the unison tuning descriptions.

Most of the descriptions I see talk of total coupling, like if the sound could be "on" or "off" and beats an absolute evil.

A single string is producing some sort of beat by itself, due to the way it behave.

That beat can be fight or used while tuning. when tuning 2 strings coupling via the bridge the tuner may modify the delay before horizontal polarization, or something similar ...

This question about the stiffness of piano wire, as it would have been when new on old pianos, brings up a broader issue for me.

Among the best scaled pianos of the world today, are they using the differing types of wire which are itemized in the Paulello Typogram as X,M,O,1,and 2 ? Isaac has replied that he personally spoke with one German company that does. This lends credibility.[Correction: Isaac corrected me to clarify: The manufacturer tended to use the same wire, but use the higher tensions with hard wire to achieve a closer approach to the yield point of that wire. That is an agreement with the theory related to yield point, but not an agreement with the value of using modified wire.]

In America, I would suspect that Mason & Hamlin might have gone into research on this as they reworked their classic designs and resurrected them for production. Does anyone have information on M&H or other companies in relation to this. The basic issue seems to be whether the very minute choices as to alloy and tempering (as they pertain to yield strength) are worthwhile in establishing the very foundation of tone.

If these fine choices are legitimate, then one would logically follow this path, if trying to change and improve an old scale: [This would not apply to Craig, since his objective is original scale and original personality of the instrument.]
1. Measure, then generate the new scale.
2. Analyze via the Paulello T.
3. String the piano with the wire giving the desired %. This would involve several types of wire across the scale.
I think that this discussion has basically concluded that the tension of the wire in that scale would not be where the change would lie. The change would be how close the strings would be, at tuned tension, to their yield point.

This seems to be the theory in application in a nutshell.

Phil no manufacturer I talk with use different stiffness wire they all use Roslau anx there is no choice, one stiffness only.

I just talked with one about the level of constrain/stress the wire is submited to and was said it was an important parameter (not the only one) .
Well I will say it, it was Mr Sauter who told me they attain 80% of the strain in the mediums, where usually we are with more moderate strain (but I have seen old scales with such strain in the mediums, depite soft wire)

Since then it is a mistery to me, as too much stress is as bad for the tone than not enough. That said the Sauter pianos have a high tension scale and their medium is very "fast" and crisp.

It is just that I will not be surprized now if I find such high level of stress on a scale. (that did not happen and I did not have the chance to see a Sauter scale so my witnessing is good for what is worth, please dont misunderstand me ...

In those older times the same thing happens, only one quality of wire was used at a time in a factory ,eventually the bass strings can be wound on another wire if the bass maker is external to the factory and he used Poehlmann wire for instance , when the factory use Roslau (both existed for some time) or other brand of wire as they where present in the beginning of 1900, slowly stopping production until one only remains (and piano wire is 3% or 5% or the production of the wire maker )

Paulello stress us to use different type of wire, insisting on the advantage of an even progression of the constrain.

It is related to relatively recent work from theoricians, but for instance Paulello wire and Roslau wire have a different spectra and enveloppe so if you begin to mix types you better use the same product on the whole piano, I have mixed types, Paulello and Roslau, the transition may happen at a plate break and I may think "registers" if not the change in spectra is somehow noticeable.

Paulello scales are very smooth in progression, having soft wire to raise the level of strain on the first plain wire (often under 40%BS if you use Roslau) is of some help.

Then I use mostly Roslau that seem to bring a different fundamental to partials ratio and a more evolving enveloppe.

If the piano you have used softer wire, you have no much choice, Roslau strings in the mediums if under 40% is really awful and nasal.

There is a nickel plated version of Paulello wire, that may have a different spectra, I understnd the W&L producer use them in their "Feurich" line of products , for the treble.

In the treble, you will see that most pianos need or may accept modern wire as Roslau or Paulello M.

As it was written above, as the scale is "set" even if you lower the tension, oor raise it, you will not change the BS percentage much.

If you want to use modern wire, you may prefer to lower the wire gauge so the tension and iH lower.

For instance theiH on the A4 of a 1920 Pleyel could be 6.5 originally (yet high) if you keep the diameter hence tension it raise to 7.2 with Roslau wire.

if you lower the tension using 2 jauge less you will attain 6.6 but the tone will be less powerful.

The BS percentage lower if you mix wire types, what change at the same time is the elongation of the wire, this parameter is very important for the tuning to stay put when the piano is submitted to heat, dryness, etc...

old pianos have often not the raise in tension we see in the modern ones, lot of tension in the low mediums as soon as possible, then a smooth slant of tension lowering. Modern pianos have a raise in tension in the last treble section (from which note ?)

Thanks for that correction, Isaac.

I edited my post to which you refer to show that clarification.