Piano World Home Page Forums Our Most Popular Forums Piano Forum FAQ - Piano Forum what is a high tension string scale?

Guitars are a little different than pianos. You don't replace all the strings in your piano on a whim to get a different sound. If a piano string breaks, you call a tech and he simply replaces and tunes it.

My point is that the tensions of piano strings can't vary too wildly from design to design since this would necessitate special proprietary materials for each type, which isn't the case. I don't have my notes with me now, but I believe that the tension of piano strings is around 150 lbs. across the gamut.

Don
Kansas City

Actually, I should have some very exciting news shortly. I am currently in the process of changing manufacturers.

QRS just couldn't keep the project on track. Every time they said it was active again, they would put it right back on the back burner. The last straw came when I received a shareholders' report that announced some of the new products that they were introducing (that had been preempting my invention). One of them was a keyboard that had black keys that "light up" when you play them...oh brother.

The vice president of a large piano company that you all know quite well (no hints just yet!) is coming out for a demonstration in the next week or so. He is very excited about the project and I'm sure he will be duly impressed with the prototype.

Ironically, since the project has sat dormant for these four years, I have continued to improve it just to fight the boredom! As of last night, I can tune a string that is 45 cents out of tune in nineteen seconds (five seconds to bend 45 cents, and fourteen more seconds to stablize). The accuracy is plus or minus one-sixth of a cent. Interestingly, I can actually measure and control the pitch much more precisely than that, but the string has a natural tendency to fluctuate by a few tenths of a cent anyway, so you really can't tune it much more accurately even if you wanted to.

After my upcoming meeting I'll make an announcement in the group. Thanks for your continued interest.

Don
Kansas City

I'll look forward to the announcement, Don. My only reservation about the idea previously was that I wished it were being implemented on a better line of pianos.

Oh, it's a better line of pianos all right...

Don
Kansas City

Sorry, but am not familiar with your work Eromlignod..what is it that you are working on?

Gpman

The self-tuning piano.


http://tech2.nytimes.com/mem/techno...3FF931A35752C0A9659C8B63&oref=slogin

http://www.npr.org/templates/story/story.php?storyId=878091

http://www.newscientist.com/article.ns?id=dn3143

Don
Kansas City

Thanks, and the best of luck with it! I wish it could be installed in an existing piano

Gpman

The relationship between scale tension and timbre is not quite as indefinable as some would have us believe.

BDB is right when he points out that many, if not most, of the string scales found on pianos in use today—including some very highly regarded pianos in current production—were developed empirically by designers who had only a vague idea of the scale tensions they were ending up with. By the time the meetings that formed the basis for the book, Piano Tone Building, however, at least some designers had a pretty good grasp of the relationship between string scale tensions and the voice of the piano. They were also aware that a change in scale tensions required appropriate changes in the soundboard design.

When considering just what effect variations in scale tension might make to the overall voice of a piano it is important to understand that changing the tensions of one or two notes in the tenor (say, by replacing the original strings with wire a half-size larger or smaller) is not going to make much difference. Restringing the whole piano and raising or lowering the average tenor string tensions by 10 or 15 lbs. probably will. (If you’re raising the tensions by that much you might also want to consider where you want to be when the plate lets go!)

Many scales in use today are hard to classify in terms of scale tensions. For example, C-52 in one version of the Steinway Model D has a tension of 166 lbs. (343.5 mm, #16 ½ wire). That is just on the high side of what I would call a low-tension scale. One octave down, (C-40) we find a tension of 179 lbs. (660 mm, #18 wire). That’s toward the high-end of what I would call a mid-tension scale. Going one more octave down, (C-28) we’re up to 206 lbs. (1291 mm, #20 wire). Now that is high-tension in most everybody’s book. So what do we call this piano? From C-52 going up to C-64 we find tensions also going up, this time to 172 lbs. (184.5 mm, #15 ½ wire). At C-76 we’re all the way up to 186 lbs (98.6 mm, #15 wire). And, finally, at C-88 we find the tension back down to 154 lbs. (49 mm, #13 ½ wire). So, this scale is all over the map. If we sum and average all of the unison tensions in the piano we come up with a total scale tension of 45,700 and an average of 188 lbs.—definitely a high-tension scale.

In terms of how scale tensions affect the voice of the piano, however, the strings that lie on the long bridge have the greatest effect. In the case of the Model D this averages 179 lbs., still quite high.

Nor can we say categorically that long pianos have high-tension scales and short pianos have low-tension scales. While a Steinway Model M would probably average out around 150¬–155 lbs., I have a 5’ 2” Knabe in my shop which averages 170 lbs., and that includes the top treble section where tensions are down in the 120 to 130 lb. range. The tensions through the tenor section of this piano average 178 lbs. Just one pound shy of the Steinway D average.

Does this mean the scaling of the 5’ 2” Knabe equal to the scaling of the 8’ 10 ½” Steinway? Not really. C-40 in the Knabe is 665 mm long and uses a #18 ½ wire for a tension of 190 lbs. Not all that much different. But in the Knabe C-28 is only 1025 mm long and uses two wrapped strings.

So, what does all this mean? In general, if all other factors are equal—string length, hammer density and resilience, soundboard design and construction, etc. the sound envelope generated by a low-tension scales will have more energy in it fundamental and lower partials. The sound envelope generated by a high-tension scale will have less energy in the fundamental and more energy in the upper partials. Also, very generally, strings at a higher tension will have the capability of storing more energy so they have the potential for more sustain.

However, all other factors are rarely, if ever, equal! String lengths can vary. A high-tension scale might use relatively short strings with large diameter wire. Or it might use relatively long strings with smaller diameter wire. And a low-tension scale also might use relatively short strings with larger wire or relatively long strings with smaller wire. Each will produce a different sound envelope.

Still, while it is impossible to categorize these things absolutely, it is possible to put forth some broad generalizations. If a designer wants to create a piano with a warm voice with broad timbre dynamics he or she will probably chose a relatively long, low-tension scale. This will work against a relatively low-mass soundboard of medium to low stiffness. If a designer wants to create a piano with great power he or she will use a higher-tension scale working against a more massive and stiffer soundboard system.

Some final thoughts.
Except in an abnormally long high treble and in a poorly designed bass scale string tensions will not approach a strings tensile proportional limit.
Generally, when a company announces that they have “increased scale tensions for more power,” they have done just that. Depending on how creative they are this will be done by going to larger wire sizes through some parts of the scale or by making the strings through all or part of the scale longer. Occasionally, both. Whether this actually gives them more power is another issue. If things are going to be kept in balance ribbing will have to be slightly stiffer, they might have to increase the thickness of the soundboard slightly and they may need a slightly more massive hammer (along with the resulting deformation of action parts). The higher tension scale will also tend to resist soundboard mobility (think of it as pushing against a stiffer spring). More than one pseudo-designer (usually someone in the marketing department) has been surprised to find out that simply increasing string tension does not yield the intended results.

Del

Del, thanks for your great explanation. What you are saying follows the logic of the string maker I was talking to.

In doing research as to what the best bass strings would be for my 1925 M&H RBB, I was told that having the exact string specs from a 1925 BB would be the most beneficial and that M&H had a certain bass string core wire and outside diameter wire in mind when they were designing the piano? I was able to find the exact specifications of a set of original strings on another 1925 BB, thanks to an owner of a 1929 BB who has the these exact size strings, core diameter, outer diameter etc. I have sent those to the string maker. For the most part, he says they are good, and he is only going to change a few of the parameters, including a smoother transition between unicords and bi-chords.
The new set of bass strings that are on my RBB now do not sound good, giving a thin, hollow sound. When my rebuilder measured the core diameter of the lowest note, it was .048. The lowest core diameter from the original set of 1925 strings was .059, that is quite a difference. In fact, all the core diameters of the current set of bass strings are smaller than the original specs...I suspect that is contributing to why the bass does not have a rich, robust sound?

I guess M&H knew what they were doing back then and knew what bass string specs would sound the best with their soundboard design.

I know there are string makers out there that design "scales" for these older pianos, changing the original specs, thinking they will sound better than the original scales...and sometimes they are not successful....interesting!

Del, have you used a lot of original bass string specs in your rebuilding of older pianos, and if so, have you found they sound optimum?

Gpman

There are a variety of ways of computing averages, including average tensions. Adding all of them up and dividing is one way, but it tends to throw things off if some numbers are way out of whack. If you graph the tension of note 88 and the tension of the first overwound string, and draw a straight line between them, and take the midpoint of the line, you probably get a better idea of what the ideal average for a given scale would be. Of course, many pianos are nothing like that.

The model that I was using is based on this straight-line method, and in this model, the line for a short piano tends to decline a bit towards the bass, while in longer pianos, it rises.

In the case of Del's Steinway D, the tension probably conforms reasonably closely to this model. In the case of the Knabe, the tension undoubtedly steps up in random jumps according to changes in wire gauge until it nose-dives to, or more likely well below, the tension of the first wound string. (Whenever I hear people say that the new Knabes use the scale of the old ones, I cringe, knowing that they could, and probably do, do better!)

A bad break more than likely comes from diverging too much from a straight line or a reasonably smooth curve. A lot of other problems do, too.

Del;
As always your responses are an education for those of use struggling to understand the intricacies of piano design. Thank you very much

DMR

To get an idea of how the scale is loading the soundboard you have to consider unison tensions. All of the tension of the individual string are added together. When I’m analyzing a scale I look at a variety of things. These include the overall tension of the scale (This is the number most often used to impress non-piano folks. The Steinway D I mentioned has a total of 45,700 lbs., the Knabe 37,850 lbs., the difference being mainly in the low tenor and bass.), the unison totals for each section, graphs of unison tension, string impedance and inharmonicity, backscale lengths, and rows and rows of numbers. All of this must then be considered in reference to the soundboard design, whether the original is being used or a new board is being installed. It is also considered in reference to how the original piano sounds—assuming the piano is in good enough condition to tell.

It is rare to find an original scale with anything like a straight line anywhere in the scale. Most are quite erratic starting fairly low in the treble, rising through peaks and valleys to top out somewhere toward the mid-tenor then dropping precipitously toward the bass/tenor break. The bass is usually equally varied, often with no discernable method to the pattern.

Of the two pianos I mentioned the Knabe actually has the more uniform scale through the tenor and treble sections. The bridge has a more uniform sweep and the scale uses half-sizes throughout. It does, however, take a huge drop from G#-36 (188 x 3 = 564 lbs.) down to G#-24 (144 x 2 = 288 lbs.). The original bass/tenor break was fairly smooth by itself, but the voice characteristic from G#-36 to G#-24 changed dramatically. None of the aggressive hammer voicing techniques evidenced in the hammers had helped. Tension then levels out some before heading back down with the mono-chords, ending up with 188 lbs. at A-1. From C-88 down to G#-36 the average unison tensions in the Knabe are very close to those of the Steinway D.

With the use of new tenor and bass bridges the scale tensions of the Knabe are being brought down to 155 -165 lb. range and the bass is being brought up to better blend with the tenor scaling. The overall tension load will be 37,260 lbs., just a bit lower than the original but with much better balance. The soundboard assembly is being designed accordingly.

Like BDB, I cringe when I hear manufacturers (not just Knabe) bragging about using original scales in some resurrected piano. We can do better and the piano buyer deserves better.

Del

Do not the force on the bridges vary, independently of the string tension, but because of height of bridge?

I thougt this was teh main issue with low or high strung pianos. But this is perhaps a different topic.

The last Knabe (5'-8") I restrung I did not bother to note the original scale, since I am aware of their deficiencies. But although Knabe did use half-sizes in the low tenor, I found that I could get a smoother scale using whole sizes. If there is an extreme foreshortening of the scale, sometimes you have to skip a half-size there, or even more. I aim for a low deviation, and in the case of this Knabe, it was less than 4 lb. The top of the scale, about the last 2 octaves, are all strung with the same size wire, showing that the designer aimed for the theoretic geometric curve in the bridge, and then screwed it all up by changing gauges.

That piano is a whole other kettle of fish or something. This piano switches to bi-chord wrapped strings fairly high up in the tenor. The new scale will be switching ever further up and they will be placed on a transition bridge.

Your point about whole wire size changes is well taken, but it really applies only to the lowest part of the tenor bridge and then only when the bridge has been foreshortened a fair amount. The rest of the scale is generally better off using half-sizes.

Del

They are interdependent.

If you replace a set of strings with higher tensions (i.e., larger diameters) on a given piano and make no other changes the string deflection angle across the bridge will be less than it was originally. The more taut string would require more force from the soundboard/bridge assembly to deflect it the same amount.

Del

Del, it sounds to me after reading your posts, that the original string scales can be improved upon? I believe that is what will happen with the new set of bass strings I will be getting for my RBB.

Thanks again for your excellent posts.

Gpman

Yes, almost always. My only warning would be to have the scaling done by someone who knows what they are doing. You want someone who is more cautious than brave. The idea is to even things out and correct for obvious scaling flaws, not try to make it into a whole other piano.

Unless, of course, you are trying to make it into a whole other piano. Then it is always safe to decrease string tensions -- rarely is it safe to increase them. For example, I am reducing the tensions through the tenor of the little Knabe I've been talking about, since the plate was obviously strong enough to hold the original higher tension scale this will be perfectly safe. (I'm also designing a new soundboard to work with the reduced scale tensions.) I would never increase the tensions on, say, a Steinway S, M, L, O or A. These plates were designed to work with relatively low-tension scales and I would not want to over stress them with a higher tension scale.

Del

Thanks Del, I belive that is exactly what will be done with my bass strings...his calculations actually follow pretty close to the original scale, but he sees the areas where he can improve in order to smooth out the breaks etc. He is definitely not going the route of changing the scale to make a "new" piano.

Interesting, I always heard that Steinway was a high tension piano..

Before doing all this research on bass strings and reading your posts, I did not realize how complicated and important stringing a piano can be. It can alter the sound quite a bit.

Gpman

Thanks, Del and others for the informative thread.