Piano World Home Page Forums Our Most Popular Forums Piano Tuner-Technicians Forum FAQ - Tuner-Technicians Forum String Tension

I'm sure many of you knowledgeable folks out there would be able to answer a few of my questions on grand piano scale design as it relates to string tension.

I have heard of "low tension" and "high tension" scale designs, but what is the real difference here? Specifically:

1) Is there a difference in sound between a high and low tension scale design?

2) Does a high tension scale design have any bearing on the longevity of the piano (i.e. more stress on the plate, bridges, pins and pin block, etc.)?

3) What are some examples of piano brands with high and low string tension? Specifically, I am considering the CW 190, AF 190, Sauter 185, and Estonia 190.

Thanks in advance for your help.

In general low tension scales sound cleaner and the tension is around 150 to 160 lbs - they tend not to have a great ammount of power. High tension scales 180 to 190 lbs are usually used in concert pianos, sound more harsh and have more power. High tension should not have any bearing on piano longevity - the plate should be designed to support the additional tension. The overall down bearing on the bridge on a high tension scale may be 3.5 lbs per wire as opposed to 3.3 lbs per wire for a low tension scale with a normal ammount of bearing angle - so the bridge should handle the difference with no problem. Tuning pins and pin block should be no problem but it may be a bit easier for a novice tuner to flagpole a tuning pin on a high tension scale.

scutch, now i am confused. my understanding is that steinway uses a low tension scale, and they certainly are among the pianos that have the most power and are used most often on the concert stage.

also, many european pianos have a high-tension scale, i.e. grotrian, and they have a cleaner sound than a steinway.

further clarification, please?

JWP,

I have in front of me a Boston piano brochure which gives the total string tension for every model. The grands range from 38,000 lbs. (GP-163) to 40,300 lbs. (GP-193). Steinway calls these "lower tension scale for longer sustaining tone and longer life of the whole piano." Perhaps someone can explain that.

You can find string tensions for Steinways on their website: www.steinway.com

The scale design for the Estonia 190 was available on the web but I don't have the URL here right now. According to my calculations the Estonia 190 has a total tension of about 35,000 lbs., actually lower than the smaller Estonia. The smaller Estonia has higher tensions throughout the piano, except for the lowest bass.

I would guess that a higher tension would be more powerful and less refined. But the Steinway L is a lower tension than the smoother (to my ears) Steinway O.

As I recall the German Steinway A is 42,000 lbs. vs. the Boston GP-193 at 40,000 lbs. and the Estonia 190 at 35,000 lbs. That is a wide range for the same size piano.

Del Fandrich has posted extensively about string tension in scale design. You might try a search.

Steinway D's use a high tension scale. When I talk about a cleaner or harsher sound it is in general terms as many other things will contribute. The scale designer may use different wire sizes to get the tension desired with drastic consequences to the inharmonicity. Example, comparing three wires of equal length one can use three different wire diameters and arrive at almost the same tension with each wire but the inharmonicity for each wire will change drastically. This scenario will contribute to either a harsh or dull sound.
When considering other factors like sustain and power you need to look at other aspects of piano construction as well like the bridge and soundboard. Energy needs to get from the hammer, into the string, through the bridge and into the soundboard and the board needs to be able to move. Much quality engineering goes into this. Not sure if this clarifies for you but hope it helps.

Steinway has posted information which is incorrect, as well as Del. Most manufacturers haven't put as much effort into designing scales for tension as they should. Yet people continue to buy and love pianos with really strange scalings, pianos which Del says (and I agree) could really benefit from rescaling. When they are rescaled, they even out, but the overall character of the piano's sound doesn't change that much.

So how much should you worry about it? I think you should worry if a salesperson tries to use words like "high-tension scale" and "low-tension scale" to try to keep you from believing your own ears. You might begin to worry if you are going to start rebuilding pianos, especially other people's pianos. But in either case, you would be far better off developing your ear than to worry about buzzwords.

Personally, I don't care if a piano is high tension or low tension. What I aim for is that there aren't any big jumps in tension from one note to the next. That is rare enough to find. Sometimes you can find bridges where the general curve of the bridge has a jump (is shorter than your eye would have it) where the wire gauges change. That is an indication that the designer spent some time thinking about tension. But even today, it is the exception, rather than the rule.

I recently had a salesman showing me the difference in sound between a low tension and high tension scale. He was not saying one was better than the other, it just depended on what I preferred. But he was doing this by hitting a key and then upon release of the key noticing the decay speed of the note being cut off by the damper. So he used a Yamaha as an example of a high tension scale design by how quickly the note was cut off after the key was released, and then a low tension scale design (Samick built Baldwin Chickering) by how much longer the note decayed after the key was release.

Was he showing me a correct methodology to demonstrate low and high tension scale designs? I could see if he let the note just naturally decay out, but by letting the damper kill the note, that seems to be more of a function of the damper.

Hububer,

Yep, you are right.

He needs to fix the dampers on the Samick.

No. That's just hype. I bet that if the pianos were similar size, the tension was similar, no matter what the salesperson said.

My engineering nose tells me that a high tension scale will hold the same character over a wider dynamic range. A low tension design will not produce the same dynamic range. Further a low tension design will have fewer upper partials.

Not proven by measurement.

Maybe that salesperson thought they were giving an accurate demo, but they were not. The scale tension has nothing to do with the quickness of the sound stopping. The effectiveness of the damper system and the how 'live' the duplex scale is on that model would be the main factors.

Don Mannino RPT

???
Something has to give, here. If the length stays the same and the wire diameter changes tension is going to change. For example, take an F33 with a length of 1,000 mm (39.4") — a rather long length for this note, but not at all unheard of — with a string size of 0.038” (#16 ½ wire) the tension will be 157 lbs., with a string diameter of 0.039” (#17 wire) the tension will be 165 lbs., with a string diameter of 0.040” (#17 ½ wire) the tension will be 174 lbs, and with a string diameter of 0.041” (#18 wire) the tension will be up to 183 lbs. Each increment will have an effect on the resultant tone quality.

The only way to keep string tension the same with a change in wire diameter (mass) is to simultaneously change string length.

And, yes, in the example given above the inharmonicity will also change but, as I’ve said before, we don’t hear inharmonicity as such. It is quite true that the harmonic structure of the energy waveform created within the string will change as the wire diameter increases and the tension increases (less energy in the fundamental and lower partials, more in the higher partials — assuming everything else stays the same) but this is not the result of any change in inharmonicity. Inharmonicity is a result, not a cause.

Del

Just a couple of comments on the practice of quoting scale tensions in a brochure or promotional piece:

Rarely, if ever, are we told anything about the tension distribution across the scale represented by these numbers. Since the Steinway D scale has already been mentioned let’s take a look at it. One version of this scale (there have been several) for example starts out (from the treble down) with about a half-octave having tensions in the 150 to 160 lb. range. It then climbs to the mid-180s by the lower end of the top treble section. In the next section down it starts out with over 180 lbs (at note #71) and then drops back down to about 165 at the lower end of the section (note #54). In the tenor section it is all over the place starting out at just under 170 lbs (note #53) and ending up at well over 200 lbs from about note #33 on down to the end of the section (note #21). The bass section is also somewhat erratic ranging from about 155 lbs. (at note 20 — times three, of course, since this is a tri-chord string set) up to about 375 lbs. at note #1.

So, what kind of scale is this? The total scale tension calculates out at just over 45,700 lbs. By most standards a relatively high-tension scale. Yet parts of the scale are decidedly low-tension or, at best, somewhere in the middle.

At best we can only make broad generalizations (as I’ve done in the past — see the archives) about these things. It would be a mistake to attempt any hard and fast rules of tonal performance based solely on overall scale tensions. At least so far as the typical “modern” piano is concerned.

Del

Del,

So reducing the wire diameter and tension would produce more fundamental and less of the higher harmonics. This might explain my fascination with the Estonia sound, but why did the old Boston Mason and Hamlin A sound very mellow. Wasn't that a high tension scale?

Has there been any trend in string tension? For example, were the "golden era" pianos lower tension? Have some makes made significant changes in string tension over the years?

In a previous post you mentioned re-scaling a Kimball grand you rebuilt. Did you go with a change in tension overall, or did you concentrate on trying to improve the break and other scaling issues?

The M & H A has used various scales over time, with a number of variations. The one that I am using these days tends to increase the tension at the break, while reducing it around middle C. These are not extreme changes, but it does seem to even it out. In particular, the higher tension at the break evens the volume in that area, so the plain strings are not softer than the wound strings. That said, the result is moderate in tension, perhaps a bit higher than is normal for a piano that size, but not as high as one would expect for a longer grand.

To my ears, are high-tension scale produces stronger, harsher tones (Grotrian, Steinway, Steingraeber) whereas those with low-tension seem somewhat less inclined to overall power and stridency of tone like (Bosendorfer, Schimmel...)

I do wonder though; I find high-tension pianos to have more string breakage. But does high tension necessarly mean the strings are closer to their breaking point?
I find Steinways (for instance) to suffer from much string breakage in the upper registers (which I might attribute some to high tension) while others like Schimmel or Bosendorfer or Bechstein seem not to be prone to this...

Manitou,

Could that string breakage occur due to something else, such as the angle over the capo bar?

As a student I attended a school which bought three new Kimball 6' 7" grands (among others) for a new practice building in 1979. These were before the "Viennese Edition." One of the piano technicians told me that they were breaking strings in the treble when trying to tune them because the strings were too long. He said that moving to larger diameter wire would not solve the problem because this would also increase the tension. He speculated that the scale was an unsuccessful attempt to copy the 6' 7" Boesendorfer. The pianos were ultimately sent back to French Lick and the scales were re-engineered. Just about everything but the rim and lid were replaced on those pianos. They re-used the plates, but moved the new bridges closer to the capo bar. I can't imagine that Steinway would have such scaling problems.

Despite the high tension in the treble, the Kimballs had a very muted tone. Maybe that was due to the plywood soundboard and bridges, hammers, or other factors.

My technician suggests that some Bechsteins were prone to plate breakage, explaining that Bechstein uses a very high tension scale. I hope this isn't too controversial a topic since the Bechstein is certainly an exceptional piano. Is the Bechstein higher in tension than others?

Marks,

String breakage can and does occur for a variety of reasons. Breaking under the capo is also common and must be seen in two ways: normal wear & tear and, design or crafstmanship defect.
In that area especially, with enough playing and force, the string will act much like a paperclip that is bent forward and back. Done long enough, it will of course break; and this is normal, if undesireable.
On the other hand, capos can have too sharp a gradient, have too soft of metal (creating grooves) and can have irregularities and burrs. These are defects that will affect everything from sustain to clarity and even string breakage.

1] Within limits, yes. Keep in mind that there are many reasons for why a piano sounds like it sounds. String tension is only one of them. The Estonia generally does have a “low-tension” scale. (Low even by low-tension standards.) But it also has a soundboard assembly matched to it. The old M&H As had an especially high-tension scale. But you could not accomplish the same result (i.e., make a M&H A sound like an Estonia) by simply replacing the wires to achieve similar tensions. To understand the sound of the early M&H you have to also consider the soundboard system design, the rest of the plate design and, especially, the hammers.

2] If there is a trend, I don’t really see it. Some go up and some down. Where I do see a trend is with hammers. The have become much more massive and much more dense over the past several decades.

3] I rarely change the overall tension scheme of a given design. I generally bring up the low areas and match that by bringing down the highs. This is especially true in the bass and tenor which is where the benefits of rescaling are most noticeable anyway.

Del

1] In general your observations about high-tension scales sounding “stronger, harsher” are typical. Part of this, however, must be attributed to the harder and more massive hammers required to drive them. The result is a relative sharp and chaotic attack with somewhat more energy being concentrated in the upper partials of the strings.

2] Within a given string length changing wire sizes will not appreciably alter the percentage of breaking strength. Some, but not a lot. In my earlier example of a 1,000 mm F-33, with a #39 wire the tension would be about 165 lbs. with a % of BS of about 38%. With a #40 wire this changes to 174 lbs. at about 38%. With a #41 wire this becomes 183 lbs. at about 39%. As the wire diameter increases so does the maximum breaking strength of the wire.

3] Most Steinway scales do not have particularly high tensions in the treble — for most of them the % of BS is close to 50% or below. String breakage is more closely related to the V-bar configuration and hammer hardness than it is to string tensions. Not that string tension can’t be a problem, just that it is not in most Steinway grands, the Model S excepted. This has to get up in the 60% range to really become a significant issue. (I’ve measured a couple of Model S grands with a speaking length at C-88 of 58 to 59 mm. This, even with the #12½ wire used, places the tensions up around 190 to 195 lbs. The % of BS is now about 68% to 70%. I once received a panic call from a technician who was trying to tune one of these pianos and had already replaced the strings up here several times. They kept breaking. I asked him to measure the speaking length of C-88. It was 63 mm. This works out to just over 220 lbs. of tension with a % of BS of 80%. This is a problem!

Del