|
'Way back when a fellow named Mark Emory Bolles wrote a program to design the strings for harps. He wrote a series of articles in the 'Folk Harp Journal' explaining the reasoning behind good string design, and giving a listing of the program (in BASIC: he wrote it on a TRS-80). He later collected all of his articles and sold them as 'The Harp Makers Notebook'. Extremely useful stuff. He talked a lot about the concepts of 'Tension to Length Ratio' (T/L), and 'Percentage of Tension' (%T).
T/L is mostly about the feel of the strings: too low and they're rubbery, and too high and they're hard. On a harp, when the T/L goes below about 1#/inch of length, it's hard to 'place' your hands well when you pluck the string. If the T/L gets much higher than about 4 (iirc) that's a sign that the string is too thick, and will vibrate more like a bar than a string, with a very inharmonic overtone series. When designing the strings for a harp, or anything else with strings of different lengths, it's really important to have the T/L ratio vary smoothly as you go across the band. A string that's much harder or softer than it's neighbors is hard to play well. Guitars, BTW, typically use T/L ratios of around .7 or so, iirc: easier to fret.
%T is a measure of how close the string is to it's theoretical breaking point. A string of a given material and length will always break at pretty much the same pitch (you can tune thin strings a little higher). The higher the %T of the string, the truer the overtone series, and all else equal, the better the sound. You can't carry 100%T on a string, though: that would not allow you to even wind it around the tuner without breaking. The piano tuner's rule is to run strings at 75%T: high enough to sound good, and low enough so that you don't break too many. The octave G string on a 12 is usually around 80%T, iirc.
On a wound string, you have to use the mass of the windings to compute an equivalent density, but all of the strength comes from the core. That's the point: you can have a thin and flexible core that will allow the higher partials to be nearly harmonic, and enough mass to keep the pitch down to where you want it.
There's one other thing to think about when designing strings: impedance. This is, more or less, a measure of how hard the string is pushing on the bridge. It's fairly easy to calculate the 'characteristic impedance' of a string: it's the square root of the tension times the mass per unit of length. It's often said that, on something like a guitar, the tension on each string should be the same. Actually, if you do that, you'll find the the impedance varies quite a lot from low to high. Usually strings are designed so that the impedances are not too far from equal, and the tension varies a little bit, but not too much. In other word, they compromise.
Guys like Steven Bennet have probably done a lot of experimenting to find string gauges that work, but with a progam like Bolles', you can do the experiments on your computer a lot quicker and cheaper. Most of the harp makers use a program like this, or get their string suppliers to run their harp designs through a similar program.
BTW: what Bennet plays is, in organological terms, a 'zither-guitar'. It has strings that run more or less parallel to the soundboard. On a real harp, the strings pull upward at an appreciable angle. I have yet to see a real harp-guitar.
|
Login
Register
FAQ
