This is an abbreviated version of ‘Workshop News’, September 2005

At certain times of year such as now a strange change – unknown to evolutionary science, but probably familiar to many clavichord makers – comes over me: I am transformed into stringwinder man, a hominid with horny pads on his fingers, a stooping gait, peering eyes and a tendency to stand on one foot. The cause of this metamorphosis is of course –

I am making over-wound strings for a new clavichord, specifically the new ‘Silbermann’. In my version the lowest 17 notes have over-wound strings: that’s 34 strings in all to be made, and it takes me more than one day to complete and fit them all. Making and fitting an over-wound string must be five or six times more work than putting on a plain wire one: why do we bother?

Well, there is no doubt that on a clavichord like the present one, where the bass is markedly foreshortened, plain wire strings would sound very dull and would be nearly impossible to tune. Put simply, that’s because they would be too slack. With a slack string, the slightest change in tension causes a big change in the pitch: not only do you have to achieve hair-accuracy when turning the tuning pin, but also any slight movement of the case, any slight change in the ambient temperature or humidity, will put those strings out of tune. Moreover both the quality and the loudness of the sound suffer. As the tension on the string is increased, the amount of pitch change that results from small changes in tension gets less and less, so that fully stretched strings are much more stable and easier to tune, as well as sounding better. If the wire is loaded with an overwinding, it will need to be tuned to a higher tension to achieve the required pitch, and so it behaves like a fully-stretched string rather than a slack one. I hope that’s clear!

The strings I am making are ‘open-wound’; that is, the covering wire winds round the core in an open spiral so that you can see the core wire itself in between each turn. A few years ago, makers used ‘close-wound’ strings as a matter of course: these are similar to the bass strings used in modern pianos, and in zithers and auto-harps. Open-wound strings were universally used in clavichords and pianos in the eighteenth century. Some say they can tell the difference between the sound of open- and close-wound strings, but I’m not entirely convinced of this. What, then, are the advantages of open-winding? I can think of three:

  1. Firstly, you have an almost complete control over the weight, and therefore the tension, of the manufactured string, because you can vary not only the diameter and material of the covering wire but also the winding pitch, that is, the distance between each turn;

  2. As the wire is pulled up to pitch on the instrument, the stretching of the core wire tends to tighten the covering wire: this effect is insignificant with close winding, but is more marked the more open the spiral is. You want the covering wire to be tight against the core, because a loose covering is a disaster. A slightly loose covering produces a dull note; a little looser and no sound at all comes from the string – a startling and puzzling effect when you encounter it for the first time;

  3. To achieve the right weight with close winding, you have to use very thin winding wire, which is apt to get damaged at the point where the wire crosses the bridge.

Open-wound strings do, however, have the problem that it is possible for the tangent to become stuck in between the windings if it happens to strike in a particular place. Some makers use extra-wide tangents to overcome this; others put a short length of close-winding just where the tangent will strike (something very troublesome to achieve and, I would argue, technically unsatisfactory because if one or two of the close-wound turns move slightly along the string, the main open-wound section is in danger of becoming loose). What I am doing on this instrument is something different: I start the winding just to the right of the tangent, so that it strikes the bare core wire. I find this avoids some of the ‘fizzle’ that wound strings sometimes produce. I have no evidence that it was done in the seventeenth and eighteenth centuries, but it does seem to work.

Mention of the seventeenth and eighteenth centuries reminds me to say that we do not really know when over-wound strings were first used on clavichords. There is some evidence that overwinding was used on lutes as early as the second half of the seventeenth century; but overwinding gut with silver wire is a very different thing from winding metal on to a metal core in such a way that it does not come loose. In fact, makers use various tricks to improve the adhesion of the covering wire: for example, I roughen the surface of the wire with quite coarse abrasive paper. It also helps to flatten the wire at the start and end of the winding by tapping it lightly with a hammer against some kind of anvil.

How high to take the covered strings? This is a matter of taste: there is a region where covered strings are essential, then, higher up, a region where plain wire or covered strings are both possible. The aim is, obviously, to achieve a transition to plain wire without any audible break in the sound: I usually find this means taking the covered wires up almost as high as possible in the compass of the instrument.

Before covered strings, it is probable that clavichord makers fitted twined strings, though hard evidence of this, too, is lacking. Making over-wound strings requires some kind of machine. Mine is driven by electricity and uses a foot-switch to start and stop, thus freeing both hands to operate on the wires, but making necessary the kind of hopping gait that I referred to at the beginning of this bulletin.

To return to Contents, click here.