A friend of ours, the guitarist Giuseppe Brancaccio from Torre del Greco (Naples), once told us he had perceived a change in his guitar timbre after replacing the original bone saddle with a carbon fibre one. This information, coming from a specially sensitive performer, prompted us to study the matter in depth: we wondered if the material from which a saddle is made could actually affect sound quality to the extent of being noticed by the player, and if the same hardware/software technologies we were familiar with, would be applicable to this kind of inquiry.
The last and most practical question is:
does it exist an indisputably superior material for building the saddle?
To answer these questions we recorded the sound of E on the first open string (whose fundamental resonance is found at 329.7 Hz. The string was plucked on three different positions: at 1/8, 1/4 and 1/2 of the scale length, measured from the saddle (results relative to 1/2 plucking will be omitted, having scarce practical relevance). The performer obviously tried to pluck the string as uniformly as possible.
So we considered the point of view of a player which checks the sound quality at different plucking positions and with saddles of different nature. As explained in the first two chapters, timbre depends on the ‘recipe’ of sound, i.e. the character of the string fundamental tone, the amplitude and decay rate (or sustain), the nature of the harmonics (for convenience we do not go beyond the second harmonic at about 989 Hz), and finally the presence (or absence) of the tuning note (the resonance of the air) and of the soundboard fundamental resonance. These parameters needed to be evaluated, in order to try and associate acoustic performance to different materials.
Moreover, as explained in the first two chapters of the book, the string acts on the resonator conveying periodic force impulses to the bridge, which follow one another at the frequency of the fundamental. These force impulses incorporate in their shape, duration and amplitude the whole information contained in the string motion: frequency and amplitude of the fundamental, as well as the information concerning the whole sequence of related harmonics; as a consequence, also information about the position and mode of plucking. In other words, force impulses summarize what we have called the ‘recipe’ of the string motion.
But between strings and bridge (ideally considered as a part of the soundboard) is interposed the saddle, playing its crucial role as
interface between the string and bridge- soundboard structure.
Going directly to results, we will see that the saddle works like a sort of
damper between string and bridge-soundboard structure; in other words, like an element connected in series between the string and the rest of the resonator which, moving at the same speed as the soundboard, modifies the characteristics of the force impulse according to its own composition. Therefore the saddle, too, behaves like an oscillator mass—spring—damping.
Based on these theoretical assumptions, we examined saddles made from six different materials: carbon, two different kinds of bone, two recycled pieces of ivory, a bar of sintered material available on the market that we will call PT. Some of these materials are ‘natural’ (bone and ivory), others are the product of chemical synthesis (carbon and PT).
We consider unnecessary to report results of all the numerous analysis executed: we will just mention hereafter the most significant examples.
The two basic resonances (of air and soundboard) are present in the recipe of the sound, though the fundamental of the string motion—at 329.7 Hz—is very distant from both. As pointed out in Chapter
, the resonance of the air is always present in the recipe of the guitar sound, at least in medium-low registers, and represents a sort of 1 basic colour on which the sound texture is interwoven. Analyses highlight a difference, however small, between different materials: on the whole, the best response comes from carbon, which provides greater amplitudes at both 1/4 and 1/8. As for the decay time, within the normal span of action—plucking position between 1/8 and 1/4—the range of variation between materials is rather narrow, and we believe that divergences measured in sound duration at basic resonances are not relevant for the instrument response.
On the contrary, the components of the acoustic spectrum of the string (the fundamental and the harmonics) manifest a much more significant response with respect to amplitude and sustain, when the string is plucked at 1/8 or 1/4 of the length. As an example, the following graphs illustrate the response in terms of amplitude and decay time of the fundamental and of the first two harmonics when the string is pressed at 1/4.
Obviously, the values of amplitude and decay time depend on the amplitude of the spectral components of the impulses that the string conveys to the resonator through saddle and bridge. These attributes are in turn affected by the plucking position. Discrepancies between theoretical and measured values tell us that
only part of the available energy passes into the resonator, the remaining part being absorbed by the saddle, according to the characteristics of the material.
Similarly, decay rates depend on losses due to viscous friction within the material, but they also depend on the force applied to the saddle: this is a
non- linear phenomenon, especially noticeable in ‘natural’ materials like bone or ivory.
From this set of measurements we deduce that noteworthy differences exist between saddles made from distinct materials but, at the same time,
there is no material superior to others under all testing conditions and playing technique
Nevertheless, we were able to determine a
mean global performance
of the different saddle compositions. We classified materials by their performance on different plucking positions, with relation to parameters that affect sound quality. The best global results come from carbon and bone, and this for different reasons:
Carbon confers excellent amplitude to high pitched sound components—the response is therefore very ‘brilliant’—but slightly inferior to bone at the fundamental and the first harmonic. Sustain is outstanding at the level of both the fundamental and the first harmonic—namely, in the middle frequency band—while inferior at higher frequencies. It is worth pointing out that this is true on both plucking positions. Bone manifests excellent amplitude response relative to the fundamental and the first harmonic, but inferior to carbon at the second harmonic. Sustain is generally shorter with respect to carbon at mid-range frequencies, therefore implying greater losses due to friction within the material. Ivory behaves poorly under all situations, without any remarkable performance. PT is even inferior.
The overall evaluation of bone is very good, which justifies its traditional employ in guitar making. Nevertheless, we must consider that bone is a natural material, whose features vary on its provenance: of the two bone samples we have analysed, one was especially good while the other was very poor in front of our global evaluation criteria. Carbon—unlike bone—manifests similar characteristics in the two conditions of percussion, as well as between various samples. In addition to that, the following attributes, as shown by the analysis, are worth noticing: optimal amplitude response at high frequencies and optimal sustain at mid-low frequencies. This is why we advise using them in preference to traditional materials.
Now, the question may arise: do these conclusions apply to the nut as well? The answer is no, since the saddle needs to transmit most of the force received from the string to the resonator, while the nut must reflect the incoming wave, absorbing the least amount of energy. Therefore, a good material for the saddle is not quite so for the nut.
Our friend Giuseppe Brancaccio, the guitarist who—as previously mentioned—promoted this investigation, has long experimented with different materials, finding out that
brass and zinc are the best materials for the nut. Garrone guitars normally feature them in this component, which in fact is far from insignificant—as it may at first appear.
Furthermore, many of these instruments have a 250 g lead mass applied under the nut, which increases the overall reflection at the nut of the incoming wave.