Bore design nr. 3 was chosen to be executed. This design follows the example set by Sax's alto 24495 as far as its general form is concerned, but is a little bit less wide.
Right from the start three necks were made for this instrument. The first one has a single conicity and so it closely follows design nr. 3. This neck is 17 centimetres of length and sets the standard. The profiles of the other necks is curved and is both wider at the entrance but narrower over the major part of its length. This type of S-shaped bore profile we find in Sax's tenor 15676 and is commonly in use nowadays in both altos and tenors.
The body of the instrument was made out of pre-existing material. To meet the requirements of the design, a steel mandril with the correct taper was made, around which the body was hammered into shape. It hardly can be seen in the pictures (as it hardly can be seen in reality) but the mandril does not have one single conicity. On the contrary, its conicity diminishes at about three quarters of its length.
The lower straight part was made out of new brass sheet material. This part too – after having been made roughly into shape – was hammered into its definitive form around a steel mandril which has the same conicity as the the lower quarter of the upper straight part right above it. So this part is a direct continuation of that last conicity. As was pointed out before, conicity is getting narrower as we get down the bore.
At this point, it appeared that the cone thus shaped, only produced its lowest note in an unstable fashion, like an involuntary multiphonic. The octave, the duodecime and the double-octave were stable though. The instability appeared to be influenced by different factors: mouthpiece volume, pitch and the taper in the last twenty or so centimetres. To establish the precise causes of this phenomenon, different experiments with the mouthpiece volume and with the cone shape were executed, using experimental cardboard cones.
This is what I'd made. The lower straight part expands slightly before opening to the bell. The experiments with the cardboard cones had indicated that this tube would produce a stable lowest note, a written low-A
You will remember: at the acoustical length of low-A the size of the scrap standard bell corresponds well with the wideness of the lower part of the tube – a rather more slender cone.
It is always easier to think of something than to really make it. Is it, because our thoughts are too straightforward for grasping the complexity of things? Anyway, the straight alto that I'd built thus far, when equipped with a key mechanism, still turned out to be unreliable in its response to its lowest note. Yes, you could make it, but not securely. There was, as mentioned before, still a strong tendency to produce an involuntary multiphonic. And, curiously enough, when overblowing it, the octave was way too sharp, more like an octave plus a semitone. Also, the choice of a mouthpiece mattered and mattered too much to my mind. While a mouthpiece with its volume distributed in the one way might work more or less, another one with the same volume distributed in a different way might not. The same held true for the choice of a reed. The instrument had to become far more idiot-proof.
And so I started to rework its lower bore profile. New sets of cardboard cones were made. Also, to bottom bow was taken up again. After all, there might be an advantage in a bow in that it creates a virtual volume, as short and wide bows do.
Finally I settled on two cardboard cones which are acoustically equivalent. Both cones go down to a Bb (I decided to drop the low-A as it turned out to be near impossible to make it stable). The one is straight while the other one has a bottom bow. To my surprise, both cones turned out to have a bell shape which coincides rather well with the bell of the Adolphe Sax alto 24495, but longer. The same kind of shape, but streched out to reach to low Bb. The wideness of both cones proved to be essential for acoustical stability of the lowest note. But there was a difference as well between both options: the cone with the bottom bow was just better, more stable and with a fuller and opener tone than the straight cone. The straight cone on the other hand felt as if there was some kind of resistance added, as if you had to blow through a barrier before the actual sound came. It was rather difficult to blow softly and the cone typically produced wind to the tone.
And so I decided not to make a straight alto, as I had originally intended, but a curved one.
the same but with a bottom bow and a bell in the familiar shape of the first Adolphe Sax saxophones! Later on the bottom bow was also trimmed to a desired profile. Experimentally, it had been used with wax-fillings.
The bell was made out of PVC, a flexible yet strong material that can easily be shaped into the appropriate form. It produces a fairly stable low Bb.
Seeing the difference between this bell and the conventional modern one, and having learned about its influence on the acoustics of the instrument, we start to understand Sax saying that it is the form of the instrument that really matters.