Those interested in the flute may find the effects of dispensing with the conventional cork and crown assembly in favour of at least some of the metal stoppers tried in this study, both surprising and intriguing.
Stopper design used in these trials
The design is that of the stoppers machined out of a plastic called Delrin by the flute maker Robert Bigio. They are cylinders 16.8mm in diameter and 8mm in length with a circumferential groove near the front face (the face nearer the embouchure) housing a neoprene rubber O-ring which stands proud of the surface of the stopper (see illustration).
The O-ring presses against the interior wall of the head and holds the stopper in the correct position - ie the front face 17.3mm from the centre of the embouchure"'. There is a circular groove cut in the front face, which improves performance.
However, for some of the stoppers in this study, the length and therefore the weight was varied from 8mm, because it was found that the weight has a critical affect on performance.
Since the whole cork and crown assembly has been dispensed with, the top of the head is left open. In these tests, a perforated cap of African blackwood held in place by an O-ring has replaced the metal crown.
Choice of metals
It was decided to make the stoppers out of metals and alloys (instead of Delrin), covering a wide range of densities, with aluminium (2.7g per cubic cm) being the least dense and tungsten (19.1g per cubic cm) the most. Why was density chosen? It appears to be important in the choice of metals for the tubing of flutes and is measurable with simple equipment.
Other plugs made partly out of metal have been made by Richard Stagg (see 'Beyond The Button' in Pan, March 1997).
*So far as I have been able to find out, the substitution of the cork by a cylindrical metal stopper held in position by a rubber O-ring was first proposed and implemented by Ernest J Eggs (1914-1983), who was a technician working for the Admiralty in the 1960s. His son, Leslie R Eggs (my informant) is a flute maker and repairer, and developed the idea. The main object was to do away with the cork, which, they believed, damaged the tone of the instrument by dampening or eliminating the vibration of the head piece walls - an opinion that is still current.
To change to a metal stopper can be quite striking. I found the first trials with titanium and lead stoppers very surprising, not to say fascinating. Each metal seems to be distinctive - varying in responsiveness, power, resonance and other tonal qualities difficult to describe. You will hear sounds from your instrument that you have not heard before.
To me (the player), the metal stoppers in general imparted a more robust and direct tone than the Delrin stopper and this contrast was even more marked compared to the traditional cork assembly. As compared with Delrin, there was perhaps a loss of delicacy.
Limitations of tests
The results given below are those discernible to the player as distinct from the listener. The tests were carried out on a silver flute but occasionally a brighter head piece was used. The effect of the stopper varies with the head used.
In many cases, the experienced listener also can detect changes in the tone of a flute when the player changes the stopper to one of a different metal.
Notes on some of the metals and alloys
Unless otherwise mentioned, the stopper length is 8mm.
Aluminium (density: 2.7g per cubic cm)An 8mm stopper with the back face hollowed out (weight: 3.4g) was very responsive but the result was wild and unfocussed. An 11mm stopper (weight: 6.3g) gave a much improved result, with a light airy tone.
Titanium (density: 4.51g per cubic cm)Results somewhat like aluminium but strikingly resonant and with a better centred sound; an 11mm stopper (weight: 9.7g) is better than 8mm (weight: 7.1g). For high aerobatics I suggest that titanium is the best solution.
Zirconium (density: 6.49g per cubic cm)The sound is clear, bell-like and very resonant throughout. An llmm stopper (weight: 14.7g) seems to be a strong candidate for the best all-round performer.
Tin 96.5% Copper 3.5% (density: 7.35g per cubic cm)Uninteresting. Tone inflexible and lacking resonance.
Brass (density about 8.55g per cubic cm) Unremarkable.
Copper (density: 8.96g per cubic cm)Bright and resonant. Careful listening reveals what I can only describe as 'brassy' sounds - reminiscent of the comet or trumpet. Very interesting. Copper discolours very quickly; gold plating is the remedy.
Sterling Silver (density: about 10.38g per cubic cm)Similar to copper but slightly more husky and less responsive. Lacks the 'brassy' quality of copper.
Lead (density: 11.84g per cubic cm)I was very surprised the first time I tried a lead stopper. The tone is unusual to say the least. The bottom octave is rich in quality - almost reedy; the second is bland; and the third has an attractive veiled quality. It is quite resistant and the tone is inflexible: using a bright sounding head improves these qualities. The metal is soft and it is easy to dent a lead stopper. Warning: lead and its compounds are poisonous and lead is leached out when the metal comes into contact with water. Although it is difficult to understand how its use in a stopper could carry a risk of poisoning, perhaps gold plating is advisable if a lead stopper is to be used regularly.
Hafnium (density: 13.1g per cubic cm)Similar to but darker in sound than zirconium. A stopper of 8mm (weight: 21.1g) gives very good results. A stopper of llmm (weight: 29.6g) is powerful in the low octave but the tone is rather dead and inflexible elsewhere. (Hafnium, which was only discovered in 1923, is commonly used in nuclear power stations where it is valued for its ability to absorb neutrons.)
Tantalum (density: 16.1g per cubic cm)Somewhat similar to hafnium, but more resistant; might suit a player who likes a dark tone (weight: 27.4g).
Gold 22K (density: 18.56g per cubic cm)This is a gold/silver alloy (weight: 28.1 g). It gives a very beautiful, rich, low octave but is otherwise rather stiff and inflexible. Somewhat disappointing (and expensive too!).
Tungsten (density: 19.1g per cubic cm)Powerful and responsive but with an arid, empty, tone (weight: 29.7g).
Tungsten 75%/Copper 25% (density: 14.8g per cubic cm)A great improvement over tungsten. Good response and a robust sound with again a touch of 'brassiness', no doubt due to the presence of copper (weight: 23.7g). Discolours quickly: gold plating is the remedy for this. Might be interesting for the jazz flutist.
Titanium, Zirconium and Hafnium
These three metals (belonging to Group 4 of the Periodic Table Of Elements) used as a material for flute stoppers seem to display a family resemblance: a clear, bell-like sound and good resonance. The tone darkens through titanium to zirconium to hafnium. (A Flute of which the tube is titanium and the keys silver has been constructed by a well-known maker in the USA; the result is said to be excellent.)
Trials on an African blackwood instrument show that different metal stoppers give different results with the piccolo, too. The stoppers used were of copper, aluminium and sterling silver. An aluminium stopper of 1O.7mm diameter and 7.7mm length (weight: 1.69g) gave improved responsiveness compared with the cork assembly provided by the maker of the instrument. The other two metals produced a more robust sound, particularly in the bottom octave. I found that the aluminium stopper made the very high notes easier to sound and hold, which should appeal strongly to all piccolo players. However, two experienced players using another piccolo did not confirm this advantage while a third player playing on another instrument did, so the position is unclear.
Influence of stopper weight and metal hardness on head characteristics
The importance of weight is exemplified by a series of stoppers made out of zirconium:
1) 4.8mm; 5.8g
2) 8.0mm; 10.7g
3) 10.0mm; 12.1g
4) 11.0mm; 14.7g
5) 12.0mm; 15.6g
6) 13.0mm; 17.1g
7) 14.0mm; 18.3g
The first and lightest stopper - surprisingly, perhaps - gave a poor, shallow sound with a weak low octave. With increasing weight (and, concomitantly, length) and resistance, the performance improved up to stoppers 4 and 5, which both gave a clear, reasonably powerful and wellbalanced tone (between the octaves). It is difficult to choose between them. With the heavier stoppers (6 and 7), the resistance is greater; the low octave increases in power but becomes muffled, and the other octaves lose resonance and flexibility.
1) For each metal/alloy used there is an optimum stopper weight.
2) The optimum stopper weight for each metal/alloy may vary between different head pieces (to complicate matters...).
3) The optimum stopper weight for each of the different metals/alloys is a function of the density - the greater the density, the greater the optimum weight.
It would be necessary to repeat something like the zirconium trial with a few metals/alloys (the more promising ones since our interest is musical rather than scientific) and a variety of heads to provide support to these theories.
It is tempting to attribute the remarkable quality imparted by a lead stopper to the softness of lead, and the powerful tone imparted by tungsten to its great hardness.
The cork is only 17.3mm from the centre of the embouchure so it is not surprising that substituting other stoppers changes the playing characteristics of the head. Whether a particular change is beneficial or not is a matter of personal preference.
Before the 1970s, one bought a Flute and stayed with it until one changed the instrument. During this decade, the idea of improving it by changing the head piece gained ground, partly as a result of the design changes to the head introduced by Albert Cooper. But now, changing the stopper (and even the crown) can change the playing characteristics of the head. (This is far from saying that you do not need a well-cut head to begin with.)
The trials described above only scratch the surface. When one considers materials such as nickel silver and platinum (to name only two) that have not been tried, the permutations possible with other alloys and the fact that small admixtures of impurities or varying methods of production can alter some of the properties of a metal, the field for experiment seems open-ended.
The shape of the stopper in these trials is cylindrical with a flat face, but other shapes and non-metallic materials have been tried by Richard Stagg (see 'Beyond The Button' in Pan, March 1997).
I have been asked which is the best metal to employ. I can answer only for myself: a zirconium stopper weighing 14.7g gives the best all-round result on my flute compared to any sort of stopper (including the cork) that I have tried so far.
Published in "Pan" the magazine of the British Flute Society" in March 2003.