Of Reeds, Shoes, and Acoustics

Note from Paul: This was an email by Dana Johnson in response to a question by newsletter subscriber Chris vonderBorch (mgccv@es.flinders.edu.au):

"Hi Dana,

I have been experimenting with concertina reed construction and tone production, prior to constructing one or 2 concertinas. I have also made a set of bellows for a Jeffries clone that I have, under the expert eyes of Peter Hyde, here in Adelaide. But I still have a long way to go, and enjoyed your article on concertinas!!

At this stage, I have a question that you may be able to answer: All the genuine Jeffries instruments that I have gutted and measured have rel. heavy reed shoes...2.2mm brass. Most others have lighter shoes. My Wheatstone and Lachenal shoes are 1.8mm. Do you think that the heavier shoes contribute to the Jeffries sound? Also, the Jeffries are heavily built, with the maple reed pans being thicker than , say, the Wheatstone. I guess the overall weight must contribute to the overall quality. Do you agree??

With thanx, Chris"

The characteristic sound of a "Jeffries" and certainly that of the best I've heard is very bright and crisp (again, there is a fair amount of variation among those instruments too). I have two nice ones that lean in that direction, and they are as you describe inside. Heavy reed shoes have a lot of inertia. The lighter steel reed weighs much less, and has a harder time trying to get the reed shoe to move with it. Imagine a child's swing set, first with the legs just standing on the ground. a child swinging on this will find the feet starting to lift off the ground as they swing higher, or when a bigger child starts to swing. now imagine the same set with large blocks of concrete cast on the legs. It now takes a much bigger child to lift them off the ground. In the first instance as the legs begin to lift the whole swing set will tilt forward, and the child will find that it is hard to keep swinging. (Try to swing on something that can roll!) The same thing applies to reeds (the child), and The swing set (the shoe). Heavy reed shoes help to reflect the energy back into the reed at all frequencies. Arthur Benade talks about wave impedance, and how vibrations reflect at boundaries where there is a significant difference in wave impedance between two connected objects.

So, reed vibrations tend not to be as easily transmitted to the reed pan. (the next part in the path of the vibration.) Lighter and or less dense reed shoes (as in Aluminum used in later wheatstones) allow more sound energy to make it into the reed pan, where it can be selectively damped. (wave impedance, like electrical impedance is frequency dependent). The added aggregate weight of all the reed shoes to the reed pan also increases its wave impedance.

Reed shoe thickness also affects how the reed vibrates, and how far it must bend to clear the bottom of the reed shoe and spill its' air. To some degree, reed shoes should be proportionally thicker for longer reeds. In my Converted Wheatstone / Hayden Duet, the low reed shoes are made of thicker stock than the high ones. How this aspect affects the sound I don't know. Also in an Oboe, the small air column that sits in the finger hole has enough mass when vibrated at musical frequencies to alter the filtering characteristics of the tone hole array. The added depth in the close confines of the reed shoe, as well as the added depth of the "port" in the reed pan beneath the reed, may also have a related influence, though in this case, the added thickness of the reed pan is so much greater than the difference in reed shoes, that it's influence would be the dominant factor here. I did some experiments with this, but had no definitive results. There are just too many factors to be able to extrapolate what happens with one reed to what happens in general.

The thickness of the reed pan also has a big effect, though in this case, weight is less an issue than the thickness itself. Thicker is stiffer. (also heavier, but the stiffness increases much faster with increasing thickness than weight does.) Stiffness is also relative to length, so of two pieces of material that are the same thickness but different lengths, the longer will be less stiff than the shorter one will be.

Now imagine the swing set again, but this time fasten the legs to a floor made of thin plywood, without any other support underneath, and the actual floor being quite large, so that the floor behaves like a large drum head. Now as the child swings, they will find the legs begin to lift on one side as the floor is pulled up, and the other legs will go down. A lot of energy will be lost in bending the floor. If you make the floor out of two layers of the same plywood, but space them apart with a light honeycomb material, so the floor is nearly the same weight, but is much stiffer, the floor will bend much less than before, and the energy lost will be much less.

This is a very frequency dependent condition though, and on something like a reed pan, it is possible to have selective frequency traps where the pan is more able to vibrate in some pattern that isn't hindered by the pattern of the chambers. (this would create something similar to the "formant" frequencies that are characteristic of vowel sounds, and makes a soprano " E" as recognizable as a Bass "E".) A stiffer reed pan will push these frequencies into a higher range, and behave more evenly below these resonance's. The added stiffness also affects progressively higher frequencies more, providing increased reflectivity, i.e.. less damping.

The net result of this is that the thicker reed pan increases the brightness of the sound, by increasing the strength of the higher overtones.

One thing to keep in mind is that reeds produce an interesting overtone spectrum. For reasons I haven't figured out yet, certain overtones are preferred, and will be stronger as a rule than others. This obviously contributes to the signature sound of reeds in the instruments we play. The two strongest tend to be the fundamental, and the second harmonic. Reasons for this are fairly clear, since each puff of air that makes it through the reed produces an overtone frequency, and the ear will perceive the puffs of different strength as part of an overtone series. The rate at which the reed decelerates as it reaches its extreme positions controls the mix of possible overtones. Since this is a "Heterodyne" process, where the frequencies produces are related to the arithmetical combinations of the frequencies that are produced directly, overtones that are produced by adding stronger overtone frequencies will be stronger than ones produced by combinations of frequencies that are of lower amplitude. The first three are likely to be strong, with the higher frequencies less so. The odd multiples 1,3,5,7 etc. tend to be favored, though construction influences can selectively strengthen or weaken them.

Reed pans that damp higher frequencies more (thinner or softer ones ) will also selectively affect the lower frequencies, (since the Heterodyne system works by subtraction as well as addition.) Conversely, lower damping can selectively increase the lower frequencies and provide a distinctive "timbre" to the system.

A general rule would be that thicker reed pans of a given material ( high damping materials are out) can produce a brighter Brassy sound as long as nothing else is working counter to that (like deep chambers or longer scale reeds).

Pay attention to chamber depth and length too. There is no "ideal" here. Each maker will have their own preferences. This has a great influence on the sound. I never was able to find any real "tuning" to the chamber length for any given note, but my own experiments bear out what Steve Dickenson told me, that there are often chamber lengths that are for some reason, just wrong. a little longer will have a big effect on the sound, as may a little shorter, where anything past that for a while will not have much influence on the tone. If you find a reed that no matter what you do to it, sounds nasty. try making the chamber a little longer. If you copy the chamber lengths of a good instrument, this shouldn't be a problem, though because the materials will never be exactly the same, what is right in one instrument won't necessarily be right for another one.

The last issue regarding Quality, If someone likes a sound, that is their measure of quality. Everyone has their preferences. I like a good Jeffries a lot, but prefer to play airs and some other Irish music on my own Anglo with it's long scale reeds and composite construction, because it sounds beautiful in the pianissimo, and wonderfully full at forte levels. I prefer my Jeffries for the bright and cheery jigs and reels, where it's added brightness helps to define the notes better. They are all to loud to be good to sing with. Other details of the action and how well the reeds are "voiced", and how well balanced etc. are things that I judge "quality " by. These are mostly independent of the parameters you asked about.

Have a great time with your instruments, and take from this what it is worth.

Sincerely,

Dana Johnson