Jump to content

why are reed tongue slot walls not parallel ?


Recommended Posts

  • Replies 36
  • Created
  • Last Reply

Top Posters In This Topic

Here's my (non-expert) understanding:

 

Free reeds actually produce tone by cutting the air stream (not through interaction of air with the vibrating reed). This means that the shape of the wave form depends on the amount of space around the reed for air to pass.

 

If the reed shoe had a perfectly parallel slot then most of the air would escape at the tip of the reed through an essentially rectangular opening.

 

If the reed is somewhat trapezoidal and the slot has a certain venting outward then more air will escape earlier in the swing cycle of the reed with air passing along the sides as well as the tip.

 

Stepping way out of my expertise I'll speculate that accordion reeds use a more parallel slot which is also thinner. Classic concertina reeds have a thicker slot with some relief in both length and width.

 

So experts--have I understood reed function correctly?

Link to comment
Share on other sites

Hi all,

 

most modern Italian Accordion reeds do have tapered slots and with trapezoidal slots and tongues. Nearly all of the mid range reed frames are made of 3,3mm Dural aluminium.

 

Best regards, Johann

Edited by Johann
Link to comment
Share on other sites

All will be revealed according to Wim:

 

http://www.concertinaconnection.com/concertina%20reeds.htm

 

I'm not an acoustic engineer but Wim's explanations make sense to me. My experience in repair and setting up concertinas to play to their fuull potential has given me lots of encounters with different reed shoes from different eras and different makers. I think the unparallel sides may also influence how quickly a reed tongue dumps its air and can continue its cycle. Perhaps a factor in response.

 

The jeffries and Wheatstone reed assemblies that most of us revere have distinctive broaching to their reed shoes.

 

Greg

Edited by Greg Jowaisas
Link to comment
Share on other sites

All will be revealed according to Wim:

 

http://www.concertinaconnection.com/concertina%20reeds.htm

 

Thanks Greg for the link to Wim’s explanation of how free reeds operate. I read this explanation, liking very much most of it, and in particular, the care with which Wim exercises in explaining details.

 

I would like, however, to make some further clarification, and my comments will perhaps arouse more interest in those who have a more esoteric drive to understand more comprehensively what’s going on, from the perspective of what Physics has to offer.

 

As I say, Wim’s explanation goes far in explaining the state of the art in the overall mechanisms, but from a fluid dynamical point of view, there’s one important omission, and that is the role of dynamics, or what amounts to the same thing, mass inertia. In Wim’s explanation of the “swing cycle” of free reed operation, he includes what he calls “tension energy,” which I gather is a direct translation from another language, and I believe he means what we normally refer to in English as “potential energy,” which is an essential part of the vibration, but what Wim omits is what we call in English, “kinetic energy.”

 

As with all oscillatory behavior, a complete understanding of the phenomenon cannot be achieved without including the interplay between potential and kinetic energies. Thus, the vibration of the reed tongue is just like any other vibration, wherein – to first approximation – total energy is conserved, with continual exchange between potential energy and kinetic energy. Take the motion of a swing. At the highest elevation, when the swing stops, all the energy is potential (in this case, due to gravitational attraction), with zero kinetic energy, and at the lowest elevation, where the swing velocity is maximum, all the energy – the total energy – is kinetic, with zero potential energy. At intermediate points, there is some mix of potential and kinetic energies. The reed vibration is similar, approximating the vibration of a cantilever. With this “first order” explanation, we now introduce the fact that a relatively small amount of energy is lost from the system, in the form of friction and sound energy, as Wim points out. Thus, motion is maintained by means of the potential energy of the pressurized air in the bellows, or the relatively pressurized air in the atmosphere, when the bellows pressure is less than atmospheric.

 

Wim seeks to explain reed tongue vibration only by reference to the static forces exerted by pressure, but we know from Newton’s Second Law that forces also arise from acceleration (or de-acceleration) of mass. This is the meaning of a dynamic system. As I try to emphasize now and then on this forum, acoustics is a dynamical phenomenon, involving time derivatives, and any complete understanding of its effects must include dynamic (time dependent) effects. Think inertia. Thus, an explanation of a dynamical event that contains only static pressure forces is necessarily incomplete. This incompleteness becomes too much to bear when we arrive at Wim’s explanation of the “4th position,” which I quote below:

 

“But, the pressure P1 build up is less than it was in the 2nd position, and because of this it will allow the reed to continue up to the first position. If the airflow obstruction would have been the same as in the 2nd position, the swing cycle could not be completed. A reed swing cycle is not symmetrical. The amount of obstruction in position 2 is larger than in position 4. Also, the rest position of the reed is not exactly parallel to the frame. The tip is slightly above the frame. This means that the tension energy in the reed helps it move through the blockage.”

 

I think it’s fair to say that this passage needs complete overhaul. Here’s how I would describe the state of the reed in Position 4, as it moves upward:

 

At Position 4, the total net pressure force acting on the reed tongue is (P1 – P2)*A, where A is the reed area, and this is the same net pressure force acting on the reed tongue in Position 2. The difference here is that the tongue is moving upward, and thus, it’s inertia carries it through Position 4, against the net pressure force and against its own spring force, just as a swing is carried from a low position to one higher, against gravity and against a small nudge, say by wind, against its motion.

 

It’s clear that, when one invokes all the dynamical elements, the explanation becomes much simpler and intuitive. The convoluted, and erroneous, explanation given by Wim results primarily from his attempt to make only static pressure forces account for all the tongue motion.

 

There are other, less important, aspects to Wim’s explanation that I disagree with. Under “Bellows pressure,” he states that the pressure applied to the bellows is the force applied by the player, “divided by the size of the bellows: P = F:S” I understand his meaning, but it would be a simple matter to be more precise, with dimensional correctness, by stating that the pressure is approximately the force divided by the end plate area: P = F/A. For a “perfect bellows,” one that is infinitely flexible and frictionless in the direction of motion and infinitely rigid in the direction normal to motion, this equation would be exact.

 

Under the explanation of the 2nd position of the swing cycle, Wim states that, when the tongue arrives at this position, “In doing so, the reed causes the pressure P1 to increase considerably.” In truth, P1 doesn’t increase. As we stated above, this pressure is determined only by the players force on the bellows, and we can assume that it’s constant. I believe this mis-statement arises simply because his explanation does not recognize that it is a pressure difference that forces vibration, and not simply P1. At Position 2, and Position 4, the net pressure force acting on the reed is, to first order, (P1 – P2), which is the maximum pressure difference attainable. At positions other than 2, or 4, the pressure difference acting on the reed tongue is (P1 – Px), where Px is intermediate between P1 and P2, determined by air flow dynamics, and can be approximated to first order by the (quasi-static) Bernoulli Equation.

 

The Bernoulli Equation is very helpful in explaining the swing cycle of the reed tongue. In fact, the Bernoulli Equation, for steady flow, is actually a statement of Conservation of Mechanical Energy, in the same way that the reed tongue vibration, to first order, conserves energy, shifting the mix of potential vs. kinetic energies. When air is in motion, it has kinetic energy, and thus it’s potential energy, measured by pressure, is decreased. That’s Bernoulli, a beautiful understanding of how Nature works.

 

Thus, when the reed tongue is in Position 3, below the slot, air flow above its top surface means that the pressure here is (from Bernoulli) somewhat less than P1, and the lowest it could be (with a maximum velocity) is P2. The important term, the pressure difference across the tongue, is (Px – P2), somewhat less than (P1 – P2). Here, a combination of pressure difference and kinetic energy of motion propels the tongue downward to its minimum position, where it's kinetic energy becomes zero.

 

In section, “Reed shapes and frequencies,” there appears:

 

“Unlike other reed instruments (e.g. clarinet, sax, etc.) the vibration of a concertina and accordion reed itself hardly produces any sound.”

 

I don’t think this is an entirely correct distinction. In all these instruments, the reed initiates a pressure pulse that results in the sound that we hear, and the distinction with the free reed is much more subtle. With a clarinet, for instance, the reed acts as a pressure valve that closes the end of a quarter-wave tube at precisely the right time, causing air molecules to collide against it, in the same way they collide against the free reed tongue, providing a pressure pulse that contributes energy to the resulting standing wave vibration in the tube, and some of this energy escapes the tube in the form of sound. The air column in the clarinet thus provides a storage mechanism whereby traveling wave pressure pulses can attain perhaps higher magnitudes than those attainable by the free reed. (I don't know if this has been proven by the way.) The free reed has no similar mechanism by which to bounce back pressure pulses in order to grow amplitude. But even this isn't the whole story, because of the relative areas involved. The wind instruments have openings to the atmosphere that have more area than the total area of the free reed. (I'm assuming this last statement.) But since the free reed in fact shuts off air flow, causing a collision with air that produces pressure pulses, why would you say that the reed "hardly produces any sound?" This is all the sound the little thing has! And many a fiddle player I know think it's too much sound! Thus, simply speaking, both reeds produce a pressure pulse by colliding with air molecules, causing pressure pulses, which result in sound that we hear. The way I would state the primary difference between the free reed and the reeds of these other (beating) reed instruments is that, in these other reed instruments, a vibrating column (tube) of air – not the natural vibration of the reed - defines the frequencies of the musical tone, whereas in the free reed, the frequencies of the musical tone are defined primarily by the natural vibrational frequency of the vibrating tongue itself. An additional distinction arises from the quantity of “useful mass” set into vibration. A vibrating air column is much more effective than is a vibrating metal tongue at converting its vibrational energy into sound waves that make it to our ears. Hence the difference in volume between a trombone, for instance, and a concertina.

 

In this same section, there also appears:

 

“The length and thickness of the reed determines its frequency. The width of the reed does not play a role in this. In fact, two reeds of the same length, material and thickness, but one of them twice as wide as the other, will swing in the same frequency. The width does however have an effect on its swing cycle, because the larger surface of the reed increases the air flow pressure (P1).”

 

I think the first part of this statement is accurate, although the last sentence is not correct, simply because the net pressure force scales as the tongue area, and the area scales the same as the other parameters (spring force and mass) that determine the vibratory motion. Thus, doubling the width doubles the area, doubles the net pressure force, doubles the mass and thus the inertia force, and doubles the spring restoring force. There are no other parameters, to first order, that determine the motion, and thus, the two tongues vibrate in theoretically identical ways, with the same swing cycle.

 

In summary, I again applaud Wim’s attempt to explain to the layman just how a free reed works, but I think we all need to bear in mind that there are many serious studies going on by devoted physicists on how these devices really work, and our total understanding of their operation is increasing. With this physical, reproducible understanding, we see a beautiful exposition of Nature’s laws, and I hope my comments are taken by all here as only an attempt by me to do what I can to entice a broader curiosity into such of Nature’s phenomenon.

 

Best regards,

Tom

www.bluesbox.biz

Edited by ttonon
Link to comment
Share on other sites

Tom, thanks for the very comprehensive explanation. The bits I understand are very elegant and give meaning to what, has been up until now, merely an instinctive understanding of the process. I remember the Bernoulli effect from a disk drive I had many years ago (funnily enough called a Bernoulli Drive) where the head floated above the magnetic platters at very high speed.

Thanks again, I am going to study it in much greater depth.

Andrew

Link to comment
Share on other sites

Hi Andrew,

Thanks for the comments, though I must say that my post really only points out where I disagree with Wim's explanation, and I have not attempted here to give my own complete description of reed operation. But in my PICA article, you can find some other perspectives, if you're interested, the link of which is: http://www.concertina.org/ica/index.php/pica/subject-index/38-articles/87-reed-cavity-design-and-resonance

Best regards,

Tom

Link to comment
Share on other sites

I have read lots of copy about why some of the English concertina reed frames are beveled on the off-side of the reed slot. I say some, and not all, because not all English reed frames were beveled on the off-side. I think most statements on this subject are honest attempts to explain something observed without the benefit of actual experimentation to prove or disprove the theory.

Today there are reeds being made that have straight reed tongues. That is to say non-trapezoidal as with the Italian accordion reed. These parallel sided reed tongues are mounted in straight sided reed slots with no bevel on the underside. They start easily, have good dynamic range and have very good pitch stability. So, while the underside bevel may improve reed function, I am not convinced that this is all that important.

Link to comment
Share on other sites

I think most statements on this subject are honest attempts to explain something observed without the benefit of actual experimentation to prove or disprove the theory.

 

Hi Harold,

 

I mostly agree, though we can elaborate. We often run into musical phenomenon that, we all agree, must have a basis in physics, but for various reasons, have never been subject to enough scientific experimentation to establish a widely accepted explanation among acousticians. In such cases, the explanation, or explanations, that win popular support among makers and other interested parties is sometimes the result of a popularity contest.

 

In the case at hand, I wouldn't go so far to say that vent beveling has no discernable acoustic effect, but I think the situation is probably more complicated than often assumed. When this question came up in the forum organized by Chris Ghent several years ago, I presented a physical analysis, based on order-of-magnitude reasoning, from basic fluid dynamics. At that time, as I recall, Dana was finding some useful effect with beveling, though I don't know how far he went with it, as I have been mostly out of contact with this group recently.

 

In that analysis, I suggested that the underlying physics does not rule out some effect, only if the pitch of the reed is sufficiently high. With my rough calculations, I found that, above a frequency of about 1 kHz, there may be some advantageous effect, but for frequencies less than that, my reasoning suggested that there probably wouldn't be. The cut off around 1 kHz is of course not as important as the conclusion that there will be some frequency range, above which the beveling will influence the propagated sound, and below which, the sound produced would have no way of knowing that the beveling even exists. In addition, with the higher pitched reeds, my analysis says nothing on whether beveling can produce an acoustical effect that the human ear can even detect.

 

From a simple point of view, an analogy is to consider the effect of speaker size on the propagated frequencies of the sound spectrum. Small speakers (tweeters) can efficiently send out sound only above a certain frequency range, and for the lower frequencies, they are not very effective.

 

Stepping back to the more general, I have great respect for the makers of musical instruments, but of course, not equally among all makers. If a maker who has impressed me as one who can think critically claims the existence of an acoustical effect, I tend to accept what he/she says, when their claim is based on their own experimenting that such and such a feature has a such and such acoustical effect. When that maker goes beyond simply stating the existence of the acoustical effect and states his/her physical reasonings behind the effect, I feel qualified enough to analyze their explanation in order to see if it jives with my own understanding of fluid dynamics. Such events can then have mixed outcomes. If my analysis leads me to disagree with the physical explanation offered by the maker (or otherwise, by the one believing in the physical effect), my comments are sometimes taken that I don't think there really exists such an acoustical effect, which may not be true. There are important distinctions among 1) the existence of an acoustical effect; 2) the validity of the physical mechanism proposed that produces the effect, and 3) whether the effect can be discerned by a listener with normal hearing ability.

 

In short, concerning beveling of the reed slot, I can very well believe that such a feature has a discernable effect on the higher pitched reeds of the concertina, probably tending to increase volume, and perhaps also increasing the higher harmonics of their resulting sound spectrum. For the lower reeds, however, based on my own physical reasonings, I question that there does exist a discernable acoustic effect: I cannot find a physical basis for it. But I don't make these statements dogmatically, either, for the simple reason that my analysis may have errors. As is normal, such controversies can only be settled through continued experimentation and discussion among a varied group of participants, each with his/her own perspective.

 

Best regards,

Tom

Link to comment
Share on other sites

Testing is very hard. If one has been incredibly careful in designing a test, making it a test of only one aspect of design, and I am not sure I have ever achieved that goal despite trying hard and despite thinking I had in fact done so on a few occasions, there remains the issue of evaluating the results. Without expensive sound evaluation equipment and the experience to read it, you are left with very subjective opinion and at the mercy of environmental factors you do not understand. The only analysis aid I have is a primitive harmonics display called Tuneit. In order to make a good comparison between concertinas/reeds you need to make sure the ear or microphone is in exactly the same place in relation to the concertina each time, and the closer to the concertina the more critical that is. You need to make sure the concertina is in exactly the same place in the room, and I mean exactly, and the closer it is to any other object the more critical this is. The room may not change either, including not having anyone else in it during one test if they weren't there before. And as someone has to operate the concertina they need to stand identically in relationship to the instrument and be wearing the same clothes. Playing the same notes and using the same bellows pressure we shall assume though the latter is hard to guarantee and if you are really serious about the tests some mechanical device for operating the concertina might be best.

 

Given these issues with creating and evaluating a test, less testing gets done on anything other than an informal basis and after a while you learn not to make casual definitive predictions based on the outcome. The question, "what am I actually testing here?" can take up a lot of time and the other question, "what am I hearing here?" is a judgement call.

 

 

In short, concerning beveling of the reed slot, I can very well believe that such a feature has a discernable effect on the higher pitched reeds of the concertina, probably tending to increase volume, and perhaps also increasing the higher harmonics of their resulting sound spectrum. For the lower reeds, however, based on my own physical reasonings, I question that there does exist a discernable acoustic effect: I cannot find a physical basis for it. But I don't make these statements dogmatically, either, for the simple reason that my analysis may have errors. As is normal, such controversies can only be settled through continued experimentation and discussion among a varied group of participants, each with his/her own perspective.

 

I wish this did not run counter to my instincts about the way these things work. In my model the effect of a vertical reed window with a parallel reed would be to channel the air towards the descending tip, creating a higher amplitude of reed swing but not necessarily any greater air flow in the air packet size per Hz, which I equate with volume. I say no greater airflow because in a reed assembly with relief the air flows around the reed once it has passed the narrowest point. I would also see the return swing as being more impeded by the air coming down the slot. When there is relief in the window the returning reed does not come under so much pressure until the moment it breasts the narrowest point. This would effect lowest reeds least because they bend far enough to clear the slot completely and come under return pressure for a shorter period of their swing. Highest reeds are always in the frame.

 

I think parallel reeds have surfaced in concertina making for other reasons than performance, more for reasons of available machinery. If you are grinding blanks to fit reed windows then parallel is much the easiest way to go. They do have slightly more sail area at the tip, and that could be a starting asset. They also need to be slightly thinner in the highest reeds. They would pass slightly more air in each packet, and is that volume?

 

A relief effect test might not be so hard to do Tom, one could make a reed with none, measure it, then increase the taper slowly by scraping away the frame, measuring all the way. I'd volunteer but for a whole bunch of reasons that are way too good I am behind in my work. This is the moment I feel like the White Rabbit; so much to do, so little time...

 

 

Chris

Edited by Chris Ghent
Link to comment
Share on other sites

I feel like the White Rabbit; so much to do, so little time...

I briefly had the thought that we should start a thread titled "The Colony of White Rabbits"... But only non-colonists would have the time to contribute, so for now that idea is down near the bottom of my To Do list. ;)

Link to comment
Share on other sites

I think parallel reeds have surfaced in concertina making for other reasons than performance, more for reasons of available machinery. If you are grinding blanks to fit reed windows then parallel is much the easiest way to go.

 

Looking at this from the other way around, I wonder if the explanation isn’t that reed frames would originally have been easier to make with a trapezoidal slot. Even if the blanks were stamped out with a press, to finish them perfectly parallel by filing would be much more difficult, and what you don’t want of course is for the tongue to jam in the slot. With a slight amount of rake, you can eliminate this possibility from the outset. Idem for the reed tongues – if they and the slot have a slight lengthwise taper, they can be more easily adjusted, because the tongue can be advanced in the slot up to the point that it fits perfectly. If you try the same thing with a parallel tongue, you’d only get one chance of getting it to the correct width before scrapping it.

In wind instrument history, there are often instances of design features like this surviving until long after later technology had effectively rendered them obsolete.

 

Adrian.

Link to comment
Share on other sites

I agree the taper in the length of the reed could be seen as an old habit, an easy clearance adjustment method, and out dated today at higher levels of technology, but I find it very handy still...

 

The relief in the slot is another matter. The degree of relief is a pointer to more than just a need to make it easy for the reed to clear the sides. You would need very little to make sure of clearance, one or two degrees. That much more was built in, with all of the extra difficulty in making a diepress to move the metal further, means they saw some need to do so more than just clearance. So on industrial archeology grounds alone the relief in the slot is about more than just industrial legacy, and experience confirms this.

Edited by Chris Ghent
Link to comment
Share on other sites

Might it be to allow the reed to return more easily?

 

On the downwind stroke the air flow is pushing the reed and supplying energy.

On the upwind stroke the reed has to return against air flow. If it returns to a parallel slot it faces close to full air pressure as soon as it reaches the bottom of the slot and loses energy in overcoming this until it clears the slot.

 

I imagine this would make it more likely to choke, and also introduce more harmonics due to asymmetry in the swing cycle.

 

With a taper, air can bleed around the reed until it is almost back to start position thus the reed loses less energy.

 

On the downwind stroke the air pressure on the reed will also drop off more rapidly once it starts to move, as air bleeds around it. This will result in a less sustained push and less energy on the downwind stroke, again improving symmetry and tendency to choke.

Link to comment
Share on other sites

My jew's-harp has a slot with an edge in the middle, and a bit of relief on both sides. My understanding for the rationale (at least partially) is that having sharp edges in the airflow (on the reed & in the slot) creates more turbulence in the stream, producing more high harmonics for a fuller, brighter tone.

Link to comment
Share on other sites

Might it be to allow the reed to return more easily?

 

On the downwind stroke the air flow is pushing the reed and supplying energy.

On the upwind stroke the reed has to return against air flow. If it returns to a parallel slot it faces close to full air pressure as soon as it reaches the bottom of the slot and loses energy in overcoming this until it clears the slot.

 

I imagine this would make it more likely to choke, and also introduce more harmonics due to asymmetry in the swing cycle.

 

With a taper, air can bleed around the reed until it is almost back to start position thus the reed loses less energy.

 

On the downwind stroke the air pressure on the reed will also drop off more rapidly once it starts to move, as air bleeds around it. This will result in a less sustained push and less energy on the downwind stroke, again improving symmetry and tendency to choke.

My jew's-harp has a slot with an edge in the middle, and a bit of relief on both sides. My understanding for the rationale (at least partially) is that having sharp edges in the airflow (on the reed & in the slot) creates more turbulence in the stream, producing more high harmonics for a fuller, brighter tone.

Is everyone here even talking about the same thing?

  1. There's "taper" from the base of the reed to its tip, i.e., being normally wider at the base (where it's fastened) than at the tip (the freely swinging end).
  2. And there's the "bevel" or "undercut", in which the slot of the reed frame is wider at its "under" side than at the "top".

I'm a bit confused, but I think you two are talking about the second, while aybee and some others are talking about the first.

 

One point that I don't think has been addressed here is whether the first has an effect on the quality of the sound. Seems to me that it does, as I understand that the rare "clarionet" concertinas -- with a sound intended to imitate a clarinet, bassoon, or other woodwind reed instrument -- have reeds that are wider at the tip than at the base, often described as "spade-shaped". And I remember seeing in the book The American Reed Organ drawings of experimental reeds of even more complex design, including one in the shape (looking from above) of a cross and one like a forked snake's tongue. No idea what those latter two sounded like, though I believe they were intended to produce unusual sound qualities.

Link to comment
Share on other sites

Is everyone here even talking about the same thing?

  1. There's "taper" from the base of the reed to its tip, i.e., being normally wider at the base (where it's fastened) than at the tip (the freely swinging end).
  2. And there's the "bevel" or "undercut", in which the slot of the reed frame is wider at its "under" side than at the "top".

I'm a bit confused, but I think you two are talking about the second, while aybee and some others are talking about the first.

 

You're right about me-- I was talking about the second. I figured both were fair game, & don't know much about the first.

Link to comment
Share on other sites

The reason both tapers are being talked about is the first you mention, taper in the length of the reed, creates the same effect in a parallel slot as the second, taper in the slot, does with a parallel reed. The two tapers are inter-related.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...