Reed slot stepped profile

[4to5to6]

Please correct me if I’m wrong or missing something (I often miss the obvious).
 

It seems that the draft in the reed slot should theoretically vary from a lot for the high pitch reeds to almost none for the low pitch reeds in order to have all the notes respond the same and have even dynamics over the range of the  instrument.  Or is this the opposite?

The hand filing method would have to be much better than machining and punching but the amount of labour and skill levels involved are astronomical.

 

I’m not sure how these results could ever be achieved with step machining and punching.
 

Other factors affect the compromise between response and dynamics like how stiff (thick) the reed is versus its length (reed scaling).  Longer scale reeds will have a stronger fundamental and respond better than shorter, heavier reeds but this is limited as you want to keep the size of the instrument small so you have less playing pressure and more bellows travel for more control. I love the range of the larger instruments but they are so hard to play and control the bellows so yet another compromise on the size of the reeds.

 

Reed slots are beveled so the tongue will respond quickly but then “dump” their air quickly so you still have a good dynamic range.  The goal is even playing pressure and volume across the range of the instrument.  Not an easy task and there are many compromises to best reach this goal… reed stiffness, tongue length and width, draft relief, air gap size on tongue edges and then final voicing or the curvature of the tongue tweaking the response versus volume.

.

Edited July 30, 2023 by 4to5to6

[Clive Thorne]

2 hours ago, 4to5to6 said:

The hand filing method would have to be much better than machining and punching but the amount of labour and skill levels involved are astronomical.

5 axis CNC machine is what you need! - could do the tapers as part of the slotting process.

 

[4to5to6]

yes.  5-axis would be nice…  Or one could just mill them out upside down and clean up the small steps left with a file   Hmmm

Edited July 31, 2023 by 4to5to6

[alex_holden]

7 hours ago, 4to5to6 said:

yes.  5-axis would be nice…  Or one could just mill them out upside down and clean up the small steps left with a file   Hmmm

 

I've thought about it (the latter: I can't afford a five axis mill). It may save some effort at the vent filing stage but it would be more difficult to program and there are a few practical issues to solve. On some early prototypes I tried doing the opposite; I stairstepped the outside profile to form a rough bevel, but it slowed the machine down a lot and didn't work as well as I'd hoped so I stuck to milling straight vertical profiles and then a lot of manual post processing.

[Little John]

19 hours ago, 4to5to6 said:

Longer scale reeds will have a stronger fundamental ...

 

Reeds don't really have a 'fundamental' in the same way that strings and wind columns have a fundamental plus harmonics based on dividing the vibrating string or air column into half, third etc. (and thus introducing harmonics at twice, three times etc. the fundamental frequency). A cantilevered beam (which is approximately what a reed is) does have different vibration modes, but they are not integer multiples of the fundamental mode. That being so, if they played any significant part in the sound of a concertina reed they would be discordant.

 

LJ

[Chris Ghent]

18 hours ago, Little John said:

 

Reeds don't really have a 'fundamental' in the same way that strings and wind columns have a fundamental plus harmonics based on dividing the vibrating string or air column into half, third etc. (and thus introducing harmonics at twice, three times etc. the fundamental frequency). A cantilevered beam (which is approximately what a reed is) does have different vibration modes, but they are not integer multiples of the fundamental mode. That being so, if they played any significant part in the sound of a concertina reed they would be discordant.

 

LJ

Little John, what is your source for this..?

[4to5to6]

Please see the chapter called “Frequency and Pitch from: The Physics of Free Reeds by Colin Pykett:

 

http://www.colinpykett.org.uk/physics-of-free-reed-wind-instruments.htm

 

The free reed tongue oscillates at or close to its fundamental natural frequency, the frequency you would get if you gently 'pinged' it. There are also higher frequency modes of oscillation, though these are inharmonic (not harmonically related to the fundamental).
.

[Little John]

4 hours ago, Chris Ghent said:

Little John, what is your source for this..?

 

Chris, I note that @4to5to6 has already given a non-mathematical source for this (which I really must read more fully). Otherwise, a search for "vibration of a cantilever beam" should bring up many sources. One I just found is this: http://emweb.unl.edu/Mechanics-Pages/Scott-Whitney/325hweb/Beams.htm#:~:text=A straight%2C horizontal cantilever beam,vibrate at its characteristic frequencies. Just over half way down is a table derived from some scary mathematics. It shows the second and third modes of vibration (for this example) are approximately 6.27 and 17.5 times the frequency of the first mode of vibration.

 

LJ

[Clive Thorne]

11 hours ago, Little John said:

 

Chris, I note that @4to5to6 has already given a non-mathematical source for this (which I really must read more fully). Otherwise, a search for "vibration of a cantilever beam" should bring up many sources. One I just found is this: http://emweb.unl.edu/Mechanics-Pages/Scott-Whitney/325hweb/Beams.htm#:~:text=A straight%2C horizontal cantilever beam,vibrate at its characteristic frequencies. Just over half way down is a table derived from some scary mathematics. It shows the second and third modes of vibration (for this example) are approximately 6.27 and 17.5 times the frequency of the first mode of vibration.

 

LJ

 I am guessing that the amplitude of these modes is much lower than the "fundamental", so as to have little effect on the frequency of the reed passing through the reed plate, i.e. little effect on the chopping of the air flow.

[4to5to6]

I am mainly interested in the draft slot and specifically a stepped slot versus a tapered slot.  

[Chris Ghent]

On 8/1/2023 at 6:59 PM, Little John said:

 

Chris, I note that @4to5to6 has already given a non-mathematical source for this (which I really must read more fully). Otherwise, a search for "vibration of a cantilever beam" should bring up many sources. One I just found is this: http://emweb.unl.edu/Mechanics-Pages/Scott-Whitney/325hweb/Beams.htm#:~:text=A straight%2C horizontal cantilever beam,vibrate at its characteristic frequencies. Just over half way down is a table derived from some scary mathematics. It shows the second and third modes of vibration (for this example) are approximately 6.27 and 17.5 times the frequency of the first mode of vibration.

 

LJ

LJ, thanks. If you can understand that paper all power and respect to you!  I was intrigued by what you said because it runs counter to the generally accepted (not the same as saying right) understanding of harmonics around here, in my experience. This would say the first harmonic should be the octave. I just ran a G3 note through a spectrum analyser and it came out with the fundamental G3 at -6cents, and the first harmonic at G4-2.9 cents. The second harmonic is D5 at 0 cents, the third G5 at 0 cents, the fourth B5 at -15, Etc. It is not until the 8th there is a serious discord, an A6. This conforms with the very general statement, the higher up the harmonics the more discordant with the fundamental, and consequently the drive to increase the volume of the fundamental as it will drown out the discordant higher harmonics. They can also be diminished by wood structure. 
 

Please don’t think I’m saying you are wrong, I have no source to back up what I am saying other than finding evidence from a cheap spectrum app that roughly supports it. It could be confirmation bias. I imagine Tom Tonon would have good info on this. 
 

[Chris Ghent]

On 8/2/2023 at 7:50 AM, 4to5to6 said:

I am mainly interested in the draft slot and specifically a stepped slot versus a tapered slot.  

456, sorry for hijacking your thread. As an act of contrition I’ll address your question directly.
 

A long time ago I made a reed assembly using a machined stepped slot because it seemed like an easy way to adjust the draught in a repeatable way (as opposed to freehand filing). My recollection is the reed was a lower one, perhaps a B one tone down from middle C. The resulting reed worked perfectly well except it used more air than a conventional reed and seemed more mellow. I did not continue with that stream of thought. 
 

I remember there used to be an accordion manufacturer who trumpeted he was the inventor of the high output piccolo reed.  The accompanying picture showed a similar effect, created by using a round cutter and not extending along the reed very far but tiny reeds tend to move more at the tip so  I think it could be directly compared with a full recess on a lower reed. It occurred to me the stepped draught could be more efficacious on a higher reed where air use is not much of an issue.  
 

Drawing Dana Johnson into this discussion could only help because he is a very practical man with a lot of good science behind him and I know we have discussed this subject at length some time ago.  He is very good on draught angles and does them differently to most others. If he doesn’t join in, an email to him would get a good response. 

[Chris Ghent]

PS. I feel like I saw a Dipper concertina a long time ago that had the stepped draught slot.

[4to5to6]

Thanks Chris.  All very interesting info.  I've come across a lot of the info on the harmonic sequences of free reeds before in regards to reed filing techniques (getting rid of these nasty inharmonicities) but very little about the draft slot except for perhaps air consumption (concertina versus accordion reeds for example) but this is the first time I have seen machined, then stamped draft vent slots shoes.  And in this case, they are not surface mounted reeds but the regular dovetail design. 

 

I am still trying to breakdown or maybe best to say "visualize" the different aspects of how a reed works and what constitutes a great concertina reed.  In this case, I've heard that the 50s Wheatstone concertinas are of lower quality as they started compromising quality over manufacturing efficiency to cut costs and wondered if this was another one of these compromises or actually an innovation.  I don't have access to the 1955 Aeola BT in question anymore but the dynamics were actually not that bad considering the stepped draft slot design but did note the most challenging low F and F# reeds were of the traditional filed draft slot design so obviously it was considered necessary to spend more time on these challenging low note reeds.

 

Looking at the photos attached again... I am also very interested in how a lot of the 50s reed shoes were made of aluminum and how this is generally to also be considered a cost cutting compromise.  In contrast some of the earlier very high end instruments (specials, torts and amboyna) during the mid 20s golden era (don't quote me this) and 30s were the first ones to use aluminum shoes calling them "alloy" and considered them to be superior.  Of interesting note here... one of my finest instruments is an Aeola built during WWII with brass shoes when very few instruments were made and almost none with brass as it was a restricted material.  It is one of my best "keeper" instruments with it's factory 8-fold bellows and factory A440 tuning so basically untouched.  I often wonder if Wheatstone kept a skeleton crew of their best craftsmen together at this time building some of their finest instruments.  It does have hook versus riveted action but I can live with this because of it's superior tone, dynamics, balance and musical expressiveness, etc.

 

So is the step reed shoe draft slot a compromise?  Probably.  I didn't notice a sudden jump in dynamics with increased playing volume.  I'll have to do some side by side experiments with all else being equal or maybe two shoes say a B and Bb installed in an instrument to compare the responsiveness and dynamics.  Draft angles in the older superior instruments are very hard to analyze as they are very subtle and hard to measure therefore hard to reverse engineer.  So much of this information is lost with the passing of these fine craftsmen.

 

Yes, it would be great if Dana would jump in here.  I am sure he's done a lot of these experiments already.

Edited August 3, 2023 by 4to5to6

[Chris Ghent]

456, I suspect keeping the better craftsmen together during the war might not have been hard because they would have been older men. They might have had to do war work in the factory though. 
 

I’ll send Dana a prod..! 

[Dana Johnson]

Marcus had reeds made similarly but with the back relief cut with a circular cutter like a woodruff key cutter.  He was trying to mimic the tapered windows on concertina reeds in his hybrid instruments.  Wheatstone mostly created the taper as part of the punching process in relatively soft brass, finishing the window sizing with small broaches.

   I experimented with different methods to reduce the volume of a low drone reed, first by making a verynarrow reed,  then changing the relief angle, and finally using an end mill to reduce the thickness at the window similarly to the reed in the image to about .020 inches.  This was the only way that produced a very noticeable reduction in volume.  What seems to happen is that reducing the thickness  limits the distance the air flow can add power to the reed.  I once made a few reed sets with bell mouth shaped windows  similar to an old Jeffries I had.  This had the effect of changing the pressure / volume curve so that small increases in pressure had greater effect on the volume.  I  used curved scrapers to creat the belling which was actually quicker than filing.  These reeds worked very well, but I stopped doing it because I wanted more control on the dynamics.  
   One thing to keep in mind is how different parts of the reeds scale.  A long reed in a window of a given thickness swings through a shallow angle before it exits the back of the window.  A short reed in the same thickness has to swing farther to exit the window.  One way to attempt to equalize this across the reed lengths is to vary the vent angle, from nearly zero at the low end to much greater for the high end.  Very low reeds may need thicker shoes to get good volume.  I have a set of bass accordion reeds  that have windows in cast plates that are about .375 inches deep at the tip end of the reed. ( the plates are tapered so they are thinner at the root to save weight I expect ).  The other way you can do this is to use the same angle throughout but change the reed stiffness so shorter reeds bend more easily at the same playing pressure.

   Practically speaking, some attention to this issue is necessary to get sufficient volume of the  mid to higher reeds  without having to use more bellows force.  Getting this right produces very noticeable effects.  I found that regarding vent angles, each reed length has an optimum vent angle above which the reed loses its maximum volume and below which  volume at a given pressure drops off.   This also holds for making reeds thinner which will respond  at lower pressures, but max out at lower pressures as well.  
    With my reeds, I change the vent angle in steps about every octave and that has made for very even response  across the range.  
‘Reeds made by someone else  would  need their own set of angles.  There is no one right angle, especially since choices in reed profiling directly affects the issue.
   The milled out reeds in question are  in my opinion, a quick and dirty attempt to deal with this issue.  That seems in keeping with the later cost cutting attempts at the Wheatstone co.  

[Geoffrey Crabb]

Apologies to those who have invited me by PM to take part in this discussion. 

But I don't think I am clever enough.😉

 

However, the attached from 2012 may be of interest.

Reed slotting and forging. GC 2012.doc

[Richard Mellish]

15 hours ago, Chris Ghent said:

LJ, thanks. If you can understand that paper all power and respect to you!  I was intrigued by what you said because it runs counter to the generally accepted (not the same as saying right) understanding of harmonics around here, in my experience. This would say the first harmonic should be the octave. I just ran a G3 note through a spectrum analyser and it came out with the fundamental G3 at -6cents, and the first harmonic at G4-2.9 cents. The second harmonic is D5 at 0 cents, the third G5 at 0 cents, the fourth B5 at -15, Etc. It is not until the 8th there is a serious discord, an A6. This conforms with the very general statement, the higher up the harmonics the more discordant with the fundamental, and consequently the drive to increase the volume of the fundamental as it will drown out the discordant higher harmonics. They can also be diminished by wood structure. 
 

Please don’t think I’m saying you are wrong, I have no source to back up what I am saying other than finding evidence from a cheap spectrum app that roughly supports it. It could be confirmation bias. I imagine Tom Tonon would have good info on this. 
 

Let's try to resolve the confusion. I'll try to avoid the terms "harmonic" and "overtone" because those may or may not mean the same thing. And I stand open to correction from anyone with deeper knowledge of the physics.

 

Many oscillating systems have multiple possible modes of vibration, with different frequencies. With a taut string, the mode with two antinodes and a note in the middle will have approximately twice the frequency - but possibly not exactly because of complexities at the ends, limited suppleness of the string, etc. A violin player (more rarely a fiddler) can excite that mode by touching the string in the middle, damping any movement at that point. With normal playing, several modes will be excited at once. In that case, I am not sure whether interaction between them affects their freqencies enough to lock the higher ones to exact multiples of the fundamental. Anyone know?

 

A free reed is a very different animal. As has been said up thread, its higher modes of vibration will not be at exact multiples of the fundamental frequency. However almost all the sound that comes out of the instrument is produced by the fluctuating air flow past the reed. Each time the reed makes one movement there and back there will be one cycle of the sound wave. The air flow does not vary sinusoidally, so the sound will include harmonics of the fundamental frequency, but they will be integral multiples, with two, three or however many cycles for each there and back movement of the reed. Even if the movement of the reed includes any of its higher frequencies and those contribute non-harmonic components to the air flow, the resulting contribution to the sound is likely to be very quiet and un-noticeable.