Jump to content

Reed Chamber Length Experiment


Recommended Posts

Adrian suggested I post this short video clip over here in the Instrument Construction forum. I've recently been designing a C/G Anglo reed pan, and was wondering about optimum chamber lengths. My understanding is that below a certain pitch, reeds start to respond sluggishly unless they are contained in (or on the bottom side of) a chamber that is longer than the minimum necessary to physically contain the reed frame. If you look at vintage reed pans, you often find that the bass reeds are set in from the outer edge of the pan (I've seen this on both parallel and radial designs). I won't pretend to have reached any conclusive answers or understand theoretically what is going on, but I was able to observe the effect by building a simple variable-length test chamber on my tuning bellows (the idea was suggested to me by Geoffrey Crabb).

The reed in the clip is the lowest C from a C/G anglo. I repeated the experiment with the next half a dozen or so pitches. I found that the benefit of extending the chamber increases rapidly as the pitch goes down. Above a certain pitch the benefit was barely detectable. I didn't notice any reduction in responsiveness from making the chamber excessively long, at least in the pitches I tested, though there did seem to be a change in tone.

Note that concertina design is an exercise in balancing compromises. It isn't necessarily practical to make every chamber the ideal length. What I learned from this experiment is to prioritise the lowest notes because they benefit the most from extra length. (As it happens, my pan designs ended up looking pretty similar to vintage ones in the end, with a few minor tweaks.)

Edited by alex_holden
  • Like 1
Link to comment
Share on other sites

Very nice jig, Alex! I made something very similar when the Button Box was designing the Morse ESB (baritone anglo model), and then took the data I gathered (on lengths that seemed to work well in practice for various pitches) and tried to come up with an algorithm that described the data. I recall getting somewhere roughly with that, but given that I'd already done the empirical research which definitely worked on a prototype reedpan, there wasn't really a need to perfect an algorithm. This (as demonstrated in instruments on three successive full prototype reedpan designs with different chamber lengths) made an enormous difference in the performance and sound of the instrument.

I was guided by a Dipper baritone that had conveniently come in for repairs just as I was starting to think about ideal chamber lengths for reeds. I measured its chambers; the one thing I remember is that there were (unsurprisingly) some differences in actual length / air volume for chambers that seemed to work with hybrid vs concertina reeds, but also (unsurprisingly) the trend in each was pretty similar in the ratios of air volumes between pitches.

Edited by wayman
  • Like 1
  • Thanks 1
Link to comment
Share on other sites

(I then used these findings to redesign the reedpan for the Georgie baritone (English) model, lengthening as many of the chambers as we could make space for, and it dramatically improved the response of the lowest half-octave of that instrument! So these experiments with chamber length are very worthwhile.)

  • Like 1
Link to comment
Share on other sites

Because lower-pitched reeds are more pressure sensitive, and the size of the chamber affects the pressure with in the chamber, a chamber which is too small will create too high a pressure and affect the way the lower reed sounds. Because you can't easily change the depth of the chamber, length is one way to reduce the air pressure within the chamber.

  • Like 2
Link to comment
Share on other sites

Hi Frank, are you saying that in your experience, on bass reeds, increasing the depth of the chamber will have a similar effect to increasing its length? It would be easy to make all the chambers in a pan deeper, but that presumably has a negative effect on the higher reeds (more so on an English than on other types, hence why they are usually sloped).

I've seen the opposite effect with the highest reeds on a 40 button Lachenal anglo - they worked fine in the open air, but inside the instrument they became very sluggish. Moving walls around to reduce the chamber volume improved matters somewhat, but I couldn't get them to respond as fast as they did on the tuning bench. I suspect the chambers were too deep.

Link to comment
Share on other sites

1 hour ago, alex_holden said:

I've seen the opposite effect with the highest reeds on a 40 button Lachenal anglo - they worked fine in the open air, but inside the instrument they became very sluggish. Moving walls around to reduce the chamber volume improved matters somewhat, but I couldn't get them to respond as fast as they did on the tuning bench. I suspect the chambers were too deep.

I've got reedpans for a large Wheatstone Maccann from the 1920s, and reedpans for a Lachenal baritone anglo from who knows when, where on the left side, the plane of the reedpan is not parallel to the plane of the action board: the chambers get progressively deeper from the end with the higher notes to the end with the lower notes, and it's a pretty big difference (the lowest chambers are probably 1/4 or 3/8 inch deeper than the highest chambers).

I've never seen this done on the right side reedpan of a Maccann or anglo, though that's perhaps because the reeds are a bit less "in pitch order" on the reedpan. But on my G/D Jeffries, a previous owner (Robin, this was you who did this, right?) shortened the chamber of one of the highest reeds on the right side by filling the back end of the chamber with blu-tac, and that does seem to increase the response of the reed! (There's a thread here, from about five or six years ago, where Robin and Adrian discuss this at some length.)

Link to comment
Share on other sites

10 hours ago, Frank Edgley said:

Because lower-pitched reeds are more pressure sensitive, and the size of the chamber affects the pressure with in the chamber, a chamber which is too small will create too high a pressure and affect the way the lower reed sounds. Because you can't easily change the depth of the chamber, length is one way to reduce the air pressure within the chamber.

Frank, is it strictly the size of the chamber (the volume of air in the chamber) that makes the difference? In my experience, if the chamber is longer than the reed, but the tip of the reed isn't all the way at the one end of the chamber and the pad hole isn't all the way at the other end of the chamber, the extra length of the chamber doesn't make nearly as much difference, no matter how much longer it is. That makes me think it's less about the volume of air in the chamber - and hence air pressure, though that may be a distinct factor - as about how far the air travels.

But because the physical dimensions of the concertina are a limiting factor (if you want to keep it to a 6 1/4 or 7 inch instrument), sometimes making the chamber deeper is the only way to enlarge the chamber, and that does make some difference, just not as great a difference.

Link to comment
Share on other sites

Now I want to take a test bellows and variable-chamber-size jig and various reeds to different altitudes (with greatly different air pressures) and collect lots more data! ?

Imagine the other mountain climbers wondering what the heck you're doing up there, when you get to the summit and set up your test bellows... ?

  • Like 1
Link to comment
Share on other sites

Thanks for getting this going here Alex. One question that immediately sprung to my mind is that given the bass reeds on Jeffries instruments are shorter scale (a shorter tongue length for a given pitch) than say Wheatstones, does it follow that these reeds can therefore get away with having a shorter chamber length? In other words, was the choice of a shorter scale to do with a limited range of dies to stamp out the reed frames, or rather to space concerns and the fact that shorter more heavily weighted reeds perhaps work better in a confined space?

You certainly seem to be able to find the "sweet spot" using this jig, which in this example seems to be at the fullest extension of the chamber? But I wonder what happens if you continue the chamber length, is there a point where the response and sound start to deteriorate?

I was also intrigued by the pad arrangement you have and it strikes me that you might be able to resolve one of the issues in the thread Wayman was alluring to above:

My feeling is that the nasal sound has a lot to do with the far from optimum position of the pad hole, which lies almost over the reed in these instances, rather than at the end of the chamber. Filling in the clamp end of the chamber does help somewhat, but the hole is in most cases, just too far along the chamber, due to mechanical constraints on the action side of the pad board. Using your set up, perhaps you might be able to observe whether the sound of a given reed (say a LH reed, as these are where you mostly hear the problem) becomes more nasal sounding as the pad hole is moved along the chamber towards the tip? I once had a concertina made for me where the maker (no names!) had mounted the inboard reeds the wrong way around, so that the pad hole was over the tip. Needless to say they sounded quite horrid with a very distorted sound, and I had to do some quite drastic surgery to enable me to turn them around, whereupon they gave the usual nasal tone.

Anyway, thanks again for starting such a interesting thread.

Adrian

 

Link to comment
Share on other sites

On the opposite end of the spectrum,  the highest note on my G/Ds lies on the press side of a chamber sized for a reed the next shoe size larger.  Mostly the reeds never are paired with ones much different in size.  I found that while the longer of the two reeds was excellent in response, the smaller was not satisfactory.  In this case, I divided the chamber and cut down its length on the higher note.  It was instantly right up there with all the other reeds.  I haven’t had a problem on any higher pitched instruments because even my wonderful personal A/E doesn’t have a large shoe size break in the high end.

   While Jeffries reed pans are oriented with the parallel chambers perpendicular to the Button rows,  I had difficulties getting my long scale reeds on the Left to fit anywhere close to the buttons they were connected to.  I found that by turning the reed pans 60 degrees clockwise and anti clockwise ( left and right pans  respectively, ).  That not only could the longest reeds be placed in the longest chambers, but my shortest lever ( 3/4 inch on a Jeffries) was now 1-1/2 inches and much more similar feel to all the others. I also was able to put the pads for the important doubled notes on a c/g like the middle D and g/a Button’s so they were placed in similar positions relative to my hand so that they had the same timbre when played.   (Also placing the air pad under my hand to quiet it a bit.)  I have done them this way now for the last 28 years.  It also really helps with the low pitch instruments, where the reeds are that much longer.

dana

 

2E6DC3ED-8967-4922-B712-C9DE6A54C7A0.jpeg

Link to comment
Share on other sites

Dana, that's brilliant! I love it!

In the 38-key Jeffries, there's a button on the left which is Bb below middle C on the draw and Bb an octave higher on the push. They never feel or sound quite the same, and I've always wished I had a solution to that. I don't really want to monkey with a Jeffries reedpan in that way, but I'll keep this in mind for future efforts of my own.

Edited by wayman
I can't count.
Link to comment
Share on other sites

I have found using a similar setup to Alex’s that the “nasal” sound occurs at certain locations of pad hole along the chamber length.  Since most concertinas even with long reeds tend to put the reed dovetail close to the outside end wall, the usual place for the pad hole at the end of the chamber is generally a safe space.  For the reeds I have tested, mid and low range, there are usually 2 nasal sounding positions along the chamber, perhaps three if the chamber is a long one.  Their distance apart seems proportional to chamber length, which is why a secondary chamber shortening partition can sometimes solve this for a nasty sounding reed.

dana

Link to comment
Share on other sites

15 hours ago, adrian brown said:

You certainly seem to be able to find the "sweet spot" using this jig, which in this example seems to be at the fullest extension of the chamber? But I wonder what happens if you continue the chamber length, is there a point where the response and sound start to deteriorate?

With the reed shown in the clip, the response was still improving when I reached the maximum extension of my jig, but the improvement per mm of extension was dropping off. If I go up a few semitones, I reach a point part way along the adjustment range after which I don't get any more noticeable change in the response, however I did notice a softening of the tone and reduction in amplitude. I stopped the experiment when I reached a note where I didn't seem to get any improvement from extending the chamber further than the length of the frame.

At some point I hope to do some more extensive tests (looking at things like depth, width, radial vs parallel shape, hole size, pad lift), but it seems like it would be easy to go overboard and spend months on the project!

  • Like 1
Link to comment
Share on other sites

15 hours ago, adrian brown said:

My feeling is that the nasal sound has a lot to do with the far from optimum position of the pad hole, which lies almost over the reed in these instances, rather than at the end of the chamber. Filling in the clamp end of the chamber does help somewhat, but the hole is in most cases, just too far along the chamber, due to mechanical constraints on the action side of the pad board. Using your set up, perhaps you might be able to observe whether the sound of a given reed (say a LH reed, as these are where you mostly hear the problem) becomes more nasal sounding as the pad hole is moved along the chamber towards the tip? I once had a concertina made for me where the maker (no names!) had mounted the inboard reeds the wrong way around, so that the pad hole was over the tip. Needless to say they sounded quite horrid with a very distorted sound, and I had to do some quite drastic surgery to enable me to turn them around, whereupon they gave the usual nasal tone.

I wonder whether reducing the size of the pad and hole might help with very high reeds, by moving the front edge of the hole further from the tip of the reed. Maybe even change the hole to a oval slot (across the end of the chamber, not along it).

  • Like 1
Link to comment
Share on other sites

14 hours ago, Dana Johnson said:

On the opposite end of the spectrum,  the highest note on my G/Ds lies on the press side of a chamber sized for a reed the next shoe size larger.  Mostly the reeds never are paired with ones much different in size.  I found that while the longer of the two reeds was excellent in response, the smaller was not satisfactory.  In this case, I divided the chamber and cut down its length on the higher note.  It was instantly right up there with all the other reeds.  I haven’t had a problem on any higher pitched instruments because even my wonderful personal A/E doesn’t have a large shoe size break in the high end.

   While Jeffries reed pans are oriented with the parallel chambers perpendicular to the Button rows,  I had difficulties getting my long scale reeds on the Left to fit anywhere close to the buttons they were connected to.  I found that by turning the reed pans 60 degrees clockwise and anti clockwise ( left and right pans  respectively, ).  That not only could the longest reeds be placed in the longest chambers, but my shortest lever ( 3/4 inch on a Jeffries) was now 1-1/2 inches and much more similar feel to all the others. I also was able to put the pads for the important doubled notes on a c/g like the middle D and g/a Button’s so they were placed in similar positions relative to my hand so that they had the same timbre when played.   (Also placing the air pad under my hand to quiet it a bit.)  I have done them this way now for the last 28 years.  It also really helps with the low pitch instruments, where the reeds are that much longer.

I see, very clever. I actually rotated my pans the other way, like in this instrument Geoff Crabb built:

http://www.concertina.net/rd_new_crabb.html

It still gives the improvements in lever length (the worst one is the left hand thumb button). The disadvantage is that, like on a Jeffries, the lowest chambers are a bit cramped for length; the advantage is that roughly speaking the lower notes are on one side of the pan and the higher on the opposite, making it possible to slope the pan to get deeper chambers on the bass side. It doesn't work out perfectly of course due to the Anglo's illogical note ordering and the position of the drone.

Link to comment
Share on other sites

17 hours ago, Dana Johnson said:

On the opposite end of the spectrum,  the highest note on my G/Ds lies on the press side of a chamber sized for a reed the next shoe size larger.  Mostly the reeds never are paired with ones much different in size.  I found that while the longer of the two reeds was excellent in response, the smaller was not satisfactory.  In this case, I divided the chamber and cut down its length on the higher note.  It was instantly right up there with all the other reeds.  I haven’t had a problem on any higher pitched instruments because even my wonderful personal A/E doesn’t have a large shoe size break in the high end.

 

2E6DC3ED-8967-4922-B712-C9DE6A54C7A0.jpeg

This is a very neat modification Dana, but what slightly puzzles me is that using this divider, the chamber volume is approximately halved, if we compare it to the chamber immediately above it in this photo, which seems to me to have the same size reed frame. Does this mean therefore that it's the length of the chamber, rather than the volume that is a determining factor in the response? How do the two reeds in question compare in tone and response?

Cheers,

Adrian

 

  • Like 1
Link to comment
Share on other sites

3 hours ago, alex_holden said:

With the reed shown in the clip, the response was still improving when I reached the maximum extension of my jig, but the improvement per mm of extension was dropping off. If I go up a few semitones, I reach a point part way along the adjustment range after which I don't get any more noticeable change in the response, however I did notice a softening of the tone and reduction in amplitude. I stopped the experiment when I reached a note where I didn't seem to get any improvement from extending the chamber further than the length of the frame.

At some point I hope to do some more extensive tests (looking at things like depth, width, radial vs parallel shape, hole size, pad lift), but it seems like it would be easy to go overboard and spend months on the project!

That's true enough, but the data you would accumulate would prove invaluable for your subsequent designs. What I admire in your approach is the "hands on - ears open" method you're using, rather than trying to get all theoretical about it - it's certainly the way I've dealt with woodwind acoustics over the last 35 years...

Good luck and thanks again,

Adrian

  • Like 1
  • Thanks 1
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...