tumbling reeds?

[ttonon]

HI Little John, you raise an interesting question that I hadn't considered, and my first response will be to rely on the differences in the mechanisms by which the two musical tones are produced.  

 

Let's first look at both cases starting with the time the tongue starts to hit the plane defined by the top surface of the underlying plate.  The beating reed hits the plate, and the free reed enters the slot.  

 

When the beating reed shuts off airflow, a rarefaction (reduced pressure) wave passes down the pipe, the resonator, at the speed of sound, hits the open end and is reflected from there as a positive pressure wave that returns to the reed, opening it, which immediately allows a puff of air on top of that returning pressure pulse, which then travels to the open end and returns as a rarefaction.  The latter hits the reed, but keeps it closed, bounces back, still as a rarefaction, up to the open end, reflects as a positive pulse that travels back to open the reed. 

 

I said all that in order to state the complete cycle, but the important thing is that the opening and closing of the reed is determined by the pressure pulses in the resonator, not because of the natural vibration of the reed itself, which for beating reeds is far above the pitch of the resonator. 

 

What do we conclude from this? Mainly that the time to establish a full cycle of reed vibration is basically the period of oscillation of the musical tone.  This time is largest with the longest pipes, having frequencies less than a 100 Hz, or on the order of ten milliseconds.  That's the very largest.  For smaller pipes, we are talking about periods much less than ten milliseconds.  

 

This is a rough analysis, but I think it gives an educated guess on why organ builders conclude that the free reed is much slower to start, having transient times many tens of milliseconds.  

 

For the free reed, much of the start transient occurs even after the tongue starts entering the slot, so we really don't have to bother with the complexities of vortex induced vibration.  Many studies have been done that prove this out.  I believe the reason lies in the fact that relatively very little additional energy can be incorporated into vibration amplitude during those starting times.  In fact, most of the swing cycle of the free reed is symmetric, in that just as much energy is put into the vibration as is taken out - except for a tiny moment in time when the tongue enters the slot from the top, where there is a net energy addition. 

 

The situation is too complex to discuss here, but to say it simply, the beating reed organ pipe has much larger energy available for the start transient.  The column of air within it is the primary vibrating element and the vibratory mechanism is more suitable for such energy levels.  That's why brass instruments can be so loud.

 

Regards,

Tom 

[Dana Johnson]

Chris mentions stainless on stainless as being prone to galling, but as always the alloy counts.  300 series stainless should not be used together with itself or others of the same series.  The nickel content is the culprit and on a molecular level will jump from one surface to the other.  You can use 300 series in combination with 400 series ( chromium only )without galling.  
   On a different note, I believe some Wheatstones were made with the best quality reeds advertised as having purposely rounded edges.

  Reed steel should be hard enough already.  I find the blue 1095 shim steel at the low end of acceptable, though it sounds just fine.  The UHB20C alloy I use ( also used for compressor valves as well as the accordion industry) is harder and retains its set much better, and shears much more cleanly than 1095, which tends to shift sides of the shear line along the cut leaving a sort of battlement shaped surface that needs to be filed smooth.

   I don’t think there is much of a fatigue problem with well made reeds without stress risers.  Shearing can cause micro cracking that may not be noticed or removed by edge filing, and these can be failure points that show up some time later.  
   Stainless alloys for compressor valves probably would do just fine as is.  I expect the tumbling operation is to effectively polish out the tiny edge defects left from punching out the valves, rather than to alter the basic properties.  Tensile strength goes up as imperfections in the crystal structure goes down.  
    Oh yes, the reed exercising is for stress relief and is as Chris says is best done in a room with a good door.  I leave the room and let it run for a few hours/side.

Dana

[Pistachio Dreamer]

6 hours ago, Dana Johnson said:

Chris mentions stainless on stainless as being prone to galling, but as always the alloy counts.  300 series stainless should not be used together with itself or others of the same series.  The nickel content is the culprit and on a molecular level will jump from one surface to the other.  You can use 300 series in combination with 400 series ( chromium only )without galling.  
   On a different note, I believe some Wheatstones were made with the best quality reeds advertised as having purposely rounded edges.

  Reed steel should be hard enough already.  I find the blue 1095 shim steel at the low end of acceptable, though it sounds just fine.  The UHB20C alloy I use ( also used for compressor valves as well as the accordion industry) is harder and retains its set much better, and shears much more cleanly than 1095, which tends to shift sides of the shear line along the cut leaving a sort of battlement shaped surface that needs to be filed smooth.

   I don’t think there is much of a fatigue problem with well made reeds without stress risers.  Shearing can cause micro cracking that may not be noticed or removed by edge filing, and these can be failure points that show up some time later.  
   Stainless alloys for compressor valves probably would do just fine as is.  I expect the tumbling operation is to effectively polish out the tiny edge defects left from punching out the valves, rather than to alter the basic properties.  Tensile strength goes up as imperfections in the crystal structure goes down.  
    Oh yes, the reed exercising is for stress relief and is as Chris says is best done in a room with a good door.  I leave the room and let it run for a few hours/side.

Dana

Good to hear from you Dana, thanks for your reply, and others in the forum too. This is the stainless grade I'm currently experimenting with: https://www.materials.sandvik/en-gb/materials-center/material-datasheets/strip-steel/sandvik-7c27mo2-flapper-valve-steel/ It really does shear very smoothly and straight, once I figured out to leave the reed strip on the LH (static) side of the shear! 

 

I think there is an equivalent from Voestalpine, including a new one called "flap-x" that looks to be harder-stronger-faster-better, but perhaps not necessarily so for concertina reeds. When I did my initial research the sandvik looked similar in properties to UHB20c, i.e. a bit better on hardness, tensile strength than 1095, which as you say I've also found doesn't hold its set as well.

 

The shoes are 3d printed in 316L, which I understand is pretty good for "marine" applications, however interested in what you and others say about galling. I chose 304 for the clamp fixings as I found a source that said the corrosion would be minimised by mixing of alloys. I have no idea though how the 316L will react with the reed steel over time though.

 

The nice thing about the 3d printed design of the shoes is that I can experiment with a variety of materials. The design of the shoe doesn't require tapping for bolts so I can try out new tech and materials such as technical ceramics.

 

As for rounding of the reeds, I'm most likely to end up with a few with rounded edges as I definitely need to work on my filing technique, so it's a comfort to hear from you and others that it may not be so critical to the overall performance. In my experience to date I'm finding that the profiling and clearances have the greatest effect on the reed performance of course, with other improvements perhaps only giving marginal gains, not to mention allowing various manufacturers to claim a benefit for commercial reasons!

[Little John]

8 hours ago, Dana Johnson said:

On a different note, I believe some Wheatstones were made with the best quality reeds advertised as having purposely rounded edges.

 

Does anyone know which edges were rounded?

[Don Taylor]

4 hours ago, Pistachio Dreamer said:

The shoes are 3d printed in 316L, which I understand is pretty good for "marine" applications, however interested in what you and others say about galling. I chose 304 for the clamp fixings as I found a source that said the corrosion would be minimised by mixing of alloys. I have no idea though how the 316L will react with the reed steel over time though.

 

My experience in mixing 316L and 304 in a marine application is that they will indeed gall together.  I used a product called Tefgel which worked for me:

https://www.tefgel.com/contain.php?param=tefgel_infor

 

[Alex West]

Dana

 

Where do you get your UHB20C steel in appropriate reed thicknesses from?  Are there any UK suppliers for small quantities?

 

Alex West

[Mike Hulme]

I can't help you with UK suppliers, but if you search for AISI SAE 1095 carbon spring steel (an exact(ish) equivalent),  there are some suppliers around who sell it in coils.

 

Mike

[alex_holden]

I think the steel I use is something similar to Dana's UHB20C, but I import it from Germany and it's more expensive than normal blue tempered shim stock. I stocked up a year ago because I'm not looking forward to trying to order more post-Brexit.

[Dana Johnson]

Alex West,
I used a US. Uddeholm distributor since that is where I live.  Since they are a European company, you ought to be able to find a distributor on the company website.  I was lucky to be able to buy a minimum order including different thicknesses, but still have a lifetime’s supply I have been sharing with a friend.  The Sandvik version sounds a reasonable choice too.  
 

I have been using (small ) 18-8 stainless nuts and bolts for the past 25 years dry with zero problems.  If you don’t have to tap anything, I’d say stick with the best corrosion resistance alloy.  That is the only place galling might occur in tight or over torqued threads.  

[Chris Ghent]

On 12/23/2020 at 9:04 PM, Pistachio Dreamer said:

 

 

As for rounding of the reeds, I'm most likely to end up with a few with rounded edges as I definitely need to work on my filing technique, so

Something to watch for is when you clean up the side of a reed created with a shear; if you do not hold the file exactly square to the side you may leave the surface which will be the bottom of the reed slightly narrower than the top. When you profile the reed you will then be making the clearance wider. 
 

An easy (but second best, best is to file it square) way to deal with this is to deliberately create a blank in which the top is narrower than the bottom. I am only talking a thou or so. The clearance at the bottom will not change when you remove material from the top.
 

Using a long file to do this job makes it easier to know how you are filing.  

[Clive Thorne]

"The Tumbling Reeds" sounds like the name of my next folk band!

[ttonon]

On 12/20/2020 at 12:25 PM, Little John said:

 

I'm surprised at this. The beating reeds used in organs have a much bigger air gap than a free reed, so one might expect it to take even longer to build up sufficient amplitude to start beating (the equivalent of entering the slot).

LIttle John, sorry for the long delay, but I just noticed your post.  I think the answer to your puzzle is the fact that the beating reeds make use of an acoustic resonator, which forces the vibration of the reed itself.  Typically, the natural frequency of vibration of the reeds in beating reeds is much higher than the frequencies of the musical tones.  This is why you sometimes hear the squeaks when a beginner plays for instance the clarinet or sax.  You're probably hearing that high frequency from the sole reed vibration, no air column vibration.  Thus the pressure pulses from the standing waves in the resonator determine the vibration of the reed, and those pulses occur at 1/f seconds, where f is the fundamental frequency of the musical note, which if you calculate is on the order of milliseconds.  

 

As I explained, the vibrational start of the free reed has entirely different physics, leading to those relatively long start transients.