Reed setting tips?

[ttonon]

2 hours ago, Theo said:

I have seen several sets of reed organs and none had as much curvature as i9n the diagram.  Some have a small twist at the free end of the reed that extends along the reed for a distance similar to the width of the reed.  The twist has the effect of raising one corner of the tongue above the plate.

This seems even more interesting to me, and I never heard of or considered such a twist.  You see, I think the sound of the free reed is boring, because of its perfect partials.  By "perfect partials," I mean that all overtones are harmonics.  "Harmonics" is a mathematical term that means multiples of two.  Thus, any overtone of the free reed are perfect multiples (or divisor) of two with any other.  The only "fly in the ointment" here are so-called transients and some cases of very loud playing situations that excite either torsional or higher vibration modes, which in general are not multiples of two of the fundamental tone.  But the "steady" sustained tone of concertina reeds produce perfect harmonics, and there are no other musical instruments that have perfect partials that I can recall at the moment, except the bowed string instruments.  The basic reason for that is the fact that all those other instruments are transient sounds, or if sustained as in the woodwinds, nonlinearities in the sound source - a vibrating air column -  cause non-harmonicity.  In fact, I believe the phenomenon is an advantage for string orchestras, because I don't think the eerie blending of the many violins playing in the string section would be possible without it.  

 

Well, with all that background, I say our little reed friends sound boring because of this perfect-harmonic feature, in a similar way as a pure sinusoidal tone (as from a tuning fork) is boring.  In addition, when more than one note is sounding, the overtones of the separate notes mesh, and we lose the identity of separate notes and the complexity of timbre - a kind of "clash" or "noise" - that say a piano or two saxophones produce. 

 

So introduce the twist.  Is it possible to excite a torsional tongue vibration mode with the twist, so that such a mode would be continually excited, along with the main bending vibration mode?  For this to occur predictably, it may be necessary to make the tongues wider.   Dana recently told me that he has had experience with some extra-wide tongues, with unpleasant tonal results.  But I'm not so sure the idea cannot be improved on, especially since the frequency of the torsional mode can be a design parameter, and some torsional pitches might blend very well, adding interesting complexities to the timbre.  

 

Here also let me suggest that the hump in the harmonium reed in the picture might be a way to force - and insure - the first vibration bending mode of the tongue.  The only other bending mode I've seen in Fourier spectrums is the second, which for a constant cross-section cantilever, is around 6.2 times the frequency of the first mode.  The second mode has a node about a third of the way from the tip of the tongue.  It may be that putting the hump there, insures that a localized pressure force acts there during the swing cycle, and that would make excitation of the second mode very unlikely.  Such an arrangement might be necessary for very long tongues.   I believe it's very unlikely to excite the second mode of a short tongue, or if it's excited, we wouldn't be able to hear it, because of its high frequency.  But for very long tongues, my guess is that the hump is perhaps necessary, if for instance harmonium makers wanted only very soft sounds, as during church service. 

 

Interesting in that I never thought of this "hump" explanation before, until I saw the twist idea.  But now, of course, the question is, why did harmonium makers put in the twist?  If we want to hold onto the idea that harmonium makers wanted softer sounds, do their tongue geometries have sufficiently small width-to-length ratios so as to rule out torsional modes, and thus the twist is just a convenient and quick way to make a starting offset?  If tongue geometries are such that torsional modes are indeed excited, my explanation above for the hump would seem to be wrong.  At this point of course, we'd have to talk to a harmonium worker that knows his business, though where can you find such a person?  

 

Best regards, and stay safe,

Tom

www.bluesbox.biz   

 

 

 

[Stephen Chambers]

14 hours ago, ttonon said:

For those interested, I'm not sure how many of you are aware that harmonium manufacturers have their own standards about "best practice."  Here's a drawing of one of their free reeds.  I'm under the impression that the curious bend is rather typical.  Can anyone here venture as to its advantages?

 

 

Only that's not a harmonium reed Tom!

 

As Gellerman states, that's an American organ reed, and they work with a suction bellows and operate on vacuum, rather than a pressure pressure bellows like a harmonium. The reeds in the American organ are considerably smaller and more curved and twisted than in the harmonium.

 

https://en.wikisource.org/wiki/A_Dictionary_of_Music_and_Musicians/American_Organ

[ttonon]

Hi Steven, thanks for pointing out the difference in terminology, which I wasn't aware of.  I found a video comparing the sound of a harmonium to that of an American reed organ, for those interested:

 

In addition, Wikipedia gives the following distinctions:
> Reed organs are operated either with pressure or with suction bellows. Pressure bellows permit a wider range to modify the volume, depending on whether the pedaling of the bellows is faster or slower. In North America and the United Kingdom, a reed organ with pressure bellows is referred to as a harmonium, whereas in continental Europe, any reed organ is called a harmonium regardless of whether it has pressure or suction bellows. As reed organs with pressure bellows were more difficult to produce and therefore more expensive, North American and British reed organs and melodeons generally use suction bellows and operate on vacuum.<

 

Lest some people get the wrong idea, it should be mentioned that the reed itself doesn't know what is positive pressure or vacuum.  What does matter is the direction of airflow through the reed and whether the mounting cavity is either upstream or downstream of the reed.  Unfortunately I couldn't find which is which in the descriptions given, indicating that the writers don't have a detailed interest in acoustics. 

 

Best regards,
Tom

[Dana Johnson]

On 4/15/2020 at 2:50 AM, alex_holden said:

Wow, I find it hard to believe a reed that shape would sound at all, never mind with any sort of efficiency.

Rich Morse had a whole set of these, though the arch at the end was more abrupt and didn’t extend too far down the reeds.  I have seen the twisted end ones as well, and it was quite substantial and not at all accidental.  That organ also had straight reeds, which were well made.  Organs have a lot of access to air and can afford to use less efficient reeds.