TONING REEDS?

[Martin Essery]

No, not a spelling mistake, I meant toning, not tuning. While the reeds seem to be reasonably well in tune on my concertina, they sound different, have a different timbre, almost like different instruments, some quite open, others somewhat choked, but others just having a different resonance. I am guessing the reeds are producing different harmonics. For instance if I play two G's together, there is a beat frequency, even though the fundamental tone is the same, according to the tuner.

So, can anything be done, or is that just the way it is with reeds? I can imagine a variety of influences at play, position of reed relative to the plate, weight of reed, size of resonant chamber. Where would one even start? I will get a better instrument when I can afford one, but I was intrigued as to what might make the difference in tone.

[Wally Carroll]

The differences in timbre between notes is often more due to the non-reed aspects of the instrument (chamber size, location on reedpan, etc.). This next statement has some caveats: Often if you switch a note that you don’t like the sound of with another of the same pitch, you won’t hear much of a difference. This assumes the notes are from the same original instrument,  neither have been damaged  or excessively tuned and they are held snugly in their slots. The location of the note can play a large role in how bright or mellow a note sounds as fretwork can be blocked over particular notes by the players hands. Also which side of the reedpan the note is on affects tone. Interestingly, if you turn one end of the instrument to point directly at your head, some of these effects go away. The player, unfortunately is in the worst position to best hear the instrument. 

[Łukasz Martynowicz]

3 hours ago, Wally Carroll said:

The differences in timbre between notes is often more due to the non-reed aspects of the instrument (chamber size, location on reedpan, etc.). This next statement has some caveats: Often if you switch a note that you don’t like the sound of with another of the same pitch, you won’t hear much of a difference. This assumes the notes are from the same original instrument,  neither have been damaged  or excessively tuned and they are held snugly in their slots. The location of the note can play a large role in how bright or mellow a note sounds as fretwork can be blocked over particular notes by the players hands. Also which side of the reedpan the note is on affects tone. Interestingly, if you turn one end of the instrument to point directly at your head, some of these effects go away. The player, unfortunately is in the worst position to best hear the instrument. 

Not only directing the end at your head, but also playing on each of the six/eight sides sounds different from the player’s perspective. 
 

As to original question: beating rhytm of two reeds with the same pitch is quite natural, unless you mean something different than „wet” musette effect. Two pitches „in tune” according to tuner readout can still be 5-6 cents apart.

 

Assuming you meant something else, first thing that comes to mind is reed orientation. When I was building my box I run into a frustrating problem of two F3 notes sounding drastically different - one was clear, the other had this saw/trumpet character to it. It was not caused by the reed itself, as the problem stayed the same side after switching reeds around. It also stayed (however to lesser extent) with the endboxes removed. What caused it was the difference in vertical orientation of the reed and turning the concertina upside down switched the side of the problem. Fretwork emphasized the problem further, but the origin was in the reedpan layout itself. I only managed to diminish the problem by altering chamber depth to counter the most offensive harmonics. And as Wally wrote - it was much more noticeable from the player’s perspective, than for the audience.

 

Another problem is the fretwork design. If some padholes are under large closed areas and others are under large open areas, notes will sound different. Obviously you can’t redo the fretwork, BUT - when I designed and cut my fretwork I had no idea about the above effects, so my box ended up with huge timbre differences between notes. The solution was to install a thin, rigid, 3D printed baffle, that had solid circles directly above each padhole. Changing the diameter of those circles changed the timbre of individual notes, and at the same time changing the overall open area changed the overall character of the box. Since my original fretwork design is really open, I had the opportunity to test all sorts of variations of such baffles, from nearly completely open up to completely closed, including asymmetrical LH/RH „voices”. I would suggest you starting with such experiments, as such baffles are completely non destructive. If your concertina is a hybrid, you could also experiment with individual chamber depths - even a mm change in depth can make a huge difference in timbre. If you have a traditional box however such modification is not possible.

 

 

[Theo]

If you are checking the pitch with an electronic tuner you need to use one that is able to measure the pitch with an accuracy of less than 1 cent. A needle type display or a simulation of it is not precise enough and the type the show green/red to indicate in tune or not typically have a tollerance of +/- 3 to 5 cents as Lukasz has mentioned. So if you have two reeds of the same pitch measured within

+/- 3 cents then one could be 3 cents sharp the other could be 3 cents flat so they differ by 6 cents.  This will produce a very strong  beat easily audible by anyone. 

[Ken_Coles]

In the case of anglo concertinas, some of the notes are necessarily under your palms and they sound different to me than the ones that don't have my hands in the way.

 

Ken

[SIMON GABRIELOW]

I suppose it comes down to the fact they are 'organic' instruments.. made of wood, metal, bits and pieces, etc... Meaning they are not electronic, but real resonating physical instruments of sound, and so will have differences in sound, each one will be unique. Even same note value will sound different on one instrument ( duplicated notes I mean)! 

 

[d.elliott]

11 hours ago, Theo said:

If you are checking the pitch with an electronic tuner you need to use one that is able to measure the pitch with an accuracy of less than 1 cent. A needle type display or a simulation of it is not precise enough and the type the show green/red to indicate in tune or not typically have a tolerance of +/- 3 to 5 cents as Lukasz has mentioned. So if you have two reeds of the same pitch measured within

+/- 3 cents then one could be 3 cents sharp the other could be 3 cents flat so they differ by 6 cents.  This will produce a very strong  beat easily audible by anyone. 

 Some years ago we had a healthy debate about tuning tolerances, this based upon audiology data, and various other factors, I (we?) settled on +/- 1.5 cents from nominal, and the use of a meter that could discriminate to no courser than 0.05 of a cent. Some repairers did not want to engage in the discussion, some players active on the  forum at the time took the view that ascribing tolerances was just lazy craftsmanship, one chap demanded nothing better than absolute perfection. The reality is that there is no such thing as perfection. we knew what the trained ear could discriminate, and what would cause a clash to that ear if two instruments tuned at different ends of the tolerance band were played in juxtaposition. Experience repairers know what level of repeatability is sensibly attainable in the tuning process.

 

Theo's point is exactly right if you are counting to portions of a cent, you need a read out that is more precise than that portion. If you want to be able to trust that reading then the instrument must be able to discriminate (if not read out) to around 0.2 times the read increments, of finer.

[Clive Thorne]

Worth remembering that the beat frequency between two notes reflects the absolute frequency difference in Hz, whereas the Cent is proportional to the frequency. So what might be an acceptable  Error, in cents, at low frequency, giving say a 2hz Beat, would give an 8 Hz beat two octaves up.

[Martin Essery]

Thank you all so much for your wonderful replies. What a wonderfully complex instrument this is! It was great to hear all the variations that can make a difference. Mine is a beginners range instrument, so I have come to not expect too much from it, but glad to hear that I am hearing the inconsistencies more clearly than an audience. I will be upgrading as soon as I can afford to, so I think I will wait till I have a better instrument and see where that one stands tuning and toning wise before I take any action.

Each gut string on my concert harp has its own personality and needs to be played with a different touch, but you cannot touch a reed the same way, other than adjusting attack with bellows pressure. I have just been playing with online spectrum analysers and tuners, and there is indeed much variation. The biggest take away from that exercise was that bellows pressure can vary the pitch by 2 Hz quite easily, so any precise tuning would need equally precise bellows pressure. It has also shown that my little Korg tuner is not up to the job.

Thank you all for your input, you have broadened my understanding.

 

[SteveS]

12 hours ago, Martin Essery said:

The biggest take away from that exercise was that bellows pressure can vary the pitch by 2 Hz quite easily, so any precise tuning would need equally precise bellows pressure.

 

Indeed bellows pressure does affect the pitch - and this is a problem when tuning reeds.  Yes as repairers we might strive for as close a tolerance as possible, as d.elliott mentioned, but its hard to tune an instrument for an individual playing style.  The pressure applied on tuning a reed will probably be very different when played by the owner.  And even getting a consistent pressure on tuning is difficult - although personally I use gravity activated bellows which does offer a degree of consistency.  After tuning an instrument I will play it for a while, making a note of any deviations from nominal, and then tune those reeds in the instrument to get as close to nominal within tolerance.

(I'm soon going to be checking the tuning on my band instruments for consistency across them ahead of recording my quartet arrangements.)

Edited January 7, 2023 by SteveS