What makes the two middle reeds sound different?

[Mike K]

Something I never really understood as I have never gotten into replacing reeds. If you have two sets of middle reeds, what is the difference between the two of them, that causes the difference in sound. Guess that would also apply to say, the lower or upper reeds too. If they are an octave different, is the bass A2 reed the same as a middle A1 reed?

 

[debra]

In an accordion using single-note reed plates (that's everything except a bayan) all treble reeds for the same frequency tend to be the same. (Bass reeds are different though.)
But... Two reeds that are supposed to play the same note (in the same octave) are not 100% identical: the voicing (opening between reed tongue and reed plate) may be slightly different. When using leather valves the valves may not behave in exactly the same way (leather is a natural material, and the pressure from metal booster springs can be different as well). Then, most importantly: the reed plates are placed on a reed block in a different position, with a different resonance chamber and possibly different resonance from the reed block. Then that reed block is placed in the accordion and when the pallet opens it turns so it doesn't open equally far for the notes on each side of the reed block... All these factors combined make that two identical reeds do not produce the exact same sound. The difference is certainly larger when comparing a note in the L register with the same note (compensating for the octave) in the M register than when comparing two M reeds in a dry-tuned accordion.

 

[Mike K]

In an accordion using single-note reed plates (that's everything except a bayan) all treble reeds for the same frequency tend to be the same. (Bass reeds are different though.)
But... Two reeds that are supposed to play the same note (in the same octave) are not 100% identical: the voicing (opening between reed tongue and reed plate) may be slightly different. When using leather valves the valves may not behave in exactly the same way (leather is a natural material, and the pressure from metal booster springs can be different as well). Then, most importantly: the reed plates are placed on a reed block in a different position, with a different resonance chamber and possibly different resonance from the reed block. Then that reed block is placed in the accordion and when the pallet opens it turns so it doesn't open equally far for the notes on each side of the reed block... All these factors combined make that two identical reeds do not produce the exact same sound. The difference is certainly larger when comparing a note in the L register with the same note (compensating for the octave) in the M register than when comparing two M reeds in a dry-tuned accordion.

My Giulietti Continental allows me to play each of the middle reeds singly. Definitely a different sound to all the reeds. That is fine with me just do not understand everything going on and if the reeds are the same or somewhat different. They are dry tuned so seperate tuning is not the cause. My other Giulietti does not give me that option. I have Clarinet and Violin vs Clarinet, Oboe and Violin on the other.

 

[Gonk]

To add to Paul's very thorough answer above... "dry" tuned, just like "wet" tuned, can actually mean a range of -/+ spreads. The likelihood of two M voices actually being tuned to exactly the same pitches is nil. "Dry" means "close enough to minimize perception of beating." If it has been recently tuned by someone very careful and thorough, it may be very close to "perfect," but true perfection is mathematically unattainable. And when you play two reeds that are tuned extremely close, in a closed system, their sympathetic vibration can actually help one another "lock in" a bit further. When you play them singly, this effect is not in operation. In addition, many Italian instruments that are now mistakenly described as "dry" were actually originally carefully tuned with about 2c of tremolo (measured at A4). This is enough to give a bit of perceptual richness and complexity while not being markedly wet.

All that said, it's even more likely to be the pallet angle and clearance, as Paul said above. The amount of engineering required to make two different reedbanks with two different pallet clearances sound identical is very high. I know of no instrument that achieves it without carefully designed cantilevers on the pallets. And I have never seen this on a piano accordion. You could theoretically achieve it with split pallets that open simultaneously from both top and bottom, using a cantilever arm for the top one. This would give similar angle and depth of clearance for the two middle openings and for the top and bottom openings, respectively. I have never seen a design like this.

A possible reason is that it's overengineering to solve a nonexistent problem. If the scale range sounds coherent, that's good, but is it good for the two M ranges to sound identical? Any difference between them will actually offer the player depth of timbre and more varied register options.

 

[Dingo40]

Now, imagine three middle reeds: say, LMMM or LMMMH.

 

[debra]

My Giulietti Continental allows me to play each of the middle reeds singly. Definitely a different sound to all the reeds. That is fine with me just do not understand everything going on and if the reeds are the same or somewhat different. They are dry tuned so seperate tuning is not the cause. My other Giulietti does not give me that option. I have Clarinet and Violin vs Clarinet, Oboe and Violin on the other.

Most accordions that allow you to select each of the middle reeds singly are instruments with a cassotto. The M reed inside cassotto sounds more mellow than the M reed outside cassotto (even though they may be tuned to exactly the same frequency). The cassotto (in combination with these registers) thus gives you more different sound (timbre) options.

 

[debra]

To add to Paul's very thorough answer above... "dry" tuned, just like "wet" tuned, can actually mean a range of -/+ spreads. The likelihood of two M voices actually being tuned to exactly the same pitches is nil. ...

With "dry" I actually meant that the two M voices are actually tuned to exactly the same pitches. The chances of achieving this when tuning individual reeds is indeed practically nil, but accordion tuners try their best anyway. When two reeds with very very close tuning play together they just sound as one, no beating at all. This is partly caused by a physical effect called "sympathetic vibration" that we talked about in an older thread. As it happens I am trying to tune the LL bass reeds in my bayan right now and sympathetic vibration is what makes the bass sound like it's just L and not LL. The problem in achieving this is that the two L reeds must be tuned close to the same frequency for the sympathetic vibration to kick in. It works well because the two reeds (on opposing sides of the reed block) share the resonance chamber. On Italian copies of the Russian bayan design you also find an LL bass configuration but with the L reeds having their own resonance chamber. Tuning is then easier but it needs to be much closer because there is hardly any sympathetic vibration.
In practice real "dry" tuning, meaning 0 cents tremolo, is quite hard to achieve and requires more precise tuning than tuning accordions with tremolo. It is also made harder because the valves (especially leather ones) cause the frequency to depend on the sound volume in a somewhat unpredictable way, so two reeds with the same desired frequency will actually only have exactly the same frequency at some specific sound volume and will start to deviate from each other when you play louder or softer. (You can sometimes clearly hear this when you watch people play on a completely dry tuned accordion.
As you say, (classical, convertor) accordions are often tuned to what is called "concert tuning" which aims at 2 cents difference (at A4). Concert tuning does sound different from completely dry tuning (0 cents). It is also hard to achieve because a slight deviation like 1.5 or 2.5 cents on different notes is also already clearly noticeable...
When I agree to tune someone's accordion I need the instrument for a week, but I tell them that if they let me have it for two weeks the result will be better. Not a lot of work is done in the second week, just playing, listening and then fine-tuning until it is absolutely right.

 

[Gonk]

Mike, does your Giulietti Continental feature a double tone chamber? That would matter quite a bit! : D

 

[godgi]

Paul my pigini mm are in unison. Would it ever be an idea to change the tuning on one to create a more slight musette sound or variation to make a more instrument for variety purposes
Or is a classical instrument always left as is: ie in unison

 

[debra]

Paul my pigini mm are in unison. Would it ever be an idea to change the tuning on one to create a more slight musette sound or variation to make a more instrument for variety purposes
Or is a classical instrument always left as is: ie in unison

(Light) tremolo or not... that has always been the question for a "classical" instrument. Several decades ago all accordions tended to have some tremolo, including the Hohner Gola (and Morino) models with melody bass. Russian bayans by default come with "concert tuning" which has about 2 cents tremolo (and most of what does is hide when the bayan is slightly out of tune). I tuned my russian bayan (and my morino) to 8 cents tremolo which is quite mild tremolo but allows you to play some styles that sound better with a bit of tremolo. It's all a matter of taste.

 

[AccordionUprising]

With "dry" I actually meant that the two M voices are actually tuned to exactly the same pitches. The chances of achieving this when tuning individual reeds is indeed practically nil, but accordion tuners try their best anyway.

What is the purpose of two identical-ish tuned reeds? Does it add volume?

 

[Ventura]

or 3 (my LMMMH excelsior)

warmer.. richer.. sweeter.. while your solo note melody line stays clear as a bell

full chords are precise

 

[debra]

What is the purpose of two identical-ish tuned reeds? Does it add volume?

It is used mostly on accordions with cassotto, to give more differently sounding register possibilities (without introducing tremolo). You can have LM, MH and LMH with M in cassotto and with M outside, you can have LMMH for the loudest non-tremolo sound.
MM without tremolo does add volume, but it also creates a sound that is less mellow than M in cassotto but still more mellow than M outside the cassotto... but in most cases it just sounds "slightly out of tune" unless the accordion was just serviced by a very precise tuner.

 

[Pedro Javier]

What an interesting post; a few months ago I finally have my accordion in my hands. At first I was quite confused that the unison voices (M inside and outside cassotto) did not sound at exactly the same pitch on all notes, despite the fact that I love the quality of their sound and that when playing any piece the difference was negligible. Since then I have wanted to compare the sound of my instrument with another perfectly tuned one, but in my country there are very few chromatic button accordions. Do you know any video or audio showing a perfectly tuned accordion for comparison?

 

[Siegmund]

What is the purpose of two identical-ish tuned reeds? Does it add volume?

You might well ask the same question about any other two instruments in unison. It does add a little bit of volume -- as a rule of thumb, 2 instruments in unison will be 3 decibels louder than one, if they play the same otherwise -- but mostly it adds, for lack of a better word, a thickening of the sound. You hear more of the fundamental tone and less of the incidental squawks and sputters and transient changes in pitch, since those won't happen at exactly the same time on both instruments.
You can go through the score of a Beethoven symphony and compare the effect of one flute vs. two, one horn vs. two, one clarinet vs. two, etc. (And then do the same with a Mahler symphony to hear what 4 flutes or horns in unison sound like - 2 to 4 is about the same size change as 1 to 2 is.)

The usual tendency was to give a melody to a single instrument to allow maximum scope for expression, to several to change the tone color at some expense of expressiveness.

 

[debra]

You might well ask the same question about any other two instruments in unison. It does add a little bit of volume -- as a rule of thumb, 2 instruments in unison will be 3 decibels louder than one, if they play the same otherwise -- but mostly it adds, for lack of a better word, a thickening of the sound. You hear more of the fundamental tone and less of the incidental squawks and sputters and transient changes in pitch, since those won't happen at exactly the same time on both instruments.
You can go through the score of a Beethoven symphony and compare the effect of one flute vs. two, one horn vs. two, one clarinet vs. two, etc. (And then do the same with a Mahler symphony to hear what 4 flutes or horns in unison sound like - 2 to 4 is about the same size change as 1 to 2 is.)

The usual tendency was to give a melody to a single instrument to allow maximum scope for expression, to several to change the tone color at some expense of expressiveness.

It's a tricky business. Most of what really changes the sound of different players playing together is that the location of the source of each of their sounds is different. As a result how you experience the mix changes as you move about just a bit. When two reeds in an accordion play together in unison their location is a constant distance away (from each other) and inside the accordion there is "sympathetic vibration" which causes the two reeds to stay more in sync with each other than when two players on different accordions both play a single reed.
I'm glad I no longer need to make a living out of using mathematics... A truly scientific explanation of why two reeds playing in unison sound different would be very complex and even with my math training I doubt I would fully understand...

 

[dak]

To add to Paul's very thorough answer above... "dry" tuned, just like "wet" tuned, can actually mean a range of -/+ spreads. The likelihood of two M voices actually being tuned to exactly the same pitches is nil. "Dry" means "close enough to minimize perception of beating." If it has been recently tuned by someone very careful and thorough, it may be very close to "perfect," but true perfection is mathematically unattainable.

That's where physics does the rest of the job. Since the reeds are sharing sounds via bellows and outside, they are slightly coupled. They will synchronize when being close enough. That also happens for LM registration. However, the phase difference will depend on their "detuning", and natural pitch is not completely pressure independent. That makes the sound quality of a multi-reed dry registration vary under pressure more than a single reed, making the sound react quite more distinctly than a mere volume control.

Of course mathematics can deal with that, but you need to put in more effort than just independent frequencies.

 

[Ikesaccordions]

(Light) tremolo or not... that has always been the question for a "classical" instrument. Several decades ago all accordions tended to have some tremolo, including the Hohner Gola (and Morino) models with melody bass. Russian bayans by default come with "concert tuning" which has about 2 cents tremolo (and most of what does is hide when the bayan is slightly out of tune). I tuned my russian bayan (and my morino) to 8 cents tremolo which is quite mild tremolo but allows you to play some styles that sound better with a bit of tremolo. It's all a matter of taste.

8 cents tremolo would that be about 2 hz at A 440? But of course as you know, 8 cts tremolo would be 4 hz at A 880.

 

[debra]

8 cents tremolo would that be about 2 hz at A 440? But of course as you know, 8 cts tremolo would be 4 hz at A 880.

Correct. But because 4Hz tremolo at 880 Hz sounds much wetter than 2 Hz at 440 Hz tremolo tuning always has the tremolo go down in cents as you go up the octaves. My rule of thumb is that one octave higher means 70% of the number of cents. (People's taste generally rests somewhere between 2/3 (66.67%) and 3/4 (75%). So my accordion then has just under 6 cents at 880Hz (A5), and around 4 cents at 1760 Hz (A6). Going down A3 (220 Hz) has about 11 cents and below that I tend to stay at that same value. (Some people even have the value go down as you go further down in frequency.