Tunings explained... but...

[debra]

That is fascinating. Why would the difference be more noticeable if the central M is in cassotto?

Yes, the central M in cassotto makes MMM produce a sound that is a bit of an acquired taste (but you do get used to it) but also without cassotto MMM sounds more "civilized" than the M-M+ combination, which some accordions call "musette vivace".

 

[debra]

Is it therefore also true that an accordion tuned at 442Hz will be louder than one tuned at 440Hz and louder still at 444Hz and, if this is true, are these differences perceptible? And why tune an accordion at anything other than 440Hz anyway?

Objectively speaking an accordion tuned at 442Hz is not louder than one tuned 440Hz (unlike a violin) but it "stands out" more among 440Hz-tuned instruments (and in my opinion just sounds bad together). Subjectively an accordion tuned 440Hz amids a group of 442Hz accordions does not stand out as much (but still sounds bad together). The main reason tuning has gone up over the decades is that string instruments objectively sound louder when tuned higher, because of the increased tension on the strings. (However, some old string instruments will just break when you try to tune them higher.)

 

[losthobos]

I agree wuth the cassotto M anomaly... I had a top range Brandoni with Binci hand made reeds... The sound of the Bassoon cassotto was awesone and the clarinet smooth as silk too.... Sadly as soon as one of the offset non cassotto M reeds was added to the mix the sound was alien to my ears... Couldn't adapt to it and sold the accordion on...
Now i only have a bassoon in cassotto and with the two Ms i get a sound i have more of an affinity with even if not such a high spec box

 

[JIM D.]

An accordion tuned 442 is no louder than a 440 tuned box of the same model.
From the 30's to the present, Italian brands of accordions have always come tuned 442. If a 440 was desired it
was necessary to order it with 440 or have it retuned. In the 30's and up to the 50's amplification of accordions
was extremely rare and accordions tuned 440 did not have a volume presence when performed with other band
instruments. Tuning them 442 gave them a presence when heard with other instruments and the 442 tuning
became a practice still used today.

Think of an orchestra with 8 violins al tuned 440. All the violins will have the same presence and volume.
Now tune just 1 violin 442 and although it's not louder than the others, but it will stick out like a sore thumb.
The 442 violin is not louder but it's presence is increased.
.

 

[debra]

An accordion tuned 442 is no louder than a 440 tuned box of the same model.
From the 30's to the present, Italian brands of accordions have always come tuned 442. If a 440 was desired it
was necessary to order it with 440 or have it retuned. In the 30's and up to the 50's amplification of accordions
was extremely rare and accordions tuned 440 did not have a volume presence when performed with other band
instruments. Tuning them 442 gave them a presence when heard with other instruments and the 442 tuning
became a practice still used today.

Think of an orchestra with 8 violins al tuned 440. All the violins will have the same presence and volume.
Now tune just 1 violin 442 and although it's not louder than the others, but it will stick out like a sore thumb.
The 442 violin is not louder but it's presence is increased.
.

Not entirely true (but almost): When you order a Bugari accordion the tuning is 440Hz by default (following the ISO standard that everyone should follow) so if you want 442 you have to specify that in the order. I would be surprised if Bugari was the only company to default to 440Hz, but several others indeed will give you 442 by default (which is wrong, now that we have an international standard which should be the default).
Hohner (not Italian, I know) also used to default to 440Hz, and that includes their Italian-built models like the Morino.

 

[Scuromondo]

Yes, the central M in cassotto makes MMM produce a sound that is a bit of an acquired taste (but you do get used to it) but also without cassotto MMM sounds more "civilized" than the M-M+ combination, which some accordions call "musette vivace".

Interesting. My understanding is that our perception of the sound from cassotto reeds as “warmer” or more pleasant is due to it effectively filtering out some higher frequency harmonics. I wonder if, in a non-cassotto instrument the presence of those in-tune harmonics is actually beneficial because they might perhaps help to “center” the detuned harmonics produced by the M+ and M- reeds.

 

[debra]

Interesting. My understanding is that our perception of the sound from cassotto reeds as “warmer” or more pleasant is due to it effectively filtering out some higher frequency harmonics. I wonder if, in a non-cassotto instrument the presence of those in-tune harmonics is actually beneficial because they might perhaps help to “center” the detuned harmonics produced by the M+ and M- reeds.

The M- and M+ "decentering" cancel each other out, so whether the center M is in cassotto or not does not influence the "overall centering". But, it is not true that the cassotto just filters out higher frequency harmonics. It also (significantly) amplifies the note's base frequency (and more so for lower than for higher notes). When you make a recording and watch the "envelope" that amplification becomes very clear. Yet, the filtering of the higher frequency harmonics makes that unless something else plays at the same time (like M+) you do not notice that amplification much. Interestingly, and I do not yet know why, where the higher notes stop having a valve that amplification seems to stop. So when you look at the envelope and the valves stop at the high C the high B looks much louder than the D above, yet subjectively they sound almost just as loud. I don't know what causes this, but it is consistent across accordions: the first note without valve is objectively less loud (according to the envelope), and that is true whether the first note without valve is a B or C or C# or D (it varies on my accordions). So it's not the absolute frequency of the note that stops the amplification but the presence of a valve.

 

[Ventura]

just tuning, equal tempramant, pythagorean, werkmeister, Warsaw, mean tone

i would devils advocate that you truly cannot JUST use mathematics to
make your plus or minus decisions for M- M_ M+

and that is because the speed of the "pulse"is determined by proportion...
if you tune everything strictly by Math, the speed of the pulses will vary
widely by note (and bear in mind the proportion between 438 and 440 is different
than the proportion between 440 and 442 because the FREQUENCIES are different)

wheras the objective is having the overall tuning share a pleasing spreadsheet of
related and intelligently varied waves

you would do this by FIRST tuning your M with incredible precision based
upon your preferred scheme, and THEN you would tune each companion
M+ note by listening to the beats and wielding your Diamond Coated Sandpaper
with minute touches until the result is both PLEASING to your brain as
well as making SENSE as it relates to the note you tuned previous to this one

then you would do the same thing with M and M-, note by note, by ear, and going for
a SLIGHTLY different speed of the pulse overall compared to what you did with M+

and THEN you put the accordion back together and hope m- m_ m+ sounding together is
simply amAZING

(this is in a perfect world, with unlimited time and patience at your disposal... children should
not attempt this without adult supervision)

 

[debra]

Fortunately the mathematical (Ventura) method is not needed because we have that "hidden" in our brain. We can just tune a few notes using exact measurements and then tune everything else "by ear". It should just be pleasing, but there are a few extra things you can do. When using the dry M when you play octaves they should be completely dry (no beating at all). When the M+ (and similarly the M-) are tuned correctly and you play octaves of M+ reeds there should be a consistent very slow beating. You may not notice a difference in MM+ beatings between adjacent notes, but with M+ octaves you can easily hear inconsistencies between adjacent notes. When the tremolo of MM+ sounds right and the M+ octaves sound right then you know for sure that the tuning will be pleasing.

 

[Dingo40]

"When the tremolo of MM+ sounds right and the M+ octaves sound right then you know for sure that the tuning will be pleasing."(Paul)
Now, repeat all that with M-, as in my Bugari Ch CAS ( 438/440/442)!?

 

[debra]

"When the tremolo of MM+ sounds right and the M+ octaves sound right then you know for sure that the tuning will be pleasing."(Paul)
Now, repeat all that with M-, as in my Bugari Ch CAS ( 438/440/432)!?

That's a slightly dangerous summary. The M+ octaves should sound consistently slightly off whereas the M octaves sound completely dry.
(And you have a typo 432 where it should say 442.)

 

[Dingo40]

Drat those pesky typos: thanks Paul!?
I was simply referring to the complexities you've just pointed out, involved in the tuning of an accordion , which wouldn't be self evident to a beginner ?

 

[debra]

Drat those pesky typos: thanks Paul!?
I was simply referring to the complexities you've just pointed out, involved in the tuning of an accordion , which wouldn't be self evident to a beginner ?

During the ACA course we all made the mistake of just checking the tremolo by ear without checking the M+ octaves. Lesson learned! To our ears the tremolo was "even" but the octaves told a different story! After adjusting them the accordion sounded clearly better than before.

 

[boxplayer4000]

I feel it is simplistic (and perhaps confusing) to talk about tunings in simple 2 reed M M+ (2 dimensional) limits. eg. 'swing' is 4 cents and 'Scottish' is 25 cents. When a 3rd reed is added eg. an M- then there are 2 more tremolos added ie. one between M- and M+ and one between
M and M-. The situation becomes much more complicated. It doesn't take long in tuning to discover that there is a big difference in sound between say -25 0 +25 and -22 0 +25.

 

[Dingo40]

indeed, as you have said boxplayer, it turns out:
"There are more things in heaven and Earth, Horatio
Than are dreamt of in your philosophy "
Complications, complications, complications!?

 

[debra]

I feel it is simplistic (and perhaps confusing) to talk about tunings in simple 2 reed M M+ (2 dimensional) limits. eg. 'swing' is 4 cents and 'Scottish' is 25 cents. When a 3rd reed is added eg. an M- then there are 2 more tremolos added ie. one between M- and M+ and one between
M and M-. The situation becomes much more complicated. It doesn't take long in tuning to discover that there is a big difference in sound between say -25 0 +25 and -22 0 +25.

Not so sure about that "big difference". Both these tunings just sound way too wet for me.
But in general, indeed symmetric or asymmetric tremolo makes a big difference. For me either -8 0 +8 (what I put on my Hohner Artiste XS) or -10 0 +10 (done on a friend's Morino) both sound good. I also did -5 0 +5 on request, but that was not much to my liking (but it was what the user wanted and he's happy with it). I believe I did put an asymmetric -10 0 +12 on my Crucianelli (no cassotto) and the asymmetric nature has not bothered me so far, but i mostly use just MM+ on that and not MMM (still too wet for my taste).

 

[Ventura]

the reason i suggest a slight difference in "beats" between M- M_ and M_ M+
is because perfect symmetry will generally over-emphasize and under-emphasize
a few frequencies

for example, it is never ever a good design to build a speaker cabinet with square dimension...
good design has the longitude different than the latitude so that different frequencies
are affected by each length (you cannot avoid frequencies being affected, so your design is
meant to minimize this... the only "perfect" speaker cabinet is literally a hole in the middle of your
Living Room Wall with a speaker mounted in it and the adjoining door between the rooms is closed)

 

[Scuromondo]

Not so sure about that "big difference". Both these tunings just sound way too wet for me.
But in general, indeed symmetric or asymmetric tremolo makes a big difference. For me either -8 0 +8 (what I put on my Hohner Artiste XS) or -10 0 +10 (done on a friend's Morino) both sound good. I also did -5 0 +5 on request, but that was not much to my liking (but it was what the user wanted and he's happy with it). I believe I did put an asymmetric -10 0 +12 on my Crucianelli (no cassotto) and the asymmetric nature has not bothered me so far, but i mostly use just MM+ on that and not MMM (still too wet for my taste).

If a player might prefer a wet sound for certain folk tunes, but want a dryer tuning for other music, could it be an advantage to him to have one of the M reed sets only slightly detuned by, say, 5 cents off while the other set is more aggressively detuned to, say, 15-16 cents? But how does that affect the quality of the MMM and LMMM registers?
What are the trade offs, as you see them, between symmetric vs. asymmetric tremolo?

 

[Ventura]

for all my MMM everyday accordions, the meusette has always been very conservative,
only the 911 (LMMM) is aggressively tuned

the 960, the FisItalia, and the Giulietti all allow for full fisted 4 and 5 finger chording
when all 3 Middles are shifted into play

so i very much prefer the asymmetrical approach as the subtle differences are
sweet in these mild tunings

(actually, the 960 is tuned as close to unison between the M- M_ as possible,
with the M+ only a very very slow pulse higher)