Melodeon.net Forums

Please login or register.

Login with username, password and session length
Advanced search  

News:

Welcome to the new melodeon.net forum

Pages: 1 2 3 [4] 5   Go Down

Author Topic: Resonance issues with reedblocks  (Read 6970 times)

0 Members and 1 Guest are viewing this topic.

mselic

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1683
Re: Resonance issues with reedblocks
« Reply #60 on: October 09, 2025, 04:50:58 AM »

If you don't mind me asking, what are your thoughts on the bluestars vs vintage Catraros? I'm still tempted to get a set of the latter for a hromka build, given their reputation with Galliano.

BTW, the new Catraro reeds (diatonic) are only available through Maxim's Group Accordions based in Castelfidardo.  According to them, the new Catraro reeds are a joint venture between them, Cagnoni and Cingolani, and look to be made in the Cagnoni factory.  I was told they share many of the same qualities as the old reeds (but who knows what that actually means?).  I inquired about a 3v, two row treble set and was quoted a price of 850 EURO!  BlueStar and Binci reeds are in the 500 EURO range for the same set, so I'll be saving my money...
Logged
C#/D player

mChavez

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 713
Re: Resonance issues with reedblocks
« Reply #61 on: October 09, 2025, 07:24:33 AM »

Ouch.
It's not really a problem if you have a customer paying for his dream custom box, but I'd find it very hard to justify spending that much to prototype a box for myself without hearing them first, especially now that I'm in the unisonoric camp.
But one of the hromka advantages is that I can shop around for a smashed high end CBA or PA and pull out a-mano reeds out of it at £10 per hundred if I'm particularly lucky.
« Last Edit: October 09, 2025, 07:27:33 AM by mChavez »
Logged

Graham W

  • Respected Sage
  • ****
  • Offline Offline
  • Posts: 277
Re: Resonance issues with reedblocks
« Reply #62 on: October 09, 2025, 03:58:03 PM »

In case it might interest, here’s a sound comparison of three different makes of reeds, one of which is a repurposed set from Catraro, all in the same instrument at different times (all MMM @ approx +-5 cents). The Catraros tend to have a fairly bright sound not unlike Bincis.

Catraro:
https://www.dropbox.com/scl/fi/1w3ev3744sz8bxlc37ye0/Catraro-reeds-comparison.m4a?rlkey=ufv157fnug6fxwgw8aj5tvt18&st=3sxbdsf3&dl=0

Binci:
https://www.dropbox.com/scl/fi/0jfyrxfgpp54pd38cec45/Binci-reeds-comparison.m4a?rlkey=3q8msmhc0x3728eajmlxi657j&st=nzp8uzf2&dl=0

Voci Armoniche TAM:
https://www.dropbox.com/scl/fi/zayyf3592jydmhzlzoyus/VA-TAM-reeds-comparison.m4a?rlkey=4ol3yw2cx8dnqfpzg47q7wh6k&st=3gtktw4q&dl=0

Thanks for the comparison. If it's of any help, I've attached spectrograms.

This and listening to the files confirms Binci are way too bright for my liking. Although they work in some configurations, it's a bit tiring to the ears after some time, I think. I like VA TAM for that, much more presence of the firsts overtones, which feels more soothing.

I'm a bit late to this, but thanks, it's interesting stuff.   

Does anyone have any comparitive experience of the three different types of 'A Mano' reed offered by Harmonikas?
Logged

Willem Feltkamp

  • Regular debater
  • ***
  • Offline Offline
  • Posts: 129
  • Tinkering and playing diatonics and CBA
Re: Resonance issues with reedblocks
« Reply #63 on: October 12, 2025, 11:38:04 AM »

I’m a bit flummoxed my post generated by AI got zero reaction.
I find the notion that the complex behaviour of air, influenced by vibrating reeds, and the interaction between air streams can cause a coupling effect leading to wah-wah effect very interesting. If anything, it seems that whatever is going on inside an accordion in terms of physics, it’s complicated stuff.
The second thing I get from the AI-generated post is that there seems to have been done quite an amount of investigation in the acoustics of accordions; ChatGPT didn’t make the information up, but rather, being the sophisticated search engine it is, it managed to find information from a variety of different sources. When asked which they were, the answer was “It drew primarily from Cottingham’s survey of free-reed acoustics, Fletcher & Rossing’s foundational physics, and general nonlinear acoustics of coupled oscillators.”
So a theory to explain persistent wah-wah even between coupled reed blocks might be that it is simply the fact that two reeds of the same make might produce sounds that are so identical that certain frequencies/overtones lock and produce other frequencies.
This would explain why swapping one of the reeds with one from another brand helps remedy the problem; the acoustic ‘fingerprint’ of the new reed differs enough from the other one to prevent the ‘locking’ of frequencies.
Logged

Theo

  • Administrator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 14475
  • Hohner Club Too
    • The Box Place
Re: Resonance issues with reedblocks
« Reply #64 on: October 12, 2025, 11:43:19 AM »

My sceptical view of A1 generated answers like this is that they may also draw information from misinformation on the web.  There is plenty of that published about free reeds!
Logged
Theo Gibb - Gateshead UK

Proprietor of The Box Place for melodeon and concertina sales and service.
See my YouTube channel  for videos on repair and tuning

IanD

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1703
  • Too many melodeons...
Re: Resonance issues with reedblocks
« Reply #65 on: October 12, 2025, 03:48:45 PM »

My sceptical view of A1 generated answers like this is that they may also draw information from misinformation on the web.  There is plenty of that published about free reeds!
True -- but what it said was pretty much correct, which is that any coupling mechanism can cause a problem between closely-tuned reeds.

However to get away from the general "anything might couple through the air" case, I doubt that this is the "bad reeds" problem being discussed, because such air coupling is similar for all reeds regardless of frequency. The problem is much more likely to be coupling via a mechanical resonance in the blocks/fondo/case at a particular frequency, because such a resonance greatly increases coupling (or decreases isolation, which is the same thing) around the resonant frequency.
Logged
Oakwood Model 4 D/G, Castagnari Dony D/G/#, Castagnari Tommy G/C, Baffetti Binci D/G, Hohner Preciosa D/G, Melos Bb/Eb, Lightwave SL5 and Kala California fretless basses

Chris Ryall

  • "doc 3-row"
  • French Interpreter
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 11153
  • Wirral UK
    • Chris Ryall
Re: Resonance issues with reedblocks
« Reply #66 on: October 12, 2025, 04:23:26 PM »



  I totally agree. Here is a simple experiment. Initially the board cannot couple the metronomes as it is immobile due to the friction of the table.

  As soon as you put the rollers underneath it look what happens.

  Physical coupling is a very powerful phenomenon in anything that vibrates.
  Particularly if the vibrational frequencies are close together.

   https://youtu.be/Aaxw4zbULMs
« Last Edit: October 12, 2025, 05:04:57 PM by Chris Ryall »
Logged
  _       _    _      _ 

Willem Feltkamp

  • Regular debater
  • ***
  • Offline Offline
  • Posts: 129
  • Tinkering and playing diatonics and CBA
Re: Resonance issues with reedblocks
« Reply #67 on: October 17, 2025, 06:40:21 AM »

(getting back to the original post) If mechanical coupling was the cause, then why did swapping one of the reeds for another brand help minimize the problem?
Logged

tirpous

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1403
Re: Resonance issues with reedblocks
« Reply #68 on: October 17, 2025, 08:34:00 PM »

Could be because the reeds have different physical properties, the blade could be thinner/thicker, narrower/wider, shorter/longer, profiled differently, made of another type of steel. The plate could be thicker or thinner, with a different weight. Etc...

Turning the argument around: if the coupling mechanism is indeed 'through the air' rather than mechanical, it's hard to imagine why a brace linking blocks (which has basically no impact on air movement inside the box) is effective to fix the problem in most cases.
Logged

Theo

  • Administrator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 14475
  • Hohner Club Too
    • The Box Place
Re: Resonance issues with reedblocks
« Reply #69 on: October 17, 2025, 09:56:14 PM »

But does it have to be either or? It could be one or other or both in different setups.
Logged
Theo Gibb - Gateshead UK

Proprietor of The Box Place for melodeon and concertina sales and service.
See my YouTube channel  for videos on repair and tuning

IanD

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1703
  • Too many melodeons...
Re: Resonance issues with reedblocks
« Reply #70 on: October 18, 2025, 10:50:49 AM »

But does it have to be either or? It could be one or other or both in different setups.
Yes it could -- except that coupling through the air is more likely to affect many reeds, and it's usually just one or two at particular frequencies which suggests mechnical coupling.

Alos things that change this (braces, swapping reed plates) often make a difference, which is less likely if the couping was acoustic.

Plus with two apparently identical boxes one can have the problem and one not; you'd expect the acoustic coupling to be thew same since all the air spaces are identical, but there will undoubtedly be mechanical resonance differences due to small variations on material/manufacture (wood, glue...).

All this suggests the mechanical coupling is mostly the problem, though of course there may be occasions where it's acoustic -- it's like the difference in corruption levels between the Tories and Labour, not all politicians are squeaky clean or guilty but I think we all know where the biggest problems lie... ;-)
Logged
Oakwood Model 4 D/G, Castagnari Dony D/G/#, Castagnari Tommy G/C, Baffetti Binci D/G, Hohner Preciosa D/G, Melos Bb/Eb, Lightwave SL5 and Kala California fretless basses

mselic

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1683
Re: Resonance issues with reedblocks
« Reply #71 on: October 19, 2025, 04:14:11 PM »

Ah. Right. Well there's the solution  (:)

Turns out that making deeper reedblock chambers may not be what mitigated the problem.  While the new block I made did indeed have deeper chambers and appeared to have mitigated most of the problem, I found that then reducing the chambers with shims did not seem to make any difference in this particular instance.  Furthermore, it would appear that my new block did not actually fix the issues; when I went with a drier tuning, it became apparent that the issues are still there.

Here's another interesting phenomenon that occurred: I decided to tune the instrument with two of the three sets of reeds tuned dry (so I went 0,0,+5 instead of -5,0,+5).  Since it was the two banks of reeds facing each other than seemed to suffer from the resonance issues, I put the two dry-tuned sets facing each other, and the + set on the other side of the block.  In theory, there should have been no resonance issues since any tremolo between reeds that weren't in close proximity to each other had not been an issue.  Using the stops to isolate reeds, I sounded the following:

0,0 = locked in unison with no tremolo
0,+5 = perfect tremolo with no resonance issues
the *other* 0 and +5 = perfect tremolo with no resonance issues

Should sound great when I put them all together, yes?  NOPE.  0,0,+5 ended up with a very unusual beating, often with noticeably increased tremolo that wasn't always consistent.  It sounded odd, unclean and wonky.  I'd go back and isolate the different banks and check the tuning, and everything was spot-on.  The issue seemed to be that the quality of the beating between one set of 0 and +5 was different from the other 0 and +5, and when put together they sounded wonky.  Now, I realize that tuning a box 0,0,+5 is not quite the same as just 0,+5 - I've done it in the past and you do get a *slight* more complex sound because you have two 0s that aren't always perfectly in unison, but never anything like what I'm getting here.  I wonder if what I'm getting here is in part due to an acoustic coupling rather than a mechanical one as mentioned in the AI post?

BTW, I have to admit I'm not a huge proponent of AI or ChatGPT - I've yet to consciously use either.  My experience with it when other people have used it for various purposes is that it tends to get a lot of information wrong which is mixed in with information that is accurate, so I've not found it to be a good resource in that sense.  However, I appreciate that Willem posted it as it presented information that I had not yet considered.
Logged
C#/D player

IanD

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1703
  • Too many melodeons...
Re: Resonance issues with reedblocks
« Reply #72 on: October 21, 2025, 11:08:13 PM »

Ah. Right. Well there's the solution  (:)

Turns out that making deeper reedblock chambers may not be what mitigated the problem.  While the new block I made did indeed have deeper chambers and appeared to have mitigated most of the problem, I found that then reducing the chambers with shims did not seem to make any difference in this particular instance.  Furthermore, it would appear that my new block did not actually fix the issues; when I went with a drier tuning, it became apparent that the issues are still there.

Here's another interesting phenomenon that occurred: I decided to tune the instrument with two of the three sets of reeds tuned dry (so I went 0,0,+5 instead of -5,0,+5).  Since it was the two banks of reeds facing each other than seemed to suffer from the resonance issues, I put the two dry-tuned sets facing each other, and the + set on the other side of the block.  In theory, there should have been no resonance issues since any tremolo between reeds that weren't in close proximity to each other had not been an issue.  Using the stops to isolate reeds, I sounded the following:

0,0 = locked in unison with no tremolo
0,+5 = perfect tremolo with no resonance issues
the *other* 0 and +5 = perfect tremolo with no resonance issues

Should sound great when I put them all together, yes?  NOPE.  0,0,+5 ended up with a very unusual beating, often with noticeably increased tremolo that wasn't always consistent.  It sounded odd, unclean and wonky.  I'd go back and isolate the different banks and check the tuning, and everything was spot-on.  The issue seemed to be that the quality of the beating between one set of 0 and +5 was different from the other 0 and +5, and when put together they sounded wonky.  Now, I realize that tuning a box 0,0,+5 is not quite the same as just 0,+5 - I've done it in the past and you do get a *slight* more complex sound because you have two 0s that aren't always perfectly in unison, but never anything like what I'm getting here.  I wonder if what I'm getting here is in part due to an acoustic coupling rather than a mechanical one as mentioned in the AI post?

BTW, I have to admit I'm not a huge proponent of AI or ChatGPT - I've yet to consciously use either.  My experience with it when other people have used it for various purposes is that it tends to get a lot of information wrong which is mixed in with information that is accurate, so I've not found it to be a good resource in that sense.  However, I appreciate that Willem posted it as it presented information that I had not yet considered.

That's not surprising; if the coupling between one 0 and +5 is different from the other 0 and +5, when you put all three together the +5 will pull each of the 0 by different amounts (and vice versa, the +5 will get pulled more with both 0 reeds in action) and you'll get some kind or inconsistent "wonky" sound, as you said. It doesn't matter whether the coupling mechanism is acoustic or mechanical, the effect will be the same.

If this is happening on several reeds (or lots of them) the coupling is probably acoustic, if it's only happening on one or a few it's more likely to be mechanical. Which is it?
Logged
Oakwood Model 4 D/G, Castagnari Dony D/G/#, Castagnari Tommy G/C, Baffetti Binci D/G, Hohner Preciosa D/G, Melos Bb/Eb, Lightwave SL5 and Kala California fretless basses

Willem Feltkamp

  • Regular debater
  • ***
  • Offline Offline
  • Posts: 129
  • Tinkering and playing diatonics and CBA
Re: Resonance issues with reedblocks
« Reply #73 on: October 25, 2025, 03:51:07 PM »

A question related to the posts above: what does it mean to say a set of reeds is tuned +5 ?
Is a specific reference note tuned 5 cents sharp?
Or should I just interpret this as a rough indication of the wetness of the tuning?
Logged

mChavez

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 713
Re: Resonance issues with reedblocks
« Reply #74 on: October 25, 2025, 04:21:55 PM »

Cents are quoted for A=440 reference.
Of course, this does not tell you the musette "taper" from bass to treble, so it's a rough indication of wetness.
Logged

IanD

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1703
  • Too many melodeons...
Re: Resonance issues with reedblocks
« Reply #75 on: October 25, 2025, 05:12:19 PM »

Cents are quoted for A=440 reference.
Of course, this does not tell you the musette "taper" from bass to treble, so it's a rough indication of wetness.

It's a pity that cents are used when discussing melodeon/accordeon tuning, where the issue is not absolute pitch accuracy (which is where cents come into their own, that's how pitch sensitivity works) but the tuning difference between reeds, because what the ear/brain hears is beats per second (Hz) not cents...

For example, depending on pitch of reeds (e.g. L/M/H) and key (e.g. G/C or D/G) and where the note is on the keyboard, the number of cents detuning for a given beat rate varies hugely, and there's no inherent understanding of what "5 cents detuning" means because it means completely different things in different circumstances.

In contrast, saying reeds are tuned 1Hz or 2Hz apart is immediately clear, and applies everywhere regardless of reed pitch or key or note position.
Logged
Oakwood Model 4 D/G, Castagnari Dony D/G/#, Castagnari Tommy G/C, Baffetti Binci D/G, Hohner Preciosa D/G, Melos Bb/Eb, Lightwave SL5 and Kala California fretless basses

mChavez

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 713
Re: Resonance issues with reedblocks
« Reply #76 on: October 25, 2025, 05:19:02 PM »

In contrast, saying reeds are tuned 1Hz or 2Hz apart is immediately clear, and applies everywhere regardless of reed pitch or key or note position.

Sorry, can't agree with that. You are describing one extreme case of a very flat taper. I do not know anyone who tunes boxes like that (fixed number of hz for every note across the whole range).
The other extreme alternative is to keep the number of beats flat for every note across the whole range. I've heard one box like that - it sounded very interesting, but not something I'd like to have as my main or even secondary box.
Most de-tunings are somewhere between these two extremes.

I do agree that cents at A=440 is not an accurate description, but even if you spend an hour discussing tapers, you are still left with other factors, like reeds and box design. Two identically de-tuned boxes (same offset and taper) can still sound completely different if one of them has cassotto, or the reeds have very different timbres.
« Last Edit: October 25, 2025, 05:21:04 PM by mChavez »
Logged

IanD

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1703
  • Too many melodeons...
Re: Resonance issues with reedblocks
« Reply #77 on: October 26, 2025, 01:08:51 PM »

I didn't say that you should use a fixed beat rate across the range -- I said that what the ear hears between two detuned reeds is bps/Hz not cents, so using this to specify reed (de)tuning makes far more sense.

For example 5 cents is a frequency difference of 0.3%, which is about what the normal human ear can detect as a frequency error on single tones at medium frequencies -- trained musicians can sopposedly detect smaller errors than this.

But 5 cents is 0.75Hz at middle C, 1.5Hz an octave above, 3Hz two octaves above, 0.37Hz an octave below and so on -- these vary from almost dry to quite wet tuning on a two-reed box.
Logged
Oakwood Model 4 D/G, Castagnari Dony D/G/#, Castagnari Tommy G/C, Baffetti Binci D/G, Hohner Preciosa D/G, Melos Bb/Eb, Lightwave SL5 and Kala California fretless basses

Aaron Wildman

  • Member
  • *
  • Offline Offline
  • Posts: 36
Re: Resonance issues with reedblocks
« Reply #78 on: October 26, 2025, 04:46:33 PM »

Forgive me if I missed this somewhere.  Is it possible that the problem lies with the receiver rather than the transmitter?  Perhaps something quirky with your ear (hyper sensitivity or something more obscure?) that creates what you are perceiving? Some sort of aural fixation or hypersensitivity or some other obscure cause?

Have you used any software to record and analyze?  Have you got any other pairs of ears to confirm your findings (getting a bigger sample size etc)? 

You did say something about having sensitive hearing, so I thought approaching that from multiple angles may provide additional information about the resonance problem, or absence of, or perhaps clarify its severity.

Logged

-Y-

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 896
  • Handry 18 (G/C), Mélodie (D), Club IIb (A/D)
    • a database of nearly 500 melodeons here
Re: Resonance issues with reedblocks
« Reply #79 on: October 26, 2025, 06:19:42 PM »

I didn't say that you should use a fixed beat rate across the range -- I said that what the ear hears between two detuned reeds is bps/Hz not cents, so using this to specify reed (de)tuning makes far more sense.

For example 5 cents is a frequency difference of 0.3%, which is about what the normal human ear can detect as a frequency error on single tones at medium frequencies -- trained musicians can sopposedly detect smaller errors than this.

But 5 cents is 0.75Hz at middle C, 1.5Hz an octave above, 3Hz two octaves above, 0.37Hz an octave below and so on -- these vary from almost dry to quite wet tuning on a two-reed box.

I find the benefit of using cents is to work on the whole range of notes. When tuning an accordion that goes from C3 to C7 for instance, the absolute scale of Hz difference is more difficult to work with when you go up. Using a relative cents scale allows to appreciate more easily the decrease in discrepancy you usually apply the higher you go.
Pages: 1 2 3 [4] 5   Go Up
 


Melodeon.net - (c) Theo Gibb; Clive Williams 2010. The access and use of this website and forum featuring these terms and conditions constitutes your acceptance of these terms and conditions.
SimplePortal 2.3.5 © 2008-2012, SimplePortal