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]   Go Down

Author Topic: tremolo  (Read 2645 times)

0 Members and 1 Guest are viewing this topic.

mselic

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1683
tremolo
« on: August 15, 2017, 04:18:05 AM »

I've done some research on this forum, but I'm still pretty unclear about a few things regarding tuning one set of reeds to get tremolo.  On my Beltuna one-row, the LMH reeds are all tuned to A440 (each note on each of the LMH rows registers as "in tune" with my tuner set to A440).  The M+ row is tuned to A441 (each note on the M+ row registers as "in tune" when my tuner is set to A441). With this, I get exactly the same amount of tremolo on each note; that is to say that the tremolo does NOT increase at all from the lowest note to the highest.  On my Hohner, for example, the M row is tuned to A442 (with each note registering as "in tune" with the tuner set to A442), and the M+ row is tuned to A446 (with each note registering as "in tune" with the tuner set to A446). However, with this, the tremolo DOES increase from the lowest note to the highest.  If the difference in tuning between the notes on both boxes is constant, why does one box have increasing tremolo, and the other not? I'm very confused!!
Logged
C#/D player

playandteach

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3526
  • Currently a music teacher in a high school.
Re: tremolo
« Reply #1 on: August 15, 2017, 07:25:39 AM »

It might seem silly for me to guess when there are others out there who know, but otherwise we'd have no discussion forum, just a group of expert consultants.
I'd guess that the differential in tuning causes interference at different rates when the pitch difference is greater. Perhaps it's a perfect storm at a closer tuning.
On another point, does it not sound more natural to have a slightly faster bleat higher up?
Logged
Serafini R2D2 GC, Serafini GC accs 18 bass

Steve_freereeder

  • Content Manager
  • Hero Member
  • ***
  • Offline Offline
  • Posts: 7511
  • MAD is inevitable. Keep Calm and Carry On
    • Lizzie Dripping
Re: tremolo
« Reply #2 on: August 15, 2017, 09:05:10 AM »

I think that what you (mselic) are noticing are some of the 'uncertainties' which can arise when tuning the reeds and relying too much on what your electronic tuner is telling you.

Yes - theoretically the beat rate is the difference between the frequencies of the two reeds. So to quote your example: with one reed tuned to A=442 and another to A=446, you will get a beat rate of 446 minus 442 = 4 Hz (beats per second). And - if all your reeds are tuned to those same pitch difference throughout the range you will get the same beat rate of 4 Hz on all the notes.

However - just to consider some of the uncertainties:

1. Are you totally sure that your reeds are tuned exactly to 442 and 446 standards throughout? You really need to measure the pitches with an accurate meter to be sure. The one I use reads to the nearest 0.1 cent. A tuning meter which at best only reads to the nearest couple of cents isn't going to be able to confirm that you have the reeds tuned exactly to 442 and 446 standards. A cent or so either way will make a noticeable difference to the beat rate of two reeds.

2. The pitch of a reed depends partly on the air pressure supplied to the reed. As a general rule, the pitch tends to be sharp at low air pressure and flat at higher air pressure. If you measure the pitch of the M and M+ reeds separately, they will be responding to a given bellows force resulting in air pressure P. However, when you sound the M and M+ reeds together (as in normal playing conditions), the same bellows force is now shared between the two reeds so the air pressure going to each reed will drop  to some value < P. This will make a small difference to the pitch of the reeds and if they are not exactly and precisely set up to respond equally, then that will result in a change in the tremolo rate. This effect is often noticeable on the highest pitch notes: squeeze the bellows a bit harder and the tremolo rate tends to drop.

3. The pitches of the M and M+ reeds will interact with one another (acoustic coupling) so that the pitch of one reed will affect the other to a certain extent. In an extreme case where the difference in pitch is small, the two reeds will lock themselves together to vibrate at the same frequency and there will be no tremolo. This is most often noticed in the higher pitched reeds, presumably because there is less mass to withstand the acoustic coupling. So a pitch difference of 2 Hz between the lowest pitched reeds will generally give a useable tremolo of 2 Hz, but going up the range there will come a point where the reeds will lock together and destroy the tremolo. The solution is to (a) add a brace between the reed blocks to try to improve the rigidity and (b) if this doesn't entirely solve the problem, accept that such a small tremolo rate is not going to be achievable, and so the reeds should be tuned slightly wetter.

4. In the case stated, a LMMH box, the other reeds will also interact with the M and M+ reeds, altering their theoretically tuned values slightly. The overall tremolo rapidly becomes very complex and difficult to predict. Even more so when you consider that reeds (especially the larger ones) do not vibrate at a single frequency, but generate a spectrum of harmonics, all of which will interact with each other, generating a very rich and complex sound.

Overall, the theoretical pitch difference between the M and M+ reeds is a good starting point for tuning to achieve a target tremolo. But it is only that - a starting point. A good human reed technician will then switch the electronic tuner off and rely solely on the biological tuner (i.e. the ears) and will make tiny adjustments to the reeds to get to the desired rate of tremolo in actual playing conditions. This is why tuning is an art, not an exact science, and needs considerable practice to get it right.
« Last Edit: August 15, 2017, 03:26:27 PM by Steve_freereeder »
Logged
Steve
Sheffield, UK.
www.lizziedripping.org.uk

mselic

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1683
Re: tremolo
« Reply #3 on: August 15, 2017, 12:41:56 PM »

Thanks for the input, Steve. I'm sure the tuner I'm using isn't *exact*, but it seems to me like this wouldn't explain why the tremolo on the Beltuna is perfectly constant, while on the Hohner it gradually and consistently increases. This is obviously by design. Even if there was minor variations in the tuning (I'm sure there are) this doesn't seem to explain why the tuning in the two boxes is overall behaving in very different manners. My only thought was that perhaps the wider the overall gap (eg 442 vs 446, rather than just 440 vs 441)  the more the tremolo would increase with higher notes? Are you saying that if one set of reeds is tuned to 442, and the other to 446, that there should be a constant beat across the notes??
Logged
C#/D player

mselic

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1683
Re: tremolo
« Reply #4 on: August 15, 2017, 01:02:39 PM »

Just a small correction: the Beltuna M+ row is tuned to A442, not A441.
Logged
C#/D player

Rob2Hook

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2278
  • Castagnaris, Hohners & Baffetti
Re: tremolo
« Reply #5 on: August 15, 2017, 01:24:17 PM »

I have lots of rather disconnected memories of tuning details preferred by various people.  IIRC Ian Dedic prefers a constant beat across the fingerboard, others like a tremolo rate increasing with pitch.

I once dropped into Roger's workshop and he showed me a book full of details and graphs of tuning profiles characteristic of major manufacturers.  So notwithstanding Steve's masterful explanation of some of the reasons and limitations involved it seems to be largely a matter of taste.

Rob.
Logged

Steve_freereeder

  • Content Manager
  • Hero Member
  • ***
  • Offline Offline
  • Posts: 7511
  • MAD is inevitable. Keep Calm and Carry On
    • Lizzie Dripping
Re: tremolo
« Reply #6 on: August 15, 2017, 03:28:25 PM »

Thanks for the input, Steve. I'm sure the tuner I'm using isn't *exact*, but it seems to me like this wouldn't explain why the tremolo on the Beltuna is perfectly constant, while on the Hohner it gradually and consistently increases. This is obviously by design. Even if there was minor variations in the tuning (I'm sure there are) this doesn't seem to explain why the tuning in the two boxes is overall behaving in very different manners. My only thought was that perhaps the wider the overall gap (eg 442 vs 446, rather than just 440 vs 441)  the more the tremolo would increase with higher notes?

Have you actually isolated each reed block, and measured and recorded the tuning of each and every one of the reeds? If you do this, I suspect you will find there is considerable variation in the actual tuning compared with what you suspect or expect it to be.

One of the first things I do when approaching tuning a melodeon (once I'm satisfied that all the reeds and valves are working as they should be) is to record the actual pitch of every reed when played in the instrument under normal bellows pressure conditions. Usually I will enter the values into a spreadsheet and plot a graph so I can visualise any deviations from what pitch the reeds should be.

Quote
Are you saying that if one set of reeds is tuned to 442, and the other to 446, that there should be a constant beat across the notes??

Yes - if that is difference in frequency between every M reed and every M+ reed, i.e. 4 Hz difference always. As previously stated, the tremolo rate (beat frequency) is the numerical difference between the frequency of the two reeds being sounded.

Tremolo rate = fM+ - fM

However, if you keep the pitch standard equal across the range, then you will definitely get an increase in tremolo rate across the range. E.g. consider two sets of reeds, one tuned to pitch standard 440 and the other tuned to 444. Then:

Note A3: M reed = 220 Hz, M+ reed = 222 Hz. Resulting tremolo will be 222 - 220 = 2 Hz. (nice)
Note A4: M reed = 440 Hz, M+ reed = 444 Hz. Resulting tremolo will be 444 - 440 = 4 Hz. (all right)
Note A5: M reed = 880 Hz, M+ reed = 888 Hz. Resulting tremolo will be 888 - 880 = 8 Hz. (getting very wet)
Note A6: M reed = 1760 Hz, M+ reed = 1776 Hz. Resulting tremolo will be 1776 - 1760 = 16 Hz. (hideously wet to the point of being unbearably out of tune!)

Hope this helps.
Logged
Steve
Sheffield, UK.
www.lizziedripping.org.uk

mselic

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1683
Re: tremolo
« Reply #7 on: August 15, 2017, 04:02:02 PM »

Steve, everything made sense until you spoke about "pitch standard", and then I got horribly lost  :P Your chart that follows lists the M reeds being tuned to 222, 444, 888, etc, and I'm not sure what you mean there! Sorry if I'm a bit thick...
Logged
C#/D player

Steve_freereeder

  • Content Manager
  • Hero Member
  • ***
  • Offline Offline
  • Posts: 7511
  • MAD is inevitable. Keep Calm and Carry On
    • Lizzie Dripping
Re: tremolo
« Reply #8 on: August 15, 2017, 07:10:55 PM »

Lester seems to have repeated what I said in my previous post...   ???

OK - pitch standard...

Musical instruments are made and tuned to an internationally recognised standard, the idea being that players in different countries can get together, in orchestras, etc., and their instruments will be in tune with each other. This is especially important with instruments which cannot easily or quickly be retuned - the melodeon, accordion, church organ etc. all come into this category. Most string instruments can and need to be tuned each time you start to play.

Without going into too much detail: historically, the pitch of instruments used to be different in different countries. The scales of those instruments were based around the note A being at, for example A=432, 435, even as high as 452. All other notes in the scale were tuned relative to the particular A tuning in use at that time and place. Result = international chaos. In 1939 by international agreement, concert pitch was defined as based on A = 440 Hz and this is still the only agreed standard in use today, with the idea being that instruments would be built to that standard. Some countries, notably Italy and Japan, are starting to push that pitch standard higher again, to A=442 or even 443, mainly in the quest for a more brilliant sound. Not a good idea in my opinion as the old chaos will return. Some professional orchestral players (notably woodwind players) are now having to equip themselves with separate instruments depending on which country they are travelling to.

More information about pitch standards here:
https://en.wikipedia.org/wiki/Concert_pitch

How does this relate to melodeons?
If we adopt the international pitch standard of A=440 Hz, then each octave higher doubles the frequency; each octave lower halves the frequency.
So the octaves for the note A are as follows:
A1=55 Hz
A2 = 110 Hz
A3 = 220 Hz
A4 = 440 Hz (our reference standard)
A5 = 880 Hz
A6 = 1760 Hz
and so on...

The other notes in the scale are found by dividing each octave into 12 equal parts (this is what's known as Equal Temperament). When we say that a melodeon is "tuned to A=440", what it means is that the A notes are tuned to the relevant multiples of 440 as shown above, and all the other notes are divided equally into 12 parts in each octave.

Many Italian-built instruments are tuned to a pitch standard of A=442 (it's that quest for a bright sound again), in which case the various octaves would be A1=55.25, A2=110.5 , A3=221, A4=442, A5=884, A6=1768, etc... and the intervening notes divided equally in each octave, as just described.

But our melodeon reeds M and M+ are usually deliberately tuned differently to give a tremolo. Whatever pitch standard the M reeds are tuned to, be it A=440 or A=442, etc., the M+ reeds must be tuned such that the difference in frequency gives the desired tremolo rate. To say that the M reeds are tuned to A=440 and the M+ reeds tuned to A=442 is a misleading simplification. Yes - you could do that, but as demonstrated in my previous post, as you went higher up the range, the tremolo would get too fast to be acceptable. In practice the M+ reeds are tuned to whatever degree of sharpness is needed to give the desired tremolo for each note.

I'm not sure I've answered your questions satisfactorily - we seem to be going round in circles to some extent. But nevertheless, ask specific questions and I will do my best to answer them.

 

Logged
Steve
Sheffield, UK.
www.lizziedripping.org.uk

mselic

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1683
Re: tremolo
« Reply #9 on: August 15, 2017, 07:57:47 PM »

Thank you both Lester and Steve - I really do appreciate your input.  I *think* I get it...

To clarify, allow me to ask a question using the Beltuna as an example (using only the MM+ voices).  All of the following measurements are accurate within a couple of tenths of a Hz.

With my electronic tuner set to A440, on the M row:

A3 = 220Hz
A4 = 440Hz
A5 = 880Hz

With my electronic tuner set to A442, on the M+ row:

A3 = 221Hz
A4 = 442Hz
A5 = 884Hz

Does this mean that, with both the M and M+ rows opened up and sounding together, that the amount of tremolo (beats per second) at A5 will be twice that of A4? And the tremolo at A3 half of what was at A4?
Logged
C#/D player

Steve_freereeder

  • Content Manager
  • Hero Member
  • ***
  • Offline Offline
  • Posts: 7511
  • MAD is inevitable. Keep Calm and Carry On
    • Lizzie Dripping
Re: tremolo
« Reply #10 on: August 15, 2017, 08:14:03 PM »

With my electronic tuner set to A440, on the M row:

A3 = 220Hz
A4 = 440Hz
A5 = 880Hz

With my electronic tuner set to A442, on the M+ row:

A3 = 221Hz
A4 = 442Hz
A5 = 884Hz

Does this mean that, with both the M and M+ rows opened up and sounding together, that the amount of tremolo (beats per second) at A5 will be twice that of A4? And the tremolo at A3 half of what was at A4?

No - you are not comparing like with like.

1. Set your tuner to A=440 and leave it there.
2. measure the pitch of the M row reeds (you've already done that for the A3, A4, A5)
3. measure the pitch of the M+ reeds (e.g.for A3, A4 and A5)

All other things being equal (as they say...), then the theoretical tremolo rate will be as follows:

For the note A3, tremolo rate = measured frequency of A3M+ minus the measured frequency of A3M
For the note A4, tremolo rate = measured frequency of A4M+ minus the measured frequency of A4M
For the note A5, tremolo rate = fmeasured frequency of A5M+ minus the measured frequency of A5M

and so on...
Logged
Steve
Sheffield, UK.
www.lizziedripping.org.uk

mselic

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1683
Re: tremolo
« Reply #11 on: August 15, 2017, 09:18:11 PM »

A-ha! I knew I wasn't understanding something...

I checked again, with my tuner set to A440:

M row:
A3 = 220
A4 = 440
A5 = 880

M+ row:
A3 = 221
A4 = 441
A5 = 882

so tremolo with both M and M+ reeds would be: A3 @ 1bps, A4 @ 1bps, A5 @ 2bps.  Listening closely to these notes (and measuring bps), this sounds about right.

Logged
C#/D player

Steve_freereeder

  • Content Manager
  • Hero Member
  • ***
  • Offline Offline
  • Posts: 7511
  • MAD is inevitable. Keep Calm and Carry On
    • Lizzie Dripping
Re: tremolo
« Reply #12 on: August 15, 2017, 09:38:57 PM »

Yes - you've got the idea now!  (:)
Logged
Steve
Sheffield, UK.
www.lizziedripping.org.uk
Pages: [1]   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