Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[diatonix]

One example that I know of for sure is that the sound holes in the bass end cover form a Helmholtz resonator with the volume of the air inside the bass end, and this acts as a lowpass filter (which you do want to remove the treble spectrum and stop it obscuring the right hand) but with a peak around cutoff -- this can lead to an unpleasant "honky" sound from the bass end.
My Dony 3-row has smaller cover hole area and more internal volume, a lower resonant frequency and a less nasal sound (but is also more muffled) -- since the reed blocks in the Model 4 were based on the ones in my 3-row I know this isn't causing the difference.

But I doubt whether many box makers are aware of what's going on, at best they probably drill more holes until they got the sound they wanted --
Ian

I may not be  aware of much of what's "going on" but even I have realized how much the bass end cover affects the sound. However, instead of drilling more holes I omit them all together (apart from necessary but relatively small openings for air and register buttons & bass strap) and place them on either side of the  buttons on the front of the casing. I find that by doing this I get a more powerful bass sound, not "honky" at all and not in the least muffled. Interestingly, but maybe not surprisingly, there is a huge difference when the cover isn't screwed on tightly, resulting in a weaker and less attractive sound.
(Not that I am planning to, but what's wrong with someone building your perfect box??)

[IanD]

One example that I know of for sure is that the sound holes in the bass end cover form a Helmholtz resonator with the volume of the air inside the bass end, and this acts as a lowpass filter (which you do want to remove the treble spectrum and stop it obscuring the right hand) but with a peak around cutoff -- this can lead to an unpleasant "honky" sound from the bass end.
My Dony 3-row has smaller cover hole area and more internal volume, a lower resonant frequency and a less nasal sound (but is also more muffled) -- since the reed blocks in the Model 4 were based on the ones in my 3-row I know this isn't causing the difference.

But I doubt whether many box makers are aware of what's going on, at best they probably drill more holes until they got the sound they wanted --
Ian

I may not be  aware of much of what's "going on" but even I have realized how much the bass end cover affects the sound. However, instead of drilling more holes I omit them all together (apart from necessary but relatively small openings for air and register buttons & bass strap) and place them on either side of the  buttons on the front of the casing. I find that by doing this I get a more powerful bass sound, not "honky" at all and not in the least muffled. Interestingly, but maybe not surprisingly, there is a huge difference when the cover isn't screwed on tightly, resulting in a weaker and less attractive sound.
(Not that I am planning to, but what's wrong about someone building your perfect box??)

Moving the holes from the cover to the front doesn't change the basic Helmholtz resonator structure, but depending on the number/diameter of the holes and the thickness of the wood it will change the tuning frequency. My guess from your description is that you've moved the tuning frequency down which makes the bass sound richer (just like my Dony), either because the wood is thicker (which increases the mass of air in the "ports") or the total hole area is lower (because there's less space) or both. If the cover then leaks the tuning will move up again.

(you can find the equations for reflex ports easily, but it gets a lot more complicated with multiple closely-spaced holes)

The answer to your last question is that if somebody is going to step up and put considerable time effort and money into my ideas to prove that they work (or don't -- that's the risk element!), it only seems fair to give them first crack of the whip at selling the resulting boxes.

Cheers

Ian

[Matthew B]

At the risk of repeating what's already been said . . . the big "resonance" impact is, then what happens beyond the pallets?  From the Concertina Connection paper, and my own casual observations, the treble end grill, and the construction of the bass end box both seem to exert a significant influence on the final sound.  This would explain the variety of choices builders make around grill design, grill material, and grill liners, or the absence thereof.  It would also account for the variety of holes that show up on both ends of various box designs, including the ones that some makers now put on the fingerboard, and the ones that point backwards towards the player.  And it would also explain the name of the growl box on a one-row, and the little trumpety jobs that show up on the bass end of Steirische boxes. 

To put it another way: it's the bling that makes it sing.

[Anahata]

it would also explain the name of the growl box on a one-row, and the little trumpety jobs that show up on the bass end of Steirische boxes. 

This is all very interesting about the bass end.
The little metal cones in the "trumpety jobs" don't make much difference to the sound on a Steirische, but there is a hole in the middle of them (usually covered with thin-weave cloth) and that probably does make a big difference.

I might try some experiments with my Pokerwork, which is very much a "spare" box: cover the holes in the bass end plate, and make new holes in the front. I guess the thing to do is make holes one at a time, record and maybe measure the response as each hole is added and stop when the optimum sound is reached...

[oggiesnr]

Interesting page on the Strasser site http://www.harmonika.com/en/wissen.htm

Steve

[IanD]

At the risk of repeating what's already been said . . . the big "resonance" impact is, then what happens beyond the pallets?  From the Concertina Connection paper, and my own casual observations, the treble end grill, and the construction of the bass end box both seem to exert a significant influence on the final sound.  This would explain the variety of choices builders make around grill design, grill material, and grill liners, or the absence thereof.  It would also account for the variety of holes that show up on both ends of various box designs, including the ones that some makers now put on the fingerboard, and the ones that point backwards towards the player.  And it would also explain the name of the growl box on a one-row, and the little trumpety jobs that show up on the bass end of Steirische boxes. 

To put it another way: it's the bling that makes it sing.

Simple answer -- yes. It's what I've been saying from the beginning, that after the reeds the thing that affects the sound most is what sits between them and the outside world, especially the shape/area/reflectivity of the air path including obstacle like grilles, baffles etc. You probably still don't want any resonances in the grille because this tends to make them vibrate and rattle (though this will be damped by cloth stuck to them), the percentage free area and thickness is what matters not the precise shape of the holes.

The trumpets on Steirische boxes don't act as horns even if they look like them, they're far too small (a horn which works at these frequencies is bigger then a Steirische!) -- what they actually are is relatively long narrow ports with a flared exit, precisely what you do in reflex speakers to lower the tuning and prevent air turbulence. But they look a lot prettier than black holes, and since the bass end is supposed to sound like a tuba they act as a miniature reminder of this :-)

It's also quite easy to think about how holes in different places distribute the sound, but one thing to remember is that anything small compared to a wavelength can't be directional. Since a wavelength at 1kHz is about 30cm (10cm at 3kHz, 3cm at 10kHz) then you can only really get much directionality at higher frequencies in the lateral plane, thought the height of the box can give more focusing in the vertical plane -- but only if the sound comes out of all the holes uniformly, which it won't do because it starts from one reed...

Holes in the keyboard are used for boxes with lots of reed banks where the ones nearest the player would otherwise be shadowed by the keyboard, preventing the sound escaping freely. Holes on the back towards the player are a dual-edged sword; they might let sound escape out the back towards the player (so it sounds better to them), but this must mean there's less sound available to come out of the holes facing the audience. Or they might help to kill any cavity resonance of the reeds with the space under the keyboard.

All this is basic acoustic theory as applied to direct radiating, reflex and horn speakers of which there is a vast amount of material on the web, including the effects of cavity and box (body) resonances, horns, tubes and so on. A melodeon might not look like a speaker but exactly the same laws of physics apply -- in fact it is more like a speaker than any other instrument if you think about it carefully... ;-)

Cheers

Ian

[IanD]

it would also explain the name of the growl box on a one-row, and the little trumpety jobs that show up on the bass end of Steirische boxes.  

This is all very interesting about the bass end.
The little metal cones in the "trumpety jobs" don't make much difference to the sound on a Steirische, but there is a hole in the middle of them (usually covered with thin-weave cloth) and that probably does make a big difference.

I might try some experiments with my Pokerwork, which is very much a "spare" box: cover the holes in the bass end plate, and make new holes in the front. I guess the thing to do is make holes one at a time, record and maybe measure the response as each hole is added and stop when the optimum sound is reached...

No "probably" about the Steirische, see my previous reply ;-)

It won't make any difference where the holes are; the number, diameter and panel thickness are what matters. If the front panel is thicker than the end plate, glue another piece of wood to the end plate instead to make it thicker then you don't have to drill holes in the casework. All the thickness this changes is the number or size of holes to tune to a specific frequency (see below).

If you're interested the formula for calculating the resonant frequency is here (you'll have to rearrange it to solve for Fb instead of Lv):

http://forum.speakerplans.com/design-of-rectangular-ports-for-a-bassreflex-box_topic50507.html

So bigger holes (or more of them) increase the resonant frequency, deeper holes decrease it.

Have fun :-)

P.S. For this to work properly the "box" (including the end panel) has to be reasonably rigid for the same reason a speaker box has to be -- if the walls move in and out too the numbers change.

[Anahata]

Interesting page on the Strasser site http://www.harmonika.com/en/wissen.htm

"the test results of universities and resonance research institutes have confirmed that the sound of a harmonica does not depend on the material of the body. We have accomplished lots of tests with different materials (Metal, plastic, diverse massive woods and plywoods) and that is why we could determine ourselves that the material of the body does not exercise any influence on the sound of the instrument."

Well, that settles that, then. 

It's also quite easy to think about how holes in different places distribute the sound, but one thing to remember is that anything small compared to a wavelength can't be directional.

If that was directed at my musings about retuning the bass cavity on my Pokerwork, fully undertood.
My intention was
(a) to try different sized holes for different resonant frequencies (this would affect the bass), and
(b) to get some higher frequency content out of the bass end directed at the audience.

2nd posting noted - thanks 

[IanD]

Interesting page on the Strasser site http://www.harmonika.com/en/wissen.htm

"the test results of universities and resonance research institutes have confirmed that the sound of a harmonica does not depend on the material of the body. We have accomplished lots of tests with different materials (Metal, plastic, diverse massive woods and plywoods) and that is why we could determine ourselves that the material of the body does not exercise any influence on the sound of the instrument."

Well, that settles that, then.  

It's also quite easy to think about how holes in different places distribute the sound, but one thing to remember is that anything small compared to a wavelength can't be directional.

If that was directed at my musings about retuning the bass cavity on my Pokerwork, fully undertood.
My intention was
(a) to try different sized holes for different resonant frequencies (this would affect the bass), and
(b) to get some higher frequency content out of the bass end directed at the audience.

2nd posting noted - thanks  

About the Strasser quote -- it's nice to be *proved* right for once... ;-)

-- and I take back what I said about no manufacturer having done this kind of investigation properly -- now it's probably just *most* that haven't...

On point (b), see how much comes out of the holes after retuning anyway and then decide if you need more of it towards the audience -- but for the frequencies a bass end will generate (even high harmonics) I doubt if you'll be able to get much directionality.

By the way, Strasser are now making super-light carbon fibre boxes, but I hate to think what the cost is :-)

http://www.harmonika.com/en/aktuelles.htm

-- answer: 1800 euros *extra* 

[Owen Woods]

We do play a strange instrument!

[Chris Ryall]

From another thread...

For  the "21st century box a la Dedic", I've suggested a bass end stop that removes the thirds and replaces them with a doubled up fifth, but an octave away from the existing fifth instead of in unison. This should give the best of both worlds, assuming it's technically possible... :-)

(but is almost certainly impossible to retrofit to existing boxes because of the extra reeds and the sliding "swap stop")

To keep the chords sounding most similar in the two cases, this means the "normal" chord has the fifth as the lowest note, the root in the middle and the third as the highest one -- the stop then swaps out the third for the fifth above it.  Ian

and another idea ...

I'm wondering if what I'd really like is stops that changed A and B between major and minor. Ideally separately.

[Cooper]

and another idea ...

I'm wondering if what I'd really like is stops that changed A and B between major and minor. Ideally separately.

the Libouton i once played had 2 buttons next to each base. One for major, and one for the minor chord.
w

[Anahata]

and another idea ...

I'm wondering if what I'd really like is stops that changed A and B between major and minor. Ideally separately.

the Libouton i once played had 2 buttons next to each base. One for major, and one for the minor chord.
w

I've heard of boxes that do that. It's probably an easier solution than adding a stop button.
The full set of D,A,E,and B in both major and minor flavours would be a big addition.

While we're on the subject of extra bass buttons though, I like the one Roger Watson had made specially with a 9th button giving C# and F# basses:

                        F#/C#
C/C      Cmaj/Cmaj       B/E      Bmaj/Emin

G/D      Gmaj/Dmaj       D/A      Dmaj/Amaj

Resulting in a full scale in both major keys on the basses. I could use that a lot!

[oggiesnr]

All of which raises the question as to whether we should just have a 36 or 48 button stradella bass.

Steve

[Theo]

All of which raises the question as to whether we should just have a 36 or 48 button stradella bass.

Steve

And a piano key keyboard

[oggiesnr]

All of which raises the question as to whether we should just have a 36 or 48 button stradella bass.

Steve

And a piano key keyboard

Nope because that loses the beauty of the melodeon "keyboard".  If I hit the next button by mistake on a melodeon it probably sounds alright, if I do it on a piano it's a discord!

Steve

[Owen Woods]

Haha, just had a wacky idea. How about valves which are a composite with a low coercivity (soft magnetic). Mount the reeds on zinc plates and have an electromagnet running under the plates the entire length of the block. Alternatively mount them on some ferrimagnetic substance and have the reed plates be the electromagnet. If you want to cut out a voice, just apply a magnetic field and all the valves should seal, as they would be attracted to the magnet.

The more I think about this, the more I realise how ridiculous it is and how it wouldn't work in any circumstances, but I thought I'd post it in order to give those with a scientific background amusement as they pick it apart. I never understood electromagnetism.

[Martin J]

Interesting page on the Strasser site http://www.harmonika.com/en/wissen.htm

"the test results of universities and resonance research institutes have confirmed that the sound of a harmonica does not depend on the material of the body. We have accomplished lots of tests with different materials (Metal, plastic, diverse massive woods and plywoods) and that is why we could determine ourselves that the material of the body does not exercise any influence on the sound of the instrument."
Well, that settles that, then. 

Whilst bowing to the superior knowledge of the above testers, my personal experience was different.  When I bought my Dony I took it to bits to see why it was so much better than my Polkawork.  All I could conclude was finish and superior fit.  The wood inside the Dony was smoother, properly finished whereas the Polkawork was rough sawn. I sanded and sealed the inside of my Polkawork and the difference was dramatic.  I agree that the material probably made no difference but having a sealed surface that reflected sound did make a difference.

So back to topic, a plastic case should produce a brighter sound than a wood one.  I've never seen inside the latest Hohner Xtreme models.  Are they perhaps plastic ?

Lastly, can anyone find the thread (it was on melnet) for the guy who built what can only be described an off road version of the Hohner 114.  He perhaps epitomises new thinking for melodeons.

[TomB]

UKBert, as it doesn't seem to take much to get a sticky valve to stop a reed sounding, I think your idea has some mileage. Problem is, in a portable instrument, how would you power the electromagnet? How about a generator worked off the bellows movement? Sort of like the contraption that harnesses the power of tidal waves. Then of course, you've got the problem of residual magnetism holding the valves down when the power is off, so you need the same power supply to service a degaussing arrangement. Simple really, you're a GENIUS 

[Owen Woods]

Or you could have a stop