My tuning bellows

[James]

This bellows has two outlets, each having a valve in it, so that one to pump air in and one to pump air out. I removed the valve from one of them, so that air goes in both ways. I then covered up the other outlet.

I could have built a larger box but I just happened to have this small box which is tight and nice built. It worked great for me. One merit is I can work with both my hands. But, it did not drive properly the very big reeds from the bass blocks.

 

[wout]

Cool bellows! Can you post a video on how it's functioning? I just started out and I use the same concept, though i thought about instead of using a footpump using an electric one (just have to figure out how to reduce the noise). I don't know how it is going to turn out but otherwise I am just going oldskool with an old bellows from some scrapaccordion.

 

[James]

Ok, I will shoot it. Electric pump? Wonder what it is like. Post a pic when you are ready?

 

[losthobos]

steampunk cool....

 

[wout]

I will definately!

 

[JerryPH]

Very interesting, thanks James!

 

[James]

There, I have made a video. Thats my mantle clock ticking away in the background. I didnt realize it could be that loud.

 

[JerryPH]

Ok, something very interesting that I was thinking about and it shows something that I was not really aware of how much of a factor it would be... and this is that it takes a lot more air over a longer period of time and at a more consistent rate so that the reed has time to stabilize, and give you more accurate reading.

I am no expert, but from what I see, the meter has not had enough time to decide exactly what it is that it's hearing to stop moving between too sharp or too flat and therefore, hard to say "yes it is exactly 2 cents sharp" or whatever, and deal with it.

I think I would want a goal to have a tuner that can give me a minimum of 6-7 seconds of steady tone from the deepest of notes (the ones that take the most air to sound properly), and from what I see on my own accordion, playing the deepest bass note requires at least a 1/2 to 3/4 amount of travel from a set of full sized bellows. Now, that or (obviously) a motorized suction or push of air at a constant velocity from a (either) a very quiet and close source or a larger long hose to keep the local ambience as quiet as possible and not "infect" what the computer or device is hearing.

Fun to speculate.

 

[James]

Hi Jerry,

Really fun to speculate. These days, after I saw your post, I have tried several tuners and found the following occurred to me:

1) The tuners. Of the tuners I tried, none of them stopped however long I played a note. I used my accordion and the testing bellows. I also played endless long notes on my Yamaha keyboard. I set it on each of the five versions of "flute" besides some other sounds.

2) Velocity of the air flow from bellows. When I pressed on my foot driven pump, the air flow from it was not constant as I could not maintain a constant pressure on it. But when I took away my foot and the bellows expanded on its own, the air flow seemed to me to be constant. But the tuners I tried still never stopped. The bellows did not last longer than 6 seconds.

3) The reeds. I noticed, be it in the accordion or on my testing bellows, the reeds I tried did vibrate at slightly different pitches.

So, conclusion? How about this: At least, to fine tune a reed on a testing bellows, it does make sense to have, at least, a constant air flow that lasts a long time. Only motorised bellows can do this job.

Fun to speculate.

 

[Mr Mark]

Interesting idea!

I'm wondering what the purpose of the (wine?) box is for though.

Does it act as an air pressure stabilizer or something to that effect?

Would it work with the black tube routed directly to the hole?

 

[kimric]

If you can set the tuner to the actual note you are going to play it will sync faster in a lot of cases. I use a peterson strobe tuner so I select the note anyway as it can lock on a harmonic fifth sometimes.
This system works pretty good but the tuning table will almost always have some harmonics of its own and will require further tuning in the accordion. It was a lot of work to get rid of the harmonics in my table.

I use a old electrolux vac as my blower and have a hose from the suction and the exhaust that goes to a shuttle valve I made that allows me to flip a leaver on the table to reverse the air flow as needed. I made the hoses out of coiled wire an leather to get rid of any motor noise, and I run the vac using a old 6-12 battery charger to keep the pressure low . You can hardly hear it, flipping the voltage and the charging amperage rate (2-6 amp) gives me four speeds. The lower two I use the most. I am thinking of changing this for a PWM motor controller for more options.

 

[Edocaster]

My experience of feeding a reed too directly from a hose is the reed is 'overblown'. Some cavity seems necessary to gently build a pressure differential.

On the subject of tuners, I've noticed that decent recording software (e.g. Audacity or Adobe Audition) can display spectral displays or have dedicated frequency analysis windows. If you record all notes played quite slowly (at least two seconds per note) and view the spectrum, you can 'see' the notes and line them up to frequencies (you'll need to find a chart online with all the note frequencies). As long as you set an FFT size very high (65536 samples or so - this is why the note needs to be at least two seconds), the accuracy is good. You can even see separate peaks for vibrato tuning. Obviously, this isn't a live solution, but is a way of verifying tuning after the event.

 

[James]

My experience of feeding a reed too directly from a hose is the reed is overblown. Some cavity seems necessary to gently build a pressure differential.

On the subject of tuners, Ive noticed that decent recording software (e.g. Audacity or Adobe Audition) can display spectral displays or have dedicated frequency analysis windows. If you record all notes played quite slowly (at least two seconds per note) and view the spectrum, you can see the notes and line them up to frequencies (youll need to find a chart online with all the note frequencies). As long as you set an FFT size very high (65536 samples or so - this is why the note needs to be at least two seconds), the accuracy is good. You can even see separate peaks for vibrato tuning. Obviously, this isnt a live solution, but is a way of verifying tuning after the event.

You are right. I did try to attach the hose to the reed plate only to find the reed overbown.