Need an air flow generating unit for automatic wind instrument

Hi,
I'm making a project that incorporates several DIY PVC recorders that are going to be played by Arduino and I need a system that generates air flow to play them. I have some spare fans from my old pc laying around but they are a bit too wide and not powerful enough (unless I make a complicated tubing system to aggregate the wide airflow into narrower and stronger one). I was thinking of hacking a mini vacuum cleaner (http://www.ebay.co.uk/itm/300817216993) cause they tend to have more condensed and stronger air flow (I'd have to invert the direction though). The only thing is that the mini vacuum cleaner might be a bit too loud. I don't mind a certain amount of noise from the fan but I have no idea how loud the mini vacuum cleaners are. I also found this tutorial on how to build a blade-less air fan (How to make a bladeless fan.wmv - YouTube) but I have no idea if the air pressure created by it would not just go back through the massive hole in the middle. My question is - is there any other way to generate air flow? Or should I stick to hacking the mini vacuum cleaner? Also, I want to be able to play several recorders with one fan if possible - does anyone have any suggestions on how to arrange them so the air flow from the fan would be equal strong to all of them?

Thanks.

what's the pressure you need? the wind speed ? volume/sec?

small compressor (air brush)

robtillaart:
what's the pressure you need? the wind speed ? volume/sec?

small compressor (air brush)

I am sorry I didn't provide more details but I have no way to measure the air pressure needed for a PVC recorder. It is not a lot - way less than for a flute or saxophone or other proper wind instrument. I am just looking into ways/components to generate air pressure so I can experiment and find a proper solution. I think an air brush might work good - I haven't thought of that. Thanks

Perhaps a bellows, like an accordion. You can make them from paper. Or maybe you can find an old bagpipe. That would be stylish.

-br

An air brush won't work, you want the pressurized air supply that powers it.

Sounds like you need what we call an "air pig" or a compressed air tank. More importantly you need a pressure regulator to vary and maintain pressure. Not to be confused with a pressure gauge which only reads the pressure, a regulator regulates the pressure. You may find this stuff at a good hardware store.

How long do you need to operate it for? I'm no musician but I've seen them played and have the impression they need quite a bit of pressure - probably several PSI - which would be hard to generate from a fan. I suspect you could operate it from a balloon for several seconds if you wanted to try that to prove the principle - then your problem would be how to get a pump/fan to generate similar pressure at the required flow rate.

PeterH:
How long do you need to operate it for? I'm no musician but I've seen them played and have the impression they need quite a bit of pressure - probably several PSI - which would be hard to generate from a fan. I suspect you could operate it from a balloon for several seconds if you wanted to try that to prove the principle - then your problem would be how to get a pump/fan to generate similar pressure at the required flow rate.

Best case scenario I'd need them to play indefinitely until I turn them off myself. The recorders I have made are of a small scale and require less pressure than regular wind instruments. I am tempted to try out the mini air compressor like this
www.google.co.uk/products/catalog?hl=en&client=firefox-a&rls=org.mozilla:en-US:official&channel=fflb&q=mini+air+compressor&bav=on.2,or.r_gc.r_pw.r_cp.r_qf.&bpcl=38625945&biw=1846&bih=987&um=1&ie=UTF-8&cid=5286243728415743297&sa=X&ei=YrCqULPdO8a90QWZh4CYAQ&ved=0CH0Q8wIwAw
it produces up to 250 psi so it is strong enough. It might be too loud and I am not sure if I can find a good way to control the pressure of air that it produces.

Here is a thread that suggests 0-10 psi is the normal human production playing the flute. It also suggests a small compressor for a similar project:

-br

Best case scenario I'd need them to play indefinitely until I turn them off myself. The recorders I have made are of a small scale and require less pressure than regular wind instruments. I am tempted to try out the mini air compressor like this
Google Shopping - Product not found
it produces up to 250 psi so it is strong enough. It might be too loud and I am not sure if I can find a good way to control the pressure of air that it produces.

That is capable of inflating something to 250psi but the air pressure coming out of it is much smaller. It might actually be not to bad for your needs. Your best bet is an air tank, with a regulator on it inflated by a compressor. A regulator is that good way to control the pressure as well as experiment with different pressures. The idea behind a regulator is that you could have tank with 250 psi in it and the regulator would limit it's output to say 10 psi. It will also deliver 10 psi until the tank's pressure dropped below that. I'd also recommend a pump that plugs into mains, the one you linked to really needs a car battery.

Thanks everyone for your help, I've done quite valuable googling based on your suggestions and found several other candidate components for my project:

Did anyone have any experience using those in projects?

Jimmy60:

Best case scenario I'd need them to play indefinitely until I turn them off myself. The recorders I have made are of a small scale and require less pressure than regular wind instruments. I am tempted to try out the mini air compressor like this
Google Shopping - Product not found
it produces up to 250 psi so it is strong enough. It might be too loud and I am not sure if I can find a good way to control the pressure of air that it produces.

That is capable of inflating something to 250psi but the air pressure coming out of it is much smaller. It might actually be not to bad for your needs. Your best bet is an air tank, with a regulator on it inflated by a compressor. A regulator is that good way to control the pressure as well as experiment with different pressures. The idea behind a regulator is that you could have tank with 250 psi in it and the regulator would limit it's output to say 10 psi. It will also deliver 10 psi until the tank's pressure dropped below that. I'd also recommend a pump that plugs into mains, the one you linked to really needs a car battery.

the major problem with those mini air compressors for car tires is that they are extremely loud - as I mentioned before I don't mind some noise but not so much that you can't hear the recorders themselves. From the videos on youtube these air compressors seem very very loud: Air Compressor and Copic Air Brush System Basics - YouTube, http://youtu.be/x9lZbmrBpIY

LOL

I just came back to mention that those car tire pumps are really loud. Most air compressors are but a tank of compressed air isn't.

Yes, car tyre inflators are always very loud, and also tend to be very low flow rate (and the higher pressure ones are even lower).

I suspect you need something in the region of a couple of liters per minute at a few psi, but it would really be a good idea to find out what pressure and flow rate you actually need. There are plenty of ways to do this, even starting with how hard you have to blow and how long a lungful of air lasts. Come on, you've got these things in front of you.

billroy:
Here is a thread that suggests 0-10 psi is the normal human production playing the flute.

2.3 PSI (120 mm Hg) is normal blood pressure and about the best that a fit person can blow in short bursts. 3.5 PSI (180 mm Hg) is dying. The 6 PSI stated by your thread sounds remarkably high.

To the OP's question, the best suggestion I can think of for an inexpensive, quiet pump would be the compressor from a refrigerator. Google will turn up a few results on how woodworkers convert them for use as vacuum presses.

Chagrin:

billroy:
Here is a thread that suggests 0-10 psi is the normal human production playing the flute.

2.3 PSI (120 mm Hg) is normal blood pressure and about the best that a fit person can blow in short bursts. 3.5 PSI (180 mm Hg) is dying. The 6 PSI stated by your thread sounds remarkably high.

To the OP's question, the best suggestion I can think of for an inexpensive, quiet pump would be the compressor from a refrigerator. Google will turn up a few results on how woodworkers convert them for use as vacuum presses.

Thanks for the info - that's very useful. The fridge compressor might work but is a bit big. For now I have ordered one of these http://www.ebay.co.uk/itm/RC-Submarine-MICRO-AIR-PUMP-DC-3v-7-2v-/280999690355?pt=US_Character_Radio_Control_Toys&hash=item416ce30073. If it doesn't work well for me I'll definitely try the fridge compressor option.

That submarine pump creates a fair bit of pressure but is lacking in cfm like the small car compressors. I don't think a fridge compressor will give you enough flow either.

Wouldn't a decent sized centrifugal fan work for this? I thought recorders don't have all that much air resistance. In fact wikipedia mentions pipe organs use pressures in the order of 0.1 psi which equates to roughly 2.75 inch-aqua. I'm pretty sure even small centrifugal fans that were used in computers years ago can create static pressures higher than that.

jabbado:
Wouldn't a decent sized centrifugal fan work for this? I thought recorders don't have all that much air resistance. In fact wikipedia mentions pipe organs use pressures in the order of 0.1 psi which equates to roughly 2.75 inch-aqua. I'm pretty sure even small centrifugal fans that were used in computers years ago can create static pressures higher than that.

The fan needs to be an impeller type and not squirrel cage. Even a ~250W furnace blower (which are squirrel cage) don't do anything for static pressure (DAMHIKT). You could use something like a vacuum cleaner motor or pool toy / mattress inflator but the problem is that they're too damn noisy.

To your question on CFM for a refrigerator compressor it looks like they're in the 1 CFM (28 liter/minute) range and capable of ~100 PSI. Surprising number of examples of "homemade air compressors" on Youtube using these pumps. They are oil piston compressors just like typical large compressors so I suppose it does make sense.

Chagrin:
The fan needs to be an impeller type and not squirrel cage. Even a ~250W furnace blower (which are squirrel cage) don't do anything for static pressure (DAMHIKT).

I used to think that. A few years ago I hooked up a 100mm centrifugal fan to a high speed mini drill motor and passed the outflow through an 8mm orifice. We were levitating ping pong balls a good few feet above it. And the noise wasn't all that bad.

To your question on CFM for a refrigerator compressor it looks like they're in the 1 CFM (28 liter/minute) range and capable of ~100 PSI. Surprising number of examples of "homemade air compressors" on Youtube using these pumps. They are oil piston compressors just like typical large compressors so I suppose it does make sense.

I don't think 1 cfm will be enough. But if it was then you'd want to build a buffer tank to smooth out the oscillations. And if you're going to do that, then just buy a cheap 2hp compressor.

BTW, even these 2.4 W fans can pump 9.5 cfm with a static pressure of 6.5 inch-aqua. And only produce 32dBA.
http://www.jaycar.com.au/productView.asp?ID=YX2530

Isn't the flow keeping the ping-pong balls aloft, not the pressure? I have seen in vacuum cleaner shops where they have a beach ball hovering above their latest model cleaner. I would think it would not require very much pressure at all, but a decent flow rate.

I'd be trying a compressor designed for airbrushing or spray tanning, they are usually quiet.

What about a CPAP machine? One would think that it's capabilities would mimic those of a human's, due to the way they function.

Just ideas for you mate!

Ash

Edit, what about inflating a car tyre beforehand and using that as a cheap reservoir for the compressed air? you would still need a regulator.

Ashduino:
Isn't the flow keeping the ping-pong balls aloft, not the pressure? I have seen in vacuum cleaner shops where they have a beach ball hovering above their latest model cleaner. I would think it would not require very much pressure at all, but a decent flow rate.

Sort of. Something has to create that high speed flow. And that is pressure. We were putting bits of blu-tack on it to make it spin and stuff. There's no way I could've made that ball stay floating with my lungs. Anyway I tried a 4.5 watt squirrel cage out of an old HP computer today. I had no recorder so made a ghetto one. It seems to work fine. Pretty sure this is more about decent flow rather than high pressure. It would be different if trying to play a trumpet or reed instrument.

What about a CPAP machine? One would think that it's capabilities would mimic those of a human's, due to the way they function.

That's a really good idea. The fixed pressure versions seem to put out enough pressure. The problem though is the cost. They're damn expensive.

I will agree with those who have said that this is more about airflow than pressure. I play clarinet and tenor sax, and I used to play the recorder. Comparatively, the recorder takes basically no air pressure or volume. A casual exhale is more than enough to get a sound on the recorder. A good breath of air (3-5 L) could play the instrument for about a minute if not more. What you will need to worry about is a very constant stream of air. Any pressure spike could cause the recorder to over-blow and thus squeak.

Don't discount the shaping of the air into the instrument either. The recorder is less sensitive to this that most wind instruments, but the angle at which the air is introduced will affect the sound.