using a microphone to measure speed

[kev 406]

My problem is that i am trying to measure the speed of a fan using a microphone with labview, any ideas??????

[PdB]

Go ahead! A seemingly crazy combination of a sensor that is not developed for a specific task can result in a very good measurement.
I read about a german company that uses microphones to measure the speed and intensity of traffic in different lanes. Or a dutch company that uses sound to measure air-temperatures.
One thing I'm sure about, it will take a lot of experimenting.

[Enrique]

kev;

An approach can be to determine a transfer function that relate the speed of the fan and the dominant frequency of the sound generated by the fan motor. (Probably is just w = 2(pi)f, but I haven't done any search on that nor I don't know what can affect the sound)

Positioning the microphone is important, as you don't want all the noise of the air hitting directly to the microphone element (Unless that's how you are suppose to do it)

If you just need to determine if the fan is in low/medium/high or any other discrete speed, the problem can be a lot more simpler.

Do you have any restrictions on how to do this, other than using a mic? What type of fan is it?

Regards;
Enrique

[Eric Inazaki]

You might also have to know the number of blades the fan has.


On Fri, 8 Mar 2002 07:41:15 -0800 (PST), Enrique wrote:

>kev;
>
>An approach can be to determine a transfer function that relate the
>speed of the fan and the dominant frequency of the sound generated by
>the fan motor. (Probably is just w = 2(pi)f, but I haven't done any
>search on that nor I don't know what can affect the sound)
>
>Positioning the microphone is important, as you don't want all the
>noise of the air hitting directly to the microphone element (Unless
>that's how you are suppose to do it)
>

[Labviewguru]

,

And I thought this would be a tough one....

Input the microphone to your sound card, Run the fan and capture the sound. Send the sound through an FFT. If the microphone is placed near the outside of the blades, facing the output of the fan, you will get a "beat frequency" of the fan blades. This will give you a very low frequency. Divide by the number of blades and 60 and you will get the RPM of the fan.

What you are trying to do is to isolate the frequency that occurs when a fan blade passes in front of the microphone, compressing the air, pushing the cel of the mic in, then the blade passes, decompressing the air (in relative space of course), pulling the cel out. This is a very strong component of the sound you will measure with the mi
crophone, and should be very easy to see in the FFT. If you want to get fancy, you can isolate this peak in the FFT, apply the math to convert it to RPM, and output the fan speed.

Good luck. 'Sounds' like fun...

[kev 406]

hi again
how do i go about reading the pulses from the sound card, does anything else have to be connected to the FFT ? will the program be capable of processing pulses that fast. thanx
garreth

[urs boegli]

kev schrieb:

> My problem is that i am trying to measure the speed of a fan using a
> microphone with labview, any ideas??????

Much easier : use light to do it. Buy a cheap reflective
photointerruptor (ie from Sharp) and mount it somewhere where it see the
blades rotating. Output will be a digital signal which could be analyzed
or which can be transformed with a low pass to a 'dc' voltage to be
measured with a meter or daq card

Good Luck

Urs Bogli
Aritron AG

[explorer]

How to setup the microphone in order to measure sound.

[LesHammer]

Go to the LabVIEW help section. Look up "sound". It gives several examples that use the SI (Sound Input) VIs.

Les.Hammer@CompleteTest.com