Microphone?

[Tim Arnold]

This is an odd request, but I'm curious if its possible / feasible to hook up a microphone to an analog input. Not for recording or anything, purely to measure volume.

Anybody heard (no pun intended) of something like this being done before, or have ideas on how to do it?

[seanl]

it sounds feasible, but i haven't tried anything like that

[Ian G]

I think you would need some sort of amplifier circuit, so the RC gets a range between 0 and 5 volts. But it would definitely be feasible, though I could see it getting a lot of noise (ha ha).

Just curious, what do you want to measure with your microphone.

[Matt H.]

I believe your best bet would be to find a pre existing sensor which suits your needs.
If you simply want to measure sound level and don't see it going much over 90decibels then you might want to pillage a Lego NXT sensor which I believe works on a 5v system.
Unless you absolutely need specific sound recognition I would recommend going with a decibel meter rather than a microphone as microphones don't simply measure "loud" but instead send a somewhat complex audio signal.

[Mr. Freeman]

Wouldn't you end up getting a sine wave (or similar)? Seems to me that you would have to sample the analog input fast enough to get a reliable set of data.

I remember talking to someone about oscilloscopes and they mentioned that your sample rate (samples per second) has to be roughly double your frequency (Hz) in order to get an accurate idea of the period of the sine wave. If you draw it out (draw a sine wave and then plot two points per period, connect those) you see that depending on where those points fall, you may not get a reliable measure of the amplitude of the sine wave.
Wikipedia says that the lowest audible frequency (for humans) is 20 Hz, meaning you would need a sample rate of at least 40 samples/second. Of course, most sounds you would want to measure are much above this.

I'm not sure how fast the robot controller can actually sample the analog inputs, so this may or may not work.

[Matt H.]

Mr. Freeman
What you say is correct if you want to analyze the actually sound wave and what is happening--if you just want to measure the intensity much less work/sampling is required. Intensity measurements can be either frequency dependent (over the human hearing range dba) or non frequency dependent (over all frequencies dbc). What they need really depends on their application.

[Kevin Watson]

Originally Posted by Tim Arnold

This is an odd request, but I'm curious if its possible / feasible to hook up a microphone to an analog input. Not for recording or anything, purely to measure volume.

Anybody heard (no pun intended) of something like this being done before, or have ideas on how to do it?

Yes, this is possible. The ADC software on my website can sample at 6400Hz, which is good enough to sample frequencies as high as 3200Hz (See here for why). You would need a filter to remove frequencies above 3200Hz to prevent aliasing and an amplifier to use as much of the five volt dynamic range as possible.

-Kevin

[Al Skierkiewicz]

You will ned an amplifier to bring up the mic level (~-50dBm ref 1 mW) up to a usable level. A simple precision rectifier would then give you an average DC level which you could then sample to get an idea of the sound level over the long term. A little fudging in software and you could get a good approximation of SPL based on your sample. Remember that most SPL measurements are weighted anyway (frequency limited). I would guess that a logrithmic approximation times 2 would be about right. Of course some serious high pass filtering in the mic preamp will be needed to keep robot crashes, um, contacts out of your measurement.
BTW, to my ears, on the field sound is at least 10-15 db quieter than in the stands. Remember the PA is aimed at the crowd, not the field.

[Qbranch]

Ok, I'll give you that you CAN do it, but i'm wondering WHY you want to do it. FIRST events are insanely loud at times (and have a high volume general murmur about them) so I'm guessing you're not using this for any kind of control...

So... are you making lights that go to the music? I have no other ideas?

-q