Microphone suggestion for speaker testing - Thank You in Advance

Status
Not open for further replies.
Can anyone tell me what advantage is gained by adding the combination of the 9v battery, 2.2K Ohm resistor, and 1 uF capacitor?

Two advantages:

1) Not all soundcard microphone inputs provide the DC bias them, the bare capsule will only work on the ones that do.

2) Many (probably most) soundcard microphone inputs don't have flat frequency response. Bear in mind that these inputs are usually meant for picking up voice (teleconferencing), not for high fidelity. Rolling off the low frequencies and the high frequencies both is advantageous for voice-only applications and is pretty common for sound card mic inputs.

So your safest bet is to use a line (not mic) input, though that probably means you'd want more gain since line ins aren't as sensitive. In which case, more than just the three parts would be needed (an opamp, and some other parts).
________

While you're downloading free programs to try, can I suggest also trying my freeware SynRTA program?

SynRTA

(No licensing, costs, or rights issues at all with it).

The advantage of SynRTA over standard RTAs is that you don't have to wait for noise to average before the curve is consistent. SynRTA is synchronous in operations and so very time efficient to use. In that it is more like an MLS system, except you can still use recorded noise signal on a CD to play through a system (rather than direct from a soundcard) if you want. Fewer cables to run around, but make sure you use only the proper "psuedo-pink" noise that the program provides, as it is keyed to the software! Otherwise it can't operate synchronously.

Though SynRTA by itself still won't solve the mic issue for you... Maybe you'll be lucky and have one of the soundcards that both provide micbias and have flat mic response.

You can check for circuit response, by the way, by using SynRTA and a resistor divider circuit (see below) running from the line out to the mic input -- you can then see if your mic input will give a usable response (if response isn't too bad, you could have SynRTA compensate it even).

Resistor divider circuit (resistor values can be anything near these):

Line Out Hot > ----[100k ohms] ------------- >Mic In Hot

Mic In Hot ---------[1k ohms]----------- Mic In Ground (shield)
 
The modified Panasonic is fine, IF it is calibrated. I like many, just bought the pre-calibrated Behringer as I have a suitable preamp. My Panasonic I built has a mini jack so it can feed a sound card direct. NO mic is at all useful if it has not been calibrated and your measurement can not accept the cal file. NONE. Do I need to say that again? It must be calibrated. Oh yea, one more thought. It must be calibrated. A mic is only slightly flatter than a speaker.
 
I disagree that the uncalibrated Panasonic mic is not useful. I calibrated mine against a B&K reference mic I have at work and it was within +/-0.5dB up to ~8kHz. I think that would be useful for the majority of crossover design work. If you want to know what your speaker is doing at very high frequencies you'll have to get something that's calibrated, but it does usually require more expensive gear to measure things at the limits of a range...

You could possibly calibrate the mic yourself, but it might be a bit tough. You can use a concept called reciprocity to calibrate the absolute sensitivity of a transducer using reversible transducers without any outside reference device. That is, you have to have transducers that can function as both speakers and microphones. I did this the other day using headset transducers - similar to tweeters. The way I did it required three transducers. The procedure and relevant equations are in Beranek's Acoustics. The tricky parts are that (at least for me), the voltages I was measuring were very small - millivolts or less and you need a voltmeter that will give you a precise enough reading. I used a meter that could resolve to thousandths of a millivolt. Also, I was doing a calibration in a sealed chamber which requires that the chamber size be <1/20 wavelength (iirc) of the frequency you're calibrating. There's a free field method as well, but at least in my case the voltages would likely have been even smaller. In theory you could do this at all the frequencies you were interested in and then you'd have a calibrated transducer. Then you could compare the response of this transducer to a sound source to the response of your measurement microphone in order to calibrate the measurement mic.

It's probably easier to find someone with a reference measurement mic and get them to calibrate your mic for you.
 
You could possibly calibrate the mic yourself, but it might be a bit tough.

That sounds as sweet as honey.

I am no expert nor professional.
My calibration goes this way..
I have an,
hd 448 head phone and ecm8000 microphone with ne5532 balanced pre amps.

Additional requirments : Good mood and patience.

I compare variety of real sounds with and without hd448 through Microphone. and I keep adjusting the equalizer in my computer until I get that almost perfect replica through Headphones.
I sometimes ask my wife to talk or get near a water tap,or near a shower ,vessels, finally the pink noise and white noise from my speaker should sound same through Headfone as I hear without them.
somehow I manage to get a near perfect setting in my equalizer with repeat ability finally I am convinced how the mic is.
but it is a tedious task and worst is People around gives a bit serious looks to me.
 
How can anybody calibrate mic as a Diy'er ?
any unique Ideas?

You can't as it requires a known laboratory standard. It only costs about $60. I have to totally disagree with the other replies. I have a Panasonic capsule, and an uncalibrated E8000. Using them I built speakers that measured well and sounded like well processed hay. Now I have a mic I can depend on. I kind of wish I had gone to a more expensive lab and gotten a phase calibration, but the E8000 is pretty flat through the midrange where it matters, so I live with that.
 
You can't as it requires a known laboratory standard. It only costs about $60. I have to totally disagree with the other replies. I have a Panasonic capsule, and an uncalibrated E8000. Using them I built speakers that measured well and sounded like well processed hay. Now I have a mic I can depend on. I kind of wish I had gone to a more expensive lab and gotten a phase calibration, but the E8000 is pretty flat through the midrange where it matters, so I live with that.

Well I haven't seen anybody calibrating microphone in in whole Indian subcontinent ..so myself too living with it.

I still hope there could be some way to achieve it to almost satisfactory.

Somebody can design a reference grade pink noise generator for 100 dollar and list it in ebay or something like that.
😀
 
Status
Not open for further replies.