software for measuring mic pres frequency response

Hi everybody.

Anywone knows if exists a software or plugin which can be used to measure the frequency response of a mic preamp?

sure there's sofware available... Rightmark audio analyzer and audiotester come to mind. But first you'll need to think about how you're going to interface you soundcard with the mic pre. Output is easy but you'll need to use a pad on the input and depending on the range of gain also a pad on the output.

greetings,

Thomas

One easy way is to take your balanced line out of your soundcard and put a 10k resistor in series with the (+) and a 10k in series with the (-) and then a 150ohm (or whatever impedence you want to load the pre with) across the 2 resistors before it goes into your pre.

I'm sure there are more scientific ways of doing it, but it seems to work ok.

thanks a lot, very helpful information, I'll give a try to those software. have to get hands on a pc with firewire to connect my interface, as i'm mac only at the moment.

The pad trick is exactly what i have to do. Is ok to do a similar thing at the output, as I don't have pad on line inputs of my interface? Anyway I think I will be able to push the gain up quite a bit before distorting the interface line input, or not?

To find the optimal level going from the software/soundcard into the mic pre, ie making like i'm sending a microphone signal, can I do any level measurement with a standard multimeter?

For exemple I am thinking about obtaining a 10mV rms level going into the mic pre, wich would be more or less a typical mic reference level. Any opinions on this?

About impedance I'll get specifications of the most used mics and put appropriate resistors.

thanks again this place is a school for me

Try Fuzz Measure on Mac.

Ben Lindell made a nice tut for it:  http://blog.benlindell.com/?p=112

Cheers,

Mike

+1 for RMAA, very capable software.

With a reasonably decent sound card I don't think you need a pad--just set output levels low enough to neither clip the preamp output nor soundcard input (measuring THD+N or SNR is another matter). Likely the soundcard has worse frequency response than the preamp, so make sure the software you're using has some sort of calibration mode where you short the soundcard input to output.

Samuel

beatnik said:

Anywone knows if exists a software or plugin which can be used to measure the frequency response of a mic preamp?

Click to expand...

Arta
http://www.fesb.hr/~mateljan/arta/

Regards,
Milan

I will give a try at fuzzmeasure first since it's mac compatible. benlindell tutorial is great explains very well the interface calibration part.

I'm still a bit insecure about the level to send into the pre. I am using a TC Konnekt 48 wich has some weird output level behavior: if I send a 0dBFS 1kHz tone out of pro tools , the output meters on the interface mixer display a -3db signal.
I am actually doing some research about this and contacting the TC support. Do you have any idea?

Anyway, what would be a good level to send to the pre, in dBFS or mV measuring with multimeter?

The 0 dBF corresponding to analog signal in the "real world" is all over the place. There isn't a set standard on what 0dBfs actually means outside of a computer. +/-3 db wouldn't be a problem.

I'm still a bit insecure about the level to send into the pre.

Click to expand...

The absolute level is irrelevant in a frequency response measurement, as long as we're well above the noise floor and below clipping (or slew-rate limiting).

Samuel

abechap024 said:

The 0 dBF corresponding to analog signal in the "real world" is all over the place. There isn't a set standard on what 0dBfs actually means outside of a computer. +/-3 db wouldn't be a problem.

Click to expand...

I don't know, it seems strange to me that the d/a converter has that -3dB attenuation. Or at least I would like to know why. I am doing some research on it, but of course any comment is appreciated.

Samuel Groner said:

The absolute level is irrelevant in a frequency response measurement, as long as we're well above the noise floor and below clipping (or slew-rate limiting).

Samuel

Click to expand...

Well, it's good to know the frequency response won't vary with the amplitude of the input signal.

About clipping and noise floor, how could I know that? In particular about the clipping, as my pres don't have a led indicator.

I imagine it varies on every kind of circuit, but is there any way I can calculate or measure it (with basic equipment)?

Sorry for my lack of knowledge and if my questions are too obvious for you, thanks a lot for the help

beatnik said:

About clipping and noise floor, how could I know that? In particular about the clipping, as my pres don't have a led indicator.

Click to expand...

For clipping:

Generate a sine wave in your DAW. 1kHz should be fine.

Drive your pre with this sine wave, record the output of the pre on one of your input channels.

Look at the recorded waveform. If you see distortion on the waveform, like clipped off tops/bottoms or something far uglier (but your converter input is below 0dBFS) your pre is clipping. If you don't see distortion yet, turn up your DAW output level some more.

Now slowly reduce your DAW output level until you can just see clipping.

Once you've found the clipping level, reduce your DAW output level by +/- 40 dB. If your pre isn't extremely noisy, this will be both sufficiently far below clipping and sufficiently high above the noise floor to do basic frequency response measurements.

JDB.
[and unless your pre is ... special, 40dB below clipping at 1kHz should still be below clipping at 20kHz]

I don't know, it seems strange to me that the d/a converter has that 3 dB attenuation.

Click to expand...

The converter has not 3 dB attenuation. There is just no nominal relation between dBFS (digital full scale level) and analogue level (measured in dBu, dBV, Vrms or whathever). If your AD-converter has a 0 dBFS level of +20 dBu (which means that an +20 dBu analogue signal produces 0 dBFS after conversion) and your DA-converter a 0 dBFS level of +17 dBu (meaning that a 0 dBFS digital signal gets reproduced at +17 dBu in the analogue world), you'll get a -3 dBFS signal if you "loop through" from the DA output to the AD input with a full scale signal. Absolutely nothing wrong with this.

Pay attention to the correct units; "dB" is just a ratio, not an absolute unit.

Samuel

thanks a lot for the explainations, very very useful.

I'll post my results when I'll do the preamp test, stay tuned

Another possible measuring software is REW (Room Eq Wizard)

It's written in Java, therefore it runs on several platforms including OSX and Linux, provided you have the Oracle/Sun Java environment.

http://www.hometheatershack.com/roomeq/

Axel

Another way to detect for clipping without looking and listening, is to increment the level up or down in 3dB steps. If the signal is going into clipping the output increase will be less than the + 3dB input increase. If moving out of clipping it will drop less than the nominal -3dB.

Note this will be more apparent with peak measurements than average but there will still be divergence between input and output tracking. Likewise, dropping a tone down into the noise floor will also fall less than the nominal signal reduction.

JR