Microphone impulse response?

K

Kristof_clg

New member
Hi everyone,

I am currently doing a project in mic impulse response, i have to program a software to emulate microphones (much like antares mic modeler), a standalone software.

I was wondering if you could shed some light in how to measure mic impulse response, and if posible, what processes would you do to get the relevant data out of the impulse response (mic characteristics).

It would be of great help.

Thanks! ;)
 
What do you mean by impulse response? Are you referring to the frequency response of the mic? Usually that's measured in an anechoic chamber, which are very complex rooms and hard to build.

Another means is to use a TEF analyzer. Also hard to come by, but you might be able to find some demo software somewhere. The good stuff is expensive.
 
PhilGood said:
What do you mean by impulse response? Are you referring to the frequency response of the mic? Usually that's measured in an anechoic chamber, which are very complex rooms and hard to build.

Another means is to use a TEF analyzer. Also hard to come by, but you might be able to find some demo software somewhere. The good stuff is expensive.
Click to expand...

I do have access to an anechoic chamber, and measuring hardware/software. What TEF analyzer do you mean?
I was refering to what method of measuring is the correct one to measure the impulse response of a microphone, or which method has anyone here used ?

Thnaks! ;)
 
Not sure I understand the question either; measurement of impulse response is simple enough, send the mic an impulse and compare input to output. Of course, you will need to understand and account for your test system's impulse response. If you have a sufficient measurement microphone you could use that as your baseline, since any microphone you want to model would likely have an inferior impulse response. You could then proceed to compare a microphone with the worst possible impulse response (a large diaphragm dynamic, I should think) to ascern whether your system has sufficient resolution to enable a useful measurement.

Perhaps the difficulty comes from the lack of a standard specification of impulse response, but any specification is a subset of the microphone's performance. If you want a successful model, you need a complete measurement.
 
The anechoic chamber will give you the best results since you can run white noise and get several snapshots.

The TEF system requires a full range speaker and setting the mic at a measured distance from the speaker. Since we know the speed of sound you set the measurement delay based on how many milliseconds it takes for the sound to reach the mic (it ignores reflections that way) and the speaker sweeps through the frequencies from 20hz to 20khz several times, then prints out a report of the mics graph. Do a Google search and you'll see what I'm talking about.
 
PhilGood said:
What do you mean by impulse response? Are you referring to the frequency response of the mic?
Click to expand...

An impulse response is a very fast transient like the firing of a starting pistol.

I seem to remember that, for measurements, the impulse is normally generated by a spark gap.

The result would look something like this:-

imp-window.gif


Though this is not really a good one as you can see the oscillations continue for a very long time - this is just a picture I searched at random.
 
I was thinking of getting the IR with a swept sine. But as a comparison:

- What method would you use?
- How would you extract the relevant data out of the IR

Thanks a lot ;)
 
Impulse response is a measurement of a room or other system with reverberation, as in its response over time the the short impulse. Microphones don't generally have ongoing reverberation in response to an impulse. They do have frequency and phase response.
 
Mics do have an impulse response; it correlates to the inertial mass of the diaphragm. This can be easily demonstrated with the 'speaker' microphones used for kick drums; they have a purposely terrible impulse response which relates to the Fs of the driver.
 
mshilarious said:
Mics do have an impulse response; it correlates to the inertial mass of the diaphragm.
Click to expand...

Of course, they do! While indeed, it correlates to the diaphragm mass, there are some other factors, such as damping, resonator systems, etc.

For starters, I'd suggest to look at this document:

http://www.aes.org/e-lib/browse.cfm?elib=14050

Also, there was an extended exchange on GS board on this very topic. There were some quite knowledgable folks participating in that discussion. Even though I don't usually reference other discussion boards, for educational sake I hope this is OK this time. There you can find some info on deriving the impulse response from the sweep (and vice vs) , info on some software, and some other goodies. Please pay attention (many even experienced folks miss this point), as I stated in that thread, for successful measurement the system should be in its linear range:

http://www.gearslutz.com/board/remo...aphram-fast-small-diaphrgam-condensers-7.html

Best, M
 
Marik said:
Of course, they do! While indeed, it correlates to the diaphragm mass, there are some other factors, such as damping, resonator systems, etc.

For starters, I'd suggest to look at this document:

http://www.aes.org/e-lib/browse.cfm?elib=14050

Also, there was an extended exchange on GS board on this very topic. There were some quite knowledgable folks participating in that discussion. Even though I don't usually reference other discussion boards, for educational sake I hope this is OK this time. There you can find some info on deriving the impulse response from the sweep (and vice vs) , info on some software, and some other goodies. Please pay attention (many even experienced folks miss this point), as I stated in that thread, for successful measurement the system should be in its linear range:

http://www.gearslutz.com/board/remo...aphram-fast-small-diaphrgam-condensers-7.html

Best, M
Click to expand...

Wow thanks!
I will read them extensively today ;)

Anything anyone would like to add will be of great help :cool:
 
You must log in or register to reply here.
Back
Top