Re: General mike diaphragm noise question
- From: "Soundhaspriority" <nowhere@xxxxxxxxxxx>
- Date: Wed, 21 Jan 2009 17:04:22 -0500
<Don Pearce> wrote in message news:497b7c1e.473287578@xxxxxxxxxxxx
On 21 Jan 2009 13:39:05 -0500, kludge@xxxxxxxxx (Scott Dorsey) wrote:
In article
<3fff9bc9-01d6-42aa-a57e-7ca667d72e8d@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>,
mchampag@xxxxxxxxx <mchampag@xxxxxxxxx> wrote:
Hello, RAPpers. Years ago I heard an engineer make a blanket statement
that microphone diaphragms have a lower noise floor the larger they
get. If this is correct, can someone please explain? Thanks!
For the most part this is correct. Most of the noise in a modern
microphone
comes from the Brownian motion of air against the capsule. As you
increase
the size of the diaphragm, the microphone sensitivity increases faster
than
the noise does.
Brownian Motion in air at NTP (normal temperature and pressure)
equates to a sound pressure level of approximately -23dB (that is
unweighted - if one uses weighting, it is more like -30dBa), so no,
Brownian motion is not responsible for the noise level of any current
microphone.
That said, there are some tricks out there that can be used to radically
reduce the noise floor on a microphone. The Sennheiser MKH-20 has a very
small diaphragm, but also an extremely low noise floor.
Those tricks allow a filter to be placed in the amplifier,
significantly reducing the amplifier hiss at higher audio frequencies.
The penalty is a rather floppy and possibly unstable diaphragm,
although Sennheiser seem to have got it nicely under control.
Fifty years ago, the contribution of the electronics was substantial too,
and the larger diaphragm meant more capacitance change from the capsule
and therefore more signal going into the electronics whose noise floor was
fixed. These days the electronics of most well-designed microphones
produce
less noise than the capsule, so that's become a non-issue.
If this were the case, pretty much all microphones of a given capsule
size would have roughly the same noise figure. They don't - they vary
widely. This is down to the design of the electronics. Some
manufacturers, notably Rode at the moment, have given a high degree of
priority to this function. Neumann are currently close behind on the
same path.
So, although it's a terrible generalization, it's more or less correct.
Like all blanket statements there are plenty of exceptions, though.
--scott
Nope, as a generalization it is a complete miss.
d
Don, you are wrong, and you owe Scott an apology. Doubtless your reference
was this: http://www.santafevisions.com/csf/html/lectures/007_hearing_II.htm
, in which the following statement is made:
"At 4 kHz, which is about the frequency of the sensitivity peak, the
pressure amplitude variations caused by the Brownian motion of air
molecules, at room temperature and over a critical bandwidth, corresponds to
a sound pressure level of about -23 dB."
You missed two things:
1. The importance of the phrase "critical bandwidth." -23dB is the total
power integrated over one critical bandwidth, not the entire audio band. For
the critical bandwidth at 4 kHz, I refer you to
http://en.wikipedia.org/wiki/Critical_band . Read off Figure 2, the ERB, a
close approximation to the critical bandwidth, is about 400 Hz. Furthermore,
the figure of -23dB is relevant only if the actual diaphragm of theory or
measurement is specified. It is not, so we have to guess.
2. But the core of the matter is something else, which you can familiarize
yourself with here:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2600454
You are proceeding on the assumption that the energy transfer to the
diaphragm is an unavoidable number that depends only upon the velocity of
the molecule. This is false. Various techniques can be used to make the
collision almost elastic. The sensitivity of a diaphragm to a pressure wave,
versus a single molecular impact, is a design characteristic that can be
optimized to preferentially detect the pressure wave.
Thus, the answer is complex: The sensitivity of a microphone is determined
by both the design of the electronics, and the design of the diaphragm,
which, as the above reference reveals, has many variations besides sheer
size. DPA also disagrees with you. Their statement can be found here:
http://www.dpamicrophones.com/Images/DM00670.pdf . Their statements are not
contradicted by the fact that B&K makes a special measurement mike that,
with a 16mm diameter, has a 5.5 dBA noise figure.
The question of microphone noise is complex and multifaceted.
Bob Morein
(310) 237-6511
.
- References:
- General mike diaphragm noise question
- From: mchampag@xxxxxxxxx
- Re: General mike diaphragm noise question
- From: Scott Dorsey
- General mike diaphragm noise question
- Prev by Date: Re: General mike diaphragm noise question
- Next by Date: Re: Nearfield active monitor shootout
- Previous by thread: Re: General mike diaphragm noise question
- Next by thread: Re: General mike diaphragm noise question
- Index(es):
Relevant Pages
|
