Comparison of ten room testing microphones
Ethan Winer
People often ask which affordable microphones I recommend for measuring
their room. To answer this question I tested ten popular microphones ranging
from an inexpensive DIY model to a Josephson with Microtech Gefell capsule
costing $1,800, and wrote up the results for this article:
http://www.realtraps.com/art_microphones.htm
Enjoy!
--Ethan
Too_Bad
news:HfqdncxgFo7DGBXV...@giganews.com:
> http://www.realtraps.com/art_microphones.htm
Too bad you didn't include one of the miniature omnidirectional back
electrets such as the Panasonic WM-063 or WM-60A, which have ruler-flat
frequency response from 30Hz to 20KHz. At one time DigiKey was selling them
for under $2.
Heinz Weissing
your results are quite in line with my own experience. To me this is no
great surprise: The basic design of a condensder microphone is so simple
that you just cannot do much wrong, so any microphone manufacterd with some
care should yield good results. The same applies to the circuitry needed for
processing the signal of the capsule.
In many cases the low-frequency roll-off of inexpensive microphones is
intentional to block spurious noise, for instance footfall noise. In
microphones with an early drop-off at high frequencies, diaphragm tension is
obviously low, resulting in a relatively low upper resonant frequency.
The deviations of your graphs from a straight line are caused not only by
room resonances - there will be some, without question. But even the best
loudspeakers available have some fluctuations in their frequemcy response,
so the sound pressure at the location of the microphone will be slightly
different at different frequencies, unless it is controlled in some way. But
in your case it does not matter, your results are really enlightening.
Just a suggestion: Why not use the measured response of the "best"
microphone as a reference and plot the difference between each microphone
and the
reference one, possible corrected for the difference at 1000 Hz? In this way
the differences between the individual types will stand out more clearly,
because the effects of room resonances and loudspeaker frequency response
cancel out.
Regards,
Heinz
"Ethan Winer" <ethanw at ethanwiner dot com> schrieb im Newsbeitrag
news:HfqdncxgFo7DGBXV...@giganews.com...
GregS
The Mitey-Mike does use the Panasonic. The orginal Panasonic capsule has been replaced.
Well, the response is not ruler flat. Some of the charts from Panasonic would
look ruler flat. The Mitey-Mike seems to be the only commercial system using it.
The mike DIY forum was busy a few years back getting the most out
of DIY systems. The Mitey-Mike uses selected capsules. ANybody looking for more info here..
http://www.pitt.edu/~szekeres/mic/mic.htm
greg
Ethan Winer
> Why not use the measured response of the "best" microphone as a reference
> and plot the difference between each microphone and the reference one
I sort of did that by repeating the Earthworks in the last two graphs. With
an informal test like this I'm not confident to proclaim which of the
"high-end" microphones is closest to flat. So I arbitrarily picked the
Earthworks as a reference.
Again, the main points are 1) the Radio Shack SPL meter is not very good
even though it's popular, and 2) the Nady and Behringer microphones seem to
be a good choice for about the same price as the Radio Shack meter.
--Ethan
Ethan Winer
> more info here..
> http://www.pitt.edu/~szekeres/mic/mic.htm
Excellent, thanks, I just added that link next to the Mitey-Mike.
--Ethan
Too_Bad
news:g6cgss$85q$1...@usenet01.srv.cis.pitt.edu:
I just measured a WM-60A against a Bruel&Kjaer 4138 1/8" reference mic
side by side, exposed simultaneously to the same sound field. The
distance between the center of the WM-60A and the center of the 4138 was
less than 1/4". The sound field was generated by a Panasonic leaf tweeter
EAS-10TH1000 which has a power bandwidth in excess of 100KHz. I have
recently run an electrostatic actuator response on the the B&K 4138 and
know that its pressure response is flat +/- 0.2dB from 100Hz to 50KHz.
The response of the WM-60A was measured relative to that of the B&K 4138
using a Bruel & Kjaer 2035 dual-channel FFT analyzer equipped with 100KHz
input modules. The two channels of the 2035 analyzer are matched to
better than +/- 0.2dB over the full 100KHz frequency range of the
analyzer. Relative to the the 4138, the WM-60A is down 0.5dB at 10KHz
and 2.0dB at 20KHz. The WM-60A remains at about -2dB between 20KHz and
30KHz, and at 32KHz begins a very steep rolloff. The response between
10KHz and 30KHz shows a slow, gradual rolloff that is devoid of any bumps
or dips. These are the numbers, and whether one chooses to refer to the
response of the WM-60A as "ruler flat" depends on one's reference and
perspective.
GregS
Without looking up the specs of current models, there was talk on the MIC DIY forum
explaning there seem to have been manufacturing problems within the last years,
with response varying between capsules.
greg
GregS
I take it the Mitey Mike is not around any more ??
At least i didn't find it.
greg
ttonon
would you please post the results? Thanks.
Tom
Too_Bad
dc4779...@79g2000hsk.googlegroups.com:
> Were you able to measure the responses in Audio range, and if so,
> would you please post the results? Thanks.
>
> Tom
Relative to 1KHz, the response at 10KHz is down 0.5dB. The response is
flat +/-0.2dB from 2.5KHz to 100Hz. In order to accurately determine the
response of the WM-60A below 100Hz, the response would need to be measured
in a sealed coupler using a reference pressure sensor and a sound source
that have response that extends to DC. While I have that capability, some
of the fixturing that is required is currently being used for a current
project and can not be disassembled.
Too_Bad
news:g6cs84$ak9$1...@usenet01.srv.cis.pitt.edu:
Perhaps it's time for you and others on the MIC DIY forum and elsewhere to
say goodbye to twenty year old technology and get up to date on current,
cutting-edge technology that is available today.
http://media.digikey.com/pdf/Data%20Sheets/Knowles%20Acoustics%
20PDFs/SPM0204UD5.pdf
While the data sheet does not show the response below 10KHz, it is flat +
0/-1dB to 300Hz and has a lower limiting frequency (-3dB) at about 30Hz.
Additionally, the unit can be easily modified to eliminate the broad bump
that is centered at 47KHz and provide a nearly flat (+0/-2dB) response out
to 100KHz.
Ethan Winer
Yes, apparently not. But it was still fun to include in my round-up.
GregS
Thanks for sharing. I would have never found it, like I did i the old days. I
don't flip pages looking at new products in the DigiKey catalog because its
too big. I Hate even putting it on the shelf. If DigiKey would carry half the products
in the catalog, I would also have a better attitute toward them.
Most of my speaker testing involved below 100 Hz, so this mic is very limited in usage.
greg
Too_Bad
news:g6kfnr$5f1$1...@usenet01.srv.cis.pitt.edu:
I didn't find it by flipping through the pages of the DigiKey catalog
looking for new products.
> If DigiKey would carry half the products in the catalog, I would also
> have a better attitute toward them.
I gave the DigiKey reference only because, to the best of my knowledge,
DigiKey is the only outfit that stocks Knowles mic capsules. While I
rarely buy components from DigiKey I have never had your experience.
Digikey has had in stock every item that I have ever ordered.
> Most of my speaker testing involved below 100 Hz, so this mic is very
> limited in usage.
> greg
Then I fail to understand your interest in this thread. It seems to me
that ir would be difficult, if not impossible, to find a mic that does
not have an acceptable response in the frequency range between 20Hz and
100Hz that would suit your purposes, whatever they might be.