Quietest anechoic chamber?

[Don Pearce]

Watching a TV programme last night I found out that Salford University
(near Manchester) has an anechoic chamber with a background noise
level of -12dB.

OK - some questions...

1. Anyone know of a quieter one (they claimed it to be the quietest in
the world)?
2. How could they confirm this level? The quietest commercial
microphone I know of has a self-noise level of +5dB. Are there really
non-commercial mics so much quieter than this? If so, how are they
made?

Anyone?

d

[Adrian Tuddenham]

Post by Don Pearce
Watching a TV programme last night I found out that Salford University
(near Manchester) has an anechoic chamber with a background noise
level of -12dB.
OK - some questions...

[...]

Post by Don Pearce
2. How could they confirm this level?

Measure the attenuation of a loud external sound, then measure the
normal outside noise level?

[Don Pearce]

On Tue, 25 Aug 2009 12:19:13 +0100,

Post by Adrian Tuddenham

Post by Don Pearce
Watching a TV programme last night I found out that Salford University
(near Manchester) has an anechoic chamber with a background noise
level of -12dB.
OK - some questions...

[...]

Post by Don Pearce
2. How could they confirm this level?

Measure the attenuation of a loud external sound, then measure the
normal outside noise level?

Maybe. But there are structure-borne routes inside as well as
airborne. I would not be happy to accept such an extrapolation on
faith.

d

[Tony]

Post by Don Pearce

Post by Adrian Tuddenham
Measure the attenuation of a loud external sound, then measure the
normal outside noise level?

Maybe. But there are structure-borne routes inside as well as
airborne. I would not be happy to accept such an extrapolation on
faith.

I would place a small bet that they are quoting the level in a filtered
band, probably 1/3 octave.
There you are, a valid acoustical discussion! Reports of death are
exaggerated (though maybe only slightly).

[Don Pearce]

Post by Tony

Post by Don Pearce

Post by Adrian Tuddenham
Measure the attenuation of a loud external sound, then measure the
normal outside noise level?

Maybe. But there are structure-borne routes inside as well as
airborne. I would not be happy to accept such an extrapolation on
faith.

I would place a small bet that they are quoting the level in a filtered
band, probably 1/3 octave.
There you are, a valid acoustical discussion! Reports of death are
exaggerated (though maybe only slightly).

Nope, just found the spec. It is -12.4dBA, and has a lower cutoff at
100Hz (having dimensions 5.4 x 4.1 x 3.3 metres). That s pretty good,
I'd say.

d

[Adrian Tuddenham]

Post by Don Pearce
On Tue, 25 Aug 2009 12:19:13 +0100,

Post by Adrian Tuddenham

Post by Don Pearce
Watching a TV programme last night I found out that Salford University
(near Manchester) has an anechoic chamber with a background noise
level of -12dB.
OK - some questions...

[...]

Post by Don Pearce
2. How could they confirm this level?

Measure the attenuation of a loud external sound, then measure the
normal outside noise level?

Maybe. But there are structure-borne routes inside as well as
airborne. I would not be happy to accept such an extrapolation on
faith.

It might not share its structure with any noise-generating building and
I think most of the tram track in that area is embedded in resilient
mastic.

Perhaps they have measured with some sort of highly-sensitised resonant
microphone over a range of spot frequencies and then combined the
results, or perhas they have excluded L.F. noise from the specification.

[Don Pearce]

On Tue, 25 Aug 2009 12:51:54 +0100,

Post by Adrian Tuddenham

Post by Don Pearce
On Tue, 25 Aug 2009 12:19:13 +0100,

Post by Adrian Tuddenham

Post by Don Pearce
Watching a TV programme last night I found out that Salford University
(near Manchester) has an anechoic chamber with a background noise
level of -12dB.
OK - some questions...

[...]

Post by Don Pearce
2. How could they confirm this level?

Measure the attenuation of a loud external sound, then measure the
normal outside noise level?

Maybe. But there are structure-borne routes inside as well as
airborne. I would not be happy to accept such an extrapolation on
faith.

It might not share its structure with any noise-generating building and
I think most of the tram track in that area is embedded in resilient
mastic.
Perhaps they have measured with some sort of highly-sensitised resonant
microphone over a range of spot frequencies and then combined the
results, or perhas they have excluded L.F. noise from the specification.

Filtering and resonant systems don't help when you are measuring a
broadband noise signal - they only work for lifting discrete spikes
out of the noise. I suspect you were right with your original
suggestion, only I hope their method had a considerable amount of
sophistication behind it.

d

[The Ghost]

Post by Don Pearce
Watching a TV programme last night I found out that Salford University
(near Manchester) has an anechoic chamber with a background noise
level of -12dB.
OK - some questions...
1. Anyone know of a quieter one (they claimed it to be the quietest in
the world)?
2. How could they confirm this level? The quietest commercial
microphone I know of has a self-noise level of +5dB. Are there really
non-commercial mics so much quieter than this? If so, how are they
made?
Anyone?
d

I don't know how the people at Salford University did it, but one way is to
apply the amplified, nth-octave filtered outputs of two closely spaced
microphones to a spectral polarity coincidence correlator which measures
the normalized correlation coefficient of the two signals. The level of
ambient noise that exists in the chamber is given by the following
equation.
SPL = 0.5 (SPLa + SPLb) + 10 log ((cos (theta)))
Where SPLa and SPLb are the SPLs measured by each of the two microphones
and theta is the meter reading on the spectral poarity coincidence
correlator, which can very between 0 and 90 degrees.

[Don Pearce]

Post by The Ghost

Post by Don Pearce
Watching a TV programme last night I found out that Salford University
(near Manchester) has an anechoic chamber with a background noise
level of -12dB.
OK - some questions...
1. Anyone know of a quieter one (they claimed it to be the quietest in
the world)?
2. How could they confirm this level? The quietest commercial
microphone I know of has a self-noise level of +5dB. Are there really
non-commercial mics so much quieter than this? If so, how are they
made?
Anyone?
d

I don't know how the people at Salford University did it, but one way is to
apply the amplified, nth-octave filtered outputs of two closely spaced
microphones to a spectral polarity coincidence correlator which measures
the normalized correlation coefficient of the two signals. The level of
ambient noise that exists in the chamber is given by the following
equation.
SPL = 0.5 (SPLa + SPLb) + 10 log ((cos (theta)))
Where SPLa and SPLb are the SPLs measured by each of the two microphones
and theta is the meter reading on the spectral poarity coincidence
correlator, which can very between 0 and 90 degrees.

OK, I see that, but the minimum possible spacing and physical size of
the microphones sets a pretty low upper limit on the frequency. Unless
the spacing is less than about 1/20 of a wavelength, theta is not
going to be a meaningful figure for anything but bore sighted noise -
which it clearly isn't.

d

[The Ghost]

Post by Don Pearce

Post by The Ghost

Post by Don Pearce
Watching a TV programme last night I found out that Salford
University (near Manchester) has an anechoic chamber with a
background noise level of -12dB.
OK - some questions...
1. Anyone know of a quieter one (they claimed it to be the quietest
in the world)?
2. How could they confirm this level? The quietest commercial
microphone I know of has a self-noise level of +5dB. Are there
really non-commercial mics so much quieter than this? If so, how are
they made?
Anyone?
d

I don't know how the people at Salford University did it, but one way
is to apply the amplified, nth-octave filtered outputs of two closely
spaced microphones to a spectral polarity coincidence correlator which
measures the normalized correlation coefficient of the two signals.
The level of ambient noise that exists in the chamber is given by the
following equation.
SPL = 0.5 (SPLa + SPLb) + 10 log ((cos (theta)))
Where SPLa and SPLb are the SPLs measured by each of the two
microphones and theta is the meter reading on the spectral poarity
coincidence correlator, which can very between 0 and 90 degrees.

OK, I see that, but the minimum possible spacing and physical size of
the microphones sets a pretty low upper limit on the frequency. Unless
the spacing is less than about 1/20 of a wavelength, theta is not
going to be a meaningful figure for anything but bore sighted noise -
which it clearly isn't.
d

I don't have a clue what so-called "bore sighted" noise is, but I doubt
that your cruiticism is a major concern under most reasonable
circumstances. Measurements in the article that described this technicque
were made in octave bands over the frequency range from 125 Hz to 8kHz.
The article was published in JASA, so I'm sure that if there were any
serious concerns about the measurement methodology, they would have been
pointed out during peer review, and commented on in the article. Also, a
face-face microphone spacing of 2mm, which is quite doable, corresponds to
1/20th of a wavelength at 8kHz, which isn't exactly what most people would
generally consider to be a pretty low upper frequency limit.

[Kari Pesonen]

Post by Don Pearce
Watching a TV programme last night I found out that Salford University
(near Manchester) has an anechoic chamber with a background noise
level of -12dB.
OK - some questions...
1. Anyone know of a quieter one (they claimed it to be the quietest in
the world)?
2. How could they confirm this level? The quietest commercial
microphone I know of has a self-noise level of +5dB. Are there really
non-commercial mics so much quieter than this? If so, how are they
made?
Anyone?
d

The Finnish state owned research institute, VTT
http://www.vtt.fi/?lang=en
has an underground acoustical laboratory (floor level ca. 70 m bellow local
ground level)
with several reverberation chambers. When laboratory ventilation is switched
off
and a chamber is floating on air springs background noise level is of same
order.
My understanding is that to reach -10...-12 dB(A?) in a reverberation
chamber is more
difficult or demanding than to reach the same level in an anechoic chamber.

Kari Pesonen
PS. I assume that -12 dB refers to A-weighted level or, at least, LF-noise
is excluded
in this value.

[Don Pearce]

On Thu, 27 Aug 2009 14:53:39 +0300, "Kari Pesonen"

Post by Kari Pesonen

Post by Don Pearce
Watching a TV programme last night I found out that Salford University
(near Manchester) has an anechoic chamber with a background noise
level of -12dB.
OK - some questions...
1. Anyone know of a quieter one (they claimed it to be the quietest in
the world)?
2. How could they confirm this level? The quietest commercial
microphone I know of has a self-noise level of +5dB. Are there really
non-commercial mics so much quieter than this? If so, how are they
made?
Anyone?
d

The Finnish state owned research institute, VTT
http://www.vtt.fi/?lang=en
has an underground acoustical laboratory (floor level ca. 70 m bellow local
ground level)
with several reverberation chambers. When laboratory ventilation is switched
off
and a chamber is floating on air springs background noise level is of same
order.
My understanding is that to reach -10...-12 dB(A?) in a reverberation
chamber is more
difficult or demanding than to reach the same level in an anechoic chamber.
Kari Pesonen
PS. I assume that -12 dB refers to A-weighted level or, at least, LF-noise
is excluded
in this value.

Yes, it is A-weighted. I've never worked out how much being
reverberant adds to noise level - the total acoustic energy entering
the space is not changed, just how quickly it is removed by
absorption.

d