AudioBanter.com - View Single Post

(Bruce J. Richman) wrote in
:

Rich Andrews wrote:

"Robert Morein" wrote in news:F-
:

"John Atkinson" wrote in message
om...
"Arny Krueger" wrote in message
...
John Atkinson wrote:
"Arny Krueger" wrote in message
...
John Atkinson wrote:
what sound pressure level was required for the 6Hz tone to
become
audible? The Fletcher-Munson or Robinson-Dadson curves imply
that
6Hz is not audible at normal levels, yet you are saying
different.

Please cite a reference that portrays minimum reliably
perceptible
acoustic levels for the 5-10 Hz range from Fletcher-Munson
Robinson-Dadson, or any other source.

The sensitivity curves in these data increasingly rise with
reducing
frequency to their lower limits. There is no reason to believe
the
curves have points of inflection below those limits.

Failure to properly answer a simple question noted. Substitution of
a
fact with questionable [relevance] isn't a proper answer.

I think you misunderstood my point, Mr. Kreuger. Whether you take the
earlier Fletcher-Munson sensitivity curves or the more recent
research
from Robinson and Dadson, simple extrapolation from these curves
indicates that the human ear continues to lose sensitivity with
reducing frequency. The implication is that by the time you reach
6Hz -- almost two octaves below the 20Hz usually stated as the
"limit"
of human hearing -- you would need extraordinary high spls to "hear"
the tone.

As you seem to be saying that you could hear the difference with and
without 6Hz content using headphones at moderate levels, all I am
doing is pointing out that it would seem unlikely you are actually
"hearing" the 6Hz content in light of the published data on the
subject.
As I conjectured, perhaps you are detecting the audioband action of
the
high-pass filter?

snip

If so, how can you report perceiving 6Hz tones using headphones?

I never said that I was perceiving 6 Hz tones.

With respect Mr. Krueger, you did appear to be saying just that. On
multiple occasions. You also criticized Scott Wheeler (S888wheel) for
saying the opposite.

My test reports related to perceiving their reduction, absence or
some by-product of common means of eliminating or reducing them.

Well yes, now you are saying that, which in turn allows for the fact
that you actually agree with me that it is perhaps the action of the
filter you are perceiving, not the presence or absence of 6Hz
content.
As I asked:

Was this a double-blind test involving a high-pass filter?

Yes.

Which is my point. That the high-pass filter is also a variable in
this
test and might well be the source of the identification you report.
Laurie Fincham, then with KEF, did some blind tests in the early
1980s
that such filters produced audible effects.

If so, then isn't it more likely that the test was detecting the
audio-band phase error of the filter, rather than the presence of
infrasonic (6Hz) information?

The filter in question is known to be a [reasonably] precise
[implementation] of the type of filter it is stated to be.

So what? My conjecture doesn't necessarily depend on the
implementation
of a high-pass filter, only that one be used. If you really wanted to
perform a test on the audibility of content in the 6Hz region, you
would
need to compensate for the filter's audio-band phase error, surely?

As I point out in a recent post, these filters are probably less
audibly
damaging than the actual filters that are built into most audio
systems,
particularly audio equipment based on tubes or audio equipment
created
for
the purpose of vinyl playback.

I agree. As I have written in Stereophile, the recording-playback
chain
consists of cascaded high-pass filters. Which is why it is very rare
for
recorded music to contain high levels of content below 30Hz, let
alone
20Hz.

John Atkinson
Editor, Stereophile

6 Hz at high SPL likely produces high frequency noise caused by air-
flow
boundary turbulence.
At this frequency, the joints of the inner ear bones, and gross
movement
of
cochlear fluid would produce similar noise, perceptible by the
cochlear
hairs as broadband noise.

There are no cochlear hairs tuned to 6 Hz. These hairs do have
harmonic
modes, as given by the the eigenvalues of the "string equation with
one
end
free", which are Hankel functions.

An interesting question would be whether an intense low frequency
signal
induces audible Doppler distortion on the eardrum. Personally, I
think
not.

It seems unlikely that a reproduction chain exists that could explore
these
effects. As Francois remarks, headphones suitable for investigational
purposes do not exist.

Personally, I would like to find a recording that has a significant
amount
of extremely low bass. Some Back organ works might do nicely if I can
find the right ones.

r

--
Nothing beats the bandwidth of a station wagon filled with DLT tapes.

While most of us could probably specify our favorite recordings (or at
least,
well-known recordings) that contain subjectively perceived low bass, who
can
actually state that they have measured it?

And if they have done so, whqt instrumentation and methodology was used
to
establish the presence of audible low bass at specified frequencies.
Given
the fact that very few subwoofers, and probably even fewer (if any)
full-range
speaker systems can reproduce extremely low bass at reasonable average
listening levels used in practice, this whole debate about "6 hz tones"
may be
of academic interest, but would appear to be of little relevance to the
average
music lover who listens at reasonable levels in their home setting.

I'd be more interested in knowing how, if at all, the average home
listener
might bo about measuring low (and/or high) frequency limits at
reasonable
listening levels in their home environment. I've done the "test CD
routine"
with test tones to get a very rough idea of what my system can do, but
of
course, that is just a rough approximation. Subjectively, with my
particular
system, all I've established is that not all tones are produced with
equal
amplitude, but I'm sure that is true of practically all non-equalized
systems.
Whether an equalizer would change this perception is an open question.

Bruce J. Richman

Bruce,

The cheapest way to obtain some sort of reasonable estimation of the
performance of a home speaker system is to buy a RS SPL meter, buy or make
a CD with suitable 1/3 octave warble tones, and then plot the results. Of
course things aren't quite as simple as they appear and this is no
exception. This method is suitably accurate IME.

r

--
Nothing beats the bandwidth of a station wagon filled with DLT tapes.