Frequently
Asked Questions
On January 10, 2002, an
international reader from Australia asked:
Hi Michael, thanks for an
informative website. My questions are in regards to
conducting a free field calibration:
Should we use dB(A) ie.
A-weighted or dB(lin) ie. unweighted when measuring and
calibrating the output at the different reference
frequencies?
Or should we just state the
corrections based on linear SPL measurements assuming
the people using the equipment will know how to apply
the appropriate corrections?
This will also affect the
choice of SLM used as an octave band filter will not be
needed when performing a calibration if A-weighted SPL's
are used (although I appreciate that octave band will
still be needed for background/ambient level
measurements).
I was also wondering if
there is as yet any standards in regards to free field
calibrations. I understand that in the US you call it
sound field calibration. Are there any European/ISO
standards in this regard? What are your recomendations
in terms of how often the free field calibration should
be conducted? The current Australian standard requires
booth backgrounds tested once every five years for
ears-covered audiometry, and as there is no standard for
free field, the general practice is also once every five
years, although audiometers and SLM's are generally
calibrated yearly.
I apologise for the long
winded nature of this email and thanks again for an
informative and useful website.
Regards,
-----------------------------------------------------------------------------------------------------------------
Michael Der's response,
I
apologize for taking so long for this reply. I've
been busy at work and have only just gotten around to
your question. When you mention the dB(lin) scale,
I am assuming you are referring to the dB(C)
weighted scale. Are you attempting to measure the
ambient room noise in a sound booth?
The
A-weighting scale is the most commonly used of the dB
scales. The reponse characteristic of a dB(a)
scale most closely matches that of human hearing. That
is, people do not have equal sensitivity to all
frequencies of sound across their dynamic range.
Humans are less sensitive to low and veyr high
frequencies, and therfore need to increase the SPL
at those frequency in order to detect the signal.
For this reason, most measurements are done in reference
to dB(a). Having said this however, it is not
clear why you would consider dB(c) scales unless it is
for reseach purposes that are not relevant to human
perception.
Regarding
your question for a standard for calibrationg sound
fields. There are no ANSI or ISO standards on how
to calibrate sound fields, so it is not surprising that
no Australian standard exists. For such a standard
to exist, sound booths would have to be standardized
essentially as a coupler (i.e a 2 cc coupler, 6 cc
coupler etc). An excellent paper which
describes how to calibrate your sound field is :
Walker
G., Dillon H., Byrne D. (1984) Sound field audiometry:
recommended stimuli and procedures. Ear & Hearing.
5(1):13-21.
however, this
is not a standard for sound field calibration.
I have written
several articles in my website in the sound fields
section, that provides details on how one can
calibrate a sound field (none of which are
standardized method, cuz one does not exist). Here
is an except from one such article:
Sound
field calibration using warble tones
Sound booths
used in sound field testing need to be calibrated so that
thresholds measurements correspond to those values
obtained using earphones.
Audiologist need to be aware that there are no
standard RETSPL
values used to calibrate the test booths but there are
recommended stimuli and procedures used in calibration.
Although these values do allow for good sound field
threshold measurements, Audiologists need to be aware that
they may also contribute to measurement variability.
The
problem of standing
waves in
sound field testing can be reduced by using FM warble
tones rather than pure tones.
Standardized RETSPL values are not currently
available although several studies have presented a
recommended procedure and set of stimuli used in
calibration and conduction of sound field tests.
Walker et al. (1984) presented what they believe to
be a good set of RETSPL for calibrating sound fields.
Despite published
reports of recommended stimuli (Walker et al 1984, Morgan
et al 1979), different manufacturers of audiometers
produce devices with slightly different FM warble tones.
In addition, sound booths and sound fields found in
clinical testing are rarely the same (acoustically and
architecturally) across clinics, which contributes to
threshold measurement variability (Staab et al 1972).
(My
website is down until Jan 16 for maintenance)
I
hope this helps.
Please write me back if further
clarification is needed.
Regards,
Michael Der
----------------------------------------------------------------------------------------------
Reader Responded,
Mike,
Thanks for your response.
I'm sorry I didn't make myself clear in my first
email. I am calibrating the output from a screening
audiometer in a free field situation.
So I am trying to calibrate
to 70dB across the frequency spectrum. I have a 3rd
octave band type 1 sound analyser and am familiar with
its use, weighting scales etc.
I have the meter setup in
3rd octave band mode. I have my colleague set the
audiometer to read 70dB at 1000Hz. This is our starting
point as no weighting correction is necessary. He then
adjusts the output from the audiometer until I read as
close as possible to 70dB at 1000Hz at the subjects ear
position.
He then changes the setting
on the audiometer to 500Hz, still reading 70dB on the
audiometer screen. My question is do I tell him to
adjust the level until my meter reads 70dB or 73dB,
which takes into account the -3.2dB correction in the
A-weighting scale at 500Hz? The same at 250Hz, should I
tell him to adjust the level until I read 70dB or 79dB
(-8.6dB at 250Hz). Taking into account that I am doing
the reading in 3rd octave band mode which are linear/unweighted
SPL's at the frequency bands. (I do appreciate that each
SLM has its own linear response.)
The reason I ask is that if
I am to tell him that adjustments are required taking
into account corrections for A-weighting then I don't
need a SLM with 3rd octave capabilities. I can just
use a type 1 integrating SLM and tell my colleague to
adjust the output at all his frequencies until my
(broadband) SLM reads 70dB(A).
Clear as mud?
Ultimately it shouldn't
matter because if the audiologist using the room has the
required knowledge base, he/she should be able to make
any corrections deemed necessary according to the manner
in which we present our results which would normally be
something along the lines of:
Frequency Hz
SPL received dB(lin) SPL adjusted
dB(lin)
125
73.4
70.1
250
68.7
69.9
500
71.8
70.2
When we present the
unweighted SPL at the test frequencies.
Or with A-weighted SPL's:
Freq
SPL recieved dB(lin) SPL adjusted
dB(lin) SPL adjusted dB(A)
125
89.4
86.0
69.9
250
79.9
78.8
70.2
.....
etc...
So basically I want to know
do you recommend the first unweighted SPL method or
second weighted SPL method.
Regards,
----------------------------------------------------------------------------------------------
Michael Der's response in progress....
Note: This site is
best viewed using IE 4.X or higher under 800x600
resolution
|
