Hi Bob, Raimund, Greg and Prataps--
For several reasons including the ones Bob notes, we know that mic to
speaker distance will make a difference in the critical step of 64 dB
amplitude calibration. I looked through three SLM manuals and
couldn't find anything about what distance to measure isolated, loud
sound sources. (In most applications, measurements are taken where
background levels are perceived). Doesn't 1 meter replicate usual mic
sensitivity test conditions? I looked for a while could not find a
succinct specification, but the distances were all on the order of 1
meter. At this short distance, if I place the sound source and mics
at 7 feet above thick grass, reflection impacts should be negligible-
especially considering my $50 Nady ASM-2 meter is +/- 2dB! I noticed
that some of the more expensive meters we've discussed are +/- 1.5dB.
It would also be interesting to play the same sound samples in one's
sound studio (same speaker) at 1 meter just to see what differences
crop up.
Its #4 on my current test list. I was thinking of using AT4022's and
my SD744T and record level "60.5 dB" at a distance of 1 meter if
anyone wants to get started. Rob D.
At 12:40 PM +0000 11/30/10, Avocet wrote:
>
>
>
>> Is there an opinion about whether the
>> speaker should be set-up at 1 meter?
>
>Rob,
>
>The mic to speaker placing could be critical. At 1 metre you would
>still have phase relationships with different parts of the speaker
>cone, box resonances, and reflex ports if any, but ground reflections
>would be minimised.
>
>As an overall check on resonances and reflections try recording a slow
>sweep tone across the whole range. My guess is that this will be very
>lumpy.
>
>The mics will also be picking up a ground reflection of the speaker
>depending on the ground surface. It may well be useful to baffle off
>the direct sound, reflecting it into the air, to see what sort of
>reflections they are picking up. Soft grass is a good absorptive
>surface for most frequencies.
>
>To minimise standing wave patterns mic testing is usually done with
>warble tones which is why frequency response curves are usually
>smoothed out. I've got a combination bandwidth limited 800Hz to 1200
>Hx white noise and warble at:
><http://www.stowford.org/sounds/bandpasswarble1000.mp3>http://www.stowford.org/sounds/bandpasswarble1000.mp3
>I made this with Audacity 1.3 which has an easy Equalization function.
>This will copy exactly end to end or you can use loop play. You can
>also shift the frequency t o save recreating bandwidtht limited
>white
>noise and/or warble tones each time.
>
>> I guess we'll see if the 1000 Hz tone produce a different RMS A
>> measurement and explore the significances. Rob D.
>
>I've also got a test tone which I originally devised for adjusting
>tape azimuth in the field, avoiding false peaks as you get with pure
>tone. It is very phase sensitive and is based on 440HZ (musical chord
>A major) with subharmonics and harmonics from 44Hz to LPF rolloff.
><http://www.stowford.org/sounds/brinitone440.wav>http://www.stowford.org/sounds/brinitone440.wav
><http://www.stowford.org/sounds/brinitonespectrum.gif>http://www.stowford.org/sounds/brinitonespectrum.gif
>This will also copy end to end or loop play.
>
>BTW if you want to see mp3 artifacts try compreessing brinitone.
>
>David
>
>David Brinicombe
>North Devon, UK
>Cogito cogito ergo cogito sum - Ambrose Bierce
>
>
--
|
