Sign in with
Sign up | Sign in
Your question

Very low frequency recording.

Last response: in Home Audio
Share
Anonymous
September 25, 2004 7:44:49 PM

Archived from groups: rec.audio.pro (More info?)

After completing the live room of my new studio, which is a floating room
construction, I thought I'd do a few tests to see how much of the outside
noise still gets through. The studio is built beside a road that gets
about one car every few minutes. It's located in the middle of a city.

I could not hear anything in the room, but was sure I could sense
something like a car went past. So, I sealed all the doors, put up an
octava omni into my quietest pre, with the mic stand on a sheet of foam,
and turned the gain way way up, and hit record.

Surprisingly, this showed a lot of activity, all of it way below 30hz, and
much more often than nearby cars going past. Some 'events' were very
quiet, very low cycles, but almost a minute in duration. (Possibly the
train line around 800 metres away, or planes?). Others were quite short
and damped. All were inaudible, though I could see the speaker cones
moving if I played it really loud.

Playing the recording ten times as fast revealed a underwater soundscape
of whale calls and clicks, strange unearthly tones. It was a unexpected
piece of ambient music.

I'd like to get a better recording of this, as the mic/pre is rather
noisy and has a limited low frequency response. Anyone got any ideas?
Geophones seem rather expensive, so a DIY solution looks likely. A
suspended coil of wire with a magnet fixed to the ground? How to get as
little noise as possible?
Anonymous
September 25, 2004 7:44:50 PM

Archived from groups: rec.audio.pro (More info?)

Sorry I can be of no help here, but could you post an mp3 clip? Sounds
interesting.

Neil R

"philicorda" <philicorda@azriel.tydrwg.org> wrote in message
news:p an.1998.01.01.04.53.50.851477@azriel.tydrwg.org...
> After completing the live room of my new studio, which is a floating room
> construction, I thought I'd do a few tests to see how much of the outside
> noise still gets through. The studio is built beside a road that gets
> about one car every few minutes. It's located in the middle of a city.
>
> I could not hear anything in the room, but was sure I could sense
> something like a car went past. So, I sealed all the doors, put up an
> octava omni into my quietest pre, with the mic stand on a sheet of foam,
> and turned the gain way way up, and hit record.
>
> Surprisingly, this showed a lot of activity, all of it way below 30hz, and
> much more often than nearby cars going past. Some 'events' were very
> quiet, very low cycles, but almost a minute in duration. (Possibly the
> train line around 800 metres away, or planes?). Others were quite short
> and damped. All were inaudible, though I could see the speaker cones
> moving if I played it really loud.
>
> Playing the recording ten times as fast revealed a underwater soundscape
> of whale calls and clicks, strange unearthly tones. It was a unexpected
> piece of ambient music.
>
> I'd like to get a better recording of this, as the mic/pre is rather
> noisy and has a limited low frequency response. Anyone got any ideas?
> Geophones seem rather expensive, so a DIY solution looks likely. A
> suspended coil of wire with a magnet fixed to the ground? How to get as
> little noise as possible?
Anonymous
September 25, 2004 8:49:57 PM

Archived from groups: rec.audio.pro (More info?)

philicorda wrote:

> A
> suspended coil of wire with a magnet fixed to the ground? How to get as
> little noise as possible?

Something simple that I'd try is a using sub-woofer as a
mic. I'd think the sensitivity could be better than a
ribbon and still give you the self noise of a metal winding,
which is pretty low.

I'd love to hear what you come up with. If I were you I'd
try a time/frequency scaling processor on it rather than
limiting it to just a speedup.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein
Related resources
Anonymous
September 26, 2004 6:33:32 AM

Archived from groups: rec.audio.pro (More info?)

One thing that you could be hearing is the low frequency noise of the
electronics. Preamps have a one-over-f noise where a JFET input will start
its one-over-f noise at a higher frequency than a bipolar input. This is
where the noise increases as the frequency goes down. At 1kHZ, it can be
very small but it can get significant at very low frequencies. Also, some
opamps (possibly older types) have a "Popcorn" noise that can add to the
mix. If you take the file and run an FFT, you should be able to see the
shape and compare it to a data sheet of an opamp to see if it is similar.
The problem is that the ADC would need to sample slow enough to get good
resolution at the low frequencies which may be difficult with PC sound
cards.

I would like to get a copy of the test file also to look at and listen to.

Just some thoughts that I had...

John





"philicorda" <philicorda@azriel.tydrwg.org> wrote in message
news:p an.1998.01.01.04.53.50.851477@azriel.tydrwg.org...
> After completing the live room of my new studio, which is a floating room
> construction, I thought I'd do a few tests to see how much of the outside
> noise still gets through. The studio is built beside a road that gets
> about one car every few minutes. It's located in the middle of a city.
>
> I could not hear anything in the room, but was sure I could sense
> something like a car went past. So, I sealed all the doors, put up an
> octava omni into my quietest pre, with the mic stand on a sheet of foam,
> and turned the gain way way up, and hit record.
>
> Surprisingly, this showed a lot of activity, all of it way below 30hz, and
> much more often than nearby cars going past. Some 'events' were very
> quiet, very low cycles, but almost a minute in duration. (Possibly the
> train line around 800 metres away, or planes?). Others were quite short
> and damped. All were inaudible, though I could see the speaker cones
> moving if I played it really loud.
>
> Playing the recording ten times as fast revealed a underwater soundscape
> of whale calls and clicks, strange unearthly tones. It was a unexpected
> piece of ambient music.
>
> I'd like to get a better recording of this, as the mic/pre is rather
> noisy and has a limited low frequency response. Anyone got any ideas?
> Geophones seem rather expensive, so a DIY solution looks likely. A
> suspended coil of wire with a magnet fixed to the ground? How to get as
> little noise as possible?
Anonymous
September 26, 2004 11:48:20 AM

Archived from groups: rec.audio.pro (More info?)

On Sat, 25 Sep 2004 18:08:01 -0400, Neil Rutman wrote
(in article <NfKdnWP1JIarccjcRVn-gg@speakeasy.net>):

> Sorry I can be of no help here, but could you post an mp3 clip? Sounds
> interesting.
>
> Neil R
>
> "philicorda" <philicorda@azriel.tydrwg.org> wrote in message
> news:p an.1998.01.01.04.53.50.851477@azriel.tydrwg.org...
>> After completing the live room of my new studio, which is a floating room
>> construction, I thought I'd do a few tests to see how much of the outside
>> noise still gets through. The studio is built beside a road that gets
>> about one car every few minutes. It's located in the middle of a city.
>>
>> I could not hear anything in the room, but was sure I could sense
>> something like a car went past. So, I sealed all the doors, put up an
>> octava omni into my quietest pre, with the mic stand on a sheet of foam,
>> and turned the gain way way up, and hit record.
>>
>> Surprisingly, this showed a lot of activity, all of it way below 30hz, and
>> much more often than nearby cars going past. Some 'events' were very
>> quiet, very low cycles, but almost a minute in duration. (Possibly the
>> train line around 800 metres away, or planes?). Others were quite short
>> and damped. All were inaudible, though I could see the speaker cones
>> moving if I played it really loud.
>>
>> Playing the recording ten times as fast revealed a underwater soundscape
>> of whale calls and clicks, strange unearthly tones. It was a unexpected
>> piece of ambient music.
>>
>> I'd like to get a better recording of this, as the mic/pre is rather
>> noisy and has a limited low frequency response. Anyone got any ideas?
>> Geophones seem rather expensive, so a DIY solution looks likely. A
>> suspended coil of wire with a magnet fixed to the ground? How to get as
>> little noise as possible?
>
>

I'm not surprised. There's a lot of stuff going on in the basement that most
of us never hear (or wish we didn't).

Try a TLM 103 and your best preamp. The TLM 103 has a selfnoise of 7dba and
is also very sensitive. Not many other mics are that quiet and that
sensitive.

I was using one on a gent a few months back and heard this clicking. I asked
if he were wearing a watch. Turns out it was his pacemaker!!

Regards,

Ty Ford



-- Ty Ford's equipment reviews, audio samples, rates and other audiocentric
stuff are at www.tyford.com
Anonymous
September 26, 2004 8:51:44 PM

Archived from groups: rec.audio.pro (More info?)

"philicorda" <philicorda@azriel.tydrwg.org> wrote in message
news:p an.1998.01.01.04.53.50.851477@azriel.tydrwg.org...

> Surprisingly, this showed a lot of activity, all of it way below
30hz, and
> much more often than nearby cars going past. Some 'events' were very
> quiet, very low cycles, but almost a minute in duration. (Possibly
the
> train line around 800 metres away, or planes?). Others were quite
short
> and damped. All were inaudible, though I could see the speaker cones
> moving if I played it really loud.
>
> Playing the recording ten times as fast revealed a underwater
soundscape
> of whale calls and clicks, strange unearthly tones. It was a
unexpected
> piece of ambient music.
>
> I'd like to get a better recording of this, as the mic/pre is rather
> noisy and has a limited low frequency response. Anyone got any
ideas?
> Geophones seem rather expensive, so a DIY solution looks likely. A
> suspended coil of wire with a magnet fixed to the ground? How to get
as
> little noise as possible?

My first thought would be a very large woofer used as a microphone.
Feed the output to a high quality output transformer with a
guaranteed -3db frequency of 5Hz. Feed the high impedance primary to
a DC coupled amplifier. In other words, everything is backwards.

You'll probably have to add lots of mass to the speaker cone to
flatten the response at low frequencies.

The best choice would be an acoustic suspension speaker in its
cabinet.

Norm Strong
Anonymous
September 26, 2004 10:58:30 PM

Archived from groups: rec.audio.pro (More info?)

Try a Sennheiser MKH-20, or even one of the MKH-104 mikes. Plenty of
subaudible response without huge amounts of 1/f noise like conventional
condenser mikes.

And, just close your eyes, take a deep breath, rest your fingertips lightly
on the desk in front of you, and feel the vibration in the room. That's why
I can't stand to be in New York for more than a couple days at a time. It
just drives me crazy.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
September 27, 2004 12:24:56 AM

Archived from groups: rec.audio.pro (More info?)

On Sun, 26 Sep 2004 16:51:44 +0000, normanstrong wrote:


> My first thought would be a very large woofer used as a microphone.
> Feed the output to a high quality output transformer with a
> guaranteed -3db frequency of 5Hz. Feed the high impedance primary to
> a DC coupled amplifier. In other words, everything is backwards.
>
> You'll probably have to add lots of mass to the speaker cone to
> flatten the response at low frequencies.
>
> The best choice would be an acoustic suspension speaker in its
> cabinet.
>
> Norm Strong

Thanks everyone for the replies.

Using a big speaker for the mic is a great idea. The soundcard is a
delta1010, but seems happy recording virtually DC.

Norm, why do you suggest using a transformer? Is it not possible to build
an amplifier that will be happy with the low impedence of a speaker output?

I want to keep the result as free from digital artifacts as possible, so
speeding it up works better than pitch/time shifting.

Another thought that came to mind is using two octavas as a stereo pair a
few feet apart. One it's recorded, invert the phase of one, add it to the
other to get the difference, and then subtract that difference from the
first. In theory, if the frequency is low enough, the waveform should
be near enough identical at the positions of both mics, and I can cancel
out the noise?

Other thoughts is that the building is just acting as a low pass filter,
and there is no reason why I can't record outside, and then low pass the
result electronicly?

Here are some excerpts of the recording. I recorded an hour, and then
sped it up to 4min by abusing a sample rate convertor. All files are
around 3-5mb in size.

Section of original untouched recording:
69.93.9.106/subcity/city_clean_original.mp3

Section of original recording sped up, 1hr down to 4min.
69.93.9.106/subcity/city_fast.mp3

Section of prettified version. Noise reduction+Eq+reverb. This sounds the
nicest. There are less chirpy noise reduction artifacts than it sounds,
but I would rather not use NR at all...

69.93.9.106/subcity/city_sub_pretty.mp3
Anonymous
September 27, 2004 12:24:57 AM

Archived from groups: rec.audio.pro (More info?)

philicorda wrote:


> Norm, why do you suggest using a transformer? Is it not possible to build
> an amplifier that will be happy with the low impedence of a speaker output?

It would be more a level matching consideration should the
signal output prove to be too low and put you too deep into
the noise of the pre-amp.

> I want to keep the result as free from digital artifacts as possible, so
> speeding it up works better than pitch/time shifting.

There are some really fine scaling products now that are
nearly artifact free, especially at low frequencies. If you
put up some of your raw recordins I'm sure some of us would
want to play with that. I've got Wavelab which has a
remarkably good time/pitch scaler.

> Another thought that came to mind is using two octavas as a stereo pair a
> few feet apart. One it's recorded, invert the phase of one, add it to the
> other to get the difference, and then subtract that difference from the
> first. In theory, if the frequency is low enough, the waveform should
> be near enough identical at the positions of both mics, and I can cancel
> out the noise?

The random noise of two mics can only be brought down by 3
dB regardless of how you combine them.

> Other thoughts is that the building is just acting as a low pass filter,
> and there is no reason why I can't record outside, and then low pass the
> result electronicly?

Absolutely! Or with something like Adobe Audition which
gives you any of the common filter types at any order.

> Here are some excerpts of the recording. I recorded an hour, and then
> sped it up to 4min by abusing a sample rate convertor. All files are
> around 3-5mb in size.

Ah, much of what you find interesting is intrinsic to the
speedup. I'd still find it interesting to play with time
and pitch scaling independantly.

> Section of original untouched recording:
> 69.93.9.106/subcity/city_clean_original.mp3

That's something to play with!

> Section of original recording sped up, 1hr down to 4min.
> 69.93.9.106/subcity/city_fast.mp3
>
> Section of prettified version. Noise reduction+Eq+reverb. This sounds the
> nicest. There are less chirpy noise reduction artifacts than it sounds,
> but I would rather not use NR at all...
>
> 69.93.9.106/subcity/city_sub_pretty.mp3

Thanks, just what I need, something else to play with. :-)


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein
Anonymous
September 27, 2004 12:44:12 AM

Archived from groups: rec.audio.pro (More info?)

normanstrong wrote:

> My first thought would be a very large woofer used as a microphone.
> Feed the output to a high quality output transformer with a
> guaranteed -3db frequency of 5Hz. Feed the high impedance primary to
> a DC coupled amplifier. In other words, everything is backwards.

I think this is a nontrivial task, and the approach suggested is many
orders of difficulty and expense beyond just using a good measurement
mic and good low noise wide bandwidth preamp. Speakers are used as mics
for effect, not transparent precision.

--
ha
Anonymous
September 27, 2004 1:09:24 AM

Archived from groups: rec.audio.pro (More info?)

"philicorda" wrote>
> Here are some excerpts of the recording. I recorded an hour, and then
> sped it up to 4min by abusing a sample rate convertor. All files are
> around 3-5mb in size.
>
> Section of original untouched recording:
> 69.93.9.106/subcity/city_clean_original.mp3
>
> Section of original recording sped up, 1hr down to 4min.
> 69.93.9.106/subcity/city_fast.mp3
>
> Section of prettified version. Noise reduction+Eq+reverb. This sounds the
> nicest. There are less chirpy noise reduction artifacts than it sounds,
> but I would rather not use NR at all...
>
> 69.93.9.106/subcity/city_sub_pretty.mp3

These are great.

While listening, I kept hearing Ali G asking "why does you fink dere's
whales inside de sewers?"

-jw
Anonymous
September 27, 2004 2:24:40 AM

Archived from groups: rec.audio.pro (More info?)

On Sun, 26 Sep 2004 20:44:12 +0000, hank alrich wrote:

> normanstrong wrote:
>
>> My first thought would be a very large woofer used as a microphone.
>> Feed the output to a high quality output transformer with a
>> guaranteed -3db frequency of 5Hz. Feed the high impedance primary to
>> a DC coupled amplifier. In other words, everything is backwards.
>
> I think this is a nontrivial task, and the approach suggested is many
> orders of difficulty and expense beyond just using a good measurement
> mic and good low noise wide bandwidth preamp. Speakers are used as mics
> for effect, not transparent precision.

At the moment I'm using an octava mc012 and a focusrite isa220. What would
you recommend trying? I'll get them from a hire company for a day or two.
Anonymous
September 27, 2004 2:53:24 AM

Archived from groups: rec.audio.pro (More info?)

philicorda wrote:

> What would you recommend trying?

What Ty said: TLM103 and a quiet preamp. For this I would choose my
Millennia pre over my Great River, to get the input transformer out of
the picture. You might rent a Millennia or Grace.

It's surprising how much subsonic stuff is out and about, and it
consumes a lot of dynamic range at capture. If you can find its
source(s) and exclude it you will get better sound than by capturing it
and then filtering it out. Consider that if it's in the air, to a
certain extent it modulates what you're wanting to record, even before
it gets to the mic(s).

--
ha
Anonymous
September 27, 2004 9:07:15 AM

Archived from groups: rec.audio.pro (More info?)

On Sun, 26 Sep 2004 22:53:24 GMT, walkinay@thegrid.net (hank alrich)
wrote:

>philicorda wrote:
>
>> What would you recommend trying?
>
>What Ty said: TLM103 and a quiet preamp. For this I would choose my
>Millennia pre over my Great River, to get the input transformer out of
>the picture. You might rent a Millennia or Grace.
>
>It's surprising how much subsonic stuff is out and about, and it

I recall a discussion of this on a seismic list. There's subsonic
(below 20 Hz, often way below) sounds from a variety of sources -
earthquakes, explosions, spacecraft re-entry, even meteors/shooting
stars.
For more on this (including transducers specifically designed for
the infrasonic frequency range), go here and type infrasound in the
searchbox:
http://psn.quake.net/searchmail.html

>consumes a lot of dynamic range at capture. If you can find its
>source(s) and exclude it you will get better sound than by capturing it
>and then filtering it out. Consider that if it's in the air, to a
>certain extent it modulates what you're wanting to record, even before
>it gets to the mic(s).

I suspect much of it is really hard to get rid of, thought I doubt
anything in the lower range, below 2 to 5 Hz or so would have any
noticable effect on a recording (unlless it's at an unusually high
amplitude, as in an earthquake). You'd have to hermetically seal the
room against changing air pressure, and that still wouldn't stop
mechanical vibrations transmitted through the building structure and
ground, generated by car/truck traffic and such.

-----
http://mindspring.com/~benbradley
Anonymous
September 27, 2004 11:17:34 AM

Archived from groups: rec.audio.pro (More info?)

On Sun, 26 Sep 2004 18:53:24 -0400, hank alrich wrote
(in article <1gkq9c9.n5x6gzsso1meN%walkinay@thegrid.net>):

> philicorda wrote:
>
>> What would you recommend trying?
>
> What Ty said: TLM103 and a quiet preamp. For this I would choose my
> Millennia pre over my Great River, to get the input transformer out of
> the picture. You might rent a Millennia or Grace.
>
> It's surprising how much subsonic stuff is out and about, and it
> consumes a lot of dynamic range at capture. If you can find its
> source(s) and exclude it you will get better sound than by capturing it
> and then filtering it out. Consider that if it's in the air, to a
> certain extent it modulates what you're wanting to record, even before
> it gets to the mic(s).
>
> --
> ha

To redouble the thought. We were listening for self noise through the studio
monitors and noticed that the bass drivers in the big Urie's were visibly
moving in and out; about 1/2 inch -- sort of fluttering and then stopping,
then moving some more.

We couldn't pin down the source, but figured the frequency was pretty damn
low if we were able to see the excursion. That would make it on the order of
1 or 2 cycles per second, right?

Regards,

Ty Ford



-- Ty Ford's equipment reviews, audio samples, rates and other audiocentric
stuff are at www.tyford.com
Anonymous
September 27, 2004 11:28:15 AM

Archived from groups: rec.audio.pro (More info?)

On Mon, 27 Sep 2004 01:07:15 -0400, Ben Bradley wrote
(in article <9d6fl0hsrhgi2avpb19nmapodq6js041ed@4ax.com>):


> I suspect much of it is really hard to get rid of, thought I doubt
> anything in the lower range, below 2 to 5 Hz or so would have any
> noticable effect on a recording (unlless it's at an unusually high
> amplitude, as in an earthquake). You'd have to hermetically seal the
> room against changing air pressure, and that still wouldn't stop
> mechanical vibrations transmitted through the building structure and
> ground, generated by car/truck traffic and such.
>
> -----
> http://mindspring.com/~benbradley

Ben,

I think most circuitry simply isn't designed to have that sort of low
frequency response. First, what would be the use? Second, what consumer gear
would reproduce it?

VLF is certainly out there....or down there. I've heard it flutter audio gate
circuits, making them swing open and closed.

From a sound design perspective, it's a whole nother area of spectrum. It's
almost like the difference between AM and FM frequency response in the
average radio receiver. We didn't worry a lot about anything over 10 kHz
until FM gave us 15 kHz to play with. Then we had to figure out what to do
with it.

With more and more sub woofers making their way into the market, we need to
be more concerned about "down there."

I had a friend send me his band's latest CD for comments. I told him the
project had been mixed on small speakers by the bass player. He was amazed by
my clairvoyance.

I told him it was just that when I played the CD on my big monitors (JBL L100
Centuries) the VLF and LF started to move my couch around.

Regards,

Ty Ford



-- Ty Ford's equipment reviews, audio samples, rates and other audiocentric
stuff are at www.tyford.com
Anonymous
September 27, 2004 11:30:53 AM

Archived from groups: rec.audio.pro (More info?)

kludge@panix.com (Scott Dorsey) wrote in message news:<cj7him$4ma$1@panix2.panix.com>...
> Try a Sennheiser MKH-20, or even one of the MKH-104 mikes. Plenty of
> subaudible response without huge amounts of 1/f noise like conventional
> condenser mikes.
>
> And, just close your eyes, take a deep breath, rest your fingertips lightly
> on the desk in front of you, and feel the vibration in the room. That's why
> I can't stand to be in New York for more than a couple days at a time. It
> just drives me crazy.
> --scott

IMO, Scott is on the right track here (or at least 1/2 way so...) by
suggesting the MKH20, which is an omni mic. The TLM 103 is good for
its low noise, but directional mics do not offer an accurate picture
of low frequencies because the LF response is dependant on distance
from the source. For extremely low frequencies, you may get
exaggerated lows, for instance, due to the physical length of the
waves being enormous (and thus proximity effect boosting the output at
those frequencies).

Several years ago, I worked on a project where it was necessary to
measure acoustic output down to 1Hz. I could only find one microphone
that would do the job and it was the Microtec Geffel measurement mic
model MK222 (for the capsule):

http://www.microtechgefell.de/eng/prod/mess/kapsel/mk22...

However, this may be lower than you need. Several of their models go
down to 3.5Hz and are probably more readily available than the 222.

Or, for more "off the shelf" mics that would reach at least down to 20
and most likely below:

Neumann KM131
Earthworks QTC1
DPA 4006

I'm sure there are others.

Measuring the problem is one thing, but solving it will require tuning
the suspension of the room to dampen specific frequencies. The
Tracking Room in Nashville was built this way, and is right next to a
highway. I think it was tuned to 5Hz, if memory serves correct, and it
is DEAD quiet inside. You may want to get ahold of Frank Wells (editor
of Pro Sound News) who used to be the chief engineer at the Tracking
Room, since he would probably know how it was done.

Regards,

Karl Winkler
Lectrosonics, Inc.
http://www.lectrosonics.com
Anonymous
September 27, 2004 2:35:50 PM

Archived from groups: rec.audio.pro (More info?)

Karl Winkler <karlwinkler66@yahoo.com> wrote:
>IMO, Scott is on the right track here (or at least 1/2 way so...) by
>suggesting the MKH20, which is an omni mic. The TLM 103 is good for
>its low noise, but directional mics do not offer an accurate picture
>of low frequencies because the LF response is dependant on distance
>from the source. For extremely low frequencies, you may get
>exaggerated lows, for instance, due to the physical length of the
>waves being enormous (and thus proximity effect boosting the output at
>those frequencies).

The other thing about the Sennheiser is that the RF discriminator circuit
has less noise at low frequencies. You aren't fighting the dominant low
frequency noise on the FET. The TLM103 is very quiet, but most of the noise
that it has is below 20 Hz.

>Several years ago, I worked on a project where it was necessary to
>measure acoustic output down to 1Hz. I could only find one microphone
>that would do the job and it was the Microtec Geffel measurement mic
>model MK222 (for the capsule):
>
>http://www.microtechgefell.de/eng/prod/mess/kapsel/mk22...
>
>However, this may be lower than you need. Several of their models go
>down to 3.5Hz and are probably more readily available than the 222.

The problem is that the noise floor of the follower electronics will kill
you. B&K, though, also makes an RF mike that is intended for sub-Hz
measurements. I don't know the number offhand, but it's in the B&K catalogue
and it's fearfully expensive.

>Measuring the problem is one thing, but solving it will require tuning
>the suspension of the room to dampen specific frequencies. The
>Tracking Room in Nashville was built this way, and is right next to a
>highway. I think it was tuned to 5Hz, if memory serves correct, and it
>is DEAD quiet inside. You may want to get ahold of Frank Wells (editor
>of Pro Sound News) who used to be the chief engineer at the Tracking
>Room, since he would probably know how it was done.

I think this is the sort of thing that you can just feel and hear directly
without needing any fancy instrumentation, but it can take some work to
get to the point of relaxation where you can feel it around you.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
September 27, 2004 8:16:52 PM

Archived from groups: rec.audio.pro (More info?)

Ty Ford wrote:

> I had a friend send me his band's latest CD for comments. I told him the
> project had been mixed on small speakers by the bass player. He was amazed by
> my clairvoyance.

> I told him it was just that when I played the CD on my big monitors (JBL L100
> Centuries) the VLF and LF started to move my couch around.

That's called "a Mr. Potatohead mix".

--
ha
Anonymous
September 27, 2004 11:37:49 PM

Archived from groups: rec.audio.pro (More info?)

--------------8<---------------------------------
>
>The problem is that the noise floor of the follower electronics will kill
>you. B&K, though, also makes an RF mike that is intended for sub-Hz
>measurements. I don't know the number offhand, but it's in the B&K catalogue
>and it's fearfully expensive.
>
-----------8<------------------------------------

>--scott

-- This is most probably the BK microphone type 4147:

"Very Low Frequency Type" -- 4147, 1/2" diameter. Pressure type
designed with special attention to the equalization arrangement to
bring the lower limiting frequency below 0,01 Hz. It is used for ultra
low frequency acoustic pulses eg. sonic boom measurements in
connjuction with Adaptor UA 0271 and Microphone Carrier System 2361."

-- This is a sealed capsule. Normally, BK microphones has a capillary
vent tube in the housing and by inserting tiny silver wires into that
vent for tuning, it is possible to achieve, at their "normal" types,
a -3 dB point at 2 Hz.

In an old Sennheiser mic catalogue ("Sennheiser Revue" Nr. 11, abt.
mid-eighties or earlier), I've found an long-discontinued condensed
microphone types MKH 110 (down to 1 Hz) and MKH 10-1 (down to 0,1 Hz).
Both go up to 20 kHz; quite flat to 9-10 kHz and then upwards.
They required their own power supplies (asymmetric) too as the
principle of operation is interesting: "The radio frequency principle
permits the frequency range to be extended down to almost 0 Hz.
<.....> To match the applications, the MKH110-1 has been reduced in
sensitivity by 20 dB by comparison with the MKH-110. In conjunction
with this, however, the modulation limit has been increased from 20 Pa
to 500 Pa".

Edi Zubovic, Crikvenica, Croatia
Anonymous
September 27, 2004 11:37:50 PM

Archived from groups: rec.audio.pro (More info?)

Edi Zubovic <edi.zubovic.[rem this]@ri.htnet.hr> wrote:
>
>In an old Sennheiser mic catalogue ("Sennheiser Revue" Nr. 11, abt.
>mid-eighties or earlier), I've found an long-discontinued condensed
>microphone types MKH 110 (down to 1 Hz) and MKH 10-1 (down to 0,1 Hz).
>Both go up to 20 kHz; quite flat to 9-10 kHz and then upwards.

I think that the MKH 110 was basically an MKH 105 mike with the time
constants in the discriminator changed. I hadn't known about the MKH 10-1
before, though!

>They required their own power supplies (asymmetric) too as the
>principle of operation is interesting: "The radio frequency principle
>permits the frequency range to be extended down to almost 0 Hz.
><.....> To match the applications, the MKH110-1 has been reduced in
>sensitivity by 20 dB by comparison with the MKH-110. In conjunction
>with this, however, the modulation limit has been increased from 20 Pa
>to 500 Pa".

Yes, the MKH-20 works the same way. This results in very, very low noise
floor, especially at lower frequencies. But it also results in some frequency
response aberrations, which are equalized in the electronics. That causes
some oddness in the lower midrange which is the only real disadvantage of
these mikes.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
September 27, 2004 11:37:50 PM

Archived from groups: rec.audio.pro (More info?)

On Mon, 27 Sep 2004 19:37:49 +0200, Edi Zubovic <> wrote:
> --------------8<---------------------------------
>>
>>The problem is that the noise floor of the follower electronics will kill
>>you. B&K, though, also makes an RF mike that is intended for sub-Hz
>>measurements. I don't know the number offhand, but it's in the B&K catalogue
>>and it's fearfully expensive.
>>
> -----------8<------------------------------------
>
>>--scott
>
> -- This is most probably the BK microphone type 4147:
>
> "Very Low Frequency Type" -- 4147, 1/2" diameter. Pressure type
> designed with special attention to the equalization arrangement to
> bring the lower limiting frequency below 0,01 Hz. It is used for ultra
> low frequency acoustic pulses eg. sonic boom measurements in
> connjuction with Adaptor UA 0271 and Microphone Carrier System 2361."
>
> -- This is a sealed capsule. Normally, BK microphones has a capillary
> vent tube in the housing and by inserting tiny silver wires into that
> vent for tuning, it is possible to achieve, at their "normal" types,
> a -3 dB point at 2 Hz.
>

If that doesn't work, try a seismophone :) 
Anonymous
September 28, 2004 2:43:47 AM

Archived from groups: rec.audio.pro (More info?)

On 27 Sep 2004 13:53:41 -0400, kludge@panix.com (Scott Dorsey) wrote:

>Edi Zubovic <edi.zubovic.[rem this]@ri.htnet.hr> wrote:
>>
>>In an old Sennheiser mic catalogue ("Sennheiser Revue" Nr. 11, abt.
>>mid-eighties or earlier), I've found an long-discontinued condensed
>>microphone types MKH 110 (down to 1 Hz) and MKH 10-1 (down to 0,1 Hz).
>>Both go up to 20 kHz; quite flat to 9-10 kHz and then upwards.
>
>I think that the MKH 110 was basically an MKH 105 mike with the time
>constants in the discriminator changed. I hadn't known about the MKH 10-1
>before, though!
---------------8<-----------------------

--- It seems you're right, Scott... In the same catalogue, there's the
MKH 106 omni with its various power supply versions. It looks the
same to the MKH 110/110-1 at the first glimpse. It's also flat to abt.
5 kHz and then it forms that "Sennheiser bulge" as I call it to 10 kHz
and then it falls back again. A very fine studio microphone but asks
for a clean and quiet place without possible feedback issues.

If you wish to, I can scan and send you the pages about this and MKH
110/110-1 types.

Edi Zubovic, Crikvenica, Croatia
Anonymous
September 28, 2004 3:00:18 AM

Archived from groups: rec.audio.pro (More info?)

On Mon, 27 Sep 2004 18:51:57 GMT, "U-CDK_CHARLES\\Charles" <"Charles
Krug"@cdksystems.com> wrote:

------8<--------------
>>
>
>If that doesn't work, try a seismophone :) 

--Actually, I'm curious on how they do it. I had a catalog of a German
"Prakla-Seismos" company (I think it doesn't exist under this name any
more) and I know that making big booms are made with either explosives
or special high pressure compressors, like those German made Junkers
250 bar or more types, and they record the responses. I don't know
what devices they're using for picking up the kaboom but I guess they
are based on accelerometers or so.

Btw in this area there was a quite loud earthquake some 20 days ago,
abt 4,5 Mercallis (I think) and the epicenter has been abt 6 Richters
strong, some 60 km from my place in a local highland area. The
epicenter depth has been abt. 9 kilometers. No significant damages in
the area, but folks in the nearby towns and in the city of Rijeka have
been pretty scared. It looked to me as if there has been a mighty
explosion of gas or explosives or alike; followed by a deeper boom
with a short shake of windows. Not quite pleasant I may say.

Edi Zubovic, Crikvenica, Croatia
Anonymous
September 28, 2004 3:16:05 AM

Archived from groups: rec.audio.pro (More info?)

Ty Ford <tyreeford@comcast.net> wrote:

> With more and more sub woofers making their way into the market, we need to
> be more concerned about "down there."

Don't get me started!

Living examples of this can be heard in TV sound from programs that
don't put much emphasis on quality audio, like the news shows.
Particularly audio recorded outside with nothing more than the foam
windscreen on the mic to cut wind, handling noise and p-pops.

I can picture TV reporters editing their own story on computers in a
big, noisy news room, making audio mixing thru whatever computer
speakers that came with the Dell...

High pass filter? What high pass filter?

A mic signal recorded, edited and transmitted all digitally without any
hi-pass filter and compression in the chain is irresponsible practice
IMHO.

You were not hearing this a few years back when editing was done by
professionnal editors in high tech editing rooms, with decent
monitoring.

Receiving such deficient, unprocessed audio in its full glory thru
digital TV, feeding my full range home theater is sometime freightening.
How much time before some catastrophic level of VLF in TV audio actually
break subwoofers across the country?

End of rant.

--
Eric (Dero) Desrochers
http://homepage.mac.com/dero72

Hiroshima 45, Tchernobyl 86, Windows 95
Anonymous
September 28, 2004 3:16:05 AM

Archived from groups: rec.audio.pro (More info?)

Ty Ford <tyreeford@comcast.net> wrote:

> We couldn't pin down the source, but figured the frequency was pretty damn
> low if we were able to see the excursion. That would make it on the order of
> 1 or 2 cycles per second, right?
>
> Regards,
>
> Ty Ford

If the eye can detect the 48 (projected) images per second in theaters,
I'd bet it can see a woofer moving at frequencies clearly in the audible
range.

--
Eric (Dero) Desrochers
http://homepage.mac.com/dero72

Hiroshima 45, Tchernobyl 86, Windows 95
Anonymous
September 28, 2004 3:59:34 AM

Archived from groups: rec.audio.pro (More info?)

On Mon, 27 Sep 2004 23:16:05 GMT, deromax@hotmail.com (Eric
Desrochers) wrote:

>> We couldn't pin down the source, but figured the frequency was pretty damn
>> low if we were able to see the excursion. That would make it on the order of
>> 1 or 2 cycles per second, right?

>If the eye can detect the 48 (projected) images per second in theaters,
>I'd bet it can see a woofer moving at frequencies clearly in the audible
>range.

A couple of random thoughts on the topic:

Our vision is more sensitive to flicker around the periphery, less
sensitive in the central portion. Observing with the peripheral
vision sounds like something from Castaneda (and is!) but works.

Woofer excursion falls at 12 dB per octave above fundamental
resonance. Below resonance, sealed-box woofers (or tweeters, etc.
of course) have constant excursion with frequency for constant
signal input. Other boxes give different and usually worse
curves.

So, as a microphone, a "sub"woofer has no advantage over a similar
size high sensitivity woofer, like a musical instrument speaker, in
same box in the subsonic range. Both are on the same slope.

Aren't you glad you didn't ask?

Chris Hornbeck
Anonymous
September 28, 2004 11:34:57 AM

Archived from groups: rec.audio.pro (More info?)

Eric Desrochers <deromax@hotmail.com> wrote:

> A mic signal recorded, edited and transmitted all digitally without any
> hi-pass filter and compression in the chain is irresponsible practice
> IMHO.

When I record an organ in a church. It is true that I don't want the low
frequecies that are from outside the church. I don't know how to filter
them out because I DO want to catch the 32' at 16 Hz. As for dynamic
range I want that big chord at full blast and I also to want just barely
hear the organist slowly turn a page in a pause. I want a big dynamic
range. I do use a digital recorder. Please advise. How would you deal
with a situation like this?

Lars


--
lars farm // http://www.farm.se
Anonymous
September 28, 2004 12:47:37 PM

Archived from groups: rec.audio.pro (More info?)

Lars Farm <mail.addr.can.be.found@www.farm.se&gt; wrote:
>Eric Desrochers <deromax@hotmail.com> wrote:
>
>> A mic signal recorded, edited and transmitted all digitally without any
>> hi-pass filter and compression in the chain is irresponsible practice
>> IMHO.
>
>When I record an organ in a church. It is true that I don't want the low
>frequecies that are from outside the church. I don't know how to filter
>them out because I DO want to catch the 32' at 16 Hz. As for dynamic
>range I want that big chord at full blast and I also to want just barely
>hear the organist slowly turn a page in a pause. I want a big dynamic
>range. I do use a digital recorder. Please advise. How would you deal
>with a situation like this?

Record at 4:00 in the morning when nobody else is in town. If it is in
a city, get a subway schedule and try to schedule things so that you
aren't in the middle of a take when one goes by.

The big dynamic range is the easy part. That's just a matter of using good
preamps and converters. Getting rid of the low end noise requires shutting
the building air conditioning down, working at night when there is no
traffic, and often placing the mikes to avoid building vibrations from the
organ blower.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
September 28, 2004 4:48:10 PM

Archived from groups: rec.audio.pro (More info?)

Next time you go for a listen, be sure to get high beforehand.
Anonymous
September 28, 2004 5:45:04 PM

Archived from groups: rec.audio.pro (More info?)

On Mon, 27 Sep 2004 23:00:18 +0200, Edi Zubovic <> wrote:
> On Mon, 27 Sep 2004 18:51:57 GMT, "U-CDK_CHARLES\\Charles" <"Charles
> Krug"@cdksystems.com> wrote:
>
> ------8<--------------
>>>
>>
>>If that doesn't work, try a seismophone :) 
>
> --Actually, I'm curious on how they do it. I had a catalog of a German
> "Prakla-Seismos" company (I think it doesn't exist under this name any
> more) and I know that making big booms are made with either explosives
> or special high pressure compressors, like those German made Junkers
> 250 bar or more types, and they record the responses. I don't know
> what devices they're using for picking up the kaboom but I guess they
> are based on accelerometers or so.
>

The old fashioned way was to hang a large mass so it can move freely.
Because most earthquakes are very small, they'd have a mirror on the
mass arranged to multiply the movement.

The whole thing is scaled logarithmically. My recollection is that
there are several different scales used for different purposes, but I've
been away from that sort of thing for a while.
Anonymous
October 1, 2004 6:08:07 AM

Archived from groups: rec.audio.pro (More info?)

"philicorda" <philicorda@azriel.tydrwg.org> wrote in message
>
> Playing the recording ten times as fast revealed a underwater
soundscape
> of whale calls and clicks, strange unearthly tones. It was a
unexpected
> piece of ambient music.
>
> I'd like to get a better recording of this, as the mic/pre is rather
> noisy and has a limited low frequency response. Anyone got any ideas?


Sounds like maybe you should be using a hydrophone.

I'd have thought you'd want to know where you can find an exterminator
who will deal with basement ducks and sewer whales.

--
"It CAN'T be too loud... some of the red lights aren't even on yet!"
- Lorin David Schultz
in the control room
making even bad news sound good

(Remove spamblock to reply)
Anonymous
October 3, 2004 5:01:21 PM

Archived from groups: rec.audio.pro (More info?)

philicorda wrote:

> I'd like to get a better recording of this, as the mic/pre is rather
> noisy and has a limited low frequency response. Anyone got any ideas?

130 volts 4006 ... or a Sennheiser MKH106 or equivalent, as I recall it
the MKH omni's - at least in measurement versioning - used go down to 1
vs. 0.1 Hz, I plain dunno how the recent Sennheiser catalogue is ....

> Geophones seem rather expensive, so a DIY solution looks likely. A
> suspended coil of wire with a magnet fixed to the ground? How to get as
> little noise as possible?

Larger membrane B&K probably .... must be rentable somewhere.


Kind regards

Peter Larsen

--
*******************************************
* My site is at: http://www.muyiovatki.dk *
*******************************************
Anonymous
October 3, 2004 5:01:23 PM

Archived from groups: rec.audio.pro (More info?)

John Phillips wrote:

> The problem is that the ADC would need to sample slow enough to get good
> resolution at the low frequencies which may be difficult with PC sound
> cards.

????????????????????????????????

> John


Kind regards

Peter Larsen

--
*******************************************
* My site is at: http://www.muyiovatki.dk *
*******************************************
Anonymous
October 3, 2004 5:01:24 PM

Archived from groups: rec.audio.pro (More info?)

Peter Larsen <SPAMSHIELD_plarsen@mail.tele.dk> wrote:
>John Phillips wrote:
>
>> The problem is that the ADC would need to sample slow enough to get good
>> resolution at the low frequencies which may be difficult with PC sound
>> cards.
>
>????????????????????????????????

It's not like tape. The low frequency corner is set by the analogue stuff
in front of the converters. Most PC soundcards are capacitively coupled with
undersized caps, but there is no reason you can't get down to DC with 44.1
sampling on the correct hardware.

My HHb portadat has a -3dB point at 2 Hz using the line inputs. Using the
mike inputs it's a bit higher mostly because of the phantom blocking caps.

But getting good bass with digital is easy at any sampling rate. Not like
the analogue world where I'd do organ recordings at 7.5 ips to pitch the
head bump down an octave.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
October 3, 2004 8:33:50 PM

Archived from groups: rec.audio.pro (More info?)

On Sun, 03 Oct 2004 13:01:45 +0200, Peter Larsen wrote:

> philicorda wrote:
>
>> Section of original untouched recording:
>
> http://69.93.9.106/subcity/city_clean_original.mp3
>
> and
>
> http://69.93.9.106/subcity/city_fast.mp3
>
> and
>
> http://69.93.9.106/subcity/city_sub_pretty.mp3
>
> don't seem to work very well for me.

Sorry, that web server appears be playing up.
Try again today, it should work ok now.
Anonymous
October 3, 2004 11:13:35 PM

Archived from groups: rec.audio.pro (More info?)

On 3 Oct 2004 10:08:18 -0400, kludge@panix.com (Scott Dorsey) wrote:

>Peter Larsen <SPAMSHIELD_plarsen@mail.tele.dk> wrote:
>>John Phillips wrote:
>>
>>> The problem is that the ADC would need to sample slow enough to get good
>>> resolution at the low frequencies which may be difficult with PC sound
>>> cards.
>>
>>????????????????????????????????
>
>It's not like tape. The low frequency corner is set by the analogue stuff
>in front of the converters. Most PC soundcards are capacitively coupled with
>undersized caps, but there is no reason you can't get down to DC with 44.1
>sampling on the correct hardware.
>
>My HHb portadat has a -3dB point at 2 Hz using the line inputs. Using the
>mike inputs it's a bit higher mostly because of the phantom blocking caps.
>
>But getting good bass with digital is easy at any sampling rate. Not like
>the analogue world where I'd do organ recordings at 7.5 ips to pitch the
>head bump down an octave.
>--scott

I'm not worried with anything below abt 15 Hz. Why, I want to get rid
of _any_ DC.
At such recordings, you wouldn't measure not a sligtest DC offset even
at pressed copies. The same filtering I apply at abt. 19 kHz -- for
44100; I record all the time with. Higher rates and bit depths bloat a
file significantly; while 24 bit is OK and desirable, 10 min. of
material recorded at 192 kHz and interpolated 64 bits take 2 GB of
hard disk space but why -- the difference to 44100/24 is an audiophile
one only. And then you should resample, filter, reduce the bitdepth
and dither so and so).

Edi Zubovic, Crikvenica, Croatia

PS. Scott, a nice trick for analog recording of very low tones indeed
:) 
Anonymous
October 4, 2004 12:20:14 AM

Archived from groups: rec.audio.pro (More info?)

I confused the issue by not completing my thought. When you do a FFT, the
lower the sample rate for a fixed number of samples will provide higher
resolution at the lower frequencies (Bin Size = Fs/#Samples). The 6 KHZ
(and probably higher) is fine for this discussions.

For fun, I resampled the mp3 file to 6 KHZ and did a FFT to look at the
noise spectrum. I do not know what the mp3 encoding did to the spectrum but
what I was saw a slow rising noise floor from 3KHZ to about 140 HZ after
which it turned up sharply as it approached DC (until it flattened at around
25 HZ). The shape does resemble the classic 1/f noise common in
electronics. The lower sample rate made it easer to find the inflection
points.

Sorry for the confusion,

John


"Scott Dorsey" <kludge@panix.com> wrote in message
news:cjp14i$raq$1@panix2.panix.com...
> Peter Larsen <SPAMSHIELD_plarsen@mail.tele.dk> wrote:
> >John Phillips wrote:
> >
> >> The problem is that the ADC would need to sample slow enough to get
good
> >> resolution at the low frequencies which may be difficult with PC sound
> >> cards.
> >
> >????????????????????????????????
>
> It's not like tape. The low frequency corner is set by the analogue stuff
> in front of the converters. Most PC soundcards are capacitively coupled
with
> undersized caps, but there is no reason you can't get down to DC with 44.1
> sampling on the correct hardware.
>
> My HHb portadat has a -3dB point at 2 Hz using the line inputs. Using the
> mike inputs it's a bit higher mostly because of the phantom blocking caps.
>
> But getting good bass with digital is easy at any sampling rate. Not like
> the analogue world where I'd do organ recordings at 7.5 ips to pitch the
> head bump down an octave.
> --scott
>
>
> --
> "C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
October 4, 2004 1:18:11 AM

Archived from groups: rec.audio.pro (More info?)

Edi Zubovic wrote:

> PS. Scott, a nice trick for analog recording of very low tones indeed
> :) 

And it saves tape. <g>

--
ha
Anonymous
October 4, 2004 2:02:33 AM

Archived from groups: rec.audio.pro (More info?)

philicorda wrote:

> Try again today, it should work ok now.

So it seems ... thanks!


Kind regards

Peter Larsen

--
*******************************************
* My site is at: http://www.muyiovatki.dk *
*******************************************
Anonymous
October 5, 2004 4:05:56 AM

Archived from groups: rec.audio.pro (More info?)

On Sun, 03 Oct 2004 21:18:11 GMT, walkinay@thegrid.net (hank alrich)
wrote:

>Edi Zubovic wrote:
>
>> PS. Scott, a nice trick for analog recording of very low tones indeed
>> :) 
>
>And it saves tape. <g>
And asks for digitizing so you could try to mend those noticeable
dropouts in eg. Sound Forge <big <g>>

Edi Zubovic, Crikvenica, Croatia
Anonymous
October 5, 2004 2:48:51 PM

Archived from groups: rec.audio.pro (More info?)

philicorda wrote:

> I want to keep the result as free from digital artifacts
> as possible, so speeding it up works better than pitch/time
> shifting.

Hmmm ... the recorded tracks seem to peak in content around 27 Hz, and
the provided sample plays back well on this here multimedia setup with
at least the lorry noises clearly identifiable.

Digital noise reduction is not likely to work well unless there is a gap
between signal and noise-floor. With signal reaching into the noise the
result will be artifacts and unnatural sound. Just what part of the city
noise is is that you call noise btw... ? ... anyway: you should probably
first try modest expansion to create or widen the gap between wanted and
unwanted.

> Another thought that came to mind is using two octavas as
> a stereo pair a few feet apart. One it's recorded, invert
> the phase of one, add it to the other to get the difference,
> and then subtract that difference from the first. In theory,
> if the frequency is low enough, the waveform should
> be near enough identical at the positions of both mics,
> and I can cancel out the noise?

That should give you comb to your hair. The concept is moderately usable
for a mic that has to discriminate between very close and generally
distant sources.

Use them as a stereo pair to record stereo and keep them as far apart as
the room reasonably permits. I have recorded fireworks using the
building my flat is in as a jecklin disk with the result being excellent
stereo panorama of the very distant sources.

> Other thoughts is that the building is just acting as a low pass filter,
> and there is no reason why I can't record outside, and then low pass the
> result electronicly?

This is when you rent hydrophones and bolt them to concrete slabs
outside.

> Here are some excerpts of the recording. I recorded an hour,
> and then sped it up to 4min by abusing a sample rate convertor.
> All files are around 3-5mb in size.

Way too much speed up. Interesting nonetheless, the song of the lorries,
the railroads and - possibly - the elephants in the nearby zoo if any.
Watch out in case the french Musique Concrete guys hear of this .... you
may have initiated a new musical genre.


Kind regards

Peter Larsen


--
*******************************************
* My site is at: http://www.muyiovatki.dk *
*******************************************
Anonymous
October 17, 2004 7:56:39 PM

Archived from groups: rec.audio.pro (More info?)

On Tue, 05 Oct 2004 10:48:51 +0200, Peter Larsen wrote:

> philicorda wrote:
>
>> I want to keep the result as free from digital artifacts
>> as possible, so speeding it up works better than pitch/time
>> shifting.
>
> Hmmm ... the recorded tracks seem to peak in content around 27 Hz, and
> the provided sample plays back well on this here multimedia setup with
> at least the lorry noises clearly identifiable.

I did the same experiment with a u87 set to omni and the results are quite
different, without some of the strange resonance that lead to the boxy
sound when the file is sped up, but with others... I'm not sure what's
going on here, but I suspect the process is revealing differences in the
mics that would otherwise be inaudible, especially about their noise
floor. It's still to noisy for me, so I'll have a bash with the Sennheiser
MKH mic mentioned (I've always wanted to have a play with one of these
anyway).

>
> Digital noise reduction is not likely to work well unless there is a gap
> between signal and noise-floor. With signal reaching into the noise the
> result will be artifacts and unnatural sound. Just what part of the city
> noise is is that you call noise btw... ? ... anyway: you should probably
> first try modest expansion to create or widen the gap between wanted and
> unwanted.

It's really the electronic hiss that I want to get rid of. Anything
else is fair game. Also, the noise of the computer is audible, even though
I can't hear it in the room. I suspect it's getting through via the 6 inch
wide cable pipe to the control room above the live room, as I have not
finalised the wiring enough to block it up yet.

>
>> Another thought that came to mind is using two octavas as a stereo pair
>> a few feet apart. One it's recorded, invert the phase of one, add it to
>> the other to get the difference, and then subtract that difference from
>> the first. In theory, if the frequency is low enough, the waveform
>> should be near enough identical at the positions of both mics, and I
>> can cancel out the noise?
>
> That should give you comb to your hair. The concept is moderately usable
> for a mic that has to discriminate between very close and generally
> distant sources.

A little googling reveals there are people doing something like this with
geophones, but it involves many sources, assuming things about the
signal I don't think would work for audio, and all sorts of horrible
convolution and maths. Not exactly the simple messing about with phase I
had imagined. Doing it the complex way does have the added benefit of
direction finding though. :) 

>
> Use them as a stereo pair to record stereo and keep them as far apart as
> the room reasonably permits. I have recorded fireworks using the
> building my flat is in as a jecklin disk with the result being excellent
> stereo panorama of the very distant sources.
>
>> Other thoughts is that the building is just acting as a low pass
>> filter, and there is no reason why I can't record outside, and then low
>> pass the result electronicly?
>
> This is when you rent hydrophones and bolt them to concrete slabs
> outside.
>
>> Here are some excerpts of the recording. I recorded an hour, and then
>> sped it up to 4min by abusing a sample rate convertor. All files are
>> around 3-5mb in size.
>
> Way too much speed up. Interesting nonetheless, the song of the lorries,
> the railroads and - possibly - the elephants in the nearby zoo if any.
> Watch out in case the french Musique Concrete guys hear of this .... you
> may have initiated a new musical genre.

Yes, It's too fast to hear much of the detail in the sounds. I think I
need a transducer with much better low frequency response as well, it's
the very low slow stuff that I find interesting when it's made audible,
and I was hunting for that by speeding it up so much.

>
>
> Kind regards
>
> Peter Larsen
Anonymous
October 17, 2004 8:05:56 PM

Archived from groups: rec.audio.pro (More info?)

philicorda <philicorda@azriel.tydrwg.org> wrote in
news:p an.1998.01.20.08.58.50.858435@azriel.tydrwg.org:

>> Digital noise reduction is not likely to work well unless there is a
>> gap between signal and noise-floor. With signal reaching into the
>> noise the result will be artifacts and unnatural sound. Just what
>> part of the city noise is is that you call noise btw... ? ... anyway:
>> you should probably first try modest expansion to create or widen the
>> gap between wanted and unwanted.
>
> It's really the electronic hiss that I want to get rid of. Anything
> else is fair game. Also, the noise of the computer is audible, even
> though I can't hear it in the room. I suspect it's getting through via
> the 6 inch wide cable pipe to the control room above the live room, as
> I have not finalised the wiring enough to block it up yet.

Bear in mind that digital NR is narrow band. In my setup, I split it to
24000 bands, smaller than 1 per Hz. The only time you'll lose signal is
when the signal level is below the noise threshold in the band being
processed.

If your content is not terribly bassy, doing intense digital NR in the low
bands will be quite effective. I frequently heavily treat organ concert
recordings on a busy downtown street by high pass filtering accompanied by
rather intense NR limited to frequencies below the content. Of course it's
a bit time consuming to change band pass as the piece moves from section to
section, but the effect can practically eliminate rumbling trucks and
thundering (distantly) jetliners.
!