In article , Dave Plowman (News)
wrote:
In article , Jim Lesurf
wrote:
In principle you can get two pressure-sensing mics and place them
slightly apart. Then connect so that their outputs 'cancel'. This
produced a setup that outputs the 'differential' in pressure along the
direction between them. So gives a cardiod-ish shape

You've confused me. The easy way to make a cardiod pattern with two mics
is to use pressure and pressure gradient types.

Sorry, I should havbe given more detail. The above jumped over too much.

Two pressure sensors spaced apart will sense the pressures in those two
*different* places. A wave propagating between them takes time to do so.
Hence if you subtract their outputs you can output the spatial differential
of the pressure. However - as per your own comments - this isn't an easy
way to do it. :-)

In fact a 'velocity' sensor is also a type of sensor for the spatial
pressure differential. It works due to the pressures on either side of the
sensor being different. Its advantage is that one bit of material (e.g. a
ribbon) is used to sense both side-pressures so avoids the difficulty of
having to pressure sensors which may have different sensitivities, etc.

So a 1st order pressure pair will give something like a dipolar pattern.
More than three pressure elements can give more like a cardiod. And a
'velocity' sensor can take the place of two pressure elements in the array.
Hence the "pressure and pressure gradient" method.

If, like Phil, you're puzzled by this, have a look at:

Directional Microphones H. F. Olsen
JAES October 1967 (Also in the old 'Microphones' book of reprints by AES)

as an example. See his first figure where he explains 1st order gradient
mics before going on to look at end-fire arrays, etc.

Slainte,

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

"Jim Lesurf"

If, like Phil, you're puzzled by this, have a look at:

Directional Microphones H. F. Olsen
JAES October 1967 (Also in the old 'Microphones' book of reprints by AES)


** What a pathetic crock of ****.

The "proof" of JLs absurd claim is not to be found anywhere.




.... Phil

On Sat, 23 Mar 2013 12:08:26 +0000 (GMT), Jim Lesurf
wrote:

In article , Dave Plowman (News)
wrote:
In article , Jim Lesurf
wrote:
In principle you can get two pressure-sensing mics and place them
slightly apart. Then connect so that their outputs 'cancel'. This
produced a setup that outputs the 'differential' in pressure along the
direction between them. So gives a cardiod-ish shape

You've confused me. The easy way to make a cardiod pattern with two mics
is to use pressure and pressure gradient types.

Sorry, I should havbe given more detail. The above jumped over too much.

Two pressure sensors spaced apart will sense the pressures in those two
*different* places. A wave propagating between them takes time to do so.
Hence if you subtract their outputs you can output the spatial differential
of the pressure. However - as per your own comments - this isn't an easy
way to do it. :-)

In fact a 'velocity' sensor is also a type of sensor for the spatial
pressure differential. It works due to the pressures on either side of the
sensor being different. Its advantage is that one bit of material (e.g. a
ribbon) is used to sense both side-pressures so avoids the difficulty of
having to pressure sensors which may have different sensitivities, etc.

So a 1st order pressure pair will give something like a dipolar pattern.
More than three pressure elements can give more like a cardiod. And a
'velocity' sensor can take the place of two pressure elements in the array.
Hence the "pressure and pressure gradient" method.

If, like Phil, you're puzzled by this, have a look at:

Directional Microphones H. F. Olsen
JAES October 1967 (Also in the old 'Microphones' book of reprints by AES)

as an example. See his first figure where he explains 1st order gradient
mics before going on to look at end-fire arrays, etc.

Slainte,

Jim

I think what you are describing in the first part is the "distance
cancelling" microphone. It responds to the pressure difference between
two capsules, hence responds strongly to close sources, and very
little to more distant ones. Its main problem is that it has to be a
compromise between cancellation and sensitivity. If the two capsules
are far enough apart to give good differential voltage, they are also
far enough apart to suffer from phase-shift problems - when the
distance is a half wavelength, they will reinforce instead of
subtract.

d

In article , Don Pearce
wrote:
On Sat, 23 Mar 2013 12:08:26 +0000 (GMT), Jim Lesurf
wrote:

In article , Dave Plowman (News)
wrote:

I think what you are describing in the first part is the "distance
cancelling" microphone. It responds to the pressure difference between
two capsules, hence responds strongly to close sources, and very little
to more distant ones.

That's a different effect to the one I (and Olsen, etc) was referring to.
he goes on to use this to explain the construction of highly directional
end-fire mic arrays. I was just pointing out you can do this, but getting
it right isn't easy.

The point is there is a time delay required for a sound wave to propagate
from one sensor element to the other. That leads to a pressure difference
given a finite wavelength, even for a plane wave which has the same peak
amplitude in both locations.

In effect, it is just like using a phased array in RF to sculpt a required
pattern of directionality.

Slainte,

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

"Jim Lesurf"

The point is there is a time delay required for a sound wave to propagate
from one sensor element to the other. That leads to a pressure difference
given a finite wavelength, even for a plane wave which has the same peak
amplitude in both locations.

** Here comes the "though experiment" - just as I fearded,


In effect, it is just like using a phased array in RF to sculpt a required
pattern of directionality.


** Utter ******** !!!!

Wot a fool.

On Sat, 23 Mar 2013 12:57:16 +0000 (GMT), Jim Lesurf
wrote:

In article , Don Pearce
wrote:
On Sat, 23 Mar 2013 12:08:26 +0000 (GMT), Jim Lesurf
wrote:

In article , Dave Plowman (News)
wrote:

I think what you are describing in the first part is the "distance
cancelling" microphone. It responds to the pressure difference between
two capsules, hence responds strongly to close sources, and very little
to more distant ones.

That's a different effect to the one I (and Olsen, etc) was referring to.
he goes on to use this to explain the construction of highly directional
end-fire mic arrays. I was just pointing out you can do this, but getting
it right isn't easy.

The point is there is a time delay required for a sound wave to propagate
from one sensor element to the other. That leads to a pressure difference
given a finite wavelength, even for a plane wave which has the same peak
amplitude in both locations.

In effect, it is just like using a phased array in RF to sculpt a required
pattern of directionality.

Slainte,

Jim

Ah, OK, I see what you are talikng about. Sennheiser use a tube in
front of the mic with a slot down the side. The tube is lightly damped
so it doesn't resonate. Anyway, the idea is that sound coming from
boresight enters the tube all the way down the slot in phase
(travelling, but in phase) and reinforces at the capsule. Any other
angle and there is sufficient length of tube for cancelling wavefronts
to enter, and reduce the level at the capsule. they manage a pretty
good shotgun this way,.

d

In article , Don Pearce
wrote:
On Sat, 23 Mar 2013 12:57:16 +0000 (GMT), Jim Lesurf
wrote:

In article , Don Pearce
wrote:
[snip]


Ah, OK, I see what you are talikng about. Sennheiser use a tube in front
of the mic with a slot down the side. The tube is lightly damped so it
doesn't resonate. Anyway, the idea is that sound coming from boresight
enters the tube all the way down the slot in phase (travelling, but in
phase) and reinforces at the capsule. Any other angle and there is
sufficient length of tube for cancelling wavefronts to enter, and reduce
the level at the capsule. they manage a pretty good shotgun this way,.

Yes, that's a practical example of the kind of system people make based on
the approach. Here each section of the gap acts as a small hole as a sensor
collector.

You could make something nominally similar with an array of actual pressure
mics with delays on their outputs before adding them.

Its similar to an end-fire RF array of elements.

If you use a pair of actual pressure sensors you can then *subtract* their
electronic outputs, leading to what I described earlier if the sensors are
far enough apart to make the acoustic propagation time large enough between
them. The behaviour is analogous to something like a ribbon that senses the
difference in pressure between its sides.

Slainte,

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

Eiron wrote:
Those electret microphone capsules are quite good but omnidirectional.
Is there any way to modify them to give a different pattern?
Modify the case, or some suitable shaped enclosure?

There are some digital recorders available with what looks,
though I haven't looked too closely, like two electret capsules
at right angles. If they are omnidirectional, they wouldn't give
much stereo effect. Anyone know what they are?


Without modifying the capsule, you can put a plate in-between them.
Pressure zone principal. You can mount the capsule flush on the plate. You
can use two plates at 90 degrees or more. I did this many years ago. The
bigger the plate, the lower the frequency of directionality.

I was looking at a capsule. It might be possible to drill a small hole in
this one. Looks like there is a rubber like seal on the back. That's to
make a rear entry point, but an extension of the rear would be needed.

http://www.pitt.edu/~szekeres/mic/wm63.jpg

Greg

On 23/03/2013 17:11, gregz wrote:
Eiron wrote:
Those electret microphone capsules are quite good but omnidirectional.
Is there any way to modify them to give a different pattern?
Modify the case, or some suitable shaped enclosure?

There are some digital recorders available with what looks,
though I haven't looked too closely, like two electret capsules
at right angles. If they are omnidirectional, they wouldn't give
much stereo effect. Anyone know what they are?


Without modifying the capsule, you can put a plate in-between them.
Pressure zone principal. You can mount the capsule flush on the plate. You
can use two plates at 90 degrees or more. I did this many years ago. The
bigger the plate, the lower the frequency of directionality.

I was looking at a capsule. It might be possible to drill a small hole in
this one. Looks like there is a rubber like seal on the back. That's to
make a rear entry point, but an extension of the rear would be needed.

http://www.pitt.edu/~szekeres/mic/wm63.jpg

Now try to put it back together!

It's probably possible to block the holes in this
http://cpc.farnell.com/jsp/level5/mo...cpc/405947.xml
with blu-tack and make it omnidirectional.

And at 50p each I'll buy a few to experiment (play) with.

--
Eiron.