Re: Sensitivity and Frequency Response

On Wed, 21 Jan 2009 13:32:29 +0000,
adrian@xxxxxxxxxxxxxxxxxxxxxxxxxxxx (Adrian Tuddenham) wrote:

On Wed, 21 Jan 2009 12:12:38 +0000,
adrian@xxxxxxxxxxxxxxxxxxxxxxxxxxxx (Adrian Tuddenham) wrote:

On 20 Jan 2009 18:01:17 -0500, kludge@xxxxxxxxx (Scott Dorsey) wrote:

In article <49724f9c.134214015@localhost>, Don Pearce <> wrote:
Kludge wrote:
i don't trust the noise figures on the data *** and I kind of doubt
the NT1 is as quiet as the manufacturer claims. But it's still very
quiet and certainly quieter than an RE-50.

I've measured one of mine, substituting an equivalent non-microphonic
capacitor for the capsule, and it met the spec quite happily -
something over 4dB. Certainly much quieter than any dynamic I ever
looked at.

If you disconnect the primary noise source in the microphone (which is
the capsule in any good modern design), of course you will get very
good noise numbers. Unfortunately these numbers are not useful and
correlate neither with the published IEC and ISO measurements nor with
the real-world performance.
--scott

The capsule isn't the noise source - what would the mechanism be?

A very high resistance shunted by a capacitance.

Exactly - not the capacitance itself, which I replaced by a
non-microphonic alternative. The capsule in a non-moving state is
purely reactive, and therefore can't be a source of energy. It is only
when it moves that it can impart some energy to the amplifier, and it
only moves in response to a stimulus - in other words an external
source of noise or vibration. It has no self-noise, that is purely
down to the electronics.

It has 'infinite' resistance so, according to simplified Johnson theory,
it generates infinte voltage noise - but from a source which is is
capable of gving zero current. Zero current produces zero voltage
across a capacitor.


No it isn't resistive - it is reactive.

I was being purist. No capacitor is a perfect reactance and when the
resistance is 'infinite' and noise rises with resistance, you have to
sit down and explain to yourself why there is no Johnson Noise from it.

...Any residual resistive
component it may have is well shunted by the capacitor, and can be
discounted.

That was the point I was trying to make - but it is still a good idea to
be aware of what you are discounting and why; otherwise, one day, it
will creep up from behind and bite you

[...]
It is possible to reduce the noise contribution of the shunt resistance
by making it part of a feedback loop. I have used this system to good
effect (noise figures around 1dB or better) with lower impedance
pre-amplifiers, but have never tried it with a capacitor capsule.

You can't feed back all the way to the acoustic coupling, so it
doesn't apply in this case.

I believe there have been attempts to do almost that (restoring the
diaphragm position by using an electronically-generated force), but no
practical equipment has yet come onto the market.


Anything that results from the diaphragm being moved by air molecules
isn't microphone noise - it is external noise, and for the purpose of
this measurement must be considered valid signal.

That's fair enough. I have heard of microphone tests being conducted in
a vacuum in order to give the most accurate results. (Not to be confused
with the hydrogen pistonphone fiasco that was hushed-up by Victor/HMV in
the late 1920s)


--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk
.

Loading