Re: Is this a good way to test pre-amp hiss of a digital recorder?


Pooh Bear wrote:

But if you want to measure the PREAMP - and that's what I thought we
were talking about, you don't want a measurement that varies with the
mic that you use with it.

But that *is* a measure of how good the preamp is. e.g a fet input pre-amp will
both measure quieter and be quieter in real use when measured using the resistor
method than a bipolar one measured into a short circuit which may appear on the
basis of shorted input noise to be better.

If you're talking about "in use" then evaluate its performance in use,
not with a resistor. Besides, if you view the preamp as a "black box"
you don't know what kind of input stage it has. You might experiment
with a shorted input and a simulated input with a known resistance and
try to determine this, and then, depending on who you are and why
you're measuring it, make the measurement in the way that gives you the
results that you want.

Typically 150R or 200R is stated together with the measuring bandwidth.

Can you point me to a piece of audio gear that might be used by readers
of this newsgroup that has such specifications? More often than not we
see just "Noise: -72 db" (an incomplete specification even as it
stands) or "EIN: -127 dB(sometimes u)."

I'm not saying that it can't be done, just that it isn't as common as
you seem to think it is. Some people want to try to duplicate
measurements at home, or make measurements that the manufacturer or a
review doesn't provide. For those people, (who are as likely as not to
use an open input) a shorted input is probably better than using a
resistor. They'll get a number that's closer to what's advertised, and
then you can tell them that in the real world, they can expect it to be
about 6 dB worse because it's being used with a real microphone.

Lab tests should be fully documented, but informal tests should be easy
to understand and compare results. Remember, the original poster was
looking to test an iRiver recorder. How is he going to separate the
input noise from any other noise given that it has stuff other than an
input stage ahead of the only output he can look at. And he doesn't
have any way of measuring the gain at any setting. Best he can do is
record a WAV file with a known dummy input and at the full gain setting
(whatever that is), and them measure the RMS amplitude of the noise in
terms of dBFS. As to what the actual noise output voltage is, that
depends on the D/A converter and whatever follows it. This could be the
headphone or line output jack of the iRiver at some playback volume
setting (which is also unknown, or at least arbitrary).

If it's common knowledge that a real
mic will raise the noise level by 6 to 12 dB, then those with the
knowledge can expect more noise than the published value.

Those with knowledge ! Ahem !!!

Hey, that's us!

In truth it's just as easy to state both figures. But then again for practical
purposes it's the one with a resistor that makes sense. The other figure is simply
specmanship for its own sake - you can't use it that way.

But it has to be stated, and we seem to disagree on how, and how often,
it's stated.

In any case it's the resistor figure that's been the norm for decades. Why worry
about it now ?

Because I"ve never worried about it before. I used to test preamps with
a resistor for the obvious reasons - that it looks something like a
microphone and the preamp input, for wahtever reasons, might care. But
then indiscussing it with DJ, who knows his microphones, I changed my
method. I don't know what other methods, other than preamp designers -
how obviously want to evaluate their own design's performance, and not
necessarily compare it with someone else's - use. I rarely if even see
it stated. Though I'll admit that, not looking for it, it might be
stated and I missed it because I wasn't looking for it.

Show me three.

.

Relevant Pages

  • Re: Confidence limits for noise measurement
    ... required to reduce the effect of chance on the measurement result ... The first thing to do is calibrate the instrument against a standard ... white noise, an attempt to measure the noise source by sampling the ...
    (rec.radio.amateur.antenna)
  • Re: Questions about resistor noise ???
    ... resistors are sources for noise in circuits. ... In a simple DC circuit (stable power source, and carbon resistor ... And yes, if the measurement bandwidth is zero, the ...
    (sci.electronics.basics)
  • Re: How to measure really small voltages
    ... >>measure this small voltage? ... One thing nobody has mentioned is the cross-correlation method. ... can be made AC by a variety of methods, but the additive Johnson noise is ... In the cross-correlation measurement, you take two voltmeters (made e.g. from ...
    (sci.electronics.design)
  • Re: Low noise IF3601/02 dont meet specs
    ... How much noise have you measured ... >> I have read that thickfilm resistors have more noise than ... The question is now about the used noise measurement method. ...
    (sci.electronics.design)
  • Re: Sun noise
    ... I have done two series of 250 measurements of audio noise voltage ... estimate of the uncertainty that I referred to in an earlier post. ... distribution, so the variance is 1/12 of the span. ... and that might also be true of measurement of noise voltage ...
    (rec.radio.amateur.antenna)