On 1/7/08, Sean Breheny <shb7@cornell.edu> wrote:
> Hi Vasile,
>
> On Jan 6, 2008 11:50 PM, Vasile Surducan <piclist9@gmail.com> wrote:
> > AFIK noise and impedance matching in audio are two different problems.
> > You may connect a 50 ohm signal source to a 10K input impedance amplifier easily
> > using either an operational amplifier (if you need bigger gain) either directly.
> > It's not RF, it's audio.
> >
> >
>
> Yes, that's true, but I think you missed the point. I have a signal
> source which has only a 50 ohm source impedance.
> If I could transform that up to something higher (it need not be
> exactly 10K), then I would have essentially "free" voltage gain.


Sean,
Depends how it affects the load your mixer or whatever you may have in
front of your audio schematic.
For example the simplest way to decrease the input impedance of a
noniverting amplifier is to hook up to the input a resistor to ground
equal with the impedance you want (this is just a grosso modo
approach).
As long the output signal is 50 ohm and the input amplifier is 10K,
connecting them toghether will equalise the impedance at 50 ohm.

>
> So, for example, if the audio noise level coming out of the mixer is
> 1uV, and the input noise level of the audio amp is 5uV, then directly
> connecting them will
> cause my signal to noise ratio to be almost 14dB worse than if I
> brought the impedances closer to matched.


Defining a signal to noise ration means you know also which is the
signal amplitude, not only the noise... For a 10mV signal and 1uV
noise do you think the 5uV noise induced by the amplifier will really
count ? (assuming you didn't set the OA gain 1000...)

I think your problem is the mixer and the filters after the mixer not
the audio chain.

>
> Yes, if my audio amp had a 1uV input noise, or if I inserted a preamp
> between them with a noise lower than the output noise of the mixer,
> then it would be just as good.
>
> In this case, that is probably what I will do. However, it just got me
> thinking about how this could be handled in situations where the noise
> levels were lower (for example, imagine that somehow you had a signal
> source with only 0.1uV noise - it is then challenging to design an
> audio amplifier which has a lower noise, especially at the low end of
> the spectrum.
>
> Sean
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist