On 1/6/08, Sean Breheny <shb7@cornell.edu> wrote:
> Hi all,
>
> I have a very low-level (microvolt) audio source with a 50 ohm
> impedance (it is a diode mixer in a direct conversion receiver). I
> also have an audio amplifier which has a gain of about 10000 in
> voltage, 5 uV input noise, flat audio response down to about 10Hz, and
> a differential audio input with a 10K impedance.
>
> I want to feed the signal from the mixer into the audio amp. I realize
> that I should provide a flat 50 ohm termination for the mixer and that
> I should also filter out the RF products coming out of the mixer.
>
> However, it would be nice to transform that 50 ohm impedance to
> something closer to 10K because it would be free (noise free) gain.


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.


>
> I could use a transformer, but I'd like to avoid it for several
> reasons all relating to the fact that it doesn't have good low
> frequency response. This not only spoils the nice low-end ability of
> the output amp, but also could be a problem because down the line a
> bit I am planning on allowing this circuit to do synchronous detection
> (the LO is produced by a DDS which is digitally tunable on the fly,
> and is controlled by a PIC. If I feed some of the audio into an ADC
> channel on the PIC, I can shift the frequency and even phase of the
> DDS to match the carrier of an AM station, providing much better
> demodulation quality. However, this needs (almost) DC response from
> the audio up to the ADC).
>
> It seems that there should be some impedance matching device which
> would work all the way down to DC (even if it has to be active and
> have a net power loss). For example, a common-base transistor amp
> naturally has a low input impedance and a high output impedance.
> However, if I just implement it in the normal way, it has overall
> power gain AND adds noise. Is there some way to reduce the
> input-referred noise (at the expense of power gain)? In other words,
> can one make an active version of a transformer which approaches the
> low noise quality of the transformer but works down to DC (or almost
> DC)?
>
> Thanks,
>
> 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