[Elecraft] USB sound dongle

[Joe Subich, W4TV]

On 8/21/2012 11:18 AM, Jim Brown wrote:
 > On 8/21/2012 7:48 AM, Barry LaZar wrote:
>> However, 24 bits of quantization will get you a better dynamic
>> range.
 >
> Not if the dynamic range of the analog electronics limits it, which
> is what Joe has pointed out. Often those extra bits are "advertising
> bits" -- they fool those who don't know any better into believing
> that more bits is better.

There are two issues here -

First most amateurs run their digital software from "line out" or the
demodulated audio from the transceiver.  The demodulated audio has
generally been through at least one A to D to A cycle and the noise
floor is simply too high to take advantage of even the theoretical 96
dB dynamic range of a 16 bit sound card.

Now, if one is looking at the "raw" I/Q RF/IF or direct downconversion
audio 24 bits "might" be usable *IF* the noise floor is low enough (a
real *IF* with off the air signals) and any mixers and amplifiers are
really linear through the theoretical 138 dB dynamic range *without* AGC.

None of this addresses the noise floor and dynamic range of any preamp
in the sound card ... it only looks at the theoretical dynamic range
of the analog to digital converter.

Then there is an additional consideration pointed out by Chen in a
private e-mail - properly designed software can achieve dynamic range
(sensitivity) *better than* the A to D converter (that is, it can
"receive into the noise") by applying proper lowpass filtering and
decimation.  I know Chen uses decimation even with 16 bit cards in
cocoaModem but it does not appear that any of the Windows centric
digital software does so since many are based on 11025 or 12000 Hz
sample rates and effectively throw away the potential processing gain
by not performing any lowpass filtering in the 44100/48000 to 11025
or 12000 sample rate downconversion.

73,

    ... Joe, W4TV


On 8/21/2012 11:18 AM, Jim Brown wrote:
> On 8/21/2012 7:48 AM, Barry LaZar wrote:
>> However, 24 bits of
>> quantization will get you a better dynamic range.
>
> Not if the dynamic range of the analog electronics limits it, which is
> what Joe has pointed out. Often those extra bits are "advertising bits"
> -- they fool those who don't know any better into believing that more
> bits is better.  Actually, it's far more important to optimize the
> signal level that the A/D converter sees. That's why I like the Tascam
> USB interface better than the Numark -- it has a much greater range of
> gain adjustment and a clip light on the front panel. That allows me to
> easily optimize the 90 dB or so of dynamic range that its 16 bits provides.
>
> Besides -- the 90 dB dynamic range provided by a decent 16-bit A/D is
> probably more than most hams will ever see at the audio output of their
> radios.  While the radio itself has much greater dynamic range, the
> WORKING dynamic range is limited by noise received by the antenna that
> is often 30-50 dB greater than the electrical noise levels inside the
> radio.
>
> 73, Jim K9YC
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:Elecraft at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>