Loading ...
Sorry, an error occurred while loading the content.

72178Re: [softrock40] Re: USB sound card dongle with stereo INPUT

Expand Messages
  • Milt Cram
    Feb 9, 2013
    • 0 Attachment
      On 2/9/2013 4:05 PM, Alan wrote:
      > ----- Original Message -----
      > Subject: Re: [softrock40] Re: USB sound card dongle with stereo INPUT
      >> I suspect that what you claim is true if you are receiving amplitude
      >> modulated signals. However, if you are receiving a digital signal, such
      >> as WSPR, won't the additional bits help in the signal processing gain?
      >> Since WSPR can detect signals with more than -20 dB S/N, I would think
      >> that 24 bit resolution would produce better results that 16 bit
      >> resolution.
      Hi Alan,

      My comments inserted below--
      > The way I understand it is that 24 bits does not give better resolution.
      > But it gives a greater dynamic range.
      > I have not heard anyone suggest that 24 bits on it's own improves the reception of weak signals.
      24 bits does give better resolution than 16 bits. Most modern A to D
      converters can digitize a signal whose peak to peak value is less than
      the supply voltage for the ADC (typically 5V). The minimum signal that
      can be quantized is 2^N times smaller than the largest signal. This is
      the "resolution" of the device. For N equal to 16 bits, the smallest
      signal is 65535 times smaller than the largest signal. For N equal to
      24 bits (most 24 bit ADC do not have a true 24 bit range, but an
      "effective number of bits" that is less--e.g. 21 bits), the smallest
      signal that can be resolved is more than 16 million times smaller than
      the largest signal. We typically refer to dynamic range as the ratio of
      the largest signal (which is frequently independent of the number of
      bits) to the smallest signal that can be quantized. Therefore the 24
      bit ADC has both improved resolution, and improved dynamic range.

      > My observations seem to show that the noise from the Softrock opamps is greater than the noise from a low noise soundcard. And
      > greater than antenna noise at higher frequencies, maybe above 21MHz.
      This is generally true, unless you are fortunate to live in an area that
      has exceptionally low environmental noise (received from the antenna).
      > Others have said that reducing the opamp gain in the Softrock does enable low noise 24 bit cards to increase the dynamic range.
      This is not increasing the dynamic range of the ADC, but is merely
      shifting the signal levels to utilize as much of the range of the ADC as
      possible. You may, however, sacrifice some of the weak signal
      capability of the ADC.
      > Without modification a Softrock will not benefit much, if anything, from a 24 bit card.
      > 73 Alan G4ZFQ
      I still believe that, when operating some of the sophisticated digital
      modes that require substantial signal processing, there is an advantage
      to using 24 bit ADCs. Even though noise is present, you need to be able
      to resolve tiny changes in the combination of signal and noise. By
      averaging (integrating) numerous samples of signal plus noise, you are
      able to suppress the effects of the noise. This, of course, assumes
      that the noise is uncorrelated with the desired signal, and that the
      noise has zero average value. (I won't comment on techniques that are
      used when the noise does not meet this criteria).

      For the typical CW or SSB user, I would agree that most would not
      experience a significant improvement when going from 16 bit to 24 bit ADCs.

      73, Milt
    • Show all 35 messages in this topic