[Lf] [Fwd: LF: Sound Cards]

[Andre Kesteloot]

Talbot Andrew wrote:

>  I built a custom LF receiver last year, filtering to 180Hz bandwidth,
> then converting the signal to a 1kHz tone which was sampled by a
> microchip MCP3204 12 bit A/D.at 4kHz.  Presumably, being made by the
> same company the A/D core is the same as that internal to their PIC
> processors (10 bit vs. 12 bit notwithstanding).    Then I and Q
> baseband signals at 1kHz were generated by taking four successive
> samples in groups (labelled S1, S2, S3 and S4) and calculating     I =
> S1 + S2 - S3 - S4   and   Q = S1 - S2 - S3 + S4.    This is
> effectively the same as multiplying the 1kHz audio by two 90 degree
> shifted square waves at 1 kHz.   Provided the input is band limited to
> the range 750 - 1250 Hz, no mixer products or alliases ought to
> appear.After writing some waterfall software to read the I/Q samples
> (at effectively 13 - 14 bit resolution) I was never too happy with the
> result. Although a dynamic range of 50dB was achieved, there were
> spurious lines present, which appeared to be related to 1kHz and
> subharmonics so I can only assume these were artifacts of the A/D
> converter itself.    Better results were achievable with the RA1792
> plus 56002 EVM sampling at 8kHz (better than most soundcards) and no
> sprog lines were present with the antenna disconnected, so I put the
> LF receiver on the shelf for a 'later' follow upWIth a 10 bit
> converter you will be limited to at most 60dB dynamic range which poor
> soundcards can match and many better, but you will have the very
> considerable advantage of being able to control the sample rate
> properly.   Furthermore you can probably control stray pick up much
> better as the audio circuits can be separated from the PC, linked with
> a properly filterd data link (RS232, USB or whatever)  With decimation
> and post processing this dynamic range will of course be increased.
> Try it and see, but don't expect too much.   I've had a serial port 8
> bit 16C71 PIC based converter going for some years (before the
> 56002EVM) , and it did a reasonable job of digging signals out of
> noise in its day, but couldn't match later 16 bit performance. Andy
> G4JNT
>
>      -----Original Message-----
>      From: DL4YHF at aol.com [mailto:DL4YHF at aol.com]
>      Sent: 2002 July 23 19:36
>      To: rsgb_lf_group at blacksheep.org
>      Subject: Re: LF: Sound Cards
>
>      Related topic: I bought some a low-cost PIC with 10-bit ADC
>      (PIC16F870) recently, planning to connect it to the serial
>      interface as replacement for my laptop's audio device (which
>      only has a lousy *mono* MIC input). Has anyone on the group
>      ever tried to use this PIC for weak-signal work and check
>      the ADC's performance, compared to a soundcard ?
>
>
>
>      Regards,
>      Wolf  DL4YHF
>
>
> "The Information contained in this E-Mail and any subsequent
> correspondence"
> "is private and is intended solely for the intended recipient(s)."
> "For those other than the recipient any disclosure, copying,
> distribution, "
> "or any action taken or omitted to be taken in reliance on such
> information is"
> "prohibited and may be unlawful."



-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://atanasoff.rf.org/pipermail/lf/attachments/20020724/683ec0c5/attachment.html