[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [linrad] Silly question or is it?
Thats not a silly question at all.
I would like to understand this detector better myself.
The 48 Khz spike is probably either 50 Hz hum or DC (carrier?).
It would be interesting to know.
I understand that there is a 20 hz wide null in the response in the middle of
the screen due to lack of sound card frequency response below 20 hz.
The detector is a double sideband detector, so the highest frequency actually
present is 48 Khz, so the sound card is not really violated in any way by seeming
to produce 96khz. Where the magic lies is cancelling the unwanted sideband (or image
frequency), in effect. This is done using phase cancellation. You are using the 48 khz
bandwidth twice. I am not sure if this answer is entirely accurate, but I think it qualifies as
simple. My apologies if it is not quite accurate.
73, Jim.
On 12 Jan 2003 at 19:39, Conrad G0RUZ wrote:
> I am about to show my ignorance but there is something I don't
> understand fully.
>
> I have been thinking about the IQ receiving scheme. I am not sure that
> I have grasped exactly what is going on. If I understand correctly the
> Nyquist theory says that the frequencies reproduced accurately are a
> little below 1/2 the sampling rate. So for a 96kHz sampling rate the
> maximum accurately reproduced frequency will be around 46kHz. The anti
> alias filters in the Delta 44 are presumably set a little below 48kHz.
> This explains how, when sampling at 96kHz, if you don't choose IQ
> mixing in Linrad you only get a 48kHz wide display on the waterfall
> graph. When you go to IQ this gets twice as wide, I am not absolutely
> sure why this is and I would like a simple explanation if possible.
>
> Also my uncalibrated system at the moment shows a big nasty spur at
> 48khz or so, does the calibration routine remove it? If so how?
>
> Sorry for the simple questions but I would really like to know what
> exactly is going on. I think that there are many who don't understand
> this.
>
> 73
>
> Conrad G0RUZ
>
>
LINRADDARNIL