开云体育

Normally yes, to sample 48KHz signals you need 96KHz sampling rate, but the I and Q are quadrature signals and are only at 24 KHz each.

7036KHz 20 KHz signal
7046KHz 10 KHz signal
7056KHz 0 KHz Signal
7066KHz 10 KHz Signal
7076KHz 20 KHz Signal

Normally you can see that a signal at 7036 and 7076 each give a 20KHz signal so you have a image problem. But because the signals are in quadrature, one signal lags the other if your below the sampling signal, and it leads if it's above the sampling frequency. The computer program looks at both signals and figures where each signal is in frequency so it can tell the lower image from the upper image.

But because each signal is only 24 KHz then 48KHz sampling does fine with each of I and Q signals.

If you sampled at 96 KHz and the integrating capacitor was not too big then you would be able to look a a 96 KHz band of signals, 48KHz below the quadrature clock and 48KHz above. You would need to modify the integrating capacitor and analog filters to allow a 48 KHz signal without attenuation.

Now a lot of sound cards can sample at 96Khz but that doesn't mean that their amplifiers are capable of passing a 48KHz signal faithfully, you need to be careful in picking your audio card.

Around 7056 you end up with a lot of garbage, because any hum and harmonics show up in that area, and if your signals are not well balanced you can end up picking some of the clock signals making it look like there is a carrier there. Good shielding and elimination of ground loops can help.

Hope that helps.

--
Cecil
KD5NWA
<www.qrpradio.com>

I fail to see why doing the same thing over and over and getting the same results every time is insanity: I've almost proved it isn't; only a few more tests now and I'm sure results will differ this time ...