> Message: 3 > Date: Tue, 30 Jun 2009 00:20:50 +0200 > From: Alberto di Bene <[email protected]> > Subject: Re: [time-nuts] GPSDO project (SDR) > To: Discussion of precise time and frequency measurement > <[email protected]> > Message-ID: <[email protected]> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > Dave Baxter wrote: > > > > Personally, I wouldn't bother with any high accuracy > clock/oscillator > > for the SDR, unless you can also synch the AD sampling clock in the > > soundcard. > > The SDR-IQ, SDR-14, Qs1r, Perseus and the HPSDR project do > not rely on a sound card ADC to sample any signals..... > The RF signal is directly sampled at RF, with sampling rates > up to 122.88 MHz (depending on the model), then it is > frequency-shifted, then downsampled by the firmware in the > accompanying FPGA. Then the downsampled stream is sent to the > PC via the USB 2.0 port. The software on the PC receives > directly digital data, no pesky sound card is needed to digitize it. > > So the _only_ factor affecting the frequency calibration is > the clock of the RF ADC. > > 73 Alberto I2PHD
In that case, it's the RF ADC clock that needs locking. No doubt that is controled by the 66.666MHz reference much discussed? (I do not have schematics / info on that device.) Either way, I stil doubt GPS locking it would be of much great benefit. With the exception of checking the accuracy of some Standard Time and Frequency transmissions, I can't think of any real reason to do it, other than "because we can"? It's also that much more stuff to potentialy go wrong at a moments notice. Thinking of the KISS principle, not exactly in the HPSDR line of things. Mind you, you should see my shack, and the over complicated system I've created to monitor HF beacons! ;-) Currently down due to an overheating CPU in one machine. 73.. Dave G0WBX. PS: I think I've got the NTP timekeeping machine configured sorted and running, now waiting for the replacement 1PPS sourcing GPS rx to arrive. _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
