On Oct 30, 2011, at 8:01 AM, Matt Zilmer wrote: > Hi Bob, > > I know for certain the the actual VFO shift is not reflected in the > REF*CAL value. Otherwise, we'd all be seeing massive frequency error > on RX and TX. The reference frequency is used in the DDS to derive > the injection freq, so the error should be reduced substantially by > that alone. The ref freq is also divided down by 6 to get the 8.23MHz > 2nd LO injection freq. If Wayne's design followed others (Motorola > for one), the shift in LO and first IF would buck (cancel) each other > to some extent.
That's correct. We don't need REF CAL to be exactly 49,380,000 Hz because we take the actual frequency of the reference into account in firmware when setting up the synthesizer. The 2nd LO is compensated at the same time, at 1/6th the rate (because of the /6 relationship between that and the 1st LO). If there's a K3EXREF installed, its microcontroller measures the frequency of the 49 MHz reference down to 1 Hz using the 10 MHz external reference as a timebase. This data is sent to the K3's main MCU, which then uses it to "capture" the normal REF CAL value. One advantage of this method (frequency-locked loop) is that no matter how noisy the external reference is--possibly due to being distributed around the shack--it can't have any effect on the radio's RX or TX noise floor. It is used only as a counting gate. 73, Wayne N6KR ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[email protected] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html
