> On Tue, Sep 27, 2011 at 12:46 PM, Leigh L Klotz Jr WA5ZNU > <[email protected]>wrote: > >> Interesting point about TX. Why do you suppose you get worse results? >> On >> my rigs, the TX goes through the widest SSB TX filter installed (2.8 KHz >> for >> example) and th sweet spot of 1500Hz ought to be the best for both TX >> and RX >> as it is the same filter. What filter architecture does you rig use? >> > Crystal filters have both passband ripple (change in attenuation across > the > passband) and change in group delay (not all frequencies in the passband > pass through the filter in the same amount of time). Both of these > characteristics introduce distortion. Generally speaking, the wider the > filter, the less there is of both of these distortions around a narrow > group > of frequencies. > > But there is no guarantee that these distortions are minimized at the > center > of the passband. They *probably* are hence 1500 Hz for a 3kHz-wide filter > is > a good guess, but it is *NOT* guaranteed to produce the best results. > FWIW, > my K2 has its best performance around 800Hz up from the low-frequency > inflection point of the filter, i.e. at around 1kHz audio input. > > I now use Flex radios where the filters are ideally defined in software. > They have very consistent behavior no matter what the width or center is > of > the filter setting and the center of the pass-band is always the best > place > to put the signal. So I no longer worry about this. >
Brian, The question here was about why Dick has found that the best spot for TX and the best spot for RX aren't the same spot. I theorized that perhaps it's not using the same filter for TX and RX. In my rig, it is. In your K2 it's using the wide filter as well. Leigh/WA5ZNU _______________________________________________ fldigi-alpha mailing list [email protected] https://lists.berlios.de/mailman/listinfo/fldigi-alpha
