It's always a good idea to have the selectivity as early as possible in a receiver. At the antenna input would be ideal, but no one yet has invented a truly wide-range tunable filter with the performance of a fixed-frequency filter at a suitable intermediate frequency (I.F.).
That's why Armstrong's superhet design "took off" in the 1920's even though it was so much more expensive to implement than the TRF receivers in use then, and it is still the standard format for virtually all high-performance tunable receivers today. Decent filter performance is available with a tunable filter over a very limited range, but I've yet to see one that matches a really good fixed-frequency filter. Still, it's an acceptable compromise in all but the most severe filtering requirements, it holds down the cost of the receiver and the results are usually better than adding another level of filtering nearer the phones or speaker such as DSP at a second I.F. (or an audio filter in an all-analog receiver). Ron AC7AC -----Original Message----- It appears that the only way the K3 achieves its high test scores in dynamic range is when the noise is outside the roofing filter's bandwidth. Only the roofing filter yields this high performance. For example, if your roofing filter is 2.8Khz, but you hear an unwanted nearby signal, and you use DSP to narrow it to 2.0khz, the reduction of the unwanted signal will not be as great as if you switched to your 2.0khz roofing filter (which you had not purchased yet). This is because the offending signal is within the roofing filter's 2.8khz bandwidth. ...Thus the interest in the "variable bandwidth crystal" roofing filter as a major money saver - if it ever comes to fuition. Howard ______________________________________________________________ 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
