FILTER SETTINGS
Locked
 
|
W4TV: >However, the 250 Hz filter is actually 370 Hz at the - 6dB points according to previous e-mail here. >With the K3, the point of diminishing returns for IMD (and close in rejection) is perhaps - 30dB instead of -60dB or more in a conventional crystal filter only radio. >If we use a straight line (linear slope) from the -6dB to -60dB points the 200 and 250 Hz filters look like this: 200 250 ------------------------------------------------- - 6dB 224 Hz 370 Hz -60dB 896 Hz 777 Hz slope 6.22 Hz/dB 3.77 Hz/dB (one side) -10dB 274 Hz 400 Hz -20dB 398 Hz 475 Hz -30dB 522 Hz 550 Hz -40dB 647 Hz 626 Hz -50dB 771 Hz 702 Hz >By this simple calculation (or by plotting on graph paper) one can see that the 200 Hz filter should do a BETTER job as the most "narrow" filter. It remains "tighter" to about -35 dB and the five pole design should result in less pulse stretching (ringing in the presence of static) than the 250 (really 350 Hz) filter. Excellent post Joe (as usual). Another reason one might want the 200 versus the 250 is the relative difference between the 400 and 500 filters. From the published data: Filter BW(-6dB) Shape Factor 200 224 4.0 250 370 2.1 400 435 2.1 500 565 3.1 Practically there is little BW difference in the 250 and 400 8-pole filters. If one wanted both a very narrow and moderately narrow filter, I would choose the 500 and 200 5-pole filters. They have at least an octave difference in total bandwidth...plus the group delay issues you cited in your last paragraph above. BTW I see very little insertion loss in either of the 200 Hz or 500 Hz filters. Using an XG1 source, I find 1 dB compensation is all that is needed for both. 73, Bill W4ZV P.S. I retract my engineers vs marketeers statement. The marketeers won in naming the filters, especially calling the 370 Hz filter above a "250 Hz"! :-) _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
Bill, > >By this simple calculation (or by plotting on graph paper) one > can see that the 200 Hz filter should do a BETTER job as the > most "narrow" filter. It remains "tighter" to about -35 dB > and the five pole design should result in less pulse stretching > (ringing in the presence of static) than the 250 (really 350 Hz) > filter. > > Excellent post Joe (as usual). Another reason one > might want the 200 versus the 250 is the relative difference > between the 400 and 500 filters. From the published data: > > Filter BW(-6dB) Shape Factor > 200 224 4.0 > 250 370 2.1 > 400 435 2.1 > 500 565 3.1 > > Practically there is little BW difference in the 250 and > 400 8-pole filters. If one wanted both a very narrow > and moderately narrow filter, I would choose the 500 > and 200 5-pole filters. They have at least an octave > difference in total bandwidth...plus the group delay > issues you cited in your last paragraph above. Expanding the 200/250 analysis to 200/250/400/500 as I did privately for someone else: 200 250 400 500 ------------------------------------------------- - 6dB 224 370 435 565 Hz -60dB 896 777 913 1751 Hz slope 6.22 3.77 4.43 10.98 Hz/dB -10dB 274 400 470 653 Hz -20dB 398 475 559 873 Hz -30dB 522 550 647 1092 Hz -40dB 647 626 736 1312 Hz -50dB 771 702 825 1531 Hz I really think the 200/500 pair is the best value. The only possible benefit of the 400 over the 500 would be an improvement of 500 Hz spaced IMD. However, as you have pointed out, with strong interfering signals at 500Hz and 1 KHz, the problem is far more likely to be radiated phase noise and clicks from those signals than receiver generated IMD. I'm looking forward to information on the variable CW filter, something like a 400 Hz to 1 KHz range would be ideal in place of either the 400 or 500 Hz filter. 73, ... Joe, W4TV _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
I should have mentioned that the 500 Hz Joe measured below was the Elecraft 5-pole...not the Inrad 8-pole.
73, Bill
|
|
|
Bill, The 500 Hz figures were an older set provided by someone else (I can't recall now - after more than three years). My subsequent measurements of four 500 Hz, 5-pole filters show slightly better performance. Those measurements average about 490 Hz at -6dB and around 1500 Hz at -60 dB. The difference is somewhat better performance for the 500 Hz, 5 pole filter ... around 800 to 850 Hz at -30 dB but the overall trends (including the lack of a significant difference between the "250 Hz" and 400 Hz filters) remain intact. 73, ... Joe, W4TV On 2/8/2011 9:23 PM, Bill W4ZV wrote: > > I should have mentioned that the 500 Hz Joe measured below was the Elecraft > 5-pole...not the Inrad 8-pole. > > 73, Bill > > > Joe Subich, W4TV wrote: >> >> >> 200 250 400 500 >> ------------------------------------------------- >> - 6dB 224 370 435 565 Hz >> -60dB 896 777 913 1751 Hz >> slope 6.22 3.77 4.43 10.98 Hz/dB >> >> -10dB 274 400 470 653 Hz >> -20dB 398 475 559 873 Hz >> -30dB 522 550 647 1092 Hz >> -40dB 647 626 736 1312 Hz >> -50dB 771 702 825 1531 Hz >> > Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
«
Return to [K3]
|
1 view|%1 views
| Free forum by Nabble | Edit this page |