K3 DSP shape factor?
Locked
 
|
There have been a few postings over the last two days
complaining of ssb splatter on receive with the K3 during the WPXSSB contest. I operated during that contest with my TS-850's and found generally that the bands were extremely crowded, and there were definitely some stations with broad signals. However, I am particularly concerned about the comments from the ham who said he preferred his TS-930 over the K3 on SSB. The question I have is "What is the effective shape factor of the K3 DSP?" Is it much steeper than the 1.5 - 2.0 shape factors of the roofing filters, such that DSP attenuation predominates? Or, is it less than 2:1, and off-frequency, the roofing filter predominates? Maybe the DSP shape factor varies as a function of bandwidth? I've done a Nabble search and searched the Elecraft Web site, but didn't find an answer. Maybe some smart person can fill me in. Thanks! Dave Hachadorian, K6LL Yuma, AZ . _______________________________________________ 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 |
|
|
> The question I have is "What is the effective shape factor of the K3
> DSP?" DSP filters usually are so-called Finite Impulse Response (FIR). FIR filters for a given complexity (length) have a nearly constant transition region width going from the passband to the stopband. For this reason, "shape factor" varies depending on the width of the filter. ( It also depends on the ultimate attenuation required, and the "windowing function" applied to deal with of the "Gibb's phenomenon" and...) The following is just an illustration. Let's say that the transition band for a set of FIR filters from 6 dB to 60 dB is 150 Hz. [ This number happens to be from a quick check I just did on a dialed-in filter width of 500 Hz, and a center of 600 Hz on my K3. This is using a 2.8 kHz roofing filter, AGC OFF, internal AFV and dBV function, and signal level at the threshold of hardware AGC action. Yes, I tuned to the higher pitch side to make the measurement :-) How to do all of this is left as an exercise for the reader. ] If the 6 dB width is 500 Hz, then the 60 dB width is (150+500+150 =) 800 Hz and the "shape factor" is 1.6:1. If the 6 dB width is 2400 Hz, then the 60 dB width is (150+2400+150 =) 2700 Hz and the "shape factor" is 1.125:1. YMMV, and all is subject to change at the whim of the firmware writers :-) 73, Lyle KK7P _______________________________________________ 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 |
|
|
In reply to this post by Dave Hachadorian
Dave,
I wonder if the problem had anything to do with the AGC SLP? Mine was at 10. Very strong signals seemed to cause some high pitch distortion. I changed it to 12. It seemed to help. I hear some have gone to 15. But I thought that was flat topping the AGC too much. N2TK, Tony -----Original Message----- From: [hidden email] [mailto:[hidden email]] On Behalf Of Dave Hachadorian Sent: Tuesday, April 01, 2008 2:09 PM To: Elecraft Reflector Subject: [Elecraft] K3 DSP shape factor? There have been a few postings over the last two days complaining of ssb splatter on receive with the K3 during the WPXSSB contest. I operated during that contest with my TS-850's and found generally that the bands were extremely crowded, and there were definitely some stations with broad signals. However, I am particularly concerned about the comments from the ham who said he preferred his TS-930 over the K3 on SSB. The question I have is "What is the effective shape factor of the K3 DSP?" Is it much steeper than the 1.5 - 2.0 shape factors of the roofing filters, such that DSP attenuation predominates? Or, is it less than 2:1, and off-frequency, the roofing filter predominates? Maybe the DSP shape factor varies as a function of bandwidth? I've done a Nabble search and searched the Elecraft Web site, but didn't find an answer. Maybe some smart person can fill me in. Thanks! Dave Hachadorian, K6LL Yuma, AZ . _______________________________________________ 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 _______________________________________________ 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 |
|
|
In reply to this post by KK7P
So for the DSP:
BW60 = BW6 + 300Hz (approx) Sounds like good numbers to me. Thanks for the answer! Dave Hachadorian, K6LL Yuma, AZ ----- Original Message ----- From: "Lyle Johnson" <[hidden email]> To: "Elecraft Reflector" <[hidden email]> Sent: Tuesday, April 01, 2008 7:10 PM Subject: Re: [Elecraft] K3 DSP shape factor? >> The question I have is "What is the effective shape >> factor of the K3 DSP?" > > DSP filters usually are so-called Finite Impulse Response > (FIR). FIR filters for a given complexity (length) have a > nearly constant transition region width going from the > passband to the stopband. For this reason, "shape factor" > varies depending on the width of the filter. ( It also > depends on the ultimate attenuation required, and the > "windowing function" applied to deal with of the "Gibb's > phenomenon" and...) > > The following is just an illustration. > > Let's say that the transition band for a set of FIR > filters from 6 dB to 60 dB is 150 Hz. [ This number > happens to be from a quick check I just did on a dialed-in > filter width of 500 Hz, and a center of 600 Hz on my K3. > This is using a 2.8 kHz roofing filter, AGC OFF, internal > AFV and dBV function, and signal level at the threshold of > hardware AGC action. Yes, I tuned to the higher pitch side > to make the measurement :-) How to do all of this is left > as an exercise for the reader. ] > > If the 6 dB width is 500 Hz, then the 60 dB width is > (150+500+150 =) 800 Hz and the "shape factor" is 1.6:1. > > If the 6 dB width is 2400 Hz, then the 60 dB width is > (150+2400+150 =) 2700 Hz and the "shape factor" is > 1.125:1. > > YMMV, and all is subject to change at the whim of the > firmware writers :-) > > 73, > > Lyle KK7P _______________________________________________ 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 |
|
|
In reply to this post by KK7P
Lyle Johnson wrote:
> <div class="moz-text-flowed" style="font-family: -moz-fixed">> The > question I have is "What is the effective shape factor of the K3 >> DSP?" > > DSP filters usually are so-called Finite Impulse Response (FIR). FIR I seem to remember that, originally, all the K3 IF filters were IIR (infinite impulse response) filters. Subsequently FIR filters were provided, as a user choice, for one end of the bandwidth range. IIR filters give a better shape factor for a given amount of processing power, but are prone to ringing (that's basically want the infinite response means) and don't have constant group delays (whereas symmetric FIR ones do). -- David Woolley "The Elecraft list is a forum for the discussion of topics related to Elecraft products and more general topics related ham radio" List Guidelines <http://www.elecraft.com/elecraft_list_guidelines.htm> _______________________________________________ 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 |
|
|
G'morning David:
Check the firmware release notes for v1.58. 73, Ken K3IU ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ David Woolley (E.L) wrote: > Lyle Johnson wrote: >> <div class="moz-text-flowed" style="font-family: -moz-fixed">> The >> question I have is "What is the effective shape factor of the K3 >>> DSP?" >> >> DSP filters usually are so-called Finite Impulse Response (FIR). FIR > > I seem to remember that, originally, all the K3 IF filters were IIR > (infinite impulse response) filters. Subsequently FIR filters were > provided, as a user choice, for one end of the bandwidth range. > > IIR filters give a better shape factor for a given amount of > processing power, but are prone to ringing (that's basically want the > infinite response means) and don't have constant group delays (whereas > symmetric FIR ones do). > 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 |
«
Return to Elecraft
|
1 view|%1 views
| Free forum by Nabble | Edit this page |