[Elecraft] K3: KRX3 question

[Joe Subich, W4TV]

Bill, 

> David you may be suggesting that "in theory" but I would pay
> *close attention* to what W0YK says.  Ed wins many RTTY contests 
> including several current world records from P49X, and results 
> sometimes speak louder to me than theory (and I'm an engineer 
> if that means anything).

Ed and I had an extensive off-line discussion concerning the K3 
filter options and RTTY.  We both made multiple measurements of 
the composite bandwidth (roofing filter, DSP and dual passband 
filter) with several settings.  The consensus is that the 200 Hz 
DSP settings work as long as the roofing filter is more than 250 
Hz wide.  

> The 200 Hz crystal filter was measured by Elecraft to be 224 Hz 
> wide with a shape factor of 4.0, so whether you use a 400, 250, 
> or 200 Hz roofing filter, that has little bearing on the final 
> BW when WIDTH is set to 200 as Ed suggested (since all XFILs 
> are wider than the DSP's BW).  

This is not the case based on careful measurements.  The problem 
is 'skirt sharpening' when multiple filters are cascaded.   Where 
a single filter might have a 6dB bandwidth of 225 Hz ("200 Hz" 
roofing fitter), when combined with the 200 Hz DSP filter the 
effective -6dB bandwidth of the 200 Hz DSP filter and the 200 Hz 
roofing filter in cascade is 150 Hz!  

Much of the narrowing is due to the very "humped" (Gaussian) 
nature of the narrow crystal filters but the best responses 
for RTTY seem to be with a roofing filter wider than 300 Hz 
(the "250" Hz/8-pole) and/or keeping the DSP filter wider than 
250 Hz with or without the dual passband filter.  Note: Ed uses 
the "250 Hz" filter with his 200 Hz DSP setting. 

73, 

   ... Joe, W4TV 
 




> -----Original Message-----
> From: elecraft-bounces at mailman.qth.net
> [mailto:elecraft-bounces at mailman.qth.net] On Behalf Of Bill W4ZV
> Sent: Monday, July 07, 2008 7:05 AM
> To: elecraft at mailman.qth.net
> Subject: Re: [Elecraft] K3: KRX3 question
> 
> 
> 
> 
> 
> David Woolley (E.L) wrote:
> > 
> > Ed Muns wrote:
> > 
> >> What DSP bandwidths were you using and where did you have your
> >> crystal filters engaging?  There is only a 65 Hz bandwidth 
> difference
> >> between the "400" and "250" crystal filters and both are
> wider than
> >> the what is needed for 170 Hz shift RTTY.  The crystal filter
> >> function is to protect the DSP
> > 
> > For clarity, 170Hz is less than the minimum required
> bandwidth for RTTY.
> >   That needs to be extended by some multiple (> 1.0) of the
> baud rate,
> > to avoid excessive truncation of significant sidebands.  I
> would suggest
> > that a filter that was significantly down at 250 Hz would
> be introducing
> > significant distortion.
> > 
> 
> David you may be suggesting that "in theory" but I would pay *close
> attention* to what W0YK says.  Ed wins many RTTY contests
> including several
> current world records from P49X, and results  sometimes speak 
> louder to me
> than theory (and I'm an engineer if that means anything).
> 
> http://www.radio-sport.net/wpxrtty08_post1.htm
> 
> FYI, the DSP filter in the K3 follows the following formula
> for shape factor
> according to Lyle KK7P:
> 
> Shape Factor = (6 dB BW) / (6 dB BW + 300 Hz)
> 
> At 200 Hz BW, the DSP shape factor computes to 2.5.  The 200
> Hz crystal
> filter was measured by Elecraft to be 224 Hz wide with a 
> shape factor of
> 4.0, so whether you use a 400, 250, or 200 Hz roofing filter, that has
> little bearing on the final BW when WIDTH is set to 200 as Ed 
> suggested
> (since all XFILs are wider than the DSP's BW).  
> 
> Many folks seem to have problems confusing the role of
> roofing filters for
> determining the final bandwidth in the K3.  That is simply 
> not the case. 
> The purpose of the roofing filter is mainly to reduce strong 
> (approximately
> S9+30) signals from over-driving the ADC in the DSP.  If they
> do that, then
> the DSP determines the final selectivity over its ~100 dB
> dynamic range.
> 
> 73,  Bill
> --
> View this message in context: 
> http://www.nabble.com/K3%3A-KRX3-question-tp18280281p18314270.html
> Sent from the Elecraft mailing list archive at Nabble.com.
> 
> _______________________________________________
> Elecraft mailing list
> Post to: Elecraft at mailman.qth.net
> 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