[Elecraft] [K3] Unanswered Questions ...

[Joe Subich, W4TV]

Lyle, 

Thanks for the information ... 

> In FM mode, capture effect helps suppress the effect of "image" 
> leakage.  AM and (E)SSB modes have far more dynamic range and the 
> 15 kHz IF image must be suppressed by some 10s of dB more than is 
> required for excellent FM operation.

Are you speaking of receive performance or transmit performance?  

The conversion "image" either transmit or receive is separated 
from the desired frequency by twice the IF (in this case 15 KHz). 
In this case the "image" would be 30 KHz away - in this case at 
8.245 MHz (or 8.185 MHz depending whether the second LO was above 
or below the IF).  

Since the specifications for the KFL3B-FM are not available, the 
closest filter I can find is the 15 KHz wide, 7.0075 MHz, 6-pole 
INRAD "front end" filter.  That filter shows a estimated shape 
factor of 2.1 (1/60 dB).  Assuming the desired signal is centered 
in the first IF (roofing) filter passband and we were dealing with 
a 6 KHz wide ESSB signal, the closest "image" frequency would be 
27 KHz from the center of the "roofing" filter.  At 27 KHz from 
center, the response of that 15 KHz/6-pole filter is down 80 to 
100 dB depending on the filter's ultimate rejection - remember, 
the filter is 30 KHz wide at the -60 dB point which is 15 KHz 
from the center of the passband.   

If we're looking at an AM signal with 5 KHz audio response, the 
nearest component is 25 KHz from the center of the filter - again 
down far more than 80 dB (and another 6 dB down compared to the 
PEP of the transmitted signal).  

It would seem to me that 70 to 80 dB suppression of the transmit 
image is far more than required by Part 97.307(d) ... and that 
something in the .01 mW PEP range for 1500 W PEP of "desired" 
signal would be more than acceptable.  

Is there something in the digital domain that I'm missing - 
aliasing, artifacts, etc. that effects this?  

Now, on receive 80 or 90 dB may not be sufficient rejection, it 
depends on conditions but still at nearly two times the -60 dB  
point on the filter skirt it would seem that the image response 
would be determined by be more one of circuit leakage and the 
filter's ultimate (stopband rejection) rather than skirt 
selectivity. 

Has the IF image response been checked with an engineering sample 
or prototype of the KFL3B-FM in either receive or transmit? 

73, 

   ... Joe, W4TV 
 

  


> -----Original Message-----
> From: elecraft-bounces at mailman.qth.net 
> [mailto:elecraft-bounces at mailman.qth.net] On Behalf Of Lyle Johnson
> Sent: Wednesday, October 24, 2007 8:01 PM
> To: Elecraft at mailman.qth.net
> Subject: Re: [Elecraft] [K3] Unanswered Questions ...
> 
> 
> Hello Joe!
> 
> Unlike most DSP-based radios, the K3 converts from the main IF (8.215 
> MHz) to the 15 kHz DSP IF without an intervening 455 kHz IF.  
> This helps 
> give the K3 a cleaner sound and better dynamic range since there are 
> fewer conversions.
> 
> However, this also means the IF crystal filter must have exceptional 
> skirt selectivity to prevent the 15 kHz "image" from getting through.
> 
> In FM mode, capture effect helps suppress the effect of 
> "image" leakage. 
>   AM and (E)SSB modes have far more dynamic range and the 15 kHz IF 
> image must be suppressed by some 10s of dB more than is required for 
> excellent FM operation.
> 
> It is for this reason that the FM filter is recommended only 
> for FM use. 
>   It may also be used for AM receive, but is not recommended for AM 
> transmit.  The 6 kHz filter has the necessary 15 kHz IF image 
> rejection 
> to be suitable for use with either AM or (E)SSB.
> 
> 73,
> 
> Lyle KK7P
> 
> _______________________________________________
> 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
> 
>