[Elecraft] RX Test (long)

[Joe Subich, W4TV]

> The K3 is very much quieter out of the passband than other 
> receivers (the K2 is close) - the response is down at least 
> 60 dB as opposed to the out of passband 'garbage' in other 
> receivers only about 30 dB down.

Here is the genesis of the complaint that the K3 is more 
fatiguing than other receivers.  The out of band noise 
(noise above 5 KHz) in the K3 is, as you say, at least 60 dB 
below the peak audio level (actually, I measure it at more 
than 90dB down above 5 KHz).  However, the stopband also has 
many leakage products (DAC artifacts) in the stopband some 
of which are as much as 50 dB ABOVE the noise floor.  

While these products are well below the desired audio in the 
passband, they are "in the clear" as far a local noise.  
This requires the ear/brain system to subconsciously work 
to reject (filter) the undesired discrete frequency products. 

With other receivers (e.g., the FT-1000MP MKV) any undesired 
discrete frequency products (and much of the general audio 
IMD) is buried (or masked) by the slow roll off of the IF 
noise which is only 50-60 dB below peak audio response well 
into the 10 KHz range.  With the masking noise, the ear/brain 
system does not need to be engaged to reject the other "signals" 
even though the overall noise level is higher.  

I can see the other "noise" effect using an audio spectrum 
analyzer but I have not been able to determine its cause - 
I suspect it may be some form of DSP related aliasing or 
overflow.  In any case it is quite visible in CW (with a 
narrow filter) and responds dramatically to reductions in 
the RF gain.  In essence, I can reduce the stopband noise 
in the 900 - 2000 Hz range significantly without reducing 
the peak audio level of the desired CW signal simply by 
reducing the RF Gain (or engaging the attenuator).  Since 
engaging the attenuator will drop the noise by more than 10 
dB, my suspicion is that it is some form of IMD but until 
I receive my lowpass filters and finish all of the audio 
updates (with appropriate before/after documentation) I will 
not speculate as to the source. 

73, 

   ... Joe, W4TV 
 


 





> -----Original Message-----
> From: elecraft-bounces at mailman.qth.net 
> [mailto:elecraft-bounces at mailman.qth.net] On Behalf Of Don Wilhelm
> Sent: Friday, January 08, 2010 9:06 PM
> To: Al Lorona
> Cc: Elecraft Reflector
> Subject: Re: [Elecraft] RX Test (long)
> 
> 
> Al,
> 
>  From the tests I have done, I discovered that the K3 has a flat 
> passband and very sharp filter skirts - in fact, very close to the 
> "ideal" filter sought by many in past years.  Other receivers tested 
> (K2, Yaesu FT-900, FT-817) showed that their passband was not 
> flat, but 
> tapered smoothly to a lower response at higher audio frequencies.
> 
> The K3 is very much quieter out of the passband than other receivers 
> (the K2 is close) - the response is down at least 60 dB as opposed to 
> the out of passband 'garbage' in other receivers only about 
> 30 dB down.
> 
> So my supposition is that we are not accustomed to hearing an "ideal" 
> filter response - we are accustomed to hearing that tapered response, 
> and as a result the K3 is judged to be "noisy", when in fact, 
> it is very 
> quiet for out of passband response and is faithfully reproducing the 
> in-passband response that is actually present.  We may not 
> like to hear 
> what is actually present, but in my mind, the K3 is faithful in 
> re-producing it without distortion - that in my mind is the 
> definition 
> of a High Fidelity receiver.
> 
> 73,
> Don W3FPR
>