[Elecraft] K3-Measuring Filter Offset

[Don Wilhelm]

Mike,

Since the goal is to make the 2 filters agree with each other, I would 
suggest using stereo input to Spectrogram - the main RX on the left 
channel and the sub RX on the right.  It should be visually apparent 
when the two filter passbands are overlapping each other to the greatest 
degree possible.

A noise generator on the antenna input with the main and sub both 
sharing the same antenna would be more helpful than attempting to use 
'band noise', but if no noise generator is available, band noise should 
still be informative.

I have not tried this (no subRX here), but I do believe it will work 
fine - Spectrogram will display both channels at once.

73,
Don W3FPR

Mike Scott wrote:
>> It's not unusual for the 8-pole filters to have small offsets, which would
>>     
> be particularly noticeable if on the bass side of the >passband.  I've heard
> of up to 80 Hz (0.08) in some CW filters.  Anything large than that,
> indicates something else may be going >on (like accidentally resetting the
> 5-pole offset to zero as K7SV noted in the original post).
>
>  
>
> Could someone describe a procedure to measure filter offset. I have
> spectragram (can't remember the real name) software on my PC.
>
>  
>
> I have a five pole 2.7 KHz in each of main and subreceiver. They are matched
> to -0.87 but I would like to measure and split the difference if there is
> one. I do notice that the main rx filter has enough of a passband slope that
> USB and LSB frequency response is notably different. I can hear a
> significant change to background noise color when I change modes. For this I
> am hoping to have mode-specific equalization some day. It is also possible
> that a different offset might improve what I am hearing.
>
>  
>
> Mike Scott
>
> AE6WA Tarzana, CA
>
> K3/100 SN508
>   
>