[Elecraft] Filter question

[Tom Hammond]

Hi Stan:

>K2 S/N 4216, assembly completed and operational. I notice that on at
>least one of the filter settings, when I switch from say FL1 to FL2 (in CW 
>mode) I notice about a 10Hz change in the received CW (nor) signal. When I 
>switch to FL3 it then goes back to the originally received tone and stays 
>there through FL4 and back to FL1. CW Rev seems ok...I do not detect a 
>shift in tone on any of the FL settings there. I have listened on LSB and 
>USB (to a CW sig) and have detected the same thing on at least two FL 
>settings. Why would the received tone freq. be shifting? I do not see a 
>change in VFO frequency, just the tone freq itself.

This is an artifact of the limited resolution of the 8-bit DACs used to 
store the BFO frequencies. There's only so much accuracy available there, 
and a 10-20 Hz error can sometimes be noticed. The good thing is that, 
while it IS present, and you will notice it in what tone you hear, it 
generally has NO significant effect upon the received signal itself (e.g. 
it's not going to jump outside the passband of the filter, or even to the 
point that it'll get too near the slope of the filter. It' just a bit of a 
peccadillo.

A number of the fellas have noted that, with a bit of patience, and a bit 
of tweaking the DAC values back' n forth (always returning to the desired 
value, but 'forcing' the K2 to resave the value) they can sometimes 
eliminate the disparity.

>I don't know if this is a problem or not, but when following the
>instructions for aligning the filters with Spectrogram, the center of the 
>filter is at different DAC values for each filter. Is this normal? For 
>instance, if the center DAC value is 140 for FL1, then FL2 center DAC 
>count might be lower, somewhere around 130. It seems to get 
>correspondingly lower as I go to FL3 and FL4. In all cases, though, my CW 
>nor and CW rev BFO frequencies are on the correct side of center...low 
>side of center for CW nor, and high side of center for CW rev. The filters 
>appeared to line up good and I was able to see the 600 Hz tone marker fall 
>right in the center of FL2-FL4. I set FL1 to where it best represented the 
>pictures provided in the literature, based on my interpretation of them.

This is normal. If you will refer to the K2 manual... section entitled 
"Crystal Filters and BFO Settings", probably around page 104 in your 
manual, you'll see a couple illustrations of the BFO frequencies for both 
SSB and CW filters FL1 and FL2. Notice that, even though FL2 is more narrow 
than FL1, it is mostly the UPPER end of the filter which is moved (lowered) 
when the filter is made more narrow. As a result, the center of the filter 
passband WILL change (in a downward direction) as you narrow the passband. 
Therefore, when you are determining the approximate center frequency of 
each filter bandwidth, first select the desired bandwidth, THEN obtain the 
approx. center frequency, and THEN perform the final BFO frequency setting.

The ONLY reason for determining filter center freq. is to ensure that you 
have your BFO set to the proper SIDE of filter center for the mode 
selected. Therefore, it's not terribly crucial that you know the EXACT 
center, just that you're either above it or below it for the mode being 
selected and adjusted.

>I have also adjusted FL1 to tailor my transmit audio somewhat...I
>have moved the LSB BFO frequency down 100 Hz and the USB BFO freq. up 100 
>Hz. I assumed that a shift in one direction for LSB must require an equal 
>shift for USB in the opposite direction to keep the audio the same for 
>both sidebands.

Probably so. Note that the CW filters were really not specifically designed 
to support SSB operation, so they may not require exactly the same 
adjustment... close, but not quite.

73,

Tom Hammond   N0SS