[Elecraft] K2: Help, I messed up the dial calibration!!!

[Steve Kallal]

Don,

I finally got to within 10 to 15 Hz at 10 MHz now. I decided to use MixW 
for the 500/600 Hz WWV tones. It works really well for that part of the 
calibration.

I had misread your web site. You do endorse the N6KR method. I did find 
the WWV Signal Format web page. Out here on the west coast, WWVH also 
mixes into the MixW waterfall as a weaker trace when only one tone is 
present from WWV.

I found the best tool for C22 tweaking is a small glasses repair kit 
screw driver. My plastic tool could do the real fine adjustments due to 
the tightness of C22. Of course the screw driver is removed before 
taking readings.

The part that wasn't clear to me at first, was the need to redo CAL PLL 
& CAL FIL after EACH C22 adjustment. I read so many comments on the 
reflector archives that mentioned simply listening to WWV and listen for 
zero beat. Obviously that is in the greater context of CAL PLL & CAL FIL 
reruns.

There must to something electrical loose in my K2. I had the CAL FCTR 
TP1 & TP2 values matching once & then they changed a few minutes later.

Using Spectrogram or MixW in LSB/USB modes easier than the CW spot 
method I used. Somewhere I read the CW spot audio signal may not to very 
accurate. Even though I am confident in my tone matching abilities my ear.

Not sure if I could do it in the time it took you to write it up, maybe 
30 to 60 minutes for me! A video on YouTube would be real nice, except 
K2 has given way to the K3 in popularity.

The bottom line: my calibration is twice as accurate (half the delta) 
against WWV, than before, although in the opposite direction.

Thanks for your help!

73,

Steve N6VL
K2/100 #2289

Don Wilhelm wrote:
> Steve,
>
> Yes, the "4 MHz" reference may have to be set a few Hz different than 
> exactly 4.00000 MHz to calibrate the dial accurately.
>
> The best method I know of is to feed the K2 audio into a computer 
> running Spectrogram.  This method compensates out any difference 
> between 4 MHz and the actual required frequency for the reference - 
> one does not even need to be concerned about the reference frequency 
> with this method.
>
> On the Spectrogram line display, set markers (pointers) for 500 and 
> 600 Hz.
> Now tune WWV in LSB (or USB if you prefer) - 10, or 15 MHz, either is 
> OK - 20MHz requires a bit of a difference in the VFO/BFO calculation 
> due the the sum of the oscillators rather than the difference.
>
> When you look at the Spectral line display of the WWV transmission, 
> you will see the carrier (only if you are mistuned), and a transmitted 
> tone (wait until the do transmit a tone).  The tone is normally either 
> 500 or 600 Hz, but there is a 440 Hz tone for one minute each hour and 
> there are some non-tone minutes.  You can look up the transmission 
> format on the web and it will tell you which minute is which tone - or 
> you can just wait and watch the tones alternate.  The short tone 
> transmitted for one second at the beginning of each hour is always1000 
> Hz, so I move the cursor to 1000 Hz (read the frequency in the lower 
> left box of the screen) and click to place a crosshair at 1000 Hz too, 
> it is a handy reference if WWV is weak ir the other tones are not clear.
>
> Now that you have the Spectrogram 'picture' of WWV and have identified 
> the relevant tones in the audio (the carrier, and one of the 
> transmitted tones), tune the VFO first listening until the voice 
> sounds rather clear - then you may see the carrier near zero Hz and 
> the tone near the 500 or 600 Hz marker that you have set.  Tune the 
> VFO until the tone is lined up exactly with the marker.  Wait until 
> the next tone minute to be certain you are tuned correctly and not 100 
> Hz off.
>
> Enter CAL FCTR, then check to see if the tones are still lined up - 
> you can fine tune a bit if necessary, but when you have the VFO set 
> correctly (for the tones, not the dial reading - the dial reading will 
> not be correct - that is why you are going to all this trouble).  Do 
> not move the VFO after this, you will next be adjusting C22.
> Now, you move the probe between TP1 and TP2 until the frequency 
> difference is exactly equal to the WWV transmitted frequency.  Since 
> WWV transmitts on 0000.000 kHz boundaries, just look for the low order 
> digits to be the same at TP1 and TP2.  If the digits are not the same, 
> adjust C22 a bit and see if the frequencies got closer.  Once C22 is 
> adjusted correcctly, the low order digits of the BFO (TP2) and the VFO 
> (TP1) will match exactly, and you can declare that C22 is now set 
> exactly right - do not touch it again.
>
> Now, switch to 40 meters and run CAL PLL followed by CAL FIL (change 
> each BFO setting even if it is correct - moving it up a bit and back 
> down to where it was constitutes a 'change' to the K2.
> Once that has happened, you are done.
> Tune WWV again as before and note the dial setting - you should find 
> it no more than 20 Hz off.  The DAC limit of tuning varies with 
> frequency, but averages about 13 Hz, so 20 Hz is within the expected 
> practical tolerance limits.
>
> It takes a LOT more time to write this than it does to do it - it is 
> easier to do than to say.
>
> 73,
> Don W3FPR