[Elecraft] K2 - VCO-Test (Alignment an d Test, Part II)

[Masleid, Michael A.]

Hi David,

Well, one good thing:  The K2 seems to be worth the effort once
you figure out how to work it.  So here we go again:

>>Put a 2.2 kOhm resistor across terminal 1 and 2 of T5 as a test.

>With the 2.2kOhm resistor the frequency is stabil at 8.9MHz.
>But I have the same effect when I tip with one end of the resitor
>at terminal 2....

In this case I'm interested in how it works with the 2.2 k resistor
trimmed and flat down against the board, over the ground plane as much
as possible.  I want to avoid any capacitive coupling from terminal 2
over to the gate of Q18.  This is see if Q18 is working too good, or
C68 is too large.

>>Or the other possibility:

>>Put 5 pF in parallel with C68 as a test.  This should increase feed =
back.
>>If that makes things better, then C68 is not enough feedback,
>>so perhaps Q18 is sick.

>I have no 5pF so I took 2.2pF -> the counter showed 7.8MHz but not =
stabil
>(+/-50kHz)
>With 12pF it's 8.9MHz and stabil!

Well, OK, so 5 pF was a shot in the dark.  I'll take 12 pF as just as =
good.
This seems to imply that C68 is open (not too likely), or Q18 is weak.
Given all the measurements and tests that you've done on bias for Q18, =
and
given the current that should flow through Q18, that is not flowing =
through
Q18, I think it is time to get a new Q18!

(As a rather strange fix, if you can't get a new Q18 right away, you =
might
try flipping Q18 around, so the Drain is where the Source should be, the
Source is where the Drain should be, and the Gate stays where it is.  =
But
don't do it if there's the smallest chance of hurting the circuit =
board.)=20

Some other details:  Yes, these are DC measurements, not RF probe
measurements.

The added capacitance at C68 will probably increase phase noise on the =
VCO
a bit, so it is not a good long term fix.

73, Michael AB9GV