[Elecraft] K2: Part II 40M checkout - ground short, or?

[Ward Willats]

I am building on K2 #3587. Thanks for everyone's help with my 
previous screw-up of R38....

Good news: got thru the Part II checkout. Values OK, oscillators 
work, signals peak and so forth. Heard faint stuff on 40m. Thought 
everything was great....

BUT, ran into the same problem as three previous folks in the 
archive, in that touching around C6 (the little 4.7pf cap that 
connects the two tanks in the 40M BPF) would bring the volume up 
_dramatically_ -- so even though everything "worked," it was not 
working right. :-(. I have the dreaded "hard of hearing" syndrome. 
Gasp!

I've changed C6, peaked up L1 and L2 "by ear" and on the scope -- all 
seems good -- but I no longer think the problem is in the band-pass, 
because...

While messing with the antenna coax I noticed that when the shield is 
NOT connected to the rig -- just the center conductor -- it works 
great too. The second the braid/shield makes contact, sigs go way 
down. (I have a BNC to PL adapter that lets me easily do this on the 
PL side -- but I also verified the behavior using jumpers on the back 
side of the K2 antenna jack...)

"Cheap and dirty" signal injection, the way I've been doing it, just 
uses the center conductor of the antenna coax (a windom) -- and that 
works loud and great everywhere I inject (from the mixer out).

I can also make everything loud by touching a screwdriver or 
soldering iron to (almost any) point in the front-end path thru the 
BPF.

There is some big clue here I am not seeing.

RIght now it seems like a bigger ground plane makes signals go down, 
a bigger RF plane makes sigs go up.

Guess I'm thinking there is some (partial or full) RF short to ground 
-- perhaps thru a cap or something, but just a guess...

Open to your thoughts and any techniques for hunting things down. I 
haven't poked around much on the low-pass filter yet, so I think I'll 
play with that next.

This RF stuff sure is slippery!

Thanks,

-- Ward / KG6HAF