[Collins] KWM-2A help (long)

[Al Parker]

Hi agn, Jerry et al,
	Well, still no joy.  My xtals are dated 8/74, and the skem I'm using is 
from a Nov. 60 manual.  Both the xtal osc., V13A & cath follower, V2A, 
are 6U8A's, on the skem & on the chassis.  I forgot to mention ystdy 
that I'd also re-tightened all ground screws, some were pretty loose.  I 
tried a different 6U8A for the osc., and a 6EA8, no dif.
	C65, the 14mc trimmer in the xtal osc. plate peaks at abt 2/3 
capacitance only.  As Jerry analyzed, that ckt won't resonate at the LO 
~3mc signal.  I do need to look into L5 & the possibility of a missing 
C26 or problem in that "balancing" ckt because it seems like there's an 
awfully big 3mc component in the output of the mixer.  But that may not 
be a causing the problem, because---
	I had earlier been looking at the signal at the grid of the 6CL6 driver 
tube, V8, being 9v p-p @ 7.1mc, & only 2v @ 14.1.  I looked back to the 
plate of the previous tube, V7 RF amp, before the C44 coupling cap, and 
find 14v & 10v respectively.  There's much more loss at 14 than at 7. 
That made me suspect the driver tube, possibly bad with more grid to 
cathode capacitance, just plain leaky, etc.  Swapped in a new 6CL6, no 
noticeable difference.  Now maybe I need to look at the tuned ckt there, 
but like C65 at the osc., C115 peaks OK, and the signal at this point is 
pure.
	I may get back into it tonite, but have another job to do in the AM, so 
it might be later tmw b4 I get at it.  One thing I do plan to do is cut 
the plate & screen voltages to the finals, and measure actual signal at 
the final grids.  It's not enuf now on 14mc to make any grid current.
	I feel close to the problem.  I have another driver 6Cl6, maybe try 
that.  The fact that all bands above 7mc are very low points away from 
individual band tuned ckts.
	Still need help, thanks for all the thoughts & suggestions.
73,

Al, W8UT
www.boatanchors.org
www.hammarlund.info

"There is nothing -- absolutely nothing -- half so much
worth doing as simply messing about in boats"
Ratty, to Mole

On 9/19/2010 2:27 PM, Dr. Gerald N. Johnson wrote:
> I'm working from the last Rockwell manual, stored on line at CCA.org. In
> the extensive list of revisions, it hints some vintages used the 6U8 for
> the crystal oscillator, then maybe 6U8A.
>
> Now I've also found my yellow book with a 1959 version of the schematic.
> I sure do wish someone would have compared schematic to radio and to
> production drawings. This vintage schematic shows V12A calibrator is 1/2
> a 6U8A, but V12B, cathode follower is a 1/2 a 6U8.
>
> The bandswitch circuits are all wrong. They show a rotor tab that
> contacts only one pole at a time and I'm sure the tab is wider
> contacting two so those open poles aren't really open. Then in the 1959
> schematic of the crystal oscillator on 20m it shows an 8-50 (C65) by
> itself and then a 47 fixed (C66) with the wires crossed on an adjacent
> switch pole. THEY SHOULD BE SHOWN CONNECTED with a dot. The last
> schematic was redrawn and neatened making it worse. I'm pretty sure the
> 8-50 won't tune down to 3 MHz because for 3.5 with the slug way deeper
> in the coil it took a 240 fixed in parallel with the 8-50.
>
> The transmit second mixer is kind of balanced for the LO from LO signal
> going to the grid of one triode and the cathode of the other. The 3 MHz
> signal is neutralized by the capacitive unbalance of the two capacitors
> inside L6 and the 6 pf capacitor C26 from mixer output to the bottom of
> that coil. If C26 is missing there would be more 3 MHz on the output.
>
> If L5 has been replaced with a choke having a series resonance in the 14
> MHz range it would kill off the 14 MHz output.
>
> These 8-50s have been known to go open. Its where the shaft hits the y
> shaped conductor on the back. DeoxIT and exercise is a solution,
> providing they will still turn. Sometimes they lock up and there is no
> modern replacement that will stand the RF voltage and current.
>
> We don't know the vintage of your M-2, but with 6U8 crystal oscillator
> its older rather than newer. Can't go by the serial number, they were
> issued in random order (by decree of Art Collins), but you can get an
> idea by the date code on the crystals and the mechanical filter and
> maybe the electrolytics if they can be decoded. Early M-2 used a metal
> cased (Y) filter, later used the plastic cased (FA) filter with a
> different mounting under the chassis.
>
>
> 73, Jerry, K0CQ, Technical Adviser to the Collins Radio Association.
>
snip