[Milsurplus] BC-9: Answer One Question, Get Another.

[Mike Morrow]

Hue wrote:

> As i recall, keying the set shifted the frequency down
> substantially.

The other bit of "key" (chuckle!) information is whether the shifted frequency remained constant as long as the key was closed,  If what I think (below) is correct, the frequency should remain offset as long as the key is closed.

> The fact that the 500 pF in the 120 volt supply has any
> effect at all on tuning means the filter F is not that
> effective.

That "imperfect" filtering is, I think, THE root source of the observed effects of keying on frequency.

The plate's LC low pass filter consists of two series 23 uH coils  with a 1000 uuF capacitor.  It has a -3 dB cutoff frequency of about 750 Hz.  Filter attenuation at 4000 kHz is likely not much better than 15 dB.  That's not much isolation of the RF signal on the plate from whatever is connected to the other end of the filter, which is:
(1) R5 and C5 and T1 and meter when the key is open,
    or
(2) C6 when the key is closed.

These two different circuit configurations will affect filter characteristics for attentuation and phase shift, which alters the impedance at the plate and changes oscillation frequency.  On key closure, the effect of R5, C5, T1, M is removed (bypassed) and the effect of C6 is substituted.  The trick is to choose a value of C6 that affects the attenuation chacteristics of the filter in a manner similar to what R5, C5, T1, M does when the key is open.  After a proper C6 is installed, little frequency change should take place key-open to key-closed.

The "tuning" of C6 is what Dave is doing now.  Of course, this arrangement is awkward.  It would all be unnecessary if an RF choke of decent value (like the common 2.5 mH RFC) and a .01 uF capacitor were substituted for the original two 23 uH chokes and 1000 mmF capacitor.  Then, nothing on the DC side of the filter would affect the RF side and the plate.  BUT...doing that would be too clever by half.  Some difference between receive and transmit frequency is necessary so that beat notes are heard between stations that just happen to be on the exact same frequency.  It appears that the component values in the plate filter and C6 were chosen with wisdom and purpose.

Anyway, that's what I think is going on.

Mike / KK5F