[R-390] 17 MHz AGC pulling problem

[Larry H]

I was curious about the 17 MH oscillator pull, so I did a simple (but I
believe adequate) test last night for this.  I let my R-390A and sig gen
warm up to the point where both had essentially no drift.  I fed a 10.9 mh
signal with no modulation into the A (above 8 mh so the 17 mh osc would not
be used) and measured the audio beat tone at different input signal levels
from 5 uv to 5,000 uv in steps.  I then did the same test using 3.9 mh, so
the 17 mh osc would be used.  Here's a table showing my results:

uv bal input:           5        50        500      5,000
agc volts:               1.2     4.2       6.9       8.8
c/l db reading:        10      35        58       70
10.9 mh bfo tone:   60      58       56        54
3.9 mh bfo tone:     50      54       59        62

As can be seen, the 10.9 mh test shows a 6 cycle downward 'pull' and the
3.9 mh test shows a 12 cycle upward 'pull'.  I'm not sure how Chuck Felton
obtained his numbers, but I don't think what I see would cause me to try to
fix this problem.

Regards, Larry



On Sat, Apr 6, 2019 at 10:51 AM Perry Sandeen via R-390 <
r-390 at mailman.qth.net> wrote:

> Yo Bubba Dudes!
> A challenging problem for solving by those with cold logic and a hot
> soldering iron.
>
> One of the puzzling mysteries of the a is the 17 crystal oscillator drift
> caused by the application of AGC to the first mixer. This can be 300 - 450
> Hz from the findings of Chuck Felton.
> Collins went to the trouble of putting the crystal in an oven so why not
> solve this problem.  Best guess.  They knew about it BUT it wasn't in the
> required stringent spec's so engineering, knowing the cost of solving the
> problem, just kept quiet.
> Now for causal AM listening Chuck noted that it probably wouldn't make any
> difference.  However e are interested in listening to the continuous sound
> of the 100 watt station in flea fart from Fiji, or something similar.
> But for us true believers in getting the ultimate performance from our
> beloved A's this intolerable technical gaffe must be remedied, if not,
> we'll become the laughing stock of those heathen rice box lovers.
>
> I started to investigating methods for solving the drift.
> The  major difficulty is that the oscillator is located on the RF chassis
> making simple access impossible.
> Also finding a 17 MHz crystal or a harmonically related prove futile
> except for SMD. Ugh
>
> An outboard metal box approach using a SS method led to using a 10 MHz and
> a 27 MHz 1/2 crystal can oscillators feeding a Minicircit passive DBM then
> using a tuned 17 MHz LC circuit.  This would be fed through a very good
> .01uF/1KV ceramic cap into to pin 5 of the 6AK5 oscillator tube V207 after
> removing B+ resistor feed R211.
> OPS! pin 5 goes to tuned circuit T207.  Square wave into LC circuit, not a
> good idea.
> Feed the new 17 MHz through P221? Perhaps.
> Plan two was to remove the 17 MHz crystal, make a new oscillator circuit
> then feed it into the grid pin 1 of V207 utilizing it as a buffer.
> Would this work? I don't know.
> This is just a starting point for others as I will never be able to pursue
> this.  Good luch to any who will even try (and post there results on the
> list.
> Regards,
> Perrier
>
>
>
>
>
>
> .
>
>
>
>
>
>
>
> ______________________________________________________________
> R-390 mailing list
> Home: http://mailman.qth.net/mailman/listinfo/r-390
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:R-390 at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html