I figured that out earlier this evening...............It looks as if this
receiver has a single loop AGC system. This causes some problems, because of
the high gain of the 1350s themselves. The I.F. gain is so high that AGC
starts on signals as small as .1uV. This means that the PIN diode attenuator
is active at very low signal levels; the result is a degraded NF as I.F.
gain is increased. It appears that the designer's solution was to
intentionally lower the system gain so that this wouldn't happen. That
didn't cut it for me, so I delayed the RF AGC by inserting a series resistor
between the PIN diode attenuator and the AGC control voltage source for the
shunt diode. For the series PIN diode, I bumped up the current to 75mA, a
much more realistic drive current for this type of device.
The I.F. strip was not optimally designed, in that V+ was applied to the
Pin2 of the 1350 (OK here), and Pin1 at the same time (not good), allowing
the voltage to flow through the primary of the I.F. transformer to Pin8
(other pin of differntial output pair, also not good). What I did was break
the V+ connection from Pin1 and rerouted this connection to the center tap
of the I.F. transformer primary. This is good, because instead of having
Pin1 of the 1350 I.F. amp at AC ground, the center tap of the transformer
can now be grounded, providing a balanced load for the MC1350. This is how
the chip was intended to be used, so this is also good. Now, when I peaked
the I.F. transformers there was a definite peak instead of a very broad
peak.
This, in combination with the RF AGC loop changes and the removal of the
high pass filter from the preselector gives the receiver a .05uV MDS over
the whole tuning range. Because of the delayed AGC, the receiver I.F. strip
clips at 15mV instead of 25mV. Once you tune the strong signal out of the
I.F. bandpass, no overload is encountered. There was some crossmodulation
from the series PIN diode, until I bumped up the diode current from 15mA to
75mA.
A 1N5767 PIN diode is ok up to 100mA, so the higher current isn't an issue.
Because of the very high AGC action, you cannot really align this receiver
by ear. What you need to do is look at the AGC voltage at the output of the
emitter follower with a VOM (analog meter), and then you have a nice signal
strength meter for doing the alignment.
I am very happy with this receiver; I will give the folks at Ten-Tec a call
about my findings.

Pete

"clifto" wrote in message
...
Pete KE9OA wrote:
I did remove the 1k swamping resistor from the primary of the 1st 455kHz
I.F. transformer. Gain came up quite a bit, so the radio sounds more like
a
651S-1 or a 75A-4.

Considering you're not getting any overloading on any band, why do you
suppose they put it there in the first place? I can only think of
reducing primary Q, which doesn't seem terribly worthwhile.