[R-390] LONG: Alignment, Overload and Distortion

[Bill Cotter]

Hi Bill, et al:

I read with interest the problems you have been dealing with, and 
they seem to be one of the tougher R-390A repair issues. I have 
been struggling with a very similar, and possibly related set of 
problems. I have an EAC that behaves in the most peculiar way and 
has given me fits in trying to correct the problem. This receiver 
is a '67 EAC sn#3669, and has been recapped, rebuilt  and aligned. 
Here's what's going on:

Pick a strong station on any band, say 17mHz (problem occurs on all 
bands). Tuning into the station and the carrier level is around 
90-100dB. With the RF gain at 10 and the AVC on, filters in the 
8kHz position the station sounds fine as long as there is no 
selective fading. When fading occurs and the signal dips down to 
say 70-80dB, distortion begins. When the level falls to 30-40dB, 
sever distortion appears until the signal rises above 70dB or more. 
Tune in a weak station around zero to 10dB and it is fine until it 
gets stronger, upwards of 30-40dB.

Ok, figure it's the AVC, right?

Go to MVC, and tune back in on the strong station. Now the signal 
jumps way up and I need to drop off the AF gain dramatically. With 
the RF gain full open, the signal is fine until fading occurs, 
signal level drops, and the distortion creeps in once more. While 
looking for that RF Gain sweet spot, I notice that as I advance the 
RFG from say 5 towards 10, an odd thing occurs. The volume through 
the speaker increases as I approach 10. Right around 8 the volume 
suddenly drops 20db or so (gain-compression??), and distortion 
begins. Increasing the control further the volume increases along 
with the distortion. The distortion is worse when the carrier meter 
is in the 10-30dB range.

Now, I haul it up on the bench.
---------------------------------------------
TEST 1 - IF Chain and AVC

Connect the HP-8640B to the IF deck J-518, HP-427C HiZ VM to the 
AVC line, Fluke DMM to the RF gain line, and a HP-332A Distortion 
analyzer to the Diode Load line. Dial in 455kHz +/- to peak with 
the filter at 100Hz, return selectivity to 8kHz, and 30% modulation 
at 1,000Hz. Controls AGC Fast, RFG wide open, and the following 
results (all AC Vrms):

AVC ON
Vinput Vdload   Vavc    Distortion
---------------------------------
10uV    0.75v   -0.6v   18.0%
100uV   1.00v   -3.5v   12.5%
1mV     1.5v    -8.4v   13.0%
10mV    2.0v    -14.9v  13.5%
100mV   2.75v   -20.6v  14%
1V      3.5     -27.0   14%

All looks fine from the AVC operational point of view.

Now to test the influence of RF Bias on the IF-AVC system, in the 
MVC mode. The RFG control is set at different positions and RF bias 
is measured at the RFG point on the rear panel. Two different input 
levels (100uV and 1000uV) are tested. The results:

AVC OFF - MVC
Vinput Vdload   Vrfbias Distortion
---------------------------------
100uV   10.0v   1.0v    14.0%
100uV   6.0v    2.0v    14.5%
100uV   3.8v    3.0v    14.0%
100uV   2.5v    4.0v    15.0%
100uV   1.9v    5.0v    14.0%
100uV   1.5v    6.0v    14.0%
100uV   1.1v    7.0v    14.5%

AVC OFF - MVC
Vinput Vdload   Vrfbias Distortion
---------------------------------
1000uV  16.5v   4.0v    35.0%
1000uV  17.0v   5.0v    22.0%
1000uV  14.5v   6.0v    18.0%
1000uV  12.0v   7.0v    16.0%
1000uV  9.5v    8.0v    15.0%
1000uV  6.0v    10.0v   14.5%

Conclusion, the IF deck seems to be operating fine below 1000uV 
input at 6vdc and higher bias. However, as the bias decreases (RF 
gain control advanced towards 10) the distortion increases. Only if 
the input was held below 1000uV, will the IF deck function 
normally. This 'sweet spot' would be found by increasing the 
generator level with the RFG wide open until the distortion sharply 
rises above the 14% average.

Does anyone know offhand the signal level upper-limit entering the 
IF deck??

---------------------------------------------

TEST 2 - Overall receiver test

Connect the HP-8640B to the antenna C-connector, HP-427C HiZ VM to 
the AVC line, Fluke DMM to the RF gain line, and a HP-332A 
Distortion analyzer to the Diode Load line. Dial in 17.5mHz, and 
30% modulation at 1,000Hz. Controls AGC Fast, RFG wide open, and 
the following results (all AC Vrms):

AVC ON
Vinput  Vavc    Dist'n  Meter
-------------------------------------
1uV     -2.2v   12.0%   15
3uV     -3.3v   5.5%    22
10uV    -4.5v   4.0%    35
30uV    -5.7v   3.5%    42
100uV   -6.8v   4.0%    52
300uV   -7.6v   4.0%    60
1mV     -8.4v   4.5%    70
3mV     -9.2v   4.5%    78
10mV    -9.8v   6.2%    80
16mV    -10.5v  25.0%   81
30mV    -10.6v  45.0%   82
40mV    -10.7v  35.0%   85
50mV    -10.8v  40.0%   89
66mV    -11.2v  22.0%   88
100mV   -11.6v  8.5%    90
300mV   -14.9v  14.0%   100+
500mV   -17.9v  ****%   100+
800mV   -21.3v  9.0%    100+
1V      -23.0v  10%     100+

As you can immediately see from the results above there is a spot 
in the input-range where the distortion rises dramatically, then 
falls off. The AGC action and the carrier meter both follow the 
received signal flawlessly through out the range.

---------------------------------------------

I have done all the usual first steps: tighten connectors, DeOxit 
sockets, swapped tubes with known good tubes, etc. I also swapped 
in a second identical EAC IF deck and had the same listening 
results. What I am wondering about now is the possibility of 
parasitics in the RF AMP, or a breakdown of a component (ie: mica 
cap, etc).

Any Suggestions would be appreciated. All results will be posted 
for the benefit of the group.

Thanks es 73 bill n4alg