[R-390] R-390A IF Module 60 Hz HUM Update

[Jon Schlegel]

Hi All,

Back again with some more info.  Most recently I've done the following.


(1)    I Built a test fixture to operate the IF module separate from 
receiver.  Definitely worth the effort.  The fixture duplicates all 
functions the IF module sees in normal operation when installed in 
receiver;  limiter, AGC speed switches.  RF Gain, Limiter level pots, 
etc.  Care was taken in fixture design to minimize/eliminate ground 
loops and other sneak current paths that could affect results.

Most notable is how I handled the two ground pins (17, 18) on 
J512.  Filament/Heater return path is from pin 18 ONLY and all other 
signal and B+ return paths are from pin 17 ONLY.  This was done to 
prevent AC filament current from being impressed upon the signal 
return paths via the fixture wiring.  The IF module operates with the 
fixture in a stable manner with no unexpected behavior.

Here are some pix of the fixture and IF module

http://www.vecnav.com/R-390A_IFmodule/IFmodFixture-001-small.jpg
     Lambda DC filament supply sitting under Heath Regulated Supply

http://www.vecnav.com/R-390A_IFmodule/IFmodFixture-002-small.jpg
     "DC Volts" Carrier Level meter reworked to 0-1 mA, 20 ohm movement.




(2)    Re-tightened all ground lug hardware.  No change in hum levels 
after this operation.


(3)    Used DC from a regulated Lambda lab supply for filament 
power.  Again and as expected, hum level became essentially 
unmeasurable.  Hum seems to be strongly related to using AC filament 
power as opposed to DC.


(4)    Sweeping the IF module (J513 to J514) showed that all four 
mech filter BW selections were better than 75 dB down in their stop 
bands.  Filter "Blowby" is probably not an issue.

Nevertheless, simply changing the selected mech filter caused a 
marked change in 60 Hz component levels.  16 kHz selection was the 
best and 8 kHz was the worst with 20 to 25 dB difference between 
them.  It seems that this would suggest that hum is getting in 
through more than a single entry point.


(5)    I'm now suspecting that currents from the AC filament supply 
within the module chassis to somehow be getting into the signal 
path.  As a trial, I fed AC to the filament string from nearly the 
opposite end of the chassis, soldering a wire to pin 9 of V506.

60 Hz levels tended to "even out".  The very low levels measured at 
the 16 kHz BW position came up and the higher levels at the other 
three BW positions tended to come down.  In a way, the more equalized 
levels seemed "More Honest."


***********

I'm now considering trying returning filament current of each tube 
through a separate wired path rather than chassis ground to get the 
current returned to the supply in a more controlled manner.  I'm 
concerned about unintended interstage coupling since filaments would 
no longer be connected to hard ground.  Many RF bypass caps would be needed.

WHAT WOULD REALLY HELP is to be able to correlate the 60 Hz hum 
levels I'm seeing with how other receivers perform.  I'm measuring 60 
Hz levels from 30 to 55 dB below a 30% modulated tone at the diode 
load output of the IF module, depending on mech filter 
selection.  This is a relative measurement that ideally should 
translate to the same measurement anywhere along the audio chain 
regardless of RF/IF path gain, etc.  If anybody has this kind of 
info, I'd be very appreciative to receive it.

Also as a note, I can see 60 HZ on the scope when the 60 Hz level 
becomes higher than about 45 dB below the 30% modulated tone.  If an 
undesired component is visible on a scope, then it's too high as far 
as I'm concerned.

Thanks to all for your help and for reading through all this stuff!

Regards,
Jon