[Lowfer] 137.78 and 185.3 kHz

[JD]

Just noticed in re-reading last night's message that the phrase "(f)or a 
while during the evening, I could see bits of EAR" was eerily apt for 
Halloween.  Must be zombies around.

Don't know if I'll get a chance to complete the PLC survey tonight or not. 
It may have to wait for the weekend because of a strong cold front 
approaching, which will bring us rain tomorrow and possibly static as early 
as tonight.  Winds are currently up to 40 MPH from the south, which is 
already making it very difficult to be in the field for long.

I had to do that this morning for a while to test a fix that I thought my 
antenna buffer amp needed.  The buffer is written up in this month's 
LOWDOWN, including both the initial attempt and a revision that improves 
frequency response in the upper LF spectrum.  Subsequent testing showed that 
the AM BCB rejection of the second version is not as good in the field as it 
looked on the bench, however, and a mixing product of our two nearest 10 kW 
radio stations now shows up at 170 kHz.  It's 60 dB down from the carriers, 
and it goes away at night when the stations switch to directional patterns 
and reduce power.  But I don't want it there at all, so I spent yesterday 
rummaging through inductors and testing different traps I might use.

It turns out that if you use a parallel LC trap with relatively low L-to-C 
ratio in series with the amplifier input, tuned at or slightly below the 
carrier frequency of the lowest station involved in the intermod, the 
combination of the trap and the two pole RC filter at the input suddenly 
takes on properties of an elliptical filter.  That old 100 uH Radio Shack 
choke is finally good for something!  The response curve is better in the LF 
band (and right up to the notch, in fact), while the attenuation on the high 
side of the notch is a few dB greater over the rest of the AM band than the 
RC filter alone.  If I integrate the trap into the RC filter itself, there's 
a chance it could do even better, but today I just wanted to test the 
concept.

Well, today's experiment was a classic good news/bad news comedy:
*The good news is, the trap attenuates the lower offending carrier by 15 or 
20 dB even with the real-world antenna.
*The bad news is, it makes little difference above the notch frequency with 
the real-world antenna.
*The good news is, that much attenuation of even one carrier should 
dramatically reduce the intermod product.
*The bad news is, it didn't.  Only a slight change.
*Some _unexpected_ news is that, while trying to determine the accuracy of 
the receiver's built-in step attenuators more precisely, I found that 
attenuating signals after the OUTPUT of the buffer with a known good 10 dB 
pad brought about a dramatic additional reduction in intermod.  The dead 
carrier that is always on 170 was still there, but no modulation products.
*Thus, the unexpected news is good news--it means most, if not all, of the 
intermod is NOT in the buffer.
*And that is bad news, because it means it's in my receiver instead.  After 
all these years, I've finally reached the point of being able to overload 
its input.

Since the receiver is now working beyond the limit where it can be trusted 
as a test instrument, further evaluation of the buffer's linearity will have 
to be done  with a moderately high order 50 ohm lowpass filter between the 
buffer and the receiver.  And I just happen to have one of those--the one 
Todd built for me to use with his upconverter.

So I've come to town to have a quick lunch, pick up the LPF, take an allergy 
tablet, and pay a couple of bills on my way out of town again.  Then we'll 
see what we see.

John