Seems to work for me here, allowing me to aim straight at Grand Island, and still be able to receive a few of the stronger of stations from the distant CN tower, which is some 60 miles or so right in line behind Grand Island Transmitters. Could never do that before adding the extra attenuation, which for me was only about a 6 dB reduction in signal strength across the board. Previously would have to do that by aiming away from Grand Island. Like putting the welders goggles on to watch that solar eclipse...
That's a good example of how to optimize the SFDR of your equipment to match the signal dynamic range at your location. But, if the difference between the strongest signals and the weakest desired signal is too great, the attenuator technique fails because you run out of weak signal. The required SFDR is too great.
As you increase attenuation, the weak signal SNR improves because the noise floor drops faster than the weak signal drops. At first glance, it seems counterintuitive that you can improve the SNR of the weak signal by making it weaker.
But as holl_ands would say could result in up to 18 dB reduction in Noise (IMD to be specific).
That is not exactly correct. If you are dealing only with IMD, then for every one dB of attenuation, the IMD spurs will drop the noise floor 3 dB, but you are also reducing your weak signal. This gives you a potential net SNR gain of 2 dB, not 3 dB.
I pushed some amplified signals into a CM7777HD preamp to produce IMD.
CM7777HD#1 Overload Test, LOW Gain Setting
Atten SONY Signal Noise* SigTo CH33Sig** CH33Sig***
dB SNR CH 45 CH43 Noise into #1 out of #1
dB dBmV dBmV dB
0 0 5.0 -6 11 10 dBmV 26 dBmV(-22.8dBm)
1 0 3.0 -9.8 12.8
2 0 1.2 -12.2 13.4
3 0 0.2 -13.8 14.0
4 0 -1.0 -14.5 13.5
5 0 -2.0 -15.8 13.8
6 14 -3.0 -17.0 14.0
7 15 -4.5 -20.8 16.3
8 16 -6.0 -22.1 16.1 strongest CH 45 stable signal
9 16 -7.0 -23.8 16.8
10 16 -8.0 -25.2 17.2
CH 45 is weakest reliable signal here
* CH 43 is nearby unused channel to measure noise floor
** CM specs say MAX input to preamp is 10 dBmV
*** CM specs say MAX output is 34 dBmV. CH 33 is strongest
local signal. It, and 3 other strong channels (16, 29, and 40)
cause the IMD.
I increased the attenuation in 1 dB steps until CH 45 was stable with an SNR on the TV of 16 dB.
As I increased the attenuation, the signal did get weaker, but the interference was reduced more rapidly. This is because the IMD is reduced 3 dB for each 1 dB of the attenuator, making the SNR net gain of 2 dB for each one dB of the attenuator. This is confirmed by the 20 dB drop in the noise for the 10 dB of attenuation.
There isn't a perfect correlation between the attenuator steps and the change in signal readings. This is because my OTA signals aren't as stable as I would like, and the noise floor here will vary because of the local electrical interference. In spite of that, I think the test is still valid.