Quote:
Originally Posted by Tower Guy
Overload can get very tricky. When calculating the cumulative level of multiple signals you add the peak voltages together (not the RMS) to get the peak voltage, and then calculate the peak power from the combined peak voltages.
For instance, the peak power of two signals of the same amplitude but different frequencies is 6 dB higher than one of the signals. It is this higher peak power that can overload the tuner.
|
Good point.
I would hope that the engineers that wrote the ATSC specs took that into consideration, but maybe they didn't.
This part was added by holl_ands, who does modeling and is an RF engineer, so I took it on faith because I'm not smart enough to verify it:
Quote:
Presumably for Two Strong Input Signals.
Max Signal Input and Output
would need to be DERATED:
3 Signals: -3.6 dB
4 Signals: -5 dB
5 Signals: -6 dB
6 Signals: -7 dB
8 Signals: -8 dB
Ignore signals more than 6 dB below the Strongest.
|
The rabbitears.info report is only a computer prediction, but we see the reality in the spectrum scan done by verder. Unfortunately, the dBm calibration is not absolute, but we can see the relative powers of the channels. He would need a signal level meter to measure the actual powers of the channels.
The power level of WDAY implied by the scan and shown in the signal report looks like it is close to the limit that could be tolerated by the tuner.
Verder gave me his address in a PM, so I was able to look at the satellite vertical view that shows trees in the signal paths from both directions that would attenuate the signals. I was also able to look at the ground photos that confirmed the trees were higher than the antenna, but he asked me not to show those images in this thread.
That's what led me to the conclusion that overload was probably not a threat unless he added a high gain preamp. I think that the intermittent pixelation that he has on the marginal channels is not from fundamental overload or IMD spurs, but from insufficient SNR of the weak signals and multipath from the signals passing through the trees. Both of those can produce errors that exceed what the FEC can correct.
This is what a signal looks like when it is going through trees on a very windy day:
https://www.dropbox.com/s/ig3va499k2...Path4.mp4?dl=0
https://www.dropbox.com/s/ig3va499k2...Path4.mp4?dl=1
In any event, verder made the correct test by adding attenuation which showed that the pixelation continued, so it probably isn't caused by overload.
A more certain test would require a single-channel custom filter that would attenuate WDAY, but that would be expensive.