View Single Post
Old 15-Jul-2016, 8:57 PM   #1
acat
Junior Member
 
Join Date: Jun 2016
Location: Boulder, Colorado
Posts: 17
Boulder Reception #2: Spectral Analysis

I have been struggling with reception in the Foothills outside Boulder, CO (elevation 6700ft) and this has been discussed elsewhere (http://forum.tvfool.com/showthread.php?t=16002) where there is a TVFool Report and considerable discussion. Many thanks to all who contributed in particular Rabbit73.

The bottom line is that I sometimes could get an adequate signal level and quality for High VHF channels 7 & 9 using either a high gain UHF or VHF antenna. It seemed to make no difference. The UHF actually seemed better than the VHF. In either case, pointing directly at the source (160 deg Magnetic) yielded no usable signal, while pointing at ~90-100 degrees often worked well.

The criteria for "usable" here is a high symbol quality (as well as strength) for a SiliconDust HDHR networked tuner. Symbol quality is the number 1 factor that determines if a recording will be watchable, or even occur.

I decided to take things to the "next level" and put together a "poor man's" Spectum Analyzer usung an RTL-SDR (software defined radio) and software from Ear-to ear Oak (http://eartoearoak.com/software/rtlsdr-scanner). This roughly $20 Spectrum Analyzer works surprisingly well but ain't no $3000 Textronix replacement.

The bottom line is that signal levels are very good with both the VHF and UHF antennas even for channels 7 & 9 for the 8-bay UHF antenna. I do not see a lot of noise overlapping the bands in which the digital data is being transmitted either. I'm forced to believe that the reception problem is some multipath-like effect where the digital data is arriving at slightly different times leading to errors and low symbol rates. The surrounding hillsides and atmospheric reflections may be the problem (it has become cloudy here and the symbol rate has gone up to 100%!) Moreover the signal levels are always pretty high, even when the symbol rate is zero! Usually the signal level is 15-30 dB above the noise floor.

Unfortunately the spectrum analyzer cannot resolve the region of the digital data very well (as far as I can tell), so I don't see how to pursue this further. If any of you have experience in this I'd love to know. So attached are screen shots of the spectral region that includes channels 7 & 9. One was for good symbol quality, the other," TWO..." low.
Attached Images
File Type: jpg ONE UHF USABLE Scan 173.0-187.0MHz.jpg (25.5 KB, 567 views)
File Type: jpg TWO Unusable UHF Ant on Roof Scan 173.0-187.0MHz.jpg (26.0 KB, 552 views)
acat is offline   Reply With Quote