|
|
Antenna Stacking Test - ATF-X300
shown with new
style Integrated Stacking Boom
and
Compact
Vertical Tilter
Description:
The main goal I
had when designing this new antenna was to build a smaller antenna that can match
the performance of much larger and heavier antennas. I also wanted to design a
new simpler stacking boom arrangement to save weight and reduce wind load. I must
have tested a hundred different element positions during the initial development
to find the optimum reception on all channels. I first tested each
antenna separately. In all my previous tests I tested the single antenna mounted
directly to the mast with its standard bracket. When tested in this
configuration both antennas tested exactly the same. In this test, I placed the
antennas on the stacking boom and ran a test on each antenna. I was surprised to
see the difference in reception I got from just a four foot change in lateral
position. I also found out that the optimum horizontal angle was different for
each antenna. There was a 2-5 degree difference in horizontal angle, so I
"toed in" the antennas about 5 degrees, this change gave me a 5%
increase in signal. I used a Spectrum Analyzer to check the waveform for
each station. The waveforms allow you to measure signal strength in decibels and
to see multipath interference in the signal. See below for results and comments.
|
Set up:
I mounted the antennas to
my new Integrated Stacking Boom. The boom assembly was then mounted to a ATF-V100
Compact
Vertical Tilter
for precise positioning. Vertical angle positioning is very critical with
stacked antennas, since the vertical beamwidth gets even tighter. The
antennas were then connected to the Low-Loss
RG-11 feed line through a Antenna
Phasing Harness. I set the spacing to 48". |
Results
Definition:
-
XX-XX =
Measured signal strength level
-
0-XX-XX =
Signal locks on briefly but keeps dropping out
-
0-50 = Signal
was detected but not able to lock
-
0 = No signal
detected
-
-- = Not tested
|
| Note:
Signal level refers to the Dish 6000 indicator which measures the signal quality
based on percentage of errors. As long as the indicator stays above 48-50%, the
signal will lock. I assume that if you have over 50% errors, the Dish 6000 does
not have enough good signal to decode properly. A reading of 100% would indicate
no errors in signal. I have found that at 47% I will see pixelation in the
picture and below 46% the picture will drop out completely. |
Dish 6000 Test

Spectrum Analyzer Waveforms:
|
|
Mt. Wilson
Stations (15 mile distance)
|
|
|
Antenna A
|
Antenna B
|
Stacked A+B
48" spacing
|
|

Channel 31 |

Channel 31 |

Channel 31 |
|

Channel 36 |

Channel 36 |

Channel 36 |
|

Channel 42 |

Channel 42 |

Channel 42 |
|

Channel 53 |

Channel 53 |

Channel 53 |
|

Channel 59 |

Channel 59 |

Channel 59 |
|

Channel 60 |

Channel 60 |

Channel 60 |
|

Channel 61 |

Channel 61 |

Channel 61 |
|

Channel 65 |

Channel 65 |

Channel 65 |
|

Channel 66 |

Channel 66 |

Channel 66 |
|
Conclusion:
- As noted above, in all my previous tests, I tested the single antennas
mounted directly to the mast in the same lateral location. When I tested the
X300 in identical locations, both antennas tested exactly the same with the
waveforms appearing the same on each antenna. However,
during this test, I placed the antennas on the stacking boom spaced 48"
apart and ran a
test on each antenna separately. I was surprised to see the difference in reception I
got from just a four foot change in lateral position. At first I thought it
was because one antenna was shielding the other, so I removed the antenna
not being tested and found it was actually the change in lateral position
that was accounting for the difference in performance. This is proof that
the placement of your antenna on the roof plays a important part in
reception, especially in multipath conditions!
- It is easy to see the level of multipath in the pictures above by the
sharp notches and bumps in the waveform. A nice, clean
signal should appear as a straight horizontal line. I have found that the flatter
this line gets, the higher the signal number gets on the Dish6000 and the better the channel
stays locked in. When the notches are deep enough (about 7dB) the signal
will not lock in. The stacked antenna arrangement greatly reduces the multipath
notches in the signal.
- Horizontal stacking decreases the vertical beamwidth, this makes the
vertical angle of the antennas very important in locking in the weakest
stations. Example- It would be very difficult to get channel 53 (ABC) without a vertical tilter.
If I change the vertical position of the antenna just one degree either up
or down, channel 53 will drop out completely. Fine tuning for strongest
signal on 53 is just a matter of +/- 0.5 degree. With channels 36 and 65, the
critical angle is +/- 2 degrees and channel 31 is +/- 3 degrees. This angle
will change according to weather and time of day.
- The horizontal tuning angle also decreases. For the weaker stations, just
+/- 2 degrees makes a big difference. For the best signal I normally
manually enter the angle on the Channel Master rotator, since it moves in 2
degree increments when you push the arrow buttons.
- Not shown in this test is the reception from San Diego 130 miles away.
When the atmospheric conditions are favorable, all San Diego channels come
in very strong. Horizontal and vertical angles are not as critical when
picking up long distance signals since multipath is rarely a problem,
increasing gain yields the best performance with long range reception.
- I have been running these two X300's now for about two months. They are so
far the best performing antennas I have used to date. I can't tell you exactly why that
is, since all my development was done through trial and error and not by
using a computer program or a laboratory. I did use a spreadsheet for
calculating the progressive element spacing. I was able to modify certain
variables to alter the spacing characteristics, then I did tests to determine
the values which seemed to work best. I then fine-tuned these variables to
get the optimum performance.
|
|