and written by
Whilst preparing this proposal we studied the following "off the shelf" antennas and rejected them for this tender because the radiation patterns were not suitable to maintain a good signal, as required in the terms, in the area of Phitshane-Molopo while at the same time minimising the signal sufficiently into South Africa.
The H-Plane marked in red is much too narrow to put any meaningful signal in and along the border area. Most of the signal would be radiated into the north of Botswana with very little to east and west.
The H-Plane marked in red is better than the previous antenna, but still falls far short of being able to maintain a strong signal east to west along the border area in Phitshane-Molopo area
This was even worse than the previous five element log antenna
The H-Plane shown in red is much better with this antenna and with improved coverage from east to west along the border. Unfortunately, it also has poor front to back ratio which results in a reasonably large signal being radiated south into South Africa.
Following the study and unsuitability of these antennas and others we have not mentioned here, we decided to design an antenna array that could provide:
After many days of testing and simulations we arrived at the following design array. This comprises of four 5 element log period antennas. Two are vertical polarised and two are horizontally polarised as shown in the diagram below.
The H-Plane is like the three element Yagi looked at previously but improved. The Yagi has a signal of -16dB radiated to the back (into South Africa). Our improved array has a back radiated signal of -22dB, this 6dB better than the Yagi, which is four times less power radiated into South Africa.
Even though there is four times less power radiated to the back, the radiated signal, east and west is not compromised being marginally better over the three element Yagi. Due to two of the individual antennas being mounted vertical and another two horizontal, this results in mixed vertical and horizontal / circular polarisation.
This is the E-Plane with a maximum gain of 5.2dBd. Because the power is divided between the horizontal and vertical equally due to the circular polarisation, then the gain for calculating the ERP is halved to 2.2dBd. Using a 1kW transmitter, 10% power loss on the cable and 2.2dB of gain, the power radiated is 1.65 kW ERP
The tower at Phitshane-Molopo (25,44'49.5"S 25,06'18.7"E) is 60m in height. It is assumed that the antennas can be mounted at 50m. A picture of the tower obtained from Google Earth (taken in 2021) shows it is relatively clear from other antennas from other users at the 50m level.
Using parameters 50m height, 1.6kW ERP and pointing the centre of the array 10 degrees left toward the west from the north (shown in the diagram) we were able to plot the signal that was most optimal for the border and overlay it on mapping provided by Google Earth.
The areas marked in red show where the signal exists at 30dBuV/m at 2m above ground
We choose the value of 30dBuV/m as this is the sensitivity of the average car radio receiver for a usable signal that can be listened to comfortably without the annoyance of noise. Most car radio receiver antennas are mounted on the car roof or across the top of the window, these are close to 2m above the ground.
The following screen shots from Google Earth show signal plot viewed North, East, South and West. Also, various zooms toward the tower
This is the Google KMZ Phitshane-Molopo-1650werp-50m-log5x4circ.png.kmz file that will allow you to view this plot in Google Earth, you can zoom in out, rotate any direction etc. Google Earth is free to download and install on your computer.