There is nothing very critical about the construction of the antenna... only a few precautions must be taken. It should be installed on the very top of the vehicle, so that it has a clear, unobstructed view of the horizon in all directions. A sheet metal "ground plane" is required, typically A 3 - foot square of aluminum, with 2X2 inch "angle aluminum" installed at the edges for structural strength.

The diameter of the antenna "circle" is not critical... smaller diameters will yields Doppler sine waves of lower amplitude. The distance between any two adjacent antennas should not exceed 1/3 wavelength, for reasons too complex to explain here. Quarter wavelength vertical antennas work well, if they have adequate ground planes beneath them in all directions.... don’t install them at the extreme edge of the ground plane. Antennas shorter than 1/4 wavelength can work well with smaller ground planes, but receiver sensitivity will suffer slightly, due to the impedance mismatch.

The length of the coaxial cables from the base of each antenna to the switching unit is not critical, but all the cables must be the same length, within a couple of inches. Cables of 1/4 wavelength are probably best, since the switching unit presents a "dead short" to the de-selected antennas, and a 1/4 - wave line section will translate this to a "dead open", at the antenna drivepoint. ( reduced "shielding" of other antennas )

PIN diodes ( Newark p/n : 09F7174 ) are best for antenna switching, but ordinary signal diodes ( such as 1N914 or 1N4148 ) work reasonably well at 2 meters, if some loss of receiver sensitivity can be tolerated. ( about 3.5 db measured, including a ten foot co - ax cable ) Schottky signal diodes can also be used... they should work better than ordinary signal diodes, and might be easier to find than PIN diodes. More information about antenna improvements is provided below.

TOP / BOTTOM ANTENNA MOUNTING

On the antenna PC board, the input labeled "090" should be connected to an antenna on the passenger side of the vehicle, ( 3 O'clock position ) and the other antennas connected in a corresponding manner. Depending on how the antenna unit is mounted, ( above or below the ground plane ) it might be more convenient to connect this input to an antenna located on the driver’s side, instead. ( with corresponding changes in other antennas )

If this is done, it will be necessary to mount the display LED’s on the solder side of the display board, instead of the component side. Otherwise, the direction of display rotation will be wrong... clockwise signal rotation will yield counter - clockwise display rotation. Signals at bearings 000 and 180 will still indicate properly, but signals for all other bearings will be ( left / right ) "mirror imaged".

ANTENNA ENHANCEMENTS

Although the basic design of the antenna works reasonably well, there is a lot of room for improvement, and ( eventually ) the existing design will be replaced with a better one. In particular, the ( ordinary ) 1N914 diodes requires a lot of current ( 10 mA ) to "effectively" switch them on, and most digital chips will not provide this much current. That explains why I chose a 74S138 to drive them... ordinary TTL chips can "sink" 16 mA of current. I actually wanted to use an ordinary 74138 chip, but ( unfortunately ) it seems that no-one makes them anymore, so I chose a "Shottky" version, instead.

It turns out that the Shottky version of the chip is so fast that it generates a fair amount of RF noise near 144 MHz, and this causes reciever sensitivity to suffer. I estimate the RF noise produced by the Shottky chip is around 0.5 uV at 144 MHz, which is enough to cause a noticeable reduction of reciever sensitivity. When I "stopped" antenna rotation and actually measured the RF loss with equipment, it was only 3.5 db, but the "effective" loss is probably closer to 10+ db.

To work properly, each diode switch must provide at least 20 db of ( on/off ) "isolation". This can be achieved with PIN diodes at much lower current levels, ( 1.0 mA ) so the ( offending ) Shottky chip can be replaced with a low power version, or even a CMOS version. If this is done, the values of the pull-up resistors ( R34 - R41 ) should be increased to 2.2K. If you are so inclined, it would also be wise to replace the resistors R42-R49 with small RF chokes, to decrease RF losses.

I use Hewlett Packard ( oops... I mean Agilent ) PIN diodes, available from Newark, and identified in the ( main ) parts list. PIN diodes are also available from Motorola, but I have not tried them. ( not yet, anyway...I think the Motorola p/n is MPN3404 ) The RF chokes that I use ( 100 uHy ) are also used in the portable / TDOA DF, ( described on this website ) and p/n's for them are included in the TDOA parts list.