PA8W Amateur Radio             

A preamped mobile UHF Array (380-470MHz)
Revised 200321

An easy to construct preamped UHF array is illustrated in the above picture.
As in the VHF array, a very good capacitive coupling to the roof metal is essential.
Due to the shorter UHF wavelength, another approach is feasible:

Simply take a sheet of aluminum as a ground surface.
Taped to the car roof it provides very good capacitive coupling, so that the car roof extends the ground surface.

This UHF array has built-in preamps in the 4 antenna feedpoints:

Giving two big advantages:

1, It has about 18dB gain, that's more than welcome on UHF bands!
2, Since the preamps are located directly at the antenna feedpoints, the coax and control lines are  less sensitive to the RF field
since they are 18dB less sensitive compared to the antennas.

Due to this, the received signal is less polluted by radiation picked up by the cables.
This improves accuracy quite a bit and makes routing of the cables much less critical.

The preamped Array's accuracy was logged in a field test @433,2MHz:
Averaged error turned out to be 2,2 degrees, with a measured Peak error of 4,5 degrees.

And that without ferrites, and simple RG58 cable.

Note that this array does a reasonable job at VHF as well, especially with slightly longer antennas on it (22cm).
But only if the ground plate is taped directly to a larger metals surface, such as the car roof.

I won't reveil the preamp type, since there is some commercial interest right now.
But I am willing to supply radio amateurs with a set of 4 antenna PCB's with all parts soldered in place, plus the combiner PCB, also with all parts soldered in place.
So you only have to build these PCB's in your array and connect control cable and coax to the combiner, plus 4 short coax cables to the preamp PCB's.

For those of you who have really payed attention reading my pages:

Yes, it is still very important that a non- selected antenna is decoupled from the system so it won't act as a parasitic element.
And yes, the preamps do just that, they are fed by the control signal through their coax, and they let go of the antenna as soon as they are switched off, virtually making the antenna "disappear".

Below some pictures to illustrate the building process:

This is the schematic of the antenna array.
For the RDF41 and RDF42 stick to this schematic.

Right-Mouse-Click on the drawing, and choose "Save as" to save the drawing to your map or desktop.
The resolution will be doubled compared to the drawing on this page. 

Take an aluminum sheet of 50x50cm, or at least 35x35cm. Thickness 1 up to 2mm.
1mm thick is sturdy enough, it flexes enough to adapt to the curvature of most car roofs.
A 2mm thick sheet is a lot stiffer, but can still be pre-bent to fit the curvature nicely.

Note that this undersized ground plate should be taped flat to the car roof, because only this way the roof will extend the ground surface.

Mark the center of the sheet, and draw a cross to the corners.
At 13 cm distance from the center, draw markings for the antenna positions.

The antenna position in the bottom left corner is already drilled in this picture.

Note that the antenna center holes are 10mm, to avoid electrical contact between antenna screw and ground plate. The fastening holes are 3mm. The fastening screws will have to provide electric contact to the ground plate.

On the back side the 3mm holes are countersunk to accept the countersunk screwheads.
The screwheads should not protrude because they will damage your car paint!

Now, the small antenna PCB's can be screwed in place.
Apply a good sealer between PCB and aluminum to prevent water from creeping in.
I countersunk the bottom side of the center hole of the PCB to make sure that the screwheads have good clearance to the below car roof.

I use a 15mm spacer bolt so the total length matches the inside height of the cups I use as enclosure.
The actual antenna goes on top of all that.

After mounting these antenna PCB's and spacer bolts, I sealed the antenna base holes on the bottom side using silicone sealer.

Then, the center PCB (combiner) is glued in place using silicone sealer.

The yellow line points at the spot where the outgoing coax core should be soldered.

Note that there's room for a MMIC there in case the combiner is used in a passive array.

The red lines indicate the spots where the antenna coaxes should be soldered.
Of course all coax shields are soldered to the nearest ground surface.
The blue lines point at the pads where the antenna control lines should be soldered.
Don't forget to connect the ground of this PCB to the ground of the RDF (the center pin of the 5 pole DIN)

So, using a 8 wire network cable, 4 wires are used for controlling the 4 antennas and the 4 remaining wires are used as ground connection.
All antenna PCB's are screwed and glued in place using silicon sealer.
The preamp IC's are so tiny that you don't see them in this picture.

A close-up of an anrtenna feedpoint with the preamp in place.
The tiny chip is about 2mm long.
Beware that soldering time should be kept very short to avoid damaging the chip.
I use old fashioned 40/60 soldering wire because of the lower melting point. 

All wired up.

All caps glued in place.

Ready for a paint job.

The actual screw-on antenna are made using identical brass spacers with a length of steel wire drilled and soldered in its top.
Overall length including all spacers should be 17cm for all 4 antennas, top to ground sheet.
I covered the steel wire in heat shrink tubing, in that case you may shorten the elements by 5%.

The picture on the left shows 3 stadia of an antenna:
Bottom: Standard brass M3 spacer .
Middle: Threaded stub removed and 2mm hole drilled into the center, about 5mm deep.
Top: Steel antenna wire soldered in and covered in heat shrink tubing.

The field test setup reveiling an average error of 2,2 degrees.

Strong signal behavior:
With the ever increasing field strengths of utility signals here in urban areas, I regularly encounter
situations where the Stabo receiver seems to be overloaded, resulting in a noisy reception, allthough the S-meter shows full signal strength.

There are two things that can lead to this phenomenon, both triggered by very strong field strengths.
First of all, the array is a wideband design, for several reasons.
Therefore, ALL signals that reach the antennas are amplified in the preamps.
So it may well be that very strong signals at other frequenty bands are causing the trouble, especially when a wide band receiver like the Stabo is used without any preselection.

This is easy to improve with a external bandfilter between array and receiver, like the one I made many years ago in maybe 20 minutes.

This filter attenuates a lot of the signals outside of the passband making it easier for the receiver.
There's plenty of documentation about this type of filters on the internet.

The worst situation however is when the fieldstrength is so high that the preamps themselves are overloaded. In that case, a bandpassfilter doesn't help since it is located between antennas and receiver. Only smaller antennas may help a bit, so often I have 4 very short antennas with me, just in case I have to hunt in very high fieldstrengths.