Maritime VHF filtered low-noise amplifier

Having built an Automatic Identification System (AIS) receiving station with a Raspberry Pi, RTL dongle, and homemade dipole antenna a few years ago, I have always wanted to add a filtered low-noise amplifier (LNA). But these have never really been available locally in South Africa and ordering one from overseas is not very affordable. Therefore, I decided to try and build one for even more money.

AIS transmitters are fitted to maritime vessels and transmit information about the vessel and its movements on the maritime Very High Frequency (VHF) (162 MHz) band. For large vessels, these transmitters are mandatory. These transmissions are used by websites like MarineTraffic and VesselFinder to plot ship locations.

Having looked at the LNA’s available from Uputronics, I thought it might be possible to build something similar. With the requirements being that the LNA is housed in an aluminium enclosure and the DC power fed over coax.

In terms of the basic building blocks, an active device and filter from Mini-Circuits were chosen. These were ordered from RF Design locally. The filter was placed before the active device to block signals from FM broadcast and other voice radio transmissions (I have noticed quite a bit of RFI from voice transmission around 162 MHz).

LNA schematic capture

The active device was powered using a 3V3 regulator instead of the 5V one indicated on the schematic. This allows the LNA to be powered from a 5V power supply. Bias tees for removing DC from the RF path and for powering the active device were built from discrete components as this was more cost effective.

I would have loved to use Altium Designer for the schematic and PCB design, but this was unfortunately not an option. As the second option, I went with KiCad, which offered all the needed functionality.

LNA PCB layout

The PCB was manufactured with PCBWay and it shipped the day after placing the order.

Manufactured PCBs

Enough PCBs and parts were ordered to build a number of boards. The PCBs were designed to fit a Hammond enclosures, and the delivered PCBs fit nicely in the ordered enclosures.

PCB in enclosure

Shown is the PCB in an enclosures without the enclosure end caps.

PCB in enclosure

The initial testing of the assembled LNAs was very rudimentary. An SDR was used for a basic test were the LNA was switched on and visually checking the spectrum.

One LNA was tested using a RF analyser, below the filter pass band can be seen and an active device gain of around 21 dB.

RF measurements

The active device has a stated noise figure of 0.75 dB, but I have not been able to test the noise figure for the whole filtered LNA. Overall system noise figure should be improved by the addition of the LNA.

At the moment, I don’t I have a nice way of visualising the difference the LNA makes to the AIS receiving station. The best way currently is the average and maximum reception distance graph that MarieTraffic generates for the receiving station. From this graph, the point at which the LNA was added to the station can be clearly seen from the graph towards the end of September 2024. This receiving station also feeds data to MarineTraffic.

MarineTraffic - average and maximum reception distance

Three LNAs were assembled and fitted to three receiving stations. Data from all three of these stations is used to generate the pages that can be seen here and here (map). One station is located in Langebaan, while two operate from Cape Town (with antennas in different locations). I’ll have to do a write-up of this multi-receiver AIS system at some point.

Populated PCB