After several failed attempts I managed to copy GB3SEE on 10368.850 MHz earlier this week with the dish and 10 GHz modules set-up on the front patio. The beacon is only around 8km from my QTH and was a good signal when I used the dish with a Bullseye LNB and SDR back in February. I’ve since found out that the beacon had been turned off for a while which would explain why I couldn’t copy anything a couple of weeks ago!

Back in January I built a W2IMU 10 GHz dual-mode feedhorn using dimensions calculated from W1GHZ’s HDL_ANT program. I used a 50mm length of 22 mm od. copper pipe for the input section and 98mm length of 54 mm od. copper pipe for the output phasing section. The flare section was cut from 1.2 mm thick copper sheet which was annealed before shaping and soldering. Some simple bench testing indicated that it was working and I also managed to carry out some limited tests with the feedhorn mounted at the dish.

I need a couple of 10 GHz bandpass filters for a down-converter to test my 10 GHz station and decided to build a evanescent mode circular waveguide filter. Evanescent mode filters work below waveguide cutoff frequency and in order to famliarise myself with their properties I started by modelling 2nd and 3rd order filters in CST Studio using designs by K5TRA and G4JNT. I also found a useful paper by Sam Mandev that runs through an example of determining distances between tuning screws.

I’ve spent the last few weeks putting together some equipment for a small 10GHz station using a mixture of pre-built and homemade modules. The block diagram shows the main components of the system. The core of the system is a simple transverter designed by W1GHZ and a DF9NP PLL local oscillator board running at 9936MHz for the 432MHz IF. It’s probably overkill but I decided to repurpose the 4 channel Ardunio sequencer I was planning to use for the 2m EME station.

I recently salvaged a 1m offset dish that I’m planning to use on 10 GHz and need to build a suitable feed. To determine the dish properties and parameters for a suitable feedhorn I used the Wifi calculations option in Parabola Calculator version 2.0 (same code as W1GHZ’s HDL_ANT program). Dual-mode feedhorns are frequently used for offset dishes because of their performance. The W2IMU Horn Design and Template option in HDL_ANT generated the following dimensions for a dish with a f/D = 0.

Background Recently I’ve started experimenting on 10GHz and as I have no test equipment above 6GHz I decided to build a noise meter that would cover 10GHz as well as possible IF’s of 432MHz and 910MHz. There are several noise meter designs using the AD8307 (DC - 500MHz) and AD8313 (0.1GHz - 2.5GHz) rf log detector devices and I’ve been using two AD8313 devices in an Arduino based RF power meter since 2021.

In October 2023 I gained DXCC award #227 for 2m - all QSO’s were via EME.

This EME station map is based on data from the call3.txt database file maintained on the MMMonVHF site. The red dot indicating the location of a station is shown at the centre of their grid. I have only included stations that have a 6 character QRA locator and an "EME" label and ignored stations marked as "Expedition". This call3.txt data is maintained by users and therefore only as good as the information they supply.

In May 2021 I changed the feed arrangement on my 144MHz XPOL EME array to transmit right hand circular polarisation (RHCP). Circular polarisation (CP) is frequently used for EME on 1296MHz and above, it’s not so common on 144MHz or 432MHz. The main advantage of CP compared with horizontal or vertical polarisation is that losses due to Faraday rotation are minimised but there is a -3dB penalty when one station is using CP and the other is using linear polarisation.

In January 2022 my SSPA shut down with a high VSWR warning when I was operating the 2m EME station. I tracked the problem down to the masthead control box that contains the antenna changeover relays and LNA’s. After testing on the work bench I found one antenna changeover relay sticking in the “on” (energised) position after control voltage was removed. The relays are setup to be energised during receive cycles so if it’s stuck “on” I would have been transmitting into an open circuit.