Before we start this story, lets rewind back to the 2016 ARRL Earth-Moon-Earth contest. Our 10GHz TWT Amplifier died shortly before the event. We still manged to get our 4.6m, Cassegrain dish on the air with a borrowed solid state amplifier. Jumping to 2017 we repaired the damaged silicone potted filament transformer in hopes to get the station on the air for the DUBUS 3cm EME contest which coincided with ARRL Field Day.
Unfortunately, after a very successful transformer repair the amplifier was still sick.
Even if the amplifier worked, we still wouldn’t have been been able to get the station on the moon. We discovered the azimuth gear drive had seized. Like the potted transformer, motor removal and replacement was not an easy task.
With the 2017 ARRL EME contest not far away, team W6YX got to work getting our 5 moon bounce bands on the air. Last time we entered this most challenging all bands - all modes section, team W6YX won first place! An impressive feat. Our California Pacific coast location means we have limited mutual moon time with Europe, where the hub of moon bounce activity is. EME operators, can you imagine having 3 hours less mutual moon time with Europe every single moon pass compared to a station on the Atlantic coast? 18 fewer hours combined mutual moon time with Europe across all three contest weekends! This is a tremendous disadvantage we consistently overcome with effective engineering choices, and efficient operating using Linrad Software Defined Radio extensively on all bands.
Personal commitments and the quantity of work prevented us from getting all 5 bands on the air for the 2017 ARRL EME contest. We decided to concentrate on only getting our 10GHz station on the air for the microwave weekend of the event. The weekend before the contest we got azimuth working and got the amplifier producing power again. However, the amplifier still had a mysterious intermittent problem. The 4kHz reference for the helix supply DC – DC converter would randomly fail to function. Removing the 200lbs (90kg) amplifier, hauling it to the work bench, hauling it back to the dish and reinstalling it is not an easy task.
Unable to reproduce the 4kHz reference problem, as it seemingly went way, we reinstalled the amplifier. It is worth mentioning a team member did quote “problems that go away on their own, come back on their own”. We enjoyed listening to our voice echos off of the moon Tuesday September 5th! Better than expected performance, given our station has yet to be optimized.
Feeling content, all we had to do is park the dish and wait three days for the contest to start. Three hours before moon rise on contest day, we were met with an unpleasant surprise. The amplifier was not producing power! Hastily in the dark, we removed the amplifier to trouble shoot.
Yes, as expected, the 4kHz reference was failing to start.
Pressure only increased, as the moon rose above the horizon while we scoped and probed the amplifier in the dark.
Lunar declination wasn’t near its maximum, meaning our time with Europe was even more limited. We hurried to repair the amplifier. A team member noticed a blip on the scope as the capacitors discharged while powering down. A desperately needed clue! The 4kHz reference was showing signs of life. We were stumped by the cause of this problem. Transformer-saturation based oscillator designs are well before our time. Unintuitive terrain, just like trying to figure out how to use oscilloscopes that lack LCD touch screens. What we affectionately refer to as the bad ‘ol days. While analyzing the schematic, we empirically discovered if we quickly blip the power to the TWTA off and on, the oscillator would sometimes start!
Not a true fix by any measure, but we can work with this. Perfection is the enemy of good enough and our moon time with Europe is limited. We soldered a wire tapping into the 4kHz reference signal and ran it out side of the amplifier housing. The plan was to reinstall the amp and blip power to it until we saw a 4kHz signal on the scope. Once this happened, we would erect the dish and get on the moon!
In our excitement, we go to move the dish towards the moon. Nothing happened. The F1EHN tracking software was not talking to our dish micro-controller. How could this subsystem fail? Everything was working great 72 hours prior. Expedited trouble shooting revealed the problem. Within three days, rodents fully chewed through our RS232 control cable!
Further analysis revealed the power/relay control cable was also chewed! This explained the mysterious short that developed between some of the conductors in this cable a few weeks earlier.
We’re not done yet. Even the transmit and receive coaxes were chewed up!
No time for a real fix. We’re losing moon time with Europe! We can solder, heat shrink and tape later. Now is the time to pry shielding braid strands from touching the exposed center conductor of the coax and twist control cable wires together. Perfection is the enemy of good enough, especially when you’re losing precious moon time with Europe!
With the amplifier working and our cabling mended, we were able to get on the moon at last!
Only to find out the last station in Europe packed it in for the night. Fortunately we still had our Northern American, Australian and Japanese friends to look forward to. A 1m diameter dish running 50W was the smallest station we worked, and he was an arm chair copy thanks to WSJT-X’s QRA64-D digital mode.
The next evening, we were ready at moon rise contacting our 10GHz European friends.
The event organizers picked a great condition weekend for microwave moon bounce. Libration spreading on 10GHz was very reasonable. We ended up with around a dozen contacts in the log and had a great time! Everyone in our team contributed significantly. The lesson learned was to never ever give up. Sometimes, you just have to choke the crane.