Category Archives: David

Before I package my 600W 1296MHz Power Amplifier into a single enclosure to be placed closer to the antenna, I thought it would be a good idea to take some pictures and draw a block diagram in case anyone wants to build something similar.

My original idea was to build a 600W PA using W6PQL boards.  However, things have become very busy at work, so realistically building the PA would have taken a very long time.  As a compromise, I decided to buy a 300W pallet and its companion controller from VHFDesign, not really knowing how that would go.

VHFDesign’s 300W 1969MHz Pallet with Heat Sink and Controller. Image Credit: VHFDesign

I was very pleasantly surprised by the VHFDesign unit.  I purchased it directly from their website using PayPal.  The unit arrived quite quickly from Ukraine.  The instructions are clear, so once I got around to it, installation went easily.  Their 2-cascade LNA has also been a pleasure to use, but I’ll leave that review for a future post.

After working at 300W for a while, I decided it was time for an upgrade, so I purchased a second unit with the intention of combining their outputs using ring hybrid couplers.  The final block diagram is shown in this figure.  The splayed-out view of the modules is shown in the following figure:

With an output of 326W for each VHFDesign pallet, each (suitable for CW and SSB), The PA produces a combined output of 600W.  IMPORTANT NOTE: See my post about phase equalization!  I’ll optimize cabling and phase matching once I assemble everything into a single enclosure.  For now, it all works nicely splayed-out indoors.

Today was a nice, sunny day, with temperatures in the 40s, so I decided to measure the power drop along the coax from my 1296MHz power amp to the feed.  The PA is at the closest indoors site to the dish.

I am measuring 520W going into the 35 feet or so of LMR900-DB that takes the signal to a junction box mounted at the antenna mast. There, I measure 420W remaining, which go into 27 feet of LMR600-UF to the feed (mast + rotor loop + cable to reach feed on 12 ft dish). After that much cable, barely 320W are left to go into the septum feed.

The numbers check out with the expected cable attenuation, so I understand that I’m simply experiencing normal cable loss. However, loosing 200W to heat the ground bothers me, so I’m looking for ways of increasing the power that reaches the septum feed.  I’m first considering ways of keeping the amp indoors, but I’m afraid that I’ll ultimately have to put it all at the mast and use a hanging cable loop straight to the feed to minimize losses.

This weekend I was able to combine the power of two 320W pallets using ring hybrid couplers (rat races) for an output of around 600W.  The strength of my own CW echoes improved dramatically.  The setup above still needs cleanup, and I may end up placing it at the base of the antenna to reduce cable losses.  On the meantime however, my system is ready for CW QSOs.

Conditions were good today, and there were quite a few stations calling CQ off the Moon.  Unfortunately, I realized how rusty my code has become, and cheating by using decoding software does not help at these signal levels, so I decided to hold off on answering until I get my code back up to speed.

I also started playing with JT65.  I can decode straight from both the IC-9700 and the AOR AR5000/SDR14.  Next weekend I’ll build a circuit so that the computer can trigger the PA sequencer to be able to TX JT65C.

Achieved today something that I’ve wanted to do since I was a teenager:  bounce a signal off the Moon.  I was able to achieve that today for the first time!

• Antenna:  12 foot dish with 1296 MHz septum feed
• Estimated power at antenna:  200W

Echoes were faint, but readable.  Next, maximize power transfer to antenna and increase power to 600W.

This weekend I exchanged the LNA isolation relay by a transmit-type relay with N connectors, and replaced a short cable and N-to-SMA adapters by just two N adapters.  This gave me one extra dB of Sun vs. cold sky.  I then swept the position of the feed very carefully, but the best I could get was 9 dB of Sun noise.

KC2TDS (Jason) and I calculated the F/D of the dish to be around 0.45, which would mean that the septum feed would be OK as-is, but the dish’s manufacturer specs state its F/D at 0.37, although with inconsistent measurements.  I decided to try a scalar ring to change the feed’s beam pattern in case that would be the reason for my low Sun measurements.

I cut out an 18″ diameter x 3″ deep “Fat Daddio” cake pan to allow the feed to protrude through the center.  I reinforced the two sides of the orifice with frames cut of the waterjet.  Two brackets hold the scalar ring firmly against the feed.

I am limited by the geometry of the struts, but was nevertheless able to gain 2 dB of Sun noise vs. cold sky (for a total of 11 dB) with this arrangement!  ON0EME now comes in at  approximately 5 dB above my noise floor, making for easy copy.

I also wired the TX port all the way to the nearby shed where the PA will reside, but didn’t get to wire the RX/TX relays, so trying to transmit and receive echoes off the Moon will have to wait until next weekend.

Sun Noise vs Cold Sky, and ON0EME copy with N2QG’s 12 ft dish

12 ft dish at N2QG with 1296 MHz septum feed and scalar ring.

After two weeks of battling with my receiver setup, I finally found the problem: The output stage of my home-brewed GPS-disciplined 10MHz reference was clipping on one side, which caused the AOR AR5000 to misbehave.  My reception cleaned up as soon as I connected the rubidium reference instead of the GPS-locked source.

The feed is not optimized because I’m receiving only 8dB of Sun noise above cold sky, but this was nevertheless good enough to receive and clearly copy by ear the ON0EME beacon bouncing off the Moon.

Next: Optimize the feed and connect the power amplifier to start transmitting my own signals towards the Moon.