While I've long thought that trying EME someday would be fun, the motivation for this project came from my daughter Elizabeth KC0OTR. I asked her shortly after she received her license a couple of years ago at age 10 what aspect of ham radio she thought it would be interesting to try. There are so many different facets of the hobby, I had no idea what she'd say. With no hesitation at all, she said "Moonbounce, Dad!".
The base is a single 10 foot section of used Rohn 55 that was also parked out behind the shed. Over several days I laboriously dug a hole for it, cutting through several inches of rock shelf and in to the mixed gravel and clay below. As I've done before on other towers, I started with pick and shovel, and ended up with my digging bar and heavy-duty Sears shop vac. Yes, I'm serious. The nature of our "soil" is decomposing granite and clay. So once the hole gets more than a couple of feet deep, by far the easiest technique I've found is to use a digging bar to break up the bottom, and then a vaccuum cleaner to suck all the loose material out. It sounds crazy, but it works great for me!
My son Robert helped me mix and pour the concrete to lock the tower into the hole. Because of the rock shelf, I only put a bit over 3 feet of the section in the ground. By belling the hole out a bit below the hole in the rock shelf, I'm confident that it's never going anywhere. The result is that we've got about 7 feet of tower sticking out of the ground.
For the elevation axis, I rapidly reworked the polar mount. The only parts I used other than what was there are a couple of lenghts of angle to move the mount point of the actuator away from the mast so that it wouldn't interfere, and a pile of big flat washers.
For the azimuth axis, I fabricated everything from scratch. The key component is a commercial gear reduction drive that John Conner NJ0C and I picked up at Dayton a few years back that's been collecting dust in my basement. I bought a couple windshield wiper motors for a few bucks each on eBay to drive the input through a Lovejoy split coupling. The output of the gear drive goes through a custom turned shaft to a 12 tooth 50 pitch chain sprocket. To mount the gearbox and wiper motor, I cut up a couple of steel blank rack panels and some 1 inch square steel tubing and welded up a custom mount. The idea is that the tubing mounts to the side of the tower with u-bolts, and the panel is bolted to the tubing through slots that allow it to be slid left and right to adjust the tension of a chain drive to the mast.
The bottom of the mast is a bearing assembly I built up using a surplus 2 inch, 4-bolt flange bearing, a 48 tooth 50 pitch steel sprocket, and some bits of pipe. It attaches to the tower using a shelf made up of 1 inch square tubing u-bolted to the legs. There is very little vertical load on this shelf, it's mostly there to keep the mast centered in the tower. The vertical load is all carried by a Timken tapered roller bearing that I bought very cheaply on eBay, that sits on a structure I welded up from angle and pipe to sit on the top of the tower. From this bearing up, the mast is used 4 inch schedule 40 pipe. There is a section of used 3.5 inch schedule 40 pipe that runs from the bottom bearing assembly up through the top bearing assembly, and which is tied to the top mast using clamping bolts.
Everything is grossly over-engineered for the current 10 foot dish... hopefully enough to be able to handle a larger dish someday without too much rework!
I purchased a kit of parts to build a septum-style feed for 1296 Mhz from Gary Abercrombie. We'll start with a surplus HP relay and DEM preamp on the receive port. I only have about 75 watts of amplifier on this band, so I'm talking to various friends about buying a higher powered amp. I'm not adverse to trying the JT heavily-coded approach, but I'd be a lot happier with 250-500 watts. One complexity is that the dish will be about 225 feet of cable run from the main operating position. I therefore would prefer to put most of the electronics in the shed at the base of the tower adjacent to the dish location to reduce feedline loss, but that shed has no environmental controls (not even insulation!), so using a water-cooled tube amp would be challenging at best.
I'm sure we'll figure it all out.
At EME2004 in New Jersey, Paul Wade gave a talk on multiple-reflector antennas that included a reference to the Axially-Displaced Ellipse antenna. I find this design fascinating, and the idea of building a 6m main reflector and associated subreflector in this configuration to use for EME, satellite, and radio astronomy purposes could be a serious challenge for my meager skills as a home-shop machinist... Some papers about this design are on the web in Brazil.