I had decided to try to put together some form of preamp for the next EME listening tests (ARRL contest in November). This was after discovering that 15m of Aircell 5 has 3dB of loss at 70cm!

The quickest and easiest thing to try was something I had already started some time ago; another bit of board sawn out of a nice commercial board with a BFG541 on it. This is a 9GHz NPN transistor with a typical noise figure of 1.3dB, so not a bad bet for a quick fix. I had used one before in the 70cm PA box, as the FT-817 I find a bit deaf at 70cm. Here’s the board:

It’s proved useful in operation, with whatever matching components are present (a pi filter by the looks of things) on the input! Here it is in context in the 70cm PA control box:

I used a surface mount relay for switching the antenna input; it’s a 3GHz type, and only has to carry the RF, not switch it, as the sequencer makes sure there’s none there when it changes over. The big relays are for the DC.

I hadn’t got another identical board, and cut out a chunk that looks likely; here it is with the input components removed, apart from a blocking capacitor.

I decided to add a tuned circuit and input trimmer to try to match this, and it does tune (rather broadband). Here are the additions:

So with that put together it was tiome to do some tests.

First of all I assembled the setup as before, using the 15m of Aircell 5 as a feeder. Here’s a screen grab showing the display as things are switched on; a very low level below the FT817 is switched on; that gives a level of about 38dB. Adding the cable to the PA box adds about 7dB, then switching on the preamp in the PA box adds a further 7dB. These were then all switched off in turn.

I then pointed the antenna at the Sun and away, with pauses on the Sun and on “cold” sky. This time I moved the antenna just up and down in altitude, and you can clearly see the first antenna lobe as we move off and on to the Sun. There are the usual short upward glitches caused by local noise. You can see three sets of moving onto the Sun, then away to cold sky. This time we seem to have about 3dB of Sun noise, a little more than last time.

I then tried the same thing with the preamp. It took a while to realise that I was clearly overlaoding the FT817, and had to reduce the receiver gain in the 817 to get a reasonable signal. However I got something that seemed reasonably sensible. I think that the system is a little more sensitive, and shows more like 4 or 5dB of Sun noise. The right hand half of this plot shows what I ended up with.

Here’s a page with some numbers:

I’m not sure what I will do at the weekend for the ARRL test. I will probably start without the preamp, especially if I put together a shorter feed. I may then try with the preamp as well. We shall see. I’m very interested to see if I can see and decode any more than I did in 2012!