Modifying the RX2
The RX2 is a purpose-designed weather satellite receiver kit, of which some 700 have been sold. I was fortunate in being able to discuss the problem and my ideas for solutions with the designer, and we agreed that the problems were most likely happening in the mixer stage.
The easiest modification to carry out is to move the mixer drain to the centre-tap of the first IF transformer. This reduced the gain, and should result in lower signal levels at the mixer drain. However, there was little or no difference to the pager interference. I therefore concluded that the problem was with too much signal going into the mixer.
The next step was trying to alter tuning of the 137MHz circuits so that the gain was maximised towards the low frequency end of the band, but as the measurements below show, it was difficult to achieve sufficient tilt in the frequency response. I ended up tuning the circuits for maximum on channel 1, and then turning just a fraction more so that the meter dropped perceptibly - estimated about 0.5 to 1dB beyond channel 1. I concluded from this that the signal-to-noise ratio did not depend strongly on correct tuning, which I should have realised at the start, I suppose. However, the amount of interference does depend on tuning, so I though it better to start measuring gain rather than simply SNR.
At this point, I became interested in how the filter had managed to solve the interference when the loft QHA was in use, so I checked the gain as well. You can see that the filter had a net insertion loss of some 2-2.5 dB, with some extra loss at the H.F. end, but not a lot. So I wondered what might happen if I reduced the RF gain by the same amount as the loss in the filter - presumably I would be better off because I would not reduce the noise factor by as much as I reduced the intermodulation.
I tried various resistors in parallel with R41, aiming to get about a 2dB gain
reduction. This, coupled with the detuning gave me about a 7dB gain loss at the high
end of the band, which result in about 14dB improvement in intermodulation rejection.
In reality now, pager signals no longer cause NOAA 14 on 137.62 MHz to break up, although they can cause bad picture degradation. RESURS on 137.85 is usable at weekends, when there is less interference, but results on NOAA 12 and 15, and Meteor are much improved, and I get many passes free from pager interference. Sadly, mains supply interference and a spurious from a badly adjusted transmitter remain to mar my enjoyment. If I could get a cavity filter that had the right characteristics, I might do so, and if I was convinced that a professional QHA would reject the interference, I might buy one. Unfortunately, the QHA will have better response near the horizon than the turnstile!
Should you retune your RX2? I would suggest only doing so if you have at
least some test equipment to check where you are, as simply detuning the coils and turning
down the gain could easily loose the satellite signals.
After getting a QHA, I did find that the problems became rather worse, as I had feared. Thus it was time to revisit the filter, and see what could be done.