The SARga continues...

Following on from my last post of the strange behavior of the SARBE-5 with the antenna input connected to the Marconi 2955 test-set, ive now tried it running on battery power, with a spare 12v 7Ah SLA battery (charged to 11.1v). The issue still exists, but its form is slightly different, sufficient for me to be able to tell the 'problem' tone apart from the 1kHz modulation. The 'problem' tone is maybe 800Hz.

I also found that the set worked properly if the BNC was 'just' disconnected from the test-set, such that the center pin was physically unconnected, effectively becoming a small capacitor.

So it certainly seems its something to do with isolation from the test-set. Now, the direct connection ive made does bypass a 10pF ceramic capacitor at the antenna, but the antenna connection is still isolated from either the Tx PA or the Rx circuits by more capacitors - the antenna connects at one end of what is essentially a 5-pole low-pass filter, with the Tx connecting at the other end, and the Rx connecting in the middle (so the Tx 'sees' a 5-pole LPF, while the Rx 'sees' a 3-pole Pi-filter. I can only imaging that without the antenna capacitor, somehow this is allowing a feedback path or loading the circuits in some bizarre way. Maybe im missing something in the circuit! This is a 2-channel unit, and I have been working from the single-channel unit schematics for clarity. Is there something going on in the 2-channel set? Maybe its something to do with the AUX channel working even without pressing the receive button?

I'll tack a 10pF capacitor into my test-rig antenna patch lead and see how I progress from there.


And later...

Well, adding the 10pF capacitor into the test-bed antenna connection has solved the issue. I still have no idea why a direct connection from the test-set to the antenna filters - with no DC present - caused such a problem. I can only surmise that it was somehow loading the filter, and causing problems with the operation of the receiver circuits.

So with the receivers responding properly to the test signal, I have found that the AUX channel, currently 282.8MHz, has an MDS (minimum discernible  signal) of around 6uV, and is all but fully quieting at the specified level of 20uV. The EMG channel however, currently 243MHz, is not quite as healthy, with an MDS of around 60uV, fully quieting by 220uV, and a distinct 'squeg' before it starts to receive. With any luck this is just a bit off-tune.

So its back now to the service manual, to brush up on the alignment procedures. I'll also check out whether there is anything froody going on as far as power switching to the AUX receiver, that might account for the constant receive regardless of the receive button!

And, if I can ever work out where my compass set is - i'll mark up and etch some battery terminal plates!