Reception fluctuations

Dave Martin <Dave.Martin@...>

just to throw in another idea - I wonder if frequency drift is playing any part here?

The B2C2 receiver cards have a Xtal osc and should be quite stable; but as I understand LNBs they don't have a Xtal oscillator but a "Dielectric Reasonant Oscillator" (DRO) which is not so stable - and, I was wondering, perhaps it is temperature sensitive? (and suffering from the onset of Autumn?)

I haven't been able to confirm anything in my native MSG setup, but, in common with the general scepticism over the "indicators" reported by the Tellique client, have carried out following experiment which led to the above suggestion :

I have tried look at things "raw" from the LNB but spectrum analyser I have which covers those LNB output frequencies wasn't stable enough, so ....

I have an ex-broadcast-relay analog receiver (Intercontinental MT900 / Agile Omni) which has both AFC and AGC which can be manually disabled. It also has metering for Carrier/Noise, Signal Strength, and Tuning; and I have used the signal strength in the past for alignment and watching rain fade. I understand that the "Tune" metering (scaled +/- 2MHz ) is some form of indication for a PLL error signal from the AFC or similar, and normally doesn't vary much - BUT as an experiment, with the receiver tuned to an analog transmission with the AFC off, I tried cooling an old LNB and I got a noticeable change - although with the AFC back on it was more than powerful enough to pull the rx back on tune.

This receiver setup / internal meter is unfortunately not able to directly meter a digital transmission - but I have found a way and it showed the same sensitivity. Even though it couldn't demodulate it, I tuned the receiver to the frequency used for the digital MSG hotbird downlink, and with the receiver bandwidth set to the max 36Mhz I picked-off the 70MHz IF to a Marconi TF2370 s/a - this only goes to 110MHz but is ideal to monitor 70MHz IF and has dual digital storage - storing one trace before cooling and superimposing a second after cooling showed a mass shift of the whole of the 'digital hump' by just over 1MHz.

Perhaps, if the link budget is already marginal, then =maybe= this is contributing - either drifting in&out, or drifting out and the card is then having to re-lock?

Dave M
This email has been scanned by the Manxnet Mail Plus anti-virus system.