[Date Prev][Date Next][Thread Prev][Thread Next] - [Date Index][Thread Index][Author Index]

Re: Converter gain and noise..

Charles Suckling wrote:

There is no problem in principle in having a high noise level in a receiving system, provided the noise figure is good and there are no excessively strong signals around.  My AO-40 receiving system noise floor runs at about "S5" and I would not be bothered if this were "S9" (or more).  On 10GHz EME, my system noise floor is at about S9+20dB.

Wayne replies:

An extremely high noise level effectively widens your IF bandwidth, allowing more noise and interference to get through relative to the desired signal.  For best signal-to-noise ratio, the noise level should be very low so that the desired signal is at an amplitude where the IF filter has its narrowest bandwidth.  Perhaps this is less of a problem with radios that use brick-wall IF DSP filtering.  But surely even those radios have crystal or ceramic filters in the first IF.

For a better explanation, read the two-part article that was in QST earlier this year.  I think it was called "Maximize your receiver's selectivity".  (May be hard to find in G-land though)  Basically, you can visualize IF selectivity vs. signal strength by using an upside down version of the IF filter's frequency response curve.  For instance, I have International Radio 8-pole crystal filters in my FT-847.  These are about the best IF filters you can get, with a shape factor of 1.5.  My 2.1 kHz SSB filter has a -6dB bandwidth of 2.175 kHz, and a -60 dB bandwidth of 3.350 kHz.  If the noise level is 60 dB above the internal IF noise level, the IF filter allows 50% more noise to pass through (relative to the signal) because the bandwidth is 50% wider.  A Collins mechanical filter has a shape factor of 1.9, which would allow 90% more noise through.  A "stock" ceramic IF filter would be much worse still.  And so far I have only been describing increases in noise.  You will also b!
e subject to more adjacent channel interference with the wider IF bandwidth.

I don't plan to cut the stub in my TranSystem downconverter.  Right now with a 2 foot dish I get an S-zero signal level on 2401 MHz.  Even at S-zero the noise level from the downconverter is about 10 dB louder than the receiver's internal nose level (and that's with AGC on because there is no way to turn off the FT-847 AGC).  That is PLENTY of signal.

I would think that a system noise floor of S9+20 dB would be very undesirable for EME because that significantly increases your IF bandwidth which by definition will reduce your signal to noise ratio.  If you had 40 dB less gain in that system, your signal would go down by 40 dB but the noise would go down more than 40 dB because of the narrower IF bandwidth.  If there's another consideration that makes an S9+20dB noise level desirable, I would like to know what it is.

Wayne Estes W9AE
Mundelein, IL, USA
Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org