[1000mp] S-meter and S/N [was Real S9] - Part 2 [somewhat long]

[Tod-ID]

In the original posting of this topic I referenced my web site which has the
details of an experiment run to determine the signal strength versus S-meter
reading for my FT1000MP following the addition of the INRAD roofing filter.

The resultant curves indicated (1) the S-meter readings did NOT increase
linearly with increases in signal; (2) the S-meter readings were
approximately one S-unit per 3 dB signal change and (3) for my radio a
setting of IF Gain = 11 provided an S9 reading for a signal of approximately
50 uVolts [-73 dBm] and an S0 signal reading of -100 dBm.

The earlier discussion on this topic on the reflector seemed to
suggest(among other things)that people were setting the IF Gain 'high' so
that they could make sure they could hear all the weak signals. My advice
was to set the IF gain so that when you added an antenna you could hear an
increase in noise in the receiver audio.


Yesterday I decided to run a follow on experiment. In this experiment I used
a signal combining transformer to add a signal from my signal generator to
the signal from my TH7DX and then send the combined signal to the radio. The
beam was aimed toward a quiet direction (South Pacific) to reduce the
antenna noise. The IF Gain was set to 11 and adding or removing the antenna
is quit noticeable even when there are no signals in the pass band.

There is about 3 dB loss for signals moving through the transformer.
Initially I tested the system with no antenna attached and only the injected
signal from the signal generator. I was able to hear without trouble a
signal of less than -130 dBm. This is in line with the measurements made of
the Minimum Detectable Signal (MDS) done by others. I then added the
antenna. The noise increase was sufficient so that the MDS in the presence
of my antenna noise was about -124 dBm. 
Finally, I tuned around for some weak 20 meter signals and then increased
the signal generator input until I felt the generator and the on the air
signal were at the same audio level. This is very dependent upon my
particular hearing but I was too lazy to stop and connect things to one of
the audio spectrum programs that are available. I figured for my purpose if
I thought they were about equal that is what counted. 

The result was that signals at -118dB or higher [less negative dBm] were
easily copyable. When the signal got to -120 dBm or less it took a very high
level of concentration to feel I had copied correctly.

I did these experiments using the 500 Hz filters and I listened for on the
air CW signals. Since the rule of thumb for bandwidth is that S/N changes
linearly but inversely with bandwidth changes, I would expect that had I
used the 2.8 kHz filters, the MDS in the presence of noise would be 5 to 7
dB less -- say -112 dBm. Adjusting the IF Shift and Width controls improved
the ability to hear the signal and could have improved the MDS by several
dB. 

All of the signals we are talking about were less than S0 on the S-meter but
definitely more than ESP signals. The IF Gain setting of 11 does not really
compromise what the radio can hear. If the atmospheric and ambient noise at
my location had been lower I would have expected to hear signals much lower
than -118 dBm without a problem. Looking at the antenna output on a spectrum
analyzer it is easy to see that the background noise of my antenna in that
direction is about -120 dBm which is a check on what I was measuring using
my hearing and the signal generator. 

I hope I have not overstepped the reflector rules by submitting such a long
message. If I have I apologize. I was simply interested in providing some
measurements to support my earlier statement that "setting the IF Gain to a
value that just provides a noise increase when you attach an antenna" is
sufficient to get good results from your FT1000X.

Tod, K0TO