Ed, W1RFI, wrote, in part:
> AGC "on" is the ONLY way to make receiver measurements
> with a receiver output that would be more than about 30 or
> 40 dB above the noise floor.
>
> In making an IP3 measurement with AGC active, the steps
> are as follows:
[I have numbered Ed's steps for future reference]
>1. Set a reference generator on channel at a level that gives
> an S5 receiver output,
"on channel" = the frequency to which the rcvr is tuned,
as an illustrative example, lets choose 14195 kHz as the
test frequency, a very popular DX calling frequency which
frequently attracts a lot of powerful signals operating "up"
during split frequency DXepdititon activity. (Yes, I know this
is not the ARRL Lab test frequency, but it will serve my
point here, hi).
>2. Log its level,
S5, presumably, or maybe the equivalent in dBm?
>3. Replace that generator with a two-tone signal spaced to
> give the desired IMD,
For example, the desired rcv frequency is 14195 kHz. Place
two "strong" signal simulating generators at frequencies such
that an IMD product from them will be at 14195 kHz; as would
signals at 14200 and 14205, as 2x14200=28400; 28400-
14205=14195. Here we have one powerful signal at 14200,
just 5 kHz above the desired signal frequency, and another
5 kHz higher again. They intermodulate and produce a
spurious signal exactly on the desired DX frequency.
Set the amplitudes of these two signals to be equal and
that their IMD product at 14195 kHz has an amplitude, as
read on the rcvr's S meter at the 5 level, or near to -97dBm
(Collin's standard for S9 and 6 dB per S unit increment change).
Is the above example typical of an ARRL Lab 5 kHz signal
spacing test for "in-band" amateur radio performance test?
Of course, for the 20 kHz strong signal spacing test, to
yield an intermod at 14195 (or whatever the chosen test
frequency) the "testing" signals would be, at say, 14215 kHz
and at 14235 kHz, and again 2f1-f2 is 14195 kHz.
>4. Adjust the level of the two-tone signal until the receiver
> gives an s5 output,
For the intermod IMD product at 14195 kHz, correct?
>5. Log the level of the two tones and do the IP3 calculation
> from those levels.
Ok, that's as far as I want to go now. First want to be
sure I have a typical test process clearly understood,
and, of course, all the above with the AGC "On".
The above example ought to be done using the 2.4 kHz
BW rig filter, as interested here in SSB IMD performance,
for this particular frequency example.
Were it to be a CW test, then do all with the test frequency
moved down to, say 14025 Khz, with the intermodulating
signals at 14030 and 14035 kHz. Either test would be good
for "evaluation" for comparative purposes of rigs "competing"
for DX or during a large CW contest weekend with many
powerful station operating in the same city, or area,
such as the SF Bay area where things can be really
busy on a contest weekend, hi. And for BW, use the rigs
500 kHz filter.
My entire purpose here is just to insure I know/understand
the test procedure and method.
73, Jim KH7M
|