Hi again,
Thanks to Ed, W1RFI, for his comments about the coming
Orion IP3 tests by the ARRL Lab.
My REAL interest in all of this is to know: how good is the
rig in a tough DX pile up, or during a messy, IMD full band
on a contest weekend. Or, not what it takes to generate
an IMD up at S5 or higher, but what is going on to cause
all that racket, especially in the SSB band segments when
the band is very busy? All of that "junk" is usually heard
down around signal levels of S1 or S2 and the junk is
covering weak signals I would like to hear/work. Or,
how good is the rig at rejecting IMD's at those levels and
what in-band signal levels are causing that stuff?
[Yes, I know, some of that junk is "splatter" and spurs
from others linears which are not linear and are way
overdriven. A high IP3 number in our rcvr will not
cure racket from other folks splattering transmitters]
The answer would come from knowing the IP3 number at
input signal levels where we know the rig is running
in the linear small signal region, before the AGC has
been activated. For the Orion, per some of Doug
Smith's writings this should be up to around input signal
levels of about S9 + 35 dB!! See:
http://www.doug-smith.net/orion.htm
where Doug writes, "automatic gain control (AGC) does not
have to be used until signal levels reach S9+35 dB."
Reading all he has to say in that piece about AGC action
and the dynamic range of the Orion is VERY interesting,
as is his somewhat more technically detailed AGC discussion
at: http://www.doug-smith.net/digitalagc.htm .
Ok, armed with this information about the Orion AGC trigger
power level (or approximately that level anyway), S9+35 dB,
is a level of -38 dBm. That is a good deal lower than the
signal levels used by either TT in their tests or that planned
for use by the ARRL Lab procedure.
So, before the Orion AGC systems are activated, what IMD's
might we be expecting? Well, let me calculate what IP3 number
would be needed to hold IMD's generated by a pair of -40 dBm
signal in the rcv'ing pass band, that should be just below the
trigger power level for the Orion AGC action to begin, if I under
stand Doug's pieces.
I would want the IMD's to be below, say -125 dBm, about
4 dB down from the S = 1 signal level of -121 dBm (Collins standard).
The difference between my two -40 dBm signals and my
required -125 dBm spurious level is 85 dB; therefore, the
IP3 number needed will be 1/2 of 85 dB added to the power
of -40 dBm. Or -40 dBm plus 43 dB. That is an IP3 number
needed of at least +3 dBm..............now that means even
the Argo V will come very close to meeting my postulated
requirement, hi. Actually, Doug argues that it would be
nice to have no IMD's above the limiting phase noise
level of the radio. For the Orion, this means that a
spur could be no stronger than just above, say -138 dBm
or some such, looking again at the "composite noise"
curve at the TT web site about the Orion. So, what
should be the IP3 number for the radio which can keep
the IMD at or below -138 dBm when two intermodulating
-40 dbm signals are within 5 kHz or so of each other?
Same arithmetic: the difference between the required IMD
number and the two equal signal's power is 98 dB; add half
of that to -40 dbm, and you get + 9 dBm for IP3.
Recall, we are doing all of this with input power levels below
the AGC action trigger level. And I am very pleased with
the results! The Orion is going to very good indeed with
overall spurious free dynamic ranges which will result per
Doug's design (well, I am sure other Ten Tec engineers
did also contribute to the Orion design, hi). And that across
the entire power range without and with signals in the
AGC "power level spectrum".
I am really looking forward to my coming Orion (2nd production
run, whenever that will come out), as my current daily
use rig is my "old" Corsair II. Per the ARRL Lab report
about it, published back in August 1987, the third order
IP is negative; e.g., 80 meters, -1 dBm, and 20 meters,
-4dBm. The Orion will be a big jump up from there!!
BTW, in my Corsair II, the AGC is fully active by the
time incoming signals are reading around S3 or 4 on the
rig's S meter; that's a whole lot earlier than will be the
case using the Orion.
73, Jim KH7M
|