Topband: In-Band IMD Test

[Bill Tippett]

Hi Tree,

N6TR wrote:
 >IMHO (getting really brave jumping in here - so emphasis on the H),
the "in band" IMD is a very important feature when using a receiver
that will be dealing with lots of noise (like Topband).

 >Think of noise as a bunch of little signals in your passband.  Think
about how the signal you are trying to receive interacts with that
noise (which is about the same strength - right?).

         I've never paid much attention to that test since it is mainly
intended as a test for internally-generated SSB splatter.  I think there
could be  2 flaws in your assumption that the test may be a good
indicator for noise floor CW interference on 160.

1.  Interfering signals are at S9 and S9+60.  These are not typical
noise floor levels unless you are trying to operate in the thunderstorm
season.  Even if you have S9 noise and a "signal about the same
strength", the IMD products will be ~30 to 50 dB down.  Thus they are
so far below the S9 noise that they are insignificant QRM for a " same
strength" signal at S9.  I don't think any rig can copy signals 30-50 dB
below the ambient noise level.

2.  These tests are done with SSB filters and 200 Hz spacings.  For
CW, "in-band" would have to fall within the standard 500 Hz BW, and
you would not normally have someone operating within +/- 250 Hz
of your TX frequency.   IMHO, the one place where this could be a
problem is in a massive simplex pileup, and then there many other
problems to contend with also (IMDDR3, phase noise, clicks, etc).

http://www2.arrl.org/members-only/prodrev/testproc.pdf

"5.15.1 The purpose of the In-Band IMD Test is to measure the 
intermodulation-distortion (IMD) products present in the audio output of 
the receiver. This test is typically performed on units that will undergo 
the expanded set of tests. The receiver will be operated in the SSB mode at 
14.200 MHz. Two input signals, spaced 200 Hz apart, are applied to the DUT 
and adjusted for equal output amplitude. (The audio tones are approximately 
900 and 1100 Hz.) The input signals are adjusted for approximately a single 
tone n S-9 level and the audio output is observed on the Spectrum Analyzer. 
The tones are set for a 0 dB level for a single tone reference. (This is 
unlike the transmit two-tone test, where the -6 dB is used for a PEP 
reference level.) The AGC fast option is used for this test."

 >I have started seeing this in the extended product reviews the ARRL
does.  For example, on page 23 of the extended product review of the
IC703 shows inband IMD - showing the typical -35 db products.

         This test has been a part of ARRL's test procedure since
1998, but only in Expanded Test Reports for "major" rigs.  I
believe it is mainly intended to be a test of audio quality
on SSB.

                                                 73,  Bill  W4ZV