On 12/27/2011 1:23 PM, Commander John wrote:
> That is about what I said. But it doesn't make the signal any wider.
> please read the previous posts as to what I was commenting on.
> john W9ZY
Let me take a shot at this:
I'm sure some one else can do this more clearly than I but...
First, lets separate the lower stages from the driver and final.
Lets say your transceiver has an audio pass band of 0.3 to 2.8 KHz or
2.5 KHz. Now lets hit it with two tones. One at 300 Hz and one at 2.8
KHz and make the foolish assumption that this is still a clean 2.8 KHz
wide signal when it gets to the final.
We now have two tones that on upper side band are your carrier
frequency plus 300 Hz and plus 2.8 KHz or still 2.5 KHz wide if all went
well. Now that final can not only pass and amplify these two signals
at 750 watts each, but if it has only - 20 db 3rd order IM we also now
have signals at 2.8 KHz +/- 300 Hz or 2.5 Khz and 3.1 KHz. 20 db down
(If I understand this correctly) we also have 300 Hz +/- 2.8 KHz or 3.1
KHz and a - 2.5 KHz 3.1 and a - 2.5 KHz below the carrier for 5.6 KHz
band width instead of 2.5. The signal is over twice as wide. What if
we have 2KHz and 2.8 KHz. Now we have 800 Hz (still in the pass band)
and 4.8 KHz but we also have 2 KHz - 2.8 KHz or a - 2 Khz (below the
carrier) and we are now up to 6.8 KHz wide. IOW in a two tone test we
have the sum and difference between *both* frequencies. In the latter
example both sums fall on the same frequency while but the differences
fall on different frequencies. Those that fall on the same frequency
depend on the phase as to whether they add, subtract, or fall some where
In either case the IM does cause the signal transmitted to be wider than
we'd hear from an amp with -50 db IMD products and I think they'd be 30
db stronger. You can keep taking the results and running the sum and
difference on all of those and then repeat with all of those results
with diminishing strengths each time.
I'll leave it to others to correct my work as I've never gone through a
math problem without making a mistake yet even though I have a minor in
It may not be correct but it should convey the general idea.
Amps mailing list