[BC] Harmonic measurements

[Phil Alexander]

Thanks Ed,

This was the reason I suggested the scan should be done
by someone fully familiar with the methods. ISTR there
is a 6 dB/octave correction involved. That multiplied
over 9 or 10 octaves would be quite significant.

I'll stand by my original advice of getting a calibrated
sweep of the station's emission made by someone qualified
before pointing blame at the station which might be
entirely clean. Secondly, I'll point out that if the IX
problem is generated within the cellular equipment then
there is nothing the station can do to resolve the issue
short of going off the air.

You bring up a very interesting example of unintended
consequences about "adaptive correction." The real problem
is, I think, semantics tending to obscure the obvious.
If you consider the true nature of "real time adaptive
correction" and call it what it really is, i.e., dynamic
***pre-distortion*** of PA linearity, then it becomes
apparent that distortion introduced to compliment load
non-linearity at the fundamental frequency probably will
exacerbate system non-linearity at the 9th harmonic in
most cases resulting in greater harmonic content in the
transmitter output. In retrospect it seems obvious that 
this would be the unintended consequence, but I wonder if 
this was given much consideration at the time when the
intentional, complementary pre-distortion idea was 
developed a few years ago.

Phil Alexander, CSRE, AMD

-----Original Message-----
>From: Edwin Bukont <ebukont at msn.com>
>
>Phil Alexander, if I may offer a clarification on your comment. 
> 
>While the measurment standard for analog FM stops after a 
>few harmonic intervals, stations running HD should do
>measurements out to the 10th harmonic. 
> 
>We have seen a fair number of instances with interference
>at the 9th harmonic, sometimes caused, rather than mitigated,
>by real time adaptive correction.  The key is to be sure 
>you are obtaining a valid sample to a measuring device with
>suitable displayed dynamic range and low enough noise floor.