[BC] "Cusp Rotation" /was/ Air-Dielectric Sections

[Phil Alexander]

This whole "phase rotation" / "cusp rotation" business 
drives me slightly nuts, because what we are really 
attempting (as best we can) is indirectly "de-rotating"
or moving the **actual** PA load at the true output of
the PA to R = x j0 at the operating frequency with the
sideband frequencies "horns up." Where "x" is the true
output resistance of the power amplifier(s). IOW it is
almost like what we did by tuning and loading a tube
box "back in the day" by operating slightly off
resonance for best load to yield best modulation b/w,
except today we must approximate it and set it in the
blind.

What is really needed is for manufacturers to include
a dynamic measurement point(s) for the **actual** 
dynamic, complex impedance across the range of operating 
frequencies as seen at the **true** PA output before 
combining/matching to the Tx output spigot. 

That is to say something approaching the idea of a 
**true** PA load sample that might be read by a RF 
vector analyzer. If someone would invent this and all 
manufacturers adopted it, then it would be a simple 
matter of adjusting the ATU or phasor input/CP matching 
network to yield a true 12 o'clock "cusp rotation"
on all PA's which is what we are now trying to do (not
always exactly) by so-called "cusp rotation." Boxes 
could then operate into loads that would make them 
very happy, and routine maintenance of IBOC systems
for optimum performance WHILE MEETING LICENSED DA 
PARAMETERS would become a relatively simple task.

Going one step further, automation of this match by
the Tx (within reasonable limits) might do for broad
band matching what complementary pre-distortion has
done for matching load non-linearity.

My concern is that although so-called "cusp rotation"
is carefully done at installation of a shiny new box
in an IBOC system, the first time the phasor has to
be tweaked to bring MP's into alignment, all that 
careful work may be compromised. OTOH if we had a
measurement means as described and the equivalent of
a "zero phase shifter" network in the Tx load, it
would be a simple matter of firing up the vector
scope and tweaking the "horns" back to 12 o'clock,
regardless of Tx brand.


---------------------------------------------
Phil Alexander, CSRE, AMD
Broadcast Engineering Services and Technology 
(a Div. of Advanced Parts Corporation) 
Ph. (317) 335-2065   FAX (317) 335-9037



On 9 Mar 2007 at 16:49, Chuck Lakaytis wrote:

> Yep, we are going to have to install some phase rotation correction in all 
> of our nine AM's before we turn on HD.
> 
> ----- Original Message ----- 
> From: "Mike McCarthy" <Towers at mre.com>
> 
> > In today's age of absolute phase rotation presented to the transmitter 
> > being important for best broadband operation, treat the cable as you would 
> > in a DA and factor it in the design of the whole plant....DA or N-DA.
> >
> > MM
> >