Owen Duffy wrote:
It is a leap to move from "can be thought of as power" or "has the
dimension of power" to your statement (which you attribute to HP AN95-1)
"The beauty of an S-Parameter analysis is that if one squares the
normalized voltages, one gets power." Did AN95-1 state clearly that which
you suggest?

The answer is "yes, they did." But, if you insist, I am willing to
change what HP said to "one gets the dimensions of power, i.e.
joules/second." As an engineer, I am content with getting close
enough but I am always agreeable to accommodating the purists.

Nowhere in Chapter 1 of AN154 do they perform alegebraic operations on
power, the chapter is full of expressions, but they do not use |Sxx|^2.

How about Chapter 2? :-)

I will have to check out that Ap Note. In the meanwhile, maybe
you should take a look at Ap Note 95-1 from which I will quote
one example (there are others) from page 17:

|s11|^2 = Power reflected from the network input divided by Power
incident on the network input. Please adjust your thinking
to agree with HP's.

So not you are superposing power to "yield" a resultant power.

My native language is American English but I cannot parse that
statement. Care to try again, maybe in the Queen's English? :-)

Power density can be added using the scalar intensity, irradiance,
or Poynting vector equations, but POWER CANNOT BE SUPERPOSED!!!
I don't know how many times I have to repeat that statement.
--
73, Cecil http://www.w5dxp.com