On Jul 16, 11:30 pm, isw wrote:
In article .com,
Keith Dysart wrote:
No, that was indeed the claim. As a demonstration, I've
attached a variant of your original LTspice simulation.
Plot Vprod and Vsum. They are on top of each other.
Plot the FFT for each. They are indistinguishable.

-- lots o' snipping goin' on --

OK. I haven't been (had the patience to keep on) following this
discussion, so I apologize if this is totally inappropriate, but

If the statements above refer to creating that set of signals by using a
bunch of signal generators, or alternately by using some sort of actual
"modulation", the answer is, there is a very significant difference.

In the case where the set is created by modulating the "carrier" with
the low frequency, there is a very specific phase relationship between
the signals which would be essentially impossible to achieve if the
signals were to be generated independently.

All true. The simulation offered previously achieves the
required phase relationship (and more, so that the sum and
product versions can be directly compared).

In fact, the only difference
between AM and FM/PM is that the phase relationship between the carrier
and the sideband set differs by 90 degrees between the two.

I am not convinced. Can you explain? AM modulation with a single
frequency produces a single sum and difference for the sidebands
while FM has an infinite number of frequencies in the sidebands.
This does not seem like a simple phase difference.

....Keith