rickman wrote:

On 7/20/2015 5:59 PM, Roger Hayter wrote:
rickman wrote:

On 7/20/2015 12:22 PM, John S wrote:
On 7/20/2015 11:09 AM, rickman wrote:

So can we all agree that a source can supply a fixed power to a load?


Well, speaking for myself, YES, but within the constraints that it
cannot supply power to a theoretical short circuit nor a theoretical
open circuit. For example, EZNEC allows a power source. Do you have and
use EZNEC? I think you know you can get a demo for free, yes?

Would you care to discuss a different scenario?

I'd just like to not discuss the minutia. There are some smart people
here and I can learn a lot, but not when people start debating how many
angels can dance on the head of a pin.

Fair enough, but do let others do it because it is sometimes the
minutiae and the limiting cases that make it easier to understand the
average case.

It's not so much the minutia I don't care for, it is the *arguing* over
minutia and then insulting one another that I don't care for. I mean,
really! How many posts were made regarding whether it was
realistic/possible/worthwhile to use an example with a source supplying
a fixed power to different loads? By the time that was discussed I
don't think anyone remembered the point actually being discussed.

But that is actually a very germane discussion. Looking at the limiting
cases clarifies that the range of usefulness of a fixed power source can
be wide enough for practical purposes. Once you accept that a fixed
power source is technically quite possible for a practical range of
loads, you can analyse the the necessary matching for it. And this
leads directly to the conclusion that fixed source impedance and
"conjugate matching" of the source is irrelevant in a real system. This
is an important conclusion as some people have been making unnecessary
difficulties for themselves assuming the opposite.

--
Roger Hayter