Richard Fry wrote:
"Roy Lewallen" wrote
Oops. I meant, WHEN RECEIVING, fiddling with the antenna or the
antenna/feedline match won't have any effect at all on the line SWR. It
will, of course, change the line SWR when transmitting.
_____________
However that doesn't mean that it's unimportant to match the feedpoint
impedance of a rx antenna to the feedline connected there.
A rx antenna with a mismatched feedline will not deliver to the rx end of
the feedline the maximum possible energy available from the fields in which
that antenna is immersed. Whatever received power that is reflected by a
mismatch at the rx antenna feedpoint is re-radiated (less losses).
What both Roy and Richard say is correct in principle, but may be
missing the point about what a receiver needs.
What's usually important is to present the RX input with the *source*
impedance it was designed for. (Most often this is 50 ohms, and let's
also assume 50-ohm line for the rest of this discussion.)
Likewise the transmitter needs to be presented with a 50-ohm load
impedance, so those two requirements coincide.
In order to achieve a 50-ohm load impedance for the transmitter, and a
50-ohm source impedance for the receiver, the antenna itself must be
matched to 50 ohms - so that's your design aim.
Now when Richard says:
Whatever received power that is reflected by a
mismatch at the rx antenna feedpoint is re-radiated (less losses),
that is true in principle, but more important is that if any energy is
reflected from the receiver input, that is perfectly OK - that energy
was "not wanted" by the receiver. The receiver *does* want a 50-ohm
source, but it only takes what it needs from that source. For example, a
simple tuned-gate FET amplifier only needs a voltage swing at the input
- it doesn't need current as well, so most of the incident power is
reflected. That type of situation is very common in receiver design, and
completely OK.
It is a myth that a receiver input is not optimized unless it presents a
50-ohm load. What it does need is a 50-ohm source impedance. The design
details about input reflection coefficient are much more complex, but
the underlying principle is simply "The RX input takes whatever it needs
from a 50-ohm source, and reflects the rest."
RX inputs *can* be designed to present a 50-ohm load impedance, even
with FETs, but this requires special design techniques that generally
involve feedback. It is usually done when some other device has to be
inserted between the feedline and the RX input, eg a filter which
requires a 50-ohm load impedance. But that device probably requires a
50-ohm source impedance too, so you still have the same requirement for
the antenna to be matched to the feedline.
--
73 from Ian G3SEK 'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.co.uk/g3sek