"Jim - NN7K" wrote in message
.. .


Jerry, just because you are wanting to recieve, instead of transmit,
doesn't mean that the V.S.W.R. mismatch isn't there-- nor its effect
on the subsequent coax loss's being multiplied because of the mismatch!
Remember- the antenna , coax, and the mismatch are all bi-directional !
The additional loss in the feedline, because of a high swr, is also
present in the recieve direction! If you have copy of ARRL handbook, or
other sources, look under "transmission lines" - these have a chart of
coax loss, and also the swr effect on that coax loss.
For example, coax with 4 dB loss, at swr of 1.5 /1 would raise the line
loss by an extra .1 dB, while an swr of 3:1 would rais it by an extra
dB, at 5:1, would add about 2.2dB loss! and- keep in mind this is
BI-DIRECTIONAL loss ! Another consideration is that the noise figure is
optimized for 50 ohm, and this (because of a mismatch), will also be
negatively affected! Best to find good impedence match for all
considerations-- Jim NN7K


Jerry Martes wrote:

SNIP
In addition, I have assumed that the impedance mismatch for receiving
antennas can be allowed to be much greater than for transmitting. This
statement is directed to the fact that transmission line loss is
increased
by VSWR, and the reciever isnt restricted to being the same impedance of
the
transmission line.

Jerry


Jim

I sure wouldnt argue with your findings. I guess this identifies my
sloppy engineering. I didnt think 4 Db was a tolerable feed line loss
under any but the most dire circumstances. I wonder what a ~3:1 VSWR would
do to add attenuation to feed line loss os, say 1 1/2 db.
To further identify my sloppyness, I cant distinguish zero db from 0.1 db.
It is even difficult for me to determine with any confidance the difference
between Zero db and 1 db.
What I'm getting at is -- I still contend that a receiver can tolerate
some mismatches that arent tolerable in a sophisticated transmitting
station. Richard Fry has pointed out how critical small mismatches are in
FM transmitters. This original post referanced a FM transmitting antenna.
It would be reasonable (to me) to assume this referanced antenna could
have some rigid constraints if it was to be used for transmitting stereo FM.
But, as a omniazimuth FM receiver antenna, the phasing might be fairly
straightforward.

Jerry

Jim - NN7K wrote:


Jerry, just because you are wanting to recieve, instead of transmit,
doesn't mean that the V.S.W.R. mismatch isn't there-- nor its effect
on the subsequent coax loss's being multiplied because of the mismatch!
Remember- the antenna , coax, and the mismatch are all bi-directional !
The additional loss in the feedline, because of a high swr, is also
present in the recieve direction!. . .

That's not quite correct.

The SWR on the line is determined solely by the transmission line
impedance and the load impedance. When transmitting, the antenna is the
load. When receiving, the receiver is the load. If the receiver and
antenna have different impedances, the line SWR will be different when
transmitting than when receiving.

Fiddling with the antenna or the antenna/feedline match won't have any
effect at all on the line SWR. It will, however, have an effect on the
strength of signal arriving at the receiver.

Roy Lewallen, W7EL

Jerry-- at your stated 1.5 dB loss coax, with a 3/1 vswr, the extra loss
would be just under .7dB, but remember that the reflected mismatch, also
will affect your recievers front end noise (factot/figure), most likely
in a negative way (no way to calculate the impedence presented to your
rcvr front end ) also, remember that fm sensitivity (because of its
nature) is not near as sensitive as cw/ssb (typical 2 meter rcvr can
detect a signal below .01 microvolt cw - fm rcvr does good around .1
microvolt- 20 dB more sensitive! Something to think about-- jim NN7K

Jerry Martes wrote:

"Jim - NN7K" wrote in message
.. .


Jerry, just because you are wanting to recieve, instead of transmit,
doesn't mean that the V.S.W.R. mismatch isn't there-- nor its effect
on the subsequent coax loss's being multiplied because of the mismatch!
Remember- the antenna , coax, and the mismatch are all bi-directional !
The additional loss in the feedline, because of a high swr, is also
present in the recieve direction! If you have copy of ARRL handbook, or
other sources, look under "transmission lines" - these have a chart of
coax loss, and also the swr effect on that coax loss.
For example, coax with 4 dB loss, at swr of 1.5 /1 would raise the line
loss by an extra .1 dB, while an swr of 3:1 would rais it by an extra
dB, at 5:1, would add about 2.2dB loss! and- keep in mind this is
BI-DIRECTIONAL loss ! Another consideration is that the noise figure is
optimized for 50 ohm, and this (because of a mismatch), will also be
negatively affected! Best to find good impedence match for all
considerations-- Jim NN7K


Jerry Martes wrote:

SNIP

In addition, I have assumed that the impedance mismatch for receiving
antennas can be allowed to be much greater than for transmitting. This
statement is directed to the fact that transmission line loss is

increased

by VSWR, and the reciever isnt restricted to being the same impedance of

the

transmission line.

Jerry



Jim

I sure wouldnt argue with your findings. I guess this identifies my
sloppy engineering. I didnt think 4 Db was a tolerable feed line loss
under any but the most dire circumstances. I wonder what a ~3:1 VSWR would
do to add attenuation to feed line loss os, say 1 1/2 db.
To further identify my sloppyness, I cant distinguish zero db from 0.1 db.
It is even difficult for me to determine with any confidance the difference
between Zero db and 1 db.
What I'm getting at is -- I still contend that a receiver can tolerate
some mismatches that arent tolerable in a sophisticated transmitting
station. Richard Fry has pointed out how critical small mismatches are in
FM transmitters. This original post referanced a FM transmitting antenna.
It would be reasonable (to me) to assume this referanced antenna could
have some rigid constraints if it was to be used for transmitting stereo FM.
But, as a omniazimuth FM receiver antenna, the phasing might be fairly
straightforward.

Jerry

Roy Lewallen wrote:
. . .
Fiddling with the antenna or the antenna/feedline match won't have any
effect at all on the line SWR. It will, however, have an effect on the
strength of signal arriving at the receiver.

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.

Roy Lewallen, W7EL

.... and, though it may not have importance at HF, any loss in the
transmission line (unless it is very cold) will add noise at the same
time that the signal is attenuated.
Once upon a time, serious consideration was given to using liquid
air (might have been Nitrogen) to cool a rather short piece of waveguide
(between feed and first receiver stage) in a really high frequency
system that was pointing out into space. Such cooling would not have
changed the attenuation a noticeable amount, but it would have improved
the SNR.
... and further: please do not think of using the
maximum-power-transfer theorem to maximize SNR. The first stage needs
to see a (small) mismatch, which might not be seen by the transmission
line.

With a low directivity antenna in the absence of close man-made
noise sources, the above issues are usually of no importance at HF and
below because the SNR is almost always (in a reasonably well designed
system) determined beyond the antenna. [Obviously, a highly directive
antenna system could dramatically affect SNR]

73 Mac N8TT
--
J. Mc Laughlin - Michigan USA
Home:

"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).

RF

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

"Ian White, G3SEK" wrote
... the underlying principle is simply "The RX input takes
whatever it needs from a 50-ohm source, and reflects the rest."
______

The above statement might be read as though there is a threshold voltage
limit above which a receiver will not deliver improved performance.

A receiver amplifies and detects whatever voltage is present at its input
terminals, if just thermal noise. Even if no current flows in the input
stage device itself, the wanted signal voltage present there should be as
high as possible above the thermal noise voltage in order to maximise SNR.

RF

Richard Fry wrote:
"Ian White, G3SEK" wrote
... the underlying principle is simply "The RX input takes
whatever it needs from a 50-ohm source, and reflects the rest."
______

The above statement might be read as though there is a threshold voltage
limit above which a receiver will not deliver improved performance.


If you truly believe that danger exists, then please don't make it worse
by quoting my statement out of the context in which it was made.


--
73 from Ian G3SEK 'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.co.uk/g3sek

Hello again,

Well, surprise, surprise...the thread deviates again! Hehe.

If i may add to the discussion:

By "reciprocity", a transmit antenna can also be used as
a receive antenna, assuming you want the same polar pattern.

A mismatch on the receive side will adversely affect the signal
to noise ratio, or would increase the noise factor of the system (which
is why you always want your low noise amplifiers as close to the
receive antenna as possible, usually mast-mounted, to avoid the
losses of a long coax).

However, at the very least, the mismatch on the receive side
will not result in catastrophic destruction of your output transistors,
which is what a mismatch on the output of a transmitter can result in.

So one mismatch is a bit more serious than the other.


Slick