Please don't consider yourself unqualified. I don't think anyone posting
on this newsgroup should, and with your background you certainly shouldn't.

My question about who "we" meant was to establish a context for the
definition you used, which you've supplied -- thanks.

The definition you use isn't a bad one, although it might not be the
most useful, provided that you restrict the analysis to free space and
are speaking only of a single plane of the 3D pattern. This is commonly
done in discussing Yagi arrays, for example. Perhaps your experience was
largely in Yagi, log periodic, or other planar arrays which lend
themselves to this simplification. The meaning of "back" is open to some
interpretation, though. Sometimes it means the precise direction that's
exactly 180 degrees from the main forward lobe. Sometimes, though, it
refers to a range of angles, even as great as the whole rear semicircle.
Let me give an example. Suppose an antenna nominally has a deep null
directly to the rear of the front lobe. But a slight asymmetry in the
antenna moves the lobe a few degrees to the side. This could easily
degrade a strictly defined ("rear" meaning exactly to the rear of the
peak of the front lobe) front/back ratio by 10 or 20 dB. It's hard to
conceive the application where it would really change the usefulness of
the antenna. But a very slightly asymmetrical antenna would look much
worse on paper. In your experience, would you consider this to be a poor
f/b ratio, or would you give the definition some slack and allow "rear"
to vary a few degrees? If you'd give it some slack, then the next
question is how much -- could the null be skewed 5 degrees? 10? more?

In my limited experience, when the second convention is used (allowing
the whole rear semicircle to count as "rear"), the "rear" figure often
comes from the largest lobe in the "rear" region. So the gain in the
precise direction opposite the front lobe doesn't matter, if there are
lobes in other directions in the rear semicircle. This definition would
be useful for amateur beam applications, because it tells you the
minimum amount of attenuation you'll get from signals coming from any
direction within the rear 180 degrees of the pattern. Who cares that you
have a 50 dB null in one very narrow direction, if a few degrees away
the response is 40 or 50 dB greater. The definition of front/back ratio
seems flexible, sometimes used to make the measure more meaningful or
useful, but sometimes, I'm sure, to obscure the quality of the pattern.

In the example I mentioned in my earlier posting, though, of the complex
pattern of an antenna over ground, the definition can get muddy indeed.
So it's often necessary to carefully define the term and state exactly
what you mean if you really want to communicate meaningful information
when you quote a "front/back" ratio.

Roy Lewallen, W7EL

Jerry Martes wrote:
Roy

I make no claim to being qualified to discuss antennas with you when we
are in disagreement. I worked as an antenna design engineer for 15 years
till 1968 when I was layed off from TRW. I was never a high level theorist
but managed to hold a decent position with designing hardware. I did work
with some highly qualified engineers from whom I thought I obtained alot of
knowledge about antennas. Thats why I was bold enough to say "we". I
still maintain a casual relationship with George Oltman who you might know
from his association with antenna groups with IEEE.
As for the F/B, I considered that to identify Front to Back of the
antenna's radiation pattern. I would consider it appropriate to identify
the radiation toward the "Front" as the max radiation to the front. Then,
it seems that the numerical level used for the "F/B ratio" should be max to
the Back.
I make no argument that this definition I use is *the* way F/B is. But,
dont we define "side lobe level" as the ratio of the main beam Max to the
side lobe Max? Be aware, I dont write to correct your thinking. I did
consider the F/B to be flawed when the main beam Max is compared with a rear
radiation Min.

I'll consider myself corrected and stop posting.

Jerry