In article , "Steve Nosko"
writes:

some snipping of a good post to avoid undue long length...

The accepted terminology is "Stripline" (think "strip transmission
line") for the line with two flat ground planes on either side of the
"center conductor" which can be thought of as this flattened coax....and
"micro strip" ( I have no memory aid) for the one that is one-sided where
the "center conductor" is on the top of a PCB with a ground plane on the
bottom..

That's quite probably true. Textbooks have the "correct" description
but some in the RF field get over-wrought about terminology and
techno-propriety. :-)

Either type of physical construction makes a TEM (Transverse
ElectroMagnetic) transmission line, just like coaxial cable.

I believe a 1/4 wave is not important here.

For an octave bandwidth and even coupling it is. Those directional
couplers show about less than +/- 1 db difference in coupling versus
frequency over that octave. If the object to obtain a low phase error
over frequency as well as magnitude, then there has to be adherence
to "traditional" shapes-configurations.

ANY conductor running close to and parallel to a transmission line
center conductor will couple something into the coupling line.
For a dual coupler arrangement in SWR checking by magnitude
alone the actual coupling values in db aren't all that important in
getting a relative reading of forward versus reverse.

Typical directional coupler coupling is 20 db down from the main line.

This is a matter of choice by the manufacturer construction. What this
means is that the power coming out the (correct end of the) secondary line
is 20dB down (1/100 th) from the power on the main line.

I'll say that it is a value chosen by the users...manufacturers make
all kinds of coupler coupling values but most seem to go to the 20 db
value because it is a compromise between coupled signal strength
and the effects on the main line from the coupled line and its varying
load impedances. With 20 db coupling (1/100th power as you noted),
the main line is hardly affected whether the coupled line is terminated
in proper resistive value or open or shorted.

While not perfect, directional
coupling differences of about 20 db are good enough to warrant the
name "directional."

This is a matter of primarily the mechanical design.

Mechanical AND electrical...that also influences the directivity.

The "directivity"
refers to the amount of power that gets coupled to the secondary line that
emerges out the "wrong" end--not the one we want.

"Directivity" values are the sum of forward coupling and reverse
coupling. If a coupler has 20 db coupling and the directivity is
45 db, the reverse coupling is down 25 db. [numerical example
only] "Single" couplers have only 3 ports. "Duals" have 4.

from the main line power. And don't forget conservation of energy. All
this "coupled" power is stolen from the main line.

At 20 db coupling, a bad coupled line termination results in a 1%
change of the main line power level. Small value, hard to measure
or hard to see on a meter.

They both should have a nice (50 ohm) impedance and be loaded with 50
ohms as well.

Only if that is the main line characteristic impedance being used. The
TV cable folks work with 75 Ohm Zo and have couplers for that main
line with no problem.

requires it] For the Moni-Match (snaked type) the secondary lines could be
some weird Zo and all is well if it is matched pretty well.

But...the amount of coupling CAN vary considerably with frequency
even though direction of coupling has an good relative agreement.
I'll base that on measuring the "snaked-through-the-large-coax" kind
(two different versions) intended for higher HF bands that had about
15 db difference between the two of them AND had quite a variation
in coupling over frequency. Did that about 40 years ago and didn't
get to keep the notes. Did the same with a rigid copper pipe coax
section having a slot for insertion of a coupling line about 30 years
ago and it was more even in coupling over high HF and reproducible
(based on the same measurement set-ups used by another to check
duplicates shortly thereafter).

To make duplicates of an article's presentation requires a slavish
devotion to copying EXACTLY as described. It's not a case of just
snaking it through the coax, running alongside the outer conductor
braid...which can result in a remarkable variance from the original due
to coax flexibility. With all the available copper piping in many sizes
in home repair stores, the choice of spending a bit more time rather
with a piece of flexible coax will be far more worth it in the long run.

By the way, you can pump power into the secondary line and have some come
out the mainline as well. This is actually done for special uses.

3 db couplers, also known as "hybrid couplers." Very good for making
wideband push-pull or push-push amplifiers out of modular amplifiers.

I'll just add that some directional couplers, both single and dual, are
made with double-sided PCB material, the coupled line on one side,
the main line on the other. Those seem to be good (as products) only
to about 4 GHz or so due to variation in dielectric constant and board
thickness variation. The tiny versions of the last decade use "hard"
substrates of ceramic, etc., and the so-called "blue line" of co-fired
construction from Mini-Circuits is an example. High dielectric constant
in substrate allows making them smaller and the material insures
stability and good control in manufacture.

Once something to be used for measurement is done, there is no
guarantee that it will work as planned. Separate testing needs to be
done to prove it out. That requires measurement with accuracy of
power levels of large dynamic range. The microcontroller display
power meter using an Analog Devices log detector is excellent for
that purpose, dynamic range as large as the basic log detector's
specs. Those seem to be popular in Denmark and Germany according
to the number of ham websites and commentary and photos therefrom.
Search under "power meter" and ignore the commercial hits to find
several ham sites having such homebuilt meters.

Len Anderson
retired (from regular hours) electronic engineer person