Dave and the gang.
You have asked for a tall order in understanding the 180 degree phase
inverter added to a transmission line. For the folks that have ever messed
with Eznec they have likely programmed a pair of vertical antennas
spaced less than a quarter wavelength.
As it turns out some very clever folks developed the log periodic antenna
in the 1950s that used a 180 degree flip added to the delay line in some
early VHF/UHF antennas. They found that adding the 180 degrees to the
delay line between elements would broaden the bandwidth .
This technique has been used several times by the military and
even showed up in ON4UNs book in 2003. Tom, W8JI popularized the
ham use of this technique in some internet postings in 2002.
If you look at the phase change of a signal as it passes two verticals
say spaced at 80 feet at an operating frequency of 1.85 MHz you will find
that the phase of the signal will change as the wave passes the two
antennas. Eznec says that the 360 degree wavelength at 1.85MHz is
531.66 feet. So lets say that as a wave passes the two antennas inline
and the phase is zero at the first antenna then the wave advances and
the phase at the second antenna will be zero degrees while the phase
at the first antenna will have changed to then be 80feet/531.66feet times
360 degrees or .1505 times 360 degrees, or -54.2 degrees later. In order
to generate a null response one must with the two antennas
(invert and add) add 180 degrees phase shift to the 54.2 degrees making
a delay of -234.2 degrees. Indeed if you model this you will find you can
form a great cardioid null pattern with a -234.2 degree delay cable at
1.850 MHz and 80 feet element spacing. If you build this -234.2 degree
cable at 1.850 MHz you will find that as you change frequency that the
cable delay values do not track the physical delay values as the
frequency changed wave passes the antennas.
Here is where we find out how this really works. Our actual -54.2
degrees at 1.85MHz between antennas as the wave passes will
change to -108.4 degrees at twice the frequency or 3.7MHz. We would
need -108.4 (2x54.2) plus the -180 or -288.4 degrees to generate the
same null and you can see that if you use these values in Eznec you
get the same pattern. But, if you look at the phase of your -234.2 cable
at 3.7 MHz instead of 1.85MHz you see a delay of -468.4 degrees.(2X)
You will find that the pattern has changed dramatically. However if you
install a 180 degree flip transformer in the delay line and subtract the
180 from the cable length you get -234.2 minus the 180 of the
transformer or an actual cable delay of -54.2. This with the 180
transformer will still give you the same -234.2 degrees and the same
pattern at 1.85 MHz. Now shift to 3.7MHz and your delay line will give
you -108.4 ( 2x 54.2) plus the -180 of the transformer or a value of
-288.4. You will find that the needed delay has tracked pretty well and
you get nearly the same pattern at 1.85 MHz as you do at 3.7MHz.
Looking back at the original delay values as the wave passes the
antennas, you will find that the needed delay values are exactly the
same and track with frequency.
There are some tracking errors as the element spacing passes
greater percentages of wave length spacing.
Sorry this is so long but did not know any other way to say it.
Lee K7TJR
_______________________________________________
UR RST IS ... ... ..9 QSB QSB - hw? BK
|