[TowerTalk] Phasing Question

[Jim Lux]

-----Original Message-----
>From: "R. David Eagle" <kb8nnu at yahoo.com>
>Sent: Jan 6, 2008 7:27 PM
>To: Tower Talk List <towertalk at contesting.com>
>Subject: [TowerTalk] Phasing Question
>
>Hello all...I have a question regarding coaxial phasing that has been totally confusing me.  Let's say I wanted to phase 2, or possibly, 4 short verticals in a swichable array by means of coax delay lines.  I understand that 90 degs phasing can correspond to 1/4 wavelength, but if you went to other intervals of about 130, 140, 150, etc...how does one calculate the length the amout of coax needed to create that phasing effect?  Assume using 75 ohm coax.  


---> Buy ON4UN John Voldevere's book "Low Band DXing".. it has lots of useful stuff on this, well beyond what you'll get in some short emails on the reflector.


But.. in a multi element array, there's more to it than just the length of the coax.. the elements interact with each other (mutual impedance), so the feed point impedance of element 1 varies depending on what you're feeding elements 2,3,4, etc. with.  Then, if the feedpoint impedance doesn't happen to exactly match that of the transmission line (i.e. 50 or 75 ohm resistive), the phase shift resulting from a length of line varies depending not only on the line length, but the feedpoint impedances. There are some "special" lengths (1/4 wavelength, 1/2 wavelength, etc.) which happen to always have certain phase shifts (90, 180, etc.), but anything else depends on the rest of the circuit. Something else is that the transmission line and mutual impedances also change the magnitudes of the currents, as well as the phase.

There ARE equations for all this (in the ARRL antenna book and in ON4UNs book,or my website, among other places), so it is possible to figure out lengths of coax that will produce a particular phasing (and sometimes magnitude), especially if you can change the length of the element (as in a SteppIR) or put a tuning network in.  There's a popular setup for producing a 90 degree phase shift between two elements 1/4 wavelength apart from Al Christman (Check out Roy Lewallen's article on this at:
http://www.eznec.com/Amateur/Articles/Simpfeed.pdf
)

I've built some Excel spreadsheets that do the calculations for arbitrary coupling, phasing, etc, and you can fool with them to try and find lengths that work, but, practically speaking, if you're not using one of the "special" spacings and phasings, it gets quite complex quite quickly.  Most folks building phased arrays for HF do stuff like 4squares or hexes, where you can get a fair number of beam directions with just a few tuned networks and a huge pile of relays.


(Mind you, as I type this, I have 5 LDG AT200PC antenna tuners and a TAPR VNA sitting in front of me all plugged into a big USB hub as I try to make the antenna tuners into computer controlled phasing boxes, so there ARE other approaches...)

Jim, W6RMK