[TowerTalk] 2 element phased vertical

[Jim Lux]

At 12:07 PM 2/3/2005, Van Fair wrote:
>Hi Jim. Thank you so very much for the time you spent. I may be in over my
>head.  I dont have or use excel and I dont know how to use NEC. I think I
>have it from the ARRL antenna handbook CD. I guess the answer is, it is
>possible to use different antennas but only if you know how.
>
>To just build it and try to make it work may be to much of a hit and miss
>proposition. Since ground which is unknown plays such a big role I guess
>there is no such thing as build it like the drawing and it will work.   How
>much of what you proposed would be required if both towers and ground fields
>were identically built and spaced exactly 1/4 wave apart for 3.800. Would
>the problem then be only determining the length of the coax when the number
>of degrees  comes from fig 11-13 in ON4UN book or is the degree based on the
>ground field.  The only thing I have to use for antennas is the MFJ antenna
>analyzer.

OK. so you can skip the NEC model to establish feasibility..
Sure, the ground losses have an effect, but NEC can approximate that, and 
when all is said and done, you'll still have to measure the real impedances.


If you've got the towers up, you can measure the impedances, and go from 
there.  The MFJ will certainly do the job as long as it's the one that 
reads X as well as R.

What is fairly certain is that you DO need at least an estimate of the 
impedances to figure out the coax length, and that figuring is a bit 
tedious to do by hand.

Maybe someone will create an online phased array calculator using java or 
javascript?  If anyone's interested, I can supply you with all the 
equations, etc.



>Thanks again for your help.  Van
>
>----- Original Message -----
>From: "Jim Lux" <jimlux at earthlink.net>
>To: "Van Fair" <giw at bellsouth.net>; <towertalk at contesting.com>
>Sent: Thursday, February 03, 2005 2:04 PM
>Subject: Re: [TowerTalk] 2 element phased vertical
>
>
> > At 10:40 AM 2/3/2005, Van Fair wrote:
> > >I just put up a Hy Gain Hy Tower which is 53 feet tall and is liner
>loaded
> > >somehow to be a 1/4 wave on 75 meters. It works quite well with a modest
> > >1000ft ground field. I am considering adding a second vertical and would
> > >use a simple Christman feed system to phase the two.
> > >
> > >My question is:  Do I have to have another very similar short vertical
> > >such as a Hy Tower or could I use a full size 1/4 wave vertical for the
> > >second antenna spaced 1/4 wave away with a second ground field just like
> > >the hy tower has.  I have some Rohn 25 and long mast which could get me
>to
> > >62 feet high and use fiberglass rod insulators at the base of the Rohn.
> >
> > You can make a Christman type feed "work" with almost any radiating
> > elements.  It's "easier" if they're identical and spaced just the right
> > distance apart.
> >
> > There are several steps to the process (ON4UN's book describes one way,
>and
> > I've got some Excel spreadsheets to help you do it, as well).
> >
> > Step 1) Know the mutual impedances beween the elements.  If you're at the
> > planning stages, you can use NEC to model it.  Gnat's eyelash isn't
> > important at the planning stage, because all you want to establish is that
> > it's feasible.
> >
> > Step 2) Once you've decided what the element currents and phases should be
> > (the classic strategy is 90 degrees apart, equal magnitudes, but others
>can
> > work), calculate the "driving point" (or "feed point") impedances that
>will
> > appear with those currents.  This is simple matrix math using the mutual
> > impedance matrix.
> >
> > This will give you a "desired voltage and current" at each antenna.
> >
> > Step 3) Use a transmission line calculator to figure out how long to make
> > the two feedlines. As you move back along the transmission line, the
> > voltages and currents change.  You pick two lengths where the voltages are
> > equal (mag and phase).
> >
> > The Excel Solver can do this pretty quickly.
> >
> > Step 4) Verify that the lengths chosen are "reasonable". You can get
> > solutions where the sum of the two feedline lengths is less than the
> > distance between elements). If not, run the analysis again with some
> > constraints (like a minimum feedline length).
> >
> > Step 5) Do a quick senstivity analysis.  How much is the optimized length
> > changed if frequency changes a bit or if the mutual impedances are
> > different?  You don't want to build a system that only works if things are
> > controlled to 0.001%.
> >
> > Step 6) Optional, you can put the transmission lines into a NEC model and
> > look at the patterns to see if they're what you want.
> >
> > At this point, you've assured yourself that if you DO build it, it will
> > probably work.
> >
> > Step 7) Build the antennas.
> > Step 8) Measure the actual mutual Z
> > Step 9) Recalculate the coax lengths (hope they aren't too different from
> > your original estimates)
> >
> > Step 10) Cut the coax
> > Step 11) See if it works.
> >
> >
> > Now.. for spreadsheets to make life easier...
> > http://home.earthlink.net/~w6rmk/antenna/phased/software/drive2.xls
> > This one will do mutual impedance estimates (for two monopoles) or reduce
> > actual measurements.  And, it will do step 2, above, calculating the
> > driving point voltage and current.
> >
> > http://home.earthlink.net/~w6rmk/antenna/phased/software/christ2d.xls
> >
> > Does the calculations for the transmission lines. If you need to install
> > the "solver" for your copy of Excel, click "Tools","Add Ins...", then
>check
> > the box for  "Solver Add-in".
> >
> > Feedback on the spreadsheets would be appreciated..
> >
> >