[Top] [All Lists]

Re: Topband: Two Wire Beverage Query...

To: "Tom Rauch" <>, <>
Subject: Re: Topband: Two Wire Beverage Query...
From: "Ford Peterson" <>
Date: Sat, 21 Aug 2004 16:54:31 -0500
List-post: <>
Tom-W8JI suggested:


> I have a short loaded vertical with almost zero radiation
> resistance. I resonate it with an inductor of known ESR and
> measure the feed impedance. What isn't coil ESR and
> radiation resistance is ground resistance. This also agrees
> with how the same rods behave when terminating a Beverage.
> >Also, re: 'surge impedance.'  How does one measure this?  I
> > was effectively able to measure the characteristic impedance
> > of the two wires to be in the 680 ohm area.  The data I
> > collected is very solid using a sweep of 1 to 8 MHz.  And a
> > sweep of a single wire over ground came in at about 450-470
> > ohms with the data not nearly as convincing in the same
> > sweep.  But how do you measure to see whether the surge
> > impedance is 300 or 600 ohms?>>>
> You already measured the effective surge impedance if you
> paralleled the wires and swept them. You did measure the
> parallel feed resistance when they were terminated to
> produce a flat response, didn't you? Just deduct the ground
> connection resistance and you have the surge impedance.
> 73 Tom

I like your approach to this.  So I grabbed a 21' piece of 1 - 5/16" fence top 
rail out of the dead antenna pile.  And wound up an set of inductors on T-130-6 
cores using #20 wire.  Using your methodology, and both an Autek VA1 and 
MFJ-269, I set out to see what the heck my ground really is.  What a bunch of 

If I understand this correctly, I measure the feedpoint of a resonant shorty 
vertical.  This piece was self resonant at about 10.67 MHz.  Using NEC2GO 
( modeling software, I modeled the antenna over perfect ground.  
Using real load values to get the model tweaked for lead length errors, etc.  
The model works perfect if I take my 2" leads on the analyzer into account.  So 
I'm using a good model over perfect ground.  In the case of both 160M and 80M, 
the model shows a "R" value of just under 5 ohms (4.9 on 160M and 4.7 on 80M).  
The no load version of the model measures 0.6 -j979 on 160M and 2.4 -j514 on 

Looking at the vertical without a load is difficult to read R and j.  Both 
instruments could read the j term but only the Autek could report an impedance 
of 32 -j920 at 1800 and 10 -j445 at 3500.

Using my loading inductors, the MFJ resonates the antenna at 1822 with a 
feedpoint of 33 +j6.  The Autek resonates it at 1839 with a feedpoint of 36 
+j6.  On 80M, the same vertical with a smaller load resonates at 3587 on the 
MFJ with a feedpoint impedance of 35 j0.  The Autek reads 3596 and 32 +j7.

The differences in resonant frequencies are due to the leads being only 2" on 
the Autek and more like 4" on the MFJ (It's a borrowed instrument).  But we are 
clearly in the ballpark.

If I understand the procedure correctly, my two 8' - 5/8" diam copper clad 
ground rods, lovingly pounded into the terra firma 10 feet apart and connected 
under the sod with #10 bare copper wire, have a ground resistance of roughly 36 
less the 0.6 ohm radiation resistance, less the 4.3 ohm load losses, or roughly 
30 ohms?  This compares to your previously referenced table of ground 
resistance.  The 
nearest comparable is the first item of Two 5' rods coming in at 85 ohms.  
After considerable heroic efforts your 7 radials and 3 - 5' rods comes in at 29 
ohms -- within spittin' distance of the two 8' rods in my soil.  Quite a 
difference huh?

Taking this information one step further, my 680 ohm characteristic impedance 
bidirectional beverage wire should be terminated at 680 less 30 or 650 ohms?  
This is a huge departure from the table in ON4UN's 3rd Ed book, which suggests 
680 ohms to 340 ohms or so.

As it is, I suspected the book's table was amiss and wound the terminations for 
680 to 450.  Would this impedance mis-match cause degradation of the pattern?  
If so, maybe I'll wind a 1:1 and try this at the termination.

Thanks for the help by the way.  It's interesting to figure out new methods of 
figuring stuff out.  Your suggestion was great.


Topband mailing list

<Prev in Thread] Current Thread [Next in Thread>