Hi Doug,
I'm responding to this as a general response, even though you sent
me an e-mail. I think this is an area that needs looked at by others.
I modeled my guylines on EZNEC, and included loads instead of
insulators. This does not include the large loops, only the
capacitance across the insulators, but it should be close.
The measured capacitance of 1/4 inch grips through 502 insulators
was 12 pF. That's 1.9 k ohms Xc at 7 MHz.
That amount of end-capacitance significantly alters the resonant
frequency of the guylines, especially at higher frequencies with
short guyline sections!
(I measured this through direct measurement by cutting the grips
just past the point where the grip-ends begin to wrap around the
guy strand. I'm certain the value, when used for coupling to the
next section, is accurate even though it does not allow for the
"fatness" of an end.)
You can model this by exciting the guyline with an external source
at the desired frequency (I used a lossless dipole several
wavelengths from the guyline to excite the guyline) and looking at
the current distribution in the guyline.
Another method is to add a current source in one guyline section
and watch for resonance and current in the adjacent wires. Model
the insulator as a load with the correct reactance for the test
frequency. It would be nice if Eznec modeled loads as capacitance
values, but it doesn't. You have to insert the correct amount of
reactance.
I hope, someday, to verify guyline currents with the model.
Before anyone just cuts guylines in the middle of a "safe range", I
suggest they look at the difference in resonant frequency AND
impedance caused by insulator capacitance to the next guy
section.
For example, modeling a 300 foot long guyline broken up every 28
feet by a 900 ohm reactance produces greatly different results than
the same line broken up by perfect insulators. If one line section is
fed like a dipole, the center impedance is a few ohms and current
in the other end-to-end sections is substantial at or anywhere
above 14 MHz!
My conclusion was to NEVER use sections approaching 1/2 wl or longer anywhere
near an antenna. In other words, a 38 foot section, even though the charts show
it "misses" all primary bands is NOT transparent, because of all the stray
capacitance, anywhere above 7-10 MHz. For example at 28 MHz the
reactance would be under 500 ohms. That would be OK if it was placed every 1/4
wl or so along the guy, but it is NOT OK at longer lengths even when chosen to
be non-resonant.
To: W8JI@contesting.com
Subject: Guy wire interaction
Interesting. Can you go one step further and suggest some "correct"
> non-resonant lengths using 1/4" guywire and 502 insulators.
>
> Doug
73, Tom W8JI
w8ji@contesting.com
--
FAQ on WWW: http://www.contesting.com/towertalkfaq.html
Submissions: towertalk@contesting.com
Administrative requests: towertalk-REQUEST@contesting.com
Problems: owner-towertalk@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|