On the #18 vs. #12 wire in the quad question:
Changing wire in the quad will likely require readjustment of all loops to
reset performance at the vendor's specifications. Increase wire diameter
in a closed loop generally calls for a larger loop for the same resonant
point (unlike our expectations of shortening with fatter wire in a linear
element).
In an antenna, the current is not a constant value throughout the wire,
but is highest near a low impedance feed and lowest a high impedance
nodes. With a full wl quad loop, there will be two high current
regions--roughly the upper and lower horizontal wires--and two low current
regions--the sides. Hence, a constant current calculation, as might be
used in circuit wiring, is not apt to antenna use.
If modeling software is available, one can model simple 1 wl loops as
symmetrical squares. Make one of #18 and the other of #12. Bring each to
resonance (+/- 1 ohm reactance at the feed). Read each gain figure. The
difference times the number of elements is about what will be gained or
lost in the wire change. The difference in loop diameters is about the
amount of dimensional change needed for the proposed wire size. These
tests are done with good accuracy on NEC-2 and can be done with free space
models for simplicity. Of course, specify copper wire as the material, so
that NEC can take the distributed wire loss into account. As a matter of
interest, also specify lossless wire to see what copper costs you in the
first place. Repeat with aluminum wire, if any element uses that
material. The importance of making resonance precise is not only to be as
accurate as possible in the amount of change of dimension required, but as
well for accuracy in the gain test. A wire that is long will show a gain
increase solely by reference to length; one that is shorter will show a
gain reduction relative to a resonant wire.
-73-
LB, W4RNL
--
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
|