The forward gain difference between my single-coil and two-coil models
is 0.02 dB. This is with no coil losses, whose difference might easily
overwhelm a small directivity difference. But I restored the coil losses
and got the same result.
Input resistance is in the mid-20s in ohms for the single coil model and
in the mid-40s for two coils. I had expected band-edge reactances to be
about the same for both models and thus SWR to significantly increase
for the single-coil model. Instead, the reactances are similar to the
input resistances, with only a small difference in band-edge SWR for the
two designs when each is matched at band center.
I neglected to say that the AWG 5 copper-clad aluminum wire weighs 37.42
lbs/1000' while pure copper is 100.2 lbs/1000'. 3/16" OD refrigeration
service soft copper (0.1875" instead of 0.1815") with a 0.03" wall
weighs 57.5 lbs/1000'. 1/8" OD is 34.7 lbs/1000'. An optimized 7.7 uH
coil using 1/8" copper is 5.6" by 3.3" with a Q of 985. (As wire
diameter decreases, the proximity effect that limits coil Q decreases,
which enables coils that are smaller, lighter, and cheaper.)
This describes an interesting trick for 40m Yagis:
http://ham-radio.com/k6sti/owa.htm
I had toyed with the idea of writing a little utility that generates
wire segments for an element bent using dacron lines that run from
somewhere near the half-element midpoints back to the boom. This would
eliminate the need for a custom bent-element mount. You could even run
the lines between driven element and reflector to bend both, which may
improve gain even more. To do this right you need the taper schedule and
tubing bending parameters. Even better would be to include the dacron
line tension in the optimization parameters and then optimize everything
simultaneously. I'll leave this for someone else.
If you think the 0.5 to 1 dB a conventional OWA costs in forward gain is
inconsequential, check this out:
http://ham-radio.com/k6sti/pileup.htm
The analysis isn't definitive. But the next time it takes forever to
bust a pileup with your OWA, you may not be able to get the probability
curves out of your head.
COIL 3.90 is here:
http://ham-radio.com/k6sti/coil.zip
Brian
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|