Its from Onion wrote:
> the sad thing:
> No modeling was done to adjust phase distortion differences
I assume you mean the time delay differences and the general phase
shift? Not some sort of narrow band dynamic effect.
> between the top arrays and the bottom arrays.
I addition the splitter circuits were not phase coherent
If you assume that you can adjust the actual radiated phase by changing
element lengths (which is the case with a SteppIR), then you might be
better off spending your time on the control algorithm than on
obssessing about a wideband power divider driving interacting loads.
This isn't like a 440 EME array where the antennas don't interact as
much and doing power dividers with simple equal length transmission
lines of the appropriate impedances.
Conceptually, it would be like taking two antennas fed in parallel, and
tuning one a bit high (so it has a R+jX feedpoint Z) and one a bit low
(so it has a R-jX feedpoint Z), which would shift the relative phases,
because of the reactive component. There's some circulating current
between the antennas through the combiner, but it can be pretty small,
so the absolute loss is small.
> and the side by side optimal spacing would not track to other bands.
Any design with fixed positions of the elements, either along the boom,
vertically stacked or horizontally displaced, is a compromise to a
certain extent. So you build what's practical.
> He would have saved a lot of money just stacking two antennas at 190 and 160
He would have saved even more money putting up a G5RV, to take it to an
> He spent $500,000 chasing the last 3-4 dB.
He spent $500K having fun, which is what it's all about. Some people
spend $100K on a cruise around the world. Some people spend $100K on a
Tesla roadster. Some people spend $100K on a ticket for a future ride
The cool thing about amateur radio is that everybody gets to try their
own ideas (it's not like there's a mob of angry shareholders and a board
of directors asking you whether you've gotten the best return on
investment here). Were I doing the same project, I might have gone to
an all driven element approach and not bothered with the elevation axis.
However, that's just me. Others might have spent a much smaller sum on
two antennas in a stack, and spent the rest of the money on buying land
for that stack in a different location.
TowerTalk mailing list