One thing was missed regarding cable lengths. The loops in the cans are part of
the equation for figuring the 1/4 wave length. I've seen that discussed here
many times in postings related to inter-cable lengths on duplexers. But the 1/4
wave length issue only applies to the inter-cabling between the cans.
It is my understanding that the antenna to duplexer lengths are irrelevant
since the T connector and the rest of the feedline are all part of the
equation. It's not like the T is some magical device that makes the rest of the
feedline disappear electrically. The only time length might be a problem is if
the entire feedline happens to be a resonant length. If by some chance that
happens, then changing the jumper a couple of inches will clear that.
Dan N8DJP
Posted
by: "n...@no6b.com"
n...@no6b.com
no6b
Date: Tue Mar 9, 2010 8:29 pm ((PST))
At 3/9/2010
20:12, you wrote:
>OK, question...
>
>If you
put a cable which is 1/4-wavelength at VHF between the T and the
>UHF
cavity, it's 3/4-wavelength at UHF. Since any odd multiple of a
>quarter
wavelength will invert the impedance, what will this really
>accomplish
on the UHF cavity side?
Doesn't matter at UHF, since the cavity
"looks" like (hopefully something
close to) 50 + j0 ohms @ UHF, so
the cable length has no effect (other than
plain ol' cable loss) @
UHF. At VHF, the short at the UHF cavity connector
(I'll take
Gary's word that it looks like a short off-resonance, though to
be
sure you'd want to put the can on a VNA to get the actual phase angle at
the connector) needs to be transformed to an open at the T so it
has no
effect & VHF. The short-to-open transformation @ VHF is
accomplished with
a 1/4 wavelength of coax @ VHF.
> The
dual-band diplexers are usually high-pass/low-pass arrangements, and
>
lose something like 0.2 dB while providing 40 dB or more isolation.
>
Assuming you get a real one, and not something made with PIM-prne
>
materials, would this not be a safer bet?
It's true you wouldn't
need to mess with cable lengths if a cross-band
diplexer were used,
but OTOH it would be another piece of hardware in the
system that
really isn't necessary, since the cavities are already
there. Plus
if you're really worried about PIM, you'd probably have to
move up
to something like a cross-band coupler from TX-RX, which IIRC runs
over
$300.
Bob NO6B
