>
> Speaking of UHF preamps, does anyone have any
> experience/recommendations
> for tower mounted preamps?
Al,
My experiences with tower-mounted preamps have been less than perfect. Good
designs will have dual amplifiers (redundant) and/or a bypass relay. Aside
from amplifier damage due to lightning, a number of tower-top preamps I've
seen fail did so because of problems with the window filters due to moisture
causing corrosion and other problems, even in well-sealed enclosures. IMHO,
in most situations you're better off in the long run with lower-loss
feedline than adding gain upstairs. Unless you *really* like to climb :-)
Since you're probably not going to want to change out your 7/8ths, here's
what I'd do. I'm assuming that this is a receive-only installation (i.e.
you're not duplexing in one or more transmitters to the antenna in
question).
First, measure the levels of the offending signals (paging Tx's and whatnot)
at the end of the hose with a spectrum analyzer. The trace peak-hold
function of most analyzers comes in handy here; let the analyzer sit there
and record for a few hours or days while you go do something else. This
will dictate how much attenuation of those signals you will need before any
gain stages.
Based on the results of those measurements, and the frequencies involved
(both desired and undesired), determine the appropriate filtering and system
design. The few most common designs would be:
a) A window filter to pass your range of interest followed by the gain stage
and power divider. This would be the simplest solution IF the offending
signals are low enough after the window filter to prevent overloading the
preamp.
b) Same as a) above, but with reject filters after the window filter to
attenuate undesired signals in the passband
c) A combination of narrower-range window filters and/or spot-frequency pass
cavities connected in a star configuration or a backbone configuration (a la
the TX-RX "T-pass" design) using critical-length cables, the output of each
filter then feeds its own preamp and power divider.
d) Depending on how close the offending signals are to desired frequencies,
you may need reject cavities after window filters or pass cavities, or
pass/reject cavities after pass cavities in c) above.
The FM problem will likely "go away" due to the rejection of the pass
cavities and/or window filters, or if not, a high-pass filter ahead of
everything would be the cure-all.
Sometimes system designs can be altered to improve performance (noise
figure) of certain bands/frequencies over others. For example, if one of
your receivers is for a point-to-point link that has a lot of signal margin,
you might be able to sacrifice performance on that frequency sans ill effect
while improving performance for other frequencies/receivers. Without going
into a lot of detail at this point, directional couplers, asymmetrical power
dividers, coupling loop adjustments/cavity Q, etc. can come into play.
Most multicoupler designs (and what I was envisioning when I wrote the
above) place the filtering BEFORE any gain stages. However, if the
offending signals are low enough in amplitude, you may be able to get away
with a high dynamic range preamp without filtering, or possibly immediately
after a wide low-loss window filter that passes everything in your range of
interest as in example a) above. Usually the trade-off with high-level
preamps is that noise figure is sacrificed for strong-signal handling
capability (higher TOI/compression points), but if the filtering that would
have been required in a pre-filtering design would have increased the system
noise figured by a greater amount, then it may be a viable option.
If you want to make some measurements and report back, I'll offer my
suggestions if you want. The solution may be simple, or complex, depending
on what you measure and the frequencies involved.
--- Jeff
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/Repeater-Builder/
<*> To unsubscribe from this group, send an email to:
[EMAIL PROTECTED]
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
