At 08:36 AM 4/1/2006, Robert McGwier wrote:
You should have seen all of this stuff starting ten years ago when I
was having to use it for various reasons. It cost Avagadro's number of
dollars and the dynamic range was about 30-40 dB.
Bob
N4HY
I think Bob is a bit pessimistic here. Going from the presentation, the
top of the line transceivers were in the $10K range, 10 years ago. Perhaps
20 years ago (mid 80s), fiber was pretty poor dynamic range, and really
expensive.
Last month at work, we bought several DC-6GHz transceivers for around $3K
each to build an optical RF delay line.
Dynamic range (just like with any other system) is mostly a matter of noise
figure and power. Something like a Fiber-span AC131 has a noise floor of
-130 dBm/Hz, and a IP3 of +29 dBm (that's a link with 1 meter of fiber
between two transceivers)
10 years ago, the performance wasn't all that much worse, it was just more
expensive, and might not have been quite as rugged.
A presentation at:
http://www.darpa.mil/mto/aosp/workshop/cox.pdf
shows Link noise figures for various efforts over the years on page 8.
One might want to look at what the Allen Telescope Array is doing. They've
put together a nice summary paper on their stuff (0.5-11GHz), and cite a
"per link" cost of $1250 (in qty:1000, exclusive of fiber cost). They're
talking 100+dB IM free dynamic range.
http://astro.berkeley.edu/ral/ata/Publications/URSI-GA02A.pdf
What's especially neat about fiber (aside from the incredibly low loss: a
few dB/km) is that because the medium itself is very broadband, there's
very little frequency dependence. And, of course, fiber is substantially
cheaper than coax. The termination costs a bit more (maybe, figure
$100/splice or connector, installed, although a lot depends on the kind of
connector and the performance you want.) Optical match performance is also
outstanding: 40 dB return loss is perfectly reasonable (dare you to get
that on any run of the mill RF connector).
What's really changed in the last 10 years is not so much the performance
(although it's gotten better), but that fiber RF links have become MUCH
easier to use. You can buy an integrated box with a coax connector on one
side and a fiber pigtail on the other. 10-15 years ago, you'd be buying
components, laying out a board, trying to get your amplifiers to be stable,
etc.
These days, doing fiber splices and/or connectors is no more difficult
than, say, doing a *good* job installing a PL-259 or N connector. All
require some amount of precision, and need the appropriate tools. Fiber is
a lot smaller, so it's less likely that you'll be able to do a good job
using a paring knife and a butane cigarette lighter, which *does* work for
RG-213 and a PL259, after a fashion.
I think that you won't see much use of RF on Fiber in HF ham applications
for while. Hams have short runs, tend to use narrow band signals, and
favor system design with everything centralized (hard to send that kilowatt
of RF down a fiber). I can see fiber being used for digital signals (say,
to a rotator or SteppIR type antenna) in the near future though, and for
short analog signal runs (where you can use cheap plastic fiber and TOSlink
style connectors). The EMI/EMC advantages are significant.
A dedicated microwave/UHF experimenter will probably be the first to use
fiber. They're already used to putting RF hardware at the antenna, and low
loss coax at 500 MHz and up isn't all that low loss, and is expensive. I
have seen surplus fiber gear starting to show up, and the raw cable is
quite cheap. Running fiber is a heck of a lot easier than rassling with
LMR600 or, even worse, waveguide.
W7RY wrote:
> Hear is your RF over fiber option.
>
> http://www.nt.hs-bremen.de/peik/rof/literatur/ROF-franz-presentation.pdf
>
> I have used a BDA (bi-directional amplifier, extends RF coverage of an
> 800 MHz radio system throughout a building) over fiber with great
> success. It is used in a large airport.
>
> 73
> Jim W7RY
>
Jim, W6RMK
