[time-nuts] Using lasers for data transmission
Dr. David Kirkby (Kirkby Microwave Ltd)
drkirkby at kirkbymicrowave.co.uk
Sun May 1 04:27:05 EDT 2016
> Hi,
> Several (many?) years ago National Geographic magazine show a picture
taken here in southern California of the state government sending red laser
signals between different mountain tops to keep track what was going on
near fault lines
> There were no technical details on what was taking place. So it can be
done.
> At a hamfest a few years ago I bought both a red and green 35 mW laser
pen for about $15 each. They do shine a long, long way.
35 mW is certainly unsafe to the eyes, so be very careful. There maybe
legal issues about doing this.
> Whether these are powerful enough, or can be properly modulated for what
is needed, I have no idea.
You can pretty much modulate any laser diode. There are two important
currents to know about
* Threshold current I_th - below which it will not lase.
* Maximum operating current I_max - above which the device will be
destroyed.
You can AM modulate them by applying a DC current
I_th + (I_max - I_th)/2.
Then superimpose the modulation which has a peak value of (I_max - I_th)/2.
Those currents ensure that the laser is always lasing, and gets you
theoretically 100% modulation. For best lifetime, run at lower levels of
peak modulation current.
Watch out for transient currents - lasers make transistors look like
antisurge fuses!
For point to point contact you want a beam which diverges as little as
possible. IIRC the divergence is something like inversely proportion to
the cavity length. For this reason a diode laser with its short cavity is
not optimal. But of course they are cheap.
A veey long time ago I used to know quite a lot about lasers, but not
using them for years I realise that I have forgotten an awful lot!
FWIW, at university we had a 10 W argon ion laser. I think it took about 50
kW to produce those 10 W. When it was disposed of, it was sold to a company
that put on light shows.
Dave
More information about the time-nuts
mailing list