J
John Larkin
Guest
On Wed, 22 Jul 2009 22:18:00 -0500,
hal-usenet@ip-64-139-1-69.sjc.megapath.net (Hal Murray) wrote:
http://www.highlandtechnology.com/DSS/J724DS.html
where the light is supposed to follow the logic input no matter what.
A lot of lasers do goofy things when you just turn them on and off.
It's hard to make a clean step of light.
improves prop delay and sometimes cleans up the edges. I think the
laser vendors buy their wafers in some back alleys in Taiwan; you
never really know how the next batch will work.
filaments so the drive transistors wouldn't die at turn-on.
John
hal-usenet@ip-64-139-1-69.sjc.megapath.net (Hal Murray) wrote:
It's for a pure digital logic transmission link,In article <17if65dmihe0fodjodepqqkbmhe1bj9ipb@4ax.com>,
John Larkin <jjlarkin@highNOTlandTHIStechnologyPART.com> writes:
We bought some 850 nm fiber-type laser diodes that acted as if they
had a microseconds-response PIN diode built into them, in series with
the actual laser. If you applied current suddenly, the voltage would
overshoot almost 2:1, as if there were an inductor in series. As long
as the current wasn't cut off for too many nanoseconds, it behaved
like a diode. Fine for telecom and maybe for burning CDs, but terrible
for sending baseband on/off digital stuff.
The telcom guys are using baseband. They use scramblers to ensure
that there are enough transitions to do clock recovery. There isn't
much low freqency energy.
What sort of data were you sending? Were there long strings of
1s or 0s?
http://www.highlandtechnology.com/DSS/J724DS.html
where the light is supposed to follow the logic input no matter what.
A lot of lasers do goofy things when you just turn them on and off.
It's hard to make a clean step of light.
We do sometimes bias them a little below the lase threshold. ThatI seem to remember something about not turning something all the
way off. I have forgotten the context and reasoning.
I think the idea was to modulate it from 10% to 100% rather
than turn it all the way off in order to avoid transients/problems
like you described.
improves prop delay and sometimes cleans up the edges. I think the
laser vendors buy their wafers in some back alleys in Taiwan; you
never really know how the next batch will work.
Sometimes keep-warm current was used to raise the resistance ofIt might have been incandescent lamps, but I don't think so. Reminds me
of the time I went into a dark machine room many many years ago.
(I was sleeping on the couch, babysitting for an all night run.)
All the red lights on the disks were on. Wow were we in trouble!
After a few seconds I figured out that it was just the keep-warm
current to avoid the turn-on transient. Nothing was visible in
normal room light.
filaments so the drive transistors wouldn't die at turn-on.
John