[Laser] Polarization modulation

[Dave]

On 2011-03-15 12:01 PM, laser-request at mailman.qth.net wrote:
> Date: Tue, 15 Mar 2011 00:32:38 -0700 (PDT)
> From: Tim Toast<toasty256 at yahoo.com>
> Subject: Re: [Laser] Polarization modulation
> To: laser at mailman.qth.net
> Message-ID:<324202.35507.qm at web37902.mail.mud.yahoo.com>
> Content-Type: text/plain; charset=us-ascii
>
> Hi Glenn and all,

Isn't polarization modulation used by some quantum cryptography
systems?  A pair of entangled photons is produced, with one being
detected locally while the other is sent to the far end.  Since the
photons are entangled, by detecting the one, you know the polarization
of the other.

Or, something like that.

> I agree, a common LCD display may be the cheapest way to get good
> polarization modulation. And they hardly use any power so they're
> battery friendly. From what I've read tonight, the only difference
> between the back-lit and reflective type LCD is the mirror backing.
> So once that is removed you'll have a back-lit type LCD. I think
> you will have to also remove the back polarizing filter or it will
> just be a light-gate and produce AM modulation. Changing the
> polarization of the beam should not vary the brightness. I don't
> know if these things are linear or not, it may just jump from one
> polarization to another when voltage is applied - experiment will
> tell :-)

Ok, more than you every wanted to know about TN-FE LCDs.  ;-)

TN-FE LCDs (Twisted Nematic-Field Effect Liquid Crystal Displays) are
VERY non-linear.  Apply a small voltage, and nothing happens.  Apply
a bit more voltage, and nothing happens.  Apply a bit more voltage, and
nothing happens.  Apply a bit more voltage, and BANG, they switch
states.  That is the very reason that multiplexing them works so well.

The reason they are so non-linear is that the permittivity of the material
is very non-isotropic.  As soon as part of the material starts to reorient,
the electric field increases due to the increased permittivity of the 
reoriented
material, which produces a cascading effect.

That doesn't mean that one can't be used for polarization modulation, 
though,
just that you are mostly stuck with only two states (which may be an 
advantage,
especially if you're doing digital data transmissions).

One drawback is that TN-FE LCDs tend to be rather slow.  The speed can
be somewhat improved by heating, but excessive heating destroys the effect
since it disorders the material (e.g., it's no longer crystalline above 
a certain
temperature).  Conversely, cooling the material slows the speed (by turning
it into more of a solid than a liquid).

As for the reflective layer, that can usually be peeled off.  The tricky 
part is
in doing this in a manner such that the display isn't destroyed.  The older
style of assembly used non-isotropic, non-homogeneous conductive elastomeric
connectors (e.g., rubber loaded with stripes of graphite), usually 
referred to in the
industry as "Zebra strips" due to the alternating light and dark bands 
resembling
a zebra.  The problem is that it's very tedious to try and reorient the 
zebra strips
to get them to make good electrical contact between the printed circuit 
board and
the LCD panel itself (e.g., the "bottle"), while clamping the bezel back 
into place.
It can be done, but it's tedious.

The newer construction style uses flexible, conductive tape (e.g., 
kapton tape with
conductive lands), usually connected to the bottle via a conductive 
epoxy (e.g., epoxy
loaded with tiny, Silver beads).  The flexible conductive tape allows 
the bottle to be
reoriented with regards to the printed circuit board.

The bottle itself is composed of two VERY flat pieces of glass.  The 
inside of the bottle
is coated with a Indium Tin Oxide (ITO) material, which is actually 
diffused into the
surface of the glass.  ITO is a transparent and conductive (but highly 
resistive) material.
The inside surfaces are then coated with an alignment layer, usually 
either Silicon Dioxide, which is sputtered, or PolyVinyl Alcohol, which 
is rubbed to align it. The two
glass plates are then glued together usually with an epoxy glue, leaving 
an opening for
filling the liquid crystal material.

The bottle is filled with TN-FE LC material and, usually, tiny glass 
spacers (usually rods,
although some manufacturers used spheres).  You can sometimes see the 
spacers if
you examine a display with high power magnification.  The usual way of 
filling the bottle
is to evacuate the air from it, and then allow capillary action to 
"suck" the liquid crystal
material inside the bottle, where it is then sealed with a blob of 
epoxy.  The spacing
between the glass plates used to be on the order of 10 micrometers, 
although that was
25 years ago when I was working in the industry, and I'm told that it's 
been reduced
since then.

The original TN-FE LCDs were constructed with the liquid crystal 
material making a 90
degree twist between the front and back surfaces of the bottle (and 
aligned with the
alignment layer that was coated onto the inside of the glass plates). 
Subsequent LCDs
were made with material which exhibited a 270 degree twist (e.g., 
"Supertwist").

Polarizers are then glued onto the front and back surfaces.  The back 
may have either a
fully reflective, or partially reflective layer glued on.

Later displays went from being "inactive" to "active" by the inclusion 
of a switching
transistor inside the display for each pixel (but, that was about the 
time I got out of the
industry).

The driving waveform for a TN-FE LCD needs to be a pure AC voltage (with 
NO DC
present, since any DC can cause an electrolytic corrosion which will 
wreck the display).
A multiplexed display has a quite complex waveform of alternating square 
waves.  A
non-multiplexed display can be driven by a simple AC square wave (e.g., 
no DC bias).

The frequency of drive isn't too critical.  It needs to be fast enough 
that the display
doesn't relax between cycles.  But, it shouldn't be fast enough that the 
capacitive
loading of the display attenuates the signal due to the highly resistive 
nature of the
ITO.  Somewhere between about 100 Hz and 1KHz is usually adequate.  At 
10KHz-
100KHz, you can start to see the capacitive loading effect on long 
displays, where
the resistance of the long row lines starts to cause the signal to 
decrease in amplitude.

> One idea i had is for a simple tool to use when working with
> polarized light projects. A "polarized light detector". It might
> take the form of a simple light controlled oscillator, like the
> "audible s-meter" circuit by Clint. You would either spin a disk
> of polarizing material in front of the detector device, or, for a
> no moving parts version, use a modified modulated LCD pixel.

Should work.  You can also get polarization effects via the Brewster
Angle:

http://en.wikipedia.org/wiki/Brewster_angle

You may also want to investigate the Pockel's Effect:

http://en.wikipedia.org/wiki/Pockels_effect

and/or the Kerr Effect:

http://en.wikipedia.org/wiki/Kerr_effect

There's also the Jeffree cell:

http://en.wikipedia.org/wiki/Jeffree_cell

> Normal unpolarized light would pass through the filter unaffected
> and produce a steady tone from the speaker. Polarized light would
> be attenuated periodically by the spinning filter or LCD and produce
> FM in the tone which should be clearly audible. Just a little gadget
> to let you hear the difference between pure light and varying
> degrees of polarized light. A small lens would give it a narrower
> well defined field of view for pointing at various objects and light
> sources.

If you really want to make your head spin, consider time varying chromatic
modulation.  ;-)

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=%2Fnetahtml%2FPTO%2Fsearch-adv.htm&r=7&f=G&l=50&d=PTXT&S1=4,715,687&OS=4,715,687&RS=4,715,687

(Yeah, that's me).  :-)

> -toast

Dave