Olympus-OM
[Top] [All Lists]

Re: [OM] Re: Polarizer crash course

Subject: Re: [OM] Re: Polarizer crash course
From: "John A. Lind" <jlind@xxxxxxxxxxx>
Date: Mon, 29 May 2000 01:38:19 +0000
Ooops!  My error   :-(   Joey (and Tom Trottier) are right.

They do rotate together; the 1/4-wave plate must remain 45 degrees out from
the linear.

Another easy way to tell between them if in doubt that doesn't require a
clear sky or a pair of polarizers.  Stand in front of a mirror and look
through the polarizer at the mirror.  With a circular, in one direction
(from the back or lens side) you will see your eye looking through what you
might expect with a ND filter.  Flip it over and look through it again
(from the front side).  The filter will appear nearly pure black and you
will not see your eye peering through the polarizer (with the rest of the
image in the mirror other than the filter as before).  With a linear
polarizer it should look the same either way.

Why?  Looking through the front, the last thing on the way to the mirror is
circularly polarized light from the 1/4-wave plate.  When it reflects off
the mirror it becomes circularly polarized in the opposite direciton (CW
comes back CCW and vice versa).  Since this is the circular equivalent to
being at right angles, it blocks out entirely.  Flipped over you have
linear polarized light reflecting back from the mirror.

BTW, the optimal angle at which you can filter nearly all reflected light
with a polarizer is called "Brewster's Angle."  It is related to the
refractive index of the two materials at which the light reflects.  With
water (and air), or glass (and air) this is just over 50 degrees to the
reflective surface.  At that angle _all_ incident light is polarized when
reflected and thus can be filtered.  It is why I mentioned optimal
performance at about right angles to the light source (you to surface to
source); actually it is a little more at about 100-115 degrees.

-- John

At 18:19 5/28/00 , Joey wrote:
>All the glass in the circular polarizer should rotate together.
>The circularizing element is actually a special anisotropic
>crystal.  When light passes through this crystal, its speed
>depends on its direction of polarization relative to an axis in
>the crystal.  In order to convert the linearly polarized light
>from the polarizer into circularly polarized light, the polarizer
>must be aligned properly with the axis of the crystal.  The
>alignment should be at 45 degrees from the axis:
>
>crystal
> axis
>   |  / polarizer axis
>   | /
>   |/
>   +---- 90-degrees from crystal axis
>
>The component of the light along the crystal axis moves faster
>than the component perpendicular to that axis.  As a result, the
>phase between these components shifts as the light travels
>through the crystal.  By making it the proper thickness, one
>can get a 1/4 wavelength shift -- such a crystal is called
>a "quarter wave plate" for this reason.
>
>Anyway, circularly polarized light is light with polarizations
>in two perpendicular directions, one 1/4 wavelength ahead of
>the other in phase (or 90 degrees ahead in phase).  That's
>exactly what the quarter wave plate does to linearly polarized
>light.
>
>If the linear polarizer is not aligned properly with the
>1/4-wave plate, then this doesn't work.  If it's parallel
>or perpendicular to the crystal axis, you just get linearly
>polarized light out.  If it's not 45-degrees between them,
>then the amplitudes of the out-of-phase components will
>not be equal and you get elliptical polariazation.
>
>
>So you probably can't tell by looking at the glass whether
>your polarizer is linear or circular.  If you have two identical
>polarizers, you can tell by looking through them both.  If
>they are linear polarizers, you will be able to see through
>them when the axes are aligned, and they will become opaque
>when the axes are at right angles.  With circular polarizers,
>the relative orientation won't matter -- you will see no
>change as they are rotated relative to each other.
>
>joey
>
>
>< This message was delivered via the Olympus Mailing List >
>< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
>< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >
>


< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >


<Prev in Thread] Current Thread [Next in Thread>
Sponsored by Tako
Impressum | Datenschutz