Re: Playing with polarisers- color effects
- From: Seán O'Leathlóbhair <jwlawler@xxxxxxxxx>
- Date: 31 May 2007 15:22:23 -0700
Joseph Miller wrote:
Seán O'Leathlóbhair wrote:
Circular polarizers are not perfectly achromatic. That is, the
quarter-wave plate is only a quarter-wave plate at one wavelength,
unless they use a very expensive achromatic quarter-wave plate, which I
doubt. (I've had them made for my research and paid $1000s.) So the
linear-to-circular conversion is only "perfect" at one wavelength, and
as you depart from that wavelength, some residual linear polarization
will remain. Even the purely linear polarizers are not completely
achromatic (unless they are very expensive) and will not give complete
cancelation when crossed for all wavelegths; often blue light leaks
through, though it can be red. The result is that, if you cross
polarizers of various kinds in various ways, you can get some funny
color effects. Two successive circular polarizers can give these color
effects for these reasons. The first one will "leak" some
wavlength-dependent linear polarized light that will interact with the
second polarizer to various degrees, depending on its orientation.
I hope this helps.
<snip>
Once, I turned one polariser around, I was quite impressed by them.
One by itself is as close to neutral grey as I can tell. Two in a row
when aligned appear the same as far as I can tell. When crossed, they
are not totally black but only very bright objects are visible such as
light bulbs. These appear a deep purple so I guess that the highest
wave lengths are leaking though.
Actually it is the other way around.
Yes, I spotted the mistake myself just after posting. I asked Google
to delete it and posted a corrected version but somehow you saw the
original incorrect version.
In the visual wavelengths
(blue-green through near-red), the polarizer is extremely good,
effectively 100%, so crossing them gets rid of all that light. But at
the shortest wavelengths the filter is less than 100% effective, so that
crossing two of them will still transmit a bit of unpolarized purple
light. Probably there is a bit of red light leaking through, too, as
these filters often lose their effectiveness somewhere in the red.
A bit hard to tell that by visual inspection.
Talking of the high wave lengths leaking through, is there any point
in stacking the polariser on top of the UV filter? I read that
digital sensors are not sensitive to UV so I guess not and the UV
would lose its protective value in this case. How about with film:
polariser alone or with UV filter? (The film question is just a
curiosity and not important)
There is no need for the UV filter, both because the sensor won't see it
and quite often the polaroid filter is a poor transmitter of UV light
itself. If I'm lazy, I don't remove my UV filter when I use a polaroid
filter, but I should. There's no point in going through an extra piece
of glass with two air-glass boundaries unless you have to. The extra
glass can reduce contrast unless it and the polaroid filter are very
well coated. As for film, it depends on the polarizer and the camera.
Many film cameras don't benefit significantly from a UV filter (the
glass used for camera lenses really doesn't transmit well in the UV),
though some do, and as I said, the polarizer itself can often be a good
UV filter.
Joe
Thanks.
--
Seán O'Leathlóbhair
.
- References:
- Re: Playing with polarisers- color effects
- From: Joseph Miller
- Re: Playing with polarisers- color effects
- Prev by Date: Re: Sending images to Costco, Wal-Mart
- Next by Date: Re: Should Photoshop open ORF files?
- Previous by thread: Re: Playing with polarisers- color effects
- Next by thread: Re: Gamma value, Levels Curves ????
- Index(es):
Relevant Pages
|
