Some notes on portions of what Michael Shupe wrote on polarisers.
> But once the light passes through the quarter wave plate, it is > effectively no longer polarized (approximately equal portions are > vibrating perpendicular to each other). And 1/4 wavelength out of phase (as Michael writes below), making circularly polarised light, which I would not call "no longer polarised". > The quarter wave plate (the additional component that makes a circular > polarizer "circular") rotates a portion of the wave (approx 50%) so that > they are out of phase from the light directed into the polrazer by a > quarter of the wave length... A circular polariser is a linear polariser followed by a 1/4 wave plate. The 1/4 wave plate splits the linearly polarised light into two components at 45 degree angles to the linear polarisation. Initially these components oscillate in phase. Due to different light speeds of the two components in the 1/4 wave plate, the two components are 1/4 period out of phase when they exit the plate. This makes circularly polarised light. It's like the horizontal and vertical position component of some point on the rim of a vertical rotating wheel: the horizontal and vertical positions both oscillate with equal amplitude but 1/4 of the rotation period out of phase. Their sum corresponds to a circular motion. At a fixed point in space behind a circular polariser, the electromagnetic fields that describe light rotate in a plane perpendicular to the direction of the light (that's why it's called a circular polariser). > essentially, the light that has passed through the polarizing sheet > at front of the polarizer will be in a vertically vibrating plane > (if the polarizer is oriented that direction), then it passed > through the quarter wave plate, half of it will be vibrating > vertically, and half will be vibrating horizontally. > (Actually, this is too great a simplification, because the light that > has passed through the polarizing sheet is not all vibrating in a single > plane, but its amplitude is greater the closer it is to the orientation > of the polaroid sheet). I think the light can be regarded as vibrating in a single plane directly after the polarising sheet. That one can detect light when you look at that light in other polarisation directions doesn't make the single plane statement untrue. > Then that light will hit the mirror in the camera and part of the > light (the vertically polarized light, I believe) Will be reflected > up through the eyepiece and to the metering system, while the > horizontally oriented portion of the light (equal to the vertically > oriented portion) will pass through the main mirror and hit the > second mirror where it will be reflected down to the AF system. I think this isn't correct. Reflection and transmission percentages depend on the polarisation, but it isn't all or nothing (non-trivial formulas). The important point for our EOS photography is that the fraction of light going to the metering system depends on the polarisation state of the light falling onto the mirror. The camera meter assumes it gets some fixed fraction of the total light to calculate proper exposure. If you use a linear polariser, this assumption can be wrong (depending on the orientation of the polariser) and this leads to an incorrect exposure indication by the camera. For a circular polariser the fraction of light going to the light meter is the same as for unpolarised light, thus a circular polariser preserves the accuracy of exposure measurement by the camera. Well, it's a long story and basically EOS independent, but I hope it can help in making better EOS photographs when polarisers are involved. Regards, Peter Wagemans * **** ******* *********************************************************** * For list instructions, including unsubscribe, see: * http://www.a1.nl/phomepag/markerink/eos_list.htm ***********************************************************
