Circular vs Linear Polarizers

this is another slightly technical question i have, but this one is about

polarizers.

 

I purchased a circular polarizer for my digital camera and it seems to behave

more like a linear polarizer.

 

i have used polarizers in a lab environment in the past and have found that my

"circular polarizer" behaves more like a linear one. light from reflections is

linearly polarized, and when i rotate my "circular polarizer," the reflections

disappear and appear (90 degrees between appearing and disappearing). this

behavior indicates the circular polarizer i have is behaving like a linear

polarizer. i have done the same thing while looking at an LCD clock display

(which also emits linearly polarized light) and find the same angularly

dependent drop in transmission.

 

circularly polarized light, when passed through a linear polarizer, does not

show this same angular dependent drop in transmission (the transmission is the

same regardless of the angle at which the linear polarizer is turned).

 

so my question is, what is the difference between circular and linear polarizing

filters, if they behave the same? is it that the circular polarizing filters

are constructed by a linear polarizer in front of a quarter wave plate so that

the light *leaving* the filter is circularly polarized?

 

thanks again for the help,

mark

To the naked eye the effect appears the same, circ or linear. The difference is that most modern auto-focus cameras have a pol element in the optical path to the AF/AE sensors and a linear pol filter can blind those sensors.

 

With an older manual camera, the Nikon FM for instance, you could use either. With a modern camera, you need the circ pol option.

 

--

Henry Posner

B&H Photo-Video

What Henry said. Both polarizer's should have the same effect when rotated. The only reason for circular is for use with autofocus cameras. Of course the quality of either might produce slightly different results when rotating.

A circular polarizer is a linear polarizer, with the addition of a quarter wave plate to "scramble" the light after being analyzed. As mentioned, the photographic effects are identical.

In actual practice, I have not seen a problem with a whole bunch of linear polarizers (many of them Spiratone Colorflow II types) that I have mounted on my Canon EOS 20D. Exposure seems unaffected, and everything works just like it did on my old film cameras.

 

Is this because the 20D AF/AE sensors are not built such that they need circular polarizers, or have I just been lucky that the settings were not such as to blind the sensors?

As others have correctly said you need to use the circular polariser with any autofocus camera; the linear polariser may give inaccurate exposure on any other than a manual focus camera body.

Your camera manual usually will state which to use on your camera, esp. if your camera was produced after the mid 1970-80s.

There is no difference in the final picture. If the exposure and auto focus work properly,

you do not need the circular. My Nikon manual says circ, but I use linears and all works

well. Just extra profit margin,

<cite>is it that the circular polarizing filters are constructed by a linear polarizer in front of a quarter wave plate so that the light *leaving* the filter is circularly polarized?</cite>

<p>

That's exactly it. Which explains why it filters incoming light just like a linear polarizer would. The front part of it IS a linear polarizer. The QWP is used to make the light which exits the polarizer unaffected by any beam splitters that may be in the metering or autofocus path.

<p>

You can verify this and do some neat experiments by using a mirror with your circular polarizer. Hold it up to the mirror with the female threads facing the mirror, and it darkens things, but you can see through the filter/mirror/filter combination. But flip the polarizer over so that the male threads face the mirror, and light is blocked by the combination in this orientation. The circularly polarized light, when reflected, won't go back through the circular polarizer.

Light reflected from non-metallic surfaces is partly polarized. This includes the image-splitters used by auto-exposure and auto-focus sensors in modern cameras. A linear polarizer will cause inconsistent performance of these components, which is why circular polarizers are specified.

>>A linear polarizer will cause inconsistent performance of these components, which is why circular polarizers are specified.<<

 

 

That's what everybody says, but I've used these old Colorflow linear polarizers a fair amount so why don't I get "inconsistent performance" on my Canon 20D? In my personal experience both AF and AE seem to work just fine with the linear polarizers. I know they aren't SUPPOSED to. Originally, I just stuck them on to see what happened, and the answer in practice was "nothing much".

JDM,

 

You have been lucky, or perhaps other inconsistencies mask the effect of using the wrong polarizer.

I admit that for my primary polarizer use, I have got circular polarizers and I would not suggest to someone seeking a polarizer for everyday use on a modern AF/AE camera should depend on a linear polarizer. But I am saying that if you, like me, have an accumulation of old linear polarizers, especially like the ColorFlow series from Spiratone, you should at least try them and see before you chuck them out.