Why does my circular polarisation filter go round?

[Chris Hale]

Just bought a Hoya Pro1 digital circular polarising filter.

As I understand, the filter first polarises light in a linear direction before passing it through to a quarter plate filter (at an angle of 45 degrees) so as to produce circular polarisation.
Why, then, is the filter designed to be turned when mounted on the lens?
My thinking is that for circular polarisation it doesn't really matter where you start from in terms of the initial linear polarisation so why make this variable?

Your thoughts most welcome.

Chris

PS - Thanks to all for your helpful replies. I am now clear about why this is necessary

[Shadowman]

Look through the viewfinder as you rotate, you'll see the difference.

[rpcrowe]

Sometimes, I just take things for granted and don't need a scientific explanation. I don't know if that is a failing or if it clears my mind for important things like composition and football scores

I don't know why both a circular polarizer and a linear polarizer are both twisted to change the polarization. I just know that when I rotate either of the polarizing filters, the polarization that I get changes. I am quite happy at that regarding my store of knowledge...

I also know from experience that linear polarizers will not work effectively with some auto exposure systems and most auto focus systems.

I have been using linear polarizers for about 50 years and circular polarizers for about 30 years. Not knowing the exact scientific principles involved have not limited my use at all.

[revi]

The polarizer works because the incoming light is partly polarised. The amount and direction of polarisation of that incoming light varies depending on the source. As the polariser only transmits light with one direction of polarisation, the effect on the incoming light will depend on the polarisation of that incoming light.

So the effect of a linear polariser will be independent of the orientation of the filter for direct sunlight, or diffuse reflections (wich have no discernable polarisation). For specular reflections on e.g. water or glass, there will be an orientation effect, as that light has a more or less pronounced polarisation.

And as you said, a circular polariser is a linear polariser followed by a quarter plate at a fixed angle (to the polariser, the two are glued together). So the light is filtered by the linear polariser, then transformed by the quarter plate. As it is in effect a linear polariser that 'selects' the light, its orientation matters.

[Dave Humphries]

Hi Chris,

Seeing is believing

I just reviewed over half a dozen short YouTube videos supposedly demonstrating them in action, this one is the best I have found so far.

Photo101: Polarizing filters, why they're so neat...and how to use them = 2 minutes 51 seconds.



It is pretty comprehensive, with a variety of good examples and appears to be fairly factually accurate (some others were not).

The thing with CPL is that they are rarely "will work/won't work", as even this video suggests. For example; imagine a scene beside a lake; if shooting towards or away from the sun, while the CPL will work far less well on the sky, it may still have a beneficial effect on reflections off the water's surface and vegetation.

They are good for revealing the colour of car/automobile paintwork, helping remove the reflective glare of the gloss polish in bright overcast conditions, unfortunately, you'll soon find this works for say, the sides of the car, or the roof, but not both at the same time. I guess one way around this would be to shoot both (changing the filters angle between), perhaps other angles to reduce reflections of other large flat areas, e.g. the boot/trunk or bonnet/hood that may be at a different angle to the camera than the roof and sides - then open each shot on a separate layer in your image editor and manually merge the different shots to obtain a result that gives the best result.

Could you do me a favour please?
Could you click Settings (right at the top),
then Edit Profile (on left),
put where you are (roughly) in the Location field,
then click the Save Changes button below and to right,
this helps everyone give you more personal and relevant answers - thanks in advance.

Cheers, Dave

[dem]

Hi Chris.

Whatever polarizer you are using, there is only one variable: its orientation with respect to the incoming light.

In the case of a circular polarizer, the quarter plate is permanently glued to the back of the linear-polarizer part. So you are right, there is no need to rotate this quarter plate with respect to the linear part (that is simply a parallel grid made of very fine wire).

Cheers,
Dem.

[jcuknz]

As for Dave's car I understand that people use two polarisers to handle the two lots of polarised reflections, or is this a myth?

[dem]

As for Dave's car I understand that people use two polarisers to handle the two lots of polarised reflections, or is this a myth?

Impossible. As Dave said, one needs to turn the polarizer, take several shots and then combine them in post. The best you can do with two CPLs is to make a variable ND filter (but the second/outer filter has to be fitted back to front).

[Dave Humphries]

As for Dave's car I understand that people use two polarisers to handle the two lots of polarised reflections, or is this a myth?

I think that's a myth - and probably an unworkable one at that

If you had the two at 90 degrees to each other; e.g. to deal with a horizontal roof surface and the vertical door surface, there wouldn't be much light coming through.

Admittedly I am thinking back to the old trick of turning two Polaroid sunglass lenses at 90 degrees, whereupon, they went black. Those would have been linear polarisers, not circular, but I'd be surprised if the effect were not the same.

(written before I saw Dem's reply)