Polarizer Effect

Hi All,

 

I’m looking into purchasing a circular polarizing filter and have a question. During my research, I came upon a review that claimed that polarizers which transmit more light don’t polarize as much as those that transmit less light. Maybe I'm misunderstanding something.

 

True?

 

See link below and look for the review of the B+W 77mm XS-Pro HTC Kaesemann Circular Polarizer with Multi-Resistant Nano Coating

 

 

https://havecamerawilltravel.com/photographer/polarizing-filter-shootout/

I think you're confusing the amount of polarization between circular & linear polarizers, with the changing amount of light passing through a given polarizers with orientation?

Circulars have a second layer & that reduces the polarizing effect as compared to linear.

Is there a metric that can measure degree of polarization. I do not know. Never thought about it except to notice degree of glare reduction and darkening of skies from certain directions in all Pola filters. I am lately using a B and W circular polarizer. Seems good to me. I look to other features when putting something in front of a lens, like the coating and ease of use, etc.. Never thought of the effectiveness of glare reduction / contrast enhancement. Expect that would be hard to measure objectively. A new subject for me.

Oddly I just went through this. I had a linear polarizer, a left over from film days. I used it briefly with my digital camera, which seemed to have zero effect. I quit using it because the Electronic v/f didn't react at all. The CCD actually lost saturation & color!

 

Then I got a new camera & bought a circular one.(slightly different size, 55 Vs 58mm), & it didn't show any effect in the EVF either! I started doing some side by side comparisons visually (no camera) & the results were surprising. The Linear definitely gave a much stronger maximum effect, both for reduced reflections & enhanced sky saturation at the correct angle. Some more research (holding filters over the lens) & it turns out my cameras all will work fine with a linear polarizer as they're mirror-less & don't use a beam splitter anywhere. The worst thing is that I absolutely can't make the viewfinder show an effect, even when the CCD captures it!:(

B+W makes some polarizers with a lower filter factor, by about one stop (1.5 v 2.5). I have examples of each. As far as I can tell, there is no difference in the amount of polarization (attenuation). Because of the rapid compensation for varying light levels, polarizing effects are difficult to see en an electronic viewfinder.

 

Both linear and circular polarizers use the same foil for polarization, but circular polarizes have a second foil (quarter wave plate) which scrambles the polarization leaving the filter. This way semi-reflective surfaces in the camera, which include AF and AE sensors, are not affected. Any attenuation of polarized light is done before entering the second foil.

 

It is possible to measure the amount of polarization, but I've never seen it applied to photographic filters. A lab instrument called a polarimeter is used to measure the angle and amount of polarization, using a birefringent crystal.

Edited September 1, 2017 by Ed_Ingold

If I remember correctly, autofocus doesn't work well or not at all in most cameras when using linear polarizers.

If they use prisms or beam-splitters then no they black out internally. If the system doesn't use them linear is fine.

My Hasselblad works fine with linear polarizers. The meters use microscopic perforations in the mirror to pass light to the sensor. Nikon uses beam splitters, which require a circular polarizer for proper operation. I prefer Kaesemann polarizers, which are sealed on the edges, keeping moisture out. AFIK, they are only offered as circular polarizers, which work on any camera.

Yes, Circular always works, just not quite as well. I found that out the hard way!

Yes, Circular always works, just not quite as well. I found that out the hard way!

What happened?

I posted on here a while back about it. I put one on my Fuji FinePix S 7000 & it desaturated the image! I am still at a loss to explain why. Several members came up with possible answers but none rang 100% "right"

 

Non-polarized image

 

polarized image.

 

I can restoe the colors by simply increasing the saturation & it's almost (but not quite) back to un-polarized!

I have never experienced anything like that with a polarizer of any kind. I recall the thread, but no solution. It looks like an exposure level issue with recovery rather than a polarization effect. The leaves are under flat lighting, so the amount of polarized reflected light would be minimal. I can't see the EXIF data. What were the respective exposures? I would expect to see about 2 stops more for the polarized example.

Its actually heavily backlit, looking upwards towards the sky. I was going for the nice luminous look of Autumn colors when they are lit that way.

 

I'm not sure if that text is readable?

top (non polarized) exposure is 1/200 @ f3.1

Bottom (polarized) is 1/100 @ 3.1

So its a 1 stop difference.

The concepts presented above are the "received text" on polarizer filters - - they interfere on some cameras with AF and AE, it is said.

 

The page is from Popular Photography 2002-September.

 

Although I buy the circular polarizers just in case, I have used older linear polarizers on many of my film and digital cameras. Have yet to see any clear indication of a serious problem.

 

Examples of a specialized "Color Flow™" polarizer at LINK with discussion on stacking different types of polarizers.

Spiratone Color Flow red/blue polarizer

Yes its because beam splitters or semi-silvered mirrors "black out" with linear polarizers at certain angles. If your equipment uses them for light metering (Canon F-1) or auto-focus take offs, (Maxxums) they become a problem.

Thank you. I can read the EXfF data. The non-polarizer image was exposed 1/200 at f/4. The polarized image at 1/100, f/4. That means the polarized image was probably 1 to 1.5 stops underexposed.

 

Regarding back-lighted autumn leaves, that's what I thought too, but with disappointing results. Autumn leaves look radiant when directly illuminated. Since these were illuminated by clear sky, the light would be polarized, causing underexposure and possible color shift.

Yes its because beam splitters or semi-silvered mirrors "black out" with linear polarizers at certain angles. If your equipment uses them for light metering (Canon F-1) or auto-focus take offs, (Maxxums) they become a problem.

That's why a circular polarize is recommended for AE/AF cameras.

But they were both lit identically, from behind (above). Its the same shot, but with & without a polarizer. I could understand if one were front lit & the other back-lit, but that's not the case.

You can't tell from the 2 images I posted but both the polarized and non polarized images were shot with auto bracketing at +/- 1 stop & there's no color change, just density. if its just a 1 stop exposure error there should be a "matching" pair, even if its OVER matching Normal or UNDER matching NORMAL!

Have you tried a different polarizer? Different brand? It's possible you have a defective or mislabeled filter. Then too, Fuji does strange things with their sensors, large and small cells, rearranged Bayer pattern. Have you tried the filter on another camera?

 

I can't imagine leaves, being highly complex, would polarize light in a regular fashion on transmission.

Yes I discovered this when checking to see if I could use a linear, or had to switch to a circular. This exact filter had worked fine on a film camera, so I don't see the bad filter being a cause.

 

Fuji's oddball double interleaved CCD array is possibly a factor, but when I used the same pair of filters on my newer FinePix S9100, which has a very similar "double decker inter-spaced" arrangement they worked. It has a larger filter size though so I switched to a different one just so it would fit.

Here are the results (same tree,. same sun angle, different season) of the tests on the 9100 with the new linear pola. You can see them on the image, but the EVF ignores them so you have to align the polarization manually then st it on the lens. Both filters have an index for full polarization, I guess one might be misaligned, but visually the polarization is strongest with the index at 90 degrees to the sun on both. Because they are indexed I can screw them in & rotate to the same position.

 

 

 

 

 

I set auto exposure on PROGRAM , because I was checking the metering for compatibility too.

How does the original camera do with the CP and other scenes? If it's not the camera, then something about open sky (polarized) as a backlight may be the key. It's worth while to reproduce that situation. I doubt that anything in the leaves would twist or scramble polarized light, but could be. There are a lot of asymmetric molecules in plants, but they would have to be crystallized or highly organized to affect polarization. This is a puzzler.

I must admit I never tried it after that test. It pretty much killed it when added to the v/f not recording the changes as a operable system.

 

I do still have all the components though so I could easily d some test series.