Linear Polarizer less effective than Circular Polarizer

gmikol · Jul 9, 2009

I have in my possession a couple of Tiffen linear polarizers, and a couple of Tiffen circular polarizers. Having been frustrated recently by not having enough neutral density to shoot some waterfalls the way I wanted to, I started playing around with the idea of making a variable neutral density filter with the polarizers.

Regardless of whether that's a good idea or not, I made a significant (to me) discovery. The CP's that I own are more effective at killing reflections (and get darker when crossed) than the LP's.

Now, my understanding of how a CP works says that the 1/4-wave retarder behind the polarizer should not affect the image through the filter, so this would suggest that a different polarizing film is being used in the CP's vs. the LP's, or that they have started using a different (less effective) film across the line, since the LP's are brand new, but the CP's are 5-7 years old.

Here's what I saw...

LP Reflection (1/15s f/4.0):

CP Reflection (1/15s f/4.0):

LP Crossed (1/6s f/2.8)

CP Crossed (1/6s f/2.8)

Does anyone have thoughts on this, or any explanation for why this is the case?

Something to keep in mind if you start to feel like your LP's aren't quite cutting the reflections as much as you think they should.

--Greg

Q.G. · Jul 10, 2009

Your own explanation will be (partly) correct: different polarizing foil.

Partly, because the same less effective foil could also be used in circular polarizers, the better also in linear polarizers.

How did you cross the circular polarizers?
With two of them pointing the same way, with the polarizing bit towards the light, the light hitting the second one is not polarized, and cannot be blocked 100% by it.
With both of them pointing the other way, same impossibility to block all of the light.
Have one reversed so that the quarter wave plates are on the inside of the 'sandwich' would have little effect too.
Have one reversed so that the quarter wave plates are on the outsides of the 'sandwich' would work. But how do you mount them like that?

cowanw · Jul 10, 2009

I think that all the light coming through the first polarizer is totally polarized. The second one mounted in the usual forward way blocks that polarized light when turned to 90 degrees. Theory may not be right but it sure works.

Rick A · Jul 10, 2009

Did I read your post correctly-that you are stacking polarizers? You only need one pola filter, and turn it until you reduce as much reflection as possible. If you cannot eliminate reflections entirely, then change your angle of view, readjust the filter until you get the desired result. Using two polas stacked is only a science lesson in how light is transmitted. I only carry one linnear polarizer(Cokin) I meter first then install my pola - turning until I get what I need then shoot. Works 99% of the time.
Rick

Steve Smith · Jul 10, 2009

At least the first polariser needs to be linear, otherwise the second polariser (regardless of type) will always be blocking some of the light from the first regardless of rotation as it has been 'randomised' by the circular polariser at the front.

Steve.

AgX · Jul 10, 2009

Looking at data sheets of polarizer foils one will see that they have varying efficiency. A typical value would be something of a transmission (of originally unpolarized light) of about 35%. In principle you could reduce that transmission somewhat by using two foils orientated parallel.

Sirius Glass · Jul 10, 2009

Q.G. said:
Your own explanation will be (partly) correct: different polarizing foil.

Partly, because the same less effective foil could also be used in circular polarizers, the better also in linear polarizers.

How did you cross the circular polarizers?
With two of them pointing the same way, with the polarizing bit towards the light, the light hitting the second one is not polarized, and cannot be blocked 100% by it.
With both of them pointing the other way, same impossibility to block all of the light.
Have one reversed so that the quarter wave plates are on the inside of the 'sandwich' would have little effect too.
Have one reversed so that the quarter wave plates are on the outsides of the 'sandwich' would work. But how do you mount them like that?

Add to that the the comparison of the LP and CP be from the same manufacture and the same product line. Anything else is not a fair comparison.

Steve

gmikol · Jul 10, 2009

OK...just to clarify. I am *NOT* stacking polarizers in normal use.

While working with the LP's, I felt like the filter was not killing enough of the reflections, so I rigged up a quickie torture test. The results of which I presented. And it seems to indicate that the Tiffen LP that I own is less effective than the Tiffen CP that I own.

The test with the crossed polarizers is just a diagnostic test. It's another way to gauge the polarization efficiency of the filters. For the CP's, I did have the 1/4 wave plates on the outside of the sandwich, and I just held the second filter (adjusted for maximum blackness) in front of the one mounted on the lens.

cowanw said:
I think that all the light coming through the first polarizer is totally polarized. The second one mounted in the usual forward way blocks that polarized light when turned to 90 degrees. Theory may not be right but it sure works.

The theory *is* right, but it assumes an ideal polarizer. In practice, it seems that the LP's that I have are admitting a significant fraction of non-polarized light, which is why there is a significant amount of image-forming light passing through.

Now, don't get me wrong...Tiffen (and every other filter manufacturer) has every right to use a less-effective foil in their less-expensive LP's. It's the nature of business. I'm just disappointed, since I didn't want to spend 2x on the CP's.

But this was as much a curiosity to me as anything else. When I figured out what was going on, I thought that other APUGgers might be interested to know that a CP and an LP might behave differently in a worst-case situation.

--Greg

Q.G. · Jul 10, 2009

gmikol said:
But this was as much a curiosity to me as anything else. When I figured out what was going on, I thought that other APUGgers might be interested to know that a CP and an LP might behave differently in a worst-case situation.

And a LP and LP. Or a CP and a CP.

Just so this point is clear: it does not depend on being either a linear or circular polarizer. It's not a linear vs circular thing, but a good vs less good thing.

cowanw · Jul 11, 2009

I was intrigued enough to get out my collection of polarizers and test the various arrangements of stacking two. And what I discovered for myself is as follows.
If a linear polarizer is on the front, it functions as a variable ND filter, no matter if the polarizer next to the lens is linear or circular.
If a circular polarizer is on the front, the combination acts as a polarizer if the front polarizer is turned and the polarizer next to the lens, circular or linear, acts as a ND filter only. Turning the polarizer next to the lens has no polarizing effect, if the outer polarizer is circular.
So for B&W, stacking two polarizers, with a circular one on front, will yield a polarizing effect with additional ND effect. (more glass surfaces and colour shifts not withstanding)
And stacking two polarizers, with a linear one in front, will yield only a continously variable ND effect.
I thought this was interesting.
No doubt others already know this.

Bob-D659 · Jul 11, 2009

cowanw said:
And stacking two polarizers, with a linear one in front, will yield only a continously variable ND effect.
I thought this was interesting.
No doubt others already know this.

If you set and hold the front one at the correct angle setting to filter out reflections etc, then you can rotate the rear one to act as a ND filter and have both effects.

polyglot · Jul 11, 2009

I built a vari-ND recently using two CPLs joined with a reversing ring. You can see the results (tested on digital) here.

Summary: it's good for about 9 stops attenuation without too much effort. Maybe more, but I ran into issues with (I think but am not sure) the bayer mask on the sensor - probably wouldn't have that problem with film.

gmikol · Jul 12, 2009

Q.G. said:
And a LP and LP. Or a CP and a CP.

Just so this point is clear: it does not depend on being either a linear or circular polarizer. It's not a linear vs circular thing, but a good vs less good thing.

...and no obvious way to tell the difference without testing it yourself, or finding a photo mag or online site to conduct a comprehensive evaluation.

Thanks, Q.G. for taking my comments beyond my issue and into a broader context.

I guess I'll stick with the CP's I own, even though I don't need them for my MF, non-AE cameras.

Thanks for the comments, everyone.

--Greg

Linear Polarizer less effective than Circular Polarizer

Privacy & Transparency

Privacy & Transparency