Polarizer angle

[Terence Murphy]

I'm shooting outdoors all day this weekend, so I've added a polarizer to my slowly growing video accessory arsenal. I understand the concept of a polarizer and how to set it, etc. What I can't seem to dredge out of my college physics (and I started off as a physics major, you'd think I'd remember this stuff) is what variables will affect the light polarity off of reflected surfaces. I'm thinking it should be related to sun angle, and orientation of the camera relative to the reflected surface. But how quickly will the optimal angle change during the day? Basically, how often do I need to re-adjust the polarizer angle? Every hour, or just morning vs. afternoon? And if I rotate my shooting position by 90 degrees, does the polarizer require small or drastic changes in angle?

-Terence

[Justine Haupt]

Originally Posted by Terence Murphy

I'm shooting outdoors all day this weekend, so I've added a polarizer to my slowly growing video accessory arsenal. I understand the concept of a polarizer and how to set it, etc. What I can't seem to dredge out of my college physics (and I started off as a physics major, you'd think I'd remember this stuff) is what variables will affect the light polarity off of reflected surfaces. I'm thinking it should be related to sun angle, and orientation of the camera relative to the reflected surface. But how quickly will the optimal angle change during the day? Basically, how often do I need to re-adjust the polarizer angle? Every hour, or just morning vs. afternoon? And if I rotate my shooting position by 90 degrees, does the polarizer require small or drastic changes in angle?

-Terence

Ok... I'm a CURRENT physics major and I think I can help. A polarizer will allow 100% of polarized light through at the angle of polarization, and 50% total of unpolarized light... the relation is simply to multiply intensity by the cosine of phi, the angle between the polarizer's polarizing axis and the axis of the polarized light (0% transmission at 90 degrees and 100% at 0 degrees, with an ideal polarizer). In short, if you're shooting East or West on a relatively flat surface at the equator during the solstice (ideal solar path... straight over head), and you're only concerned with the landscape that draws a path exactly from east to west than you should't have to adjust the polarizer at all, whereas if you were faceing north or south in the same scenario, you'd have to adjust it often (at least once every half hour). Haha, I can't believe I dared preface that with "in shot..."

You have to find the angle between the horizontal and a line connecting the sun and your subject. The further to the left or to the right of the frame is the sun, the larger the angle between a point directly in front of the camera and the sun, and the polarizer should be rotated 90 degrees to that angle.

An extreme case: Sunset. The sun is on the horizon to the west. You have your camera pointing north. The polarizer's polarizing axis should be verticle, because the incident light will be reflecting from parallel to the horizon, and any light reflected at that angle will be polarized 0 degrees to the horizon, and 90 degrees to the horizon would be verticle.

If the sun is directly in line with the camera's field of view (no matter what inclination), than the polarizer should be horizontal, as all incident light will be coming from the verticle.

Or... just twist until it works ;)

[Terence Murphy]

Thanks Justin. You made me realize one (of probably many) fallicies in my thinking. I was trying to figure out how the polarity of light off the same reflected surface would change during the day (holding the camera position constant). But I'll be dealing with different reflected surfaces as the day progresses, each with a different polarity. At least the polarity of the sky haze will change very slowly, so I don't have to worry about shots having drastically different sky coloration if I haven't re-adjusted the polarizer in a while.

-Terence

[Jarrod Whaley]

The simple, math-free way to figure this out is simply to adjust the filter whenever it needs to be adjusted. :)

(Oh, like Justin said at the end)

[Andre De Clercq]

Polarisation by reflection does exist whenever a non metalic object shows specular reflection and is always occuring even in absense of sunlight. The highest amount(100%) of polarization of the reflected light occurs when the reflecting area (plane) is seen under the Brewster angle (+/- 55 degree). At that angle the polarizer when rotated in the right poisition can kill virtually all specualar reflection components. Only optimal "sky darkening" is dependent on the sun's position and camera orientation.

[A. J. deLange]

Sunlight scattered by the atmosphere tends to be linearly polarized (no idea why) and this effect is most pronounced when shooting perpendicular to the line between the camera and the sun. Light reflected from clouds is not so polarized so the polarizing filter can make clouds pop right out largely by darkening the blue sky. The polarization doesn't shift with time (AFAIK) so the advice already given (tweak it until you like it) is good. You shouldn't need to readjust but there is a very good chance that the setting will get bumped as you move the camera around so it never hurts to tweak the filter as you set up each shot.

In a specular reflection horizontally polarized and nearly horizontally polarized waves enter the medium and are not reflected beyond the Brewster angle (which depends on the indices of refraction of the two media one of which is presumably air). The practical advice here is again rotate the filter for best effect but also try changing the angle between the camera and the reflecting surface to see if you can get more suppression (assumng that's what you want).

[Andre De Clercq]

Indeed the clear sky looks blue because of the partly polarised "reflection" (in fact Rayleigh reflection) when the light from the sun hits the gas molecules (mainly nitrogen and oxigen) in our atmosphere. Without this reflection, the sky would look perfectly black (that's what astronauts see). By taking away the the polarised part of the light, the blue sky gets darker, and the clouds which do not polarize the sunlight are not dimmed. The largest effect is indeed seen at the angle perpendicular to the sun/camera axis.
The polarization axis of unpolarized light which becomes polarized by reflection depends on the orientation of the reflection plane(parallel). The largest effect occurs at the Brewster angle and lowers when the viewing angle gets off of this optimal angle.
Tip: never think on the above and rotate your polarfilter until you get what you want.

[Terence Murphy]

This has been an incredibly useful lesson, and I think I even understand about half of it (!). Its also a good reminder for why I switched from majoring in Physics to Molecular Biology....

My filming went pretty well, and I think the polarizer did help a little bit, although not as much as I would have liked. I had hoped to cut down on glare from some reflected surfaces, but most of those surfaces are curved or painted with a rough surface so I suspect they are not reflecting with even polarity. Consequently, I just set my polarizer angle based on the background sky (and sure enough, I saw very little effect with the sun behind me, but more effect when the sun was at 90 degrees). I rarely had any sky in the shots, and it was often a little hazy, so I don't see any dramatic effects. I was able to let the reflected surfaces blow out a little without it being too obvious in the final footage, so I could keep the overall brightness up at a nice level.

Thanks for all the advice. My technique is constantly improving, largely because of all the great information posted on this site.

-Terence