What do You guys do about polarization?

[Kaspar Kallas]

Hi

Very simple question, how deal with polarization from mirror rig?
Any good optical solution?

[Tim Dashwood]

I will typically try to avoid angles with surfaces that have reflections and will be affected by polarization. This can be as simple as moving a shiny prop or using some dulling spray. However I've even had issues in the past with black hair that picked up a beautiful sheen from a backlight on one camera and nothing on the other.
Issues with polarization or sometimes difficult to identify. You basically have to look through one eye, then the other, and spot the differences in reflections.

We have also experimented with films of λ/4 wave plate retarder on the glass, which has shown to be very promising.

[Kaspar Kallas]

Hi

Thanks Tim, could You explain bit more in detail λ/4 wave plate retarder, how it works and how are the results?

Thank You
-Kaspar

[Tim Dashwood]

The 45° mirror itself has a polarization effect on the light, therefore the intent is to nullify the polarization effect.
A 1/4 waveplate combined with the glass may help with this in most situations. 1/4 waveplate will turn circular polarization into linear and vise-versa, which is I think the primary reason it actually works.

The tricky thing right now is finding beamsplitter glass that already has it fused to it. I can buy it in panel form from these guys but an added panel has some other diverse effects on the quality of the image so there is a give-and-take aspect to this.

On a side note: 1/2 waveplate will reverse the clock-wiseness of circular polarization. This is what we use in front of one of our JVC SX-21 projectors to make it counter-clockwise circular polarized. The almost-perfect 1/2 waveplate retarder ended up being a piece of clear plastic wrap that had just the right properties.
If you walk around your house with a polarizing filter in your hand you will find all kinds of day-to-day items that affect polarization. Play with two polarizing filters and you will have hours of fun, especially with LCD screens.
Fun with Polarized Light

[Charles W. Hull]

I'll show my ignorance about mirror boxes, but in most cases couldn't you not use a beam splitter - but shoot straight through with one camera and use a 45 degree front coated mirror with the other? This would avoid the polarization issue; and give double the light for each camera.

Looking at my XF100 you should be able to get down to an interocular distance of about 30mm. Wouldn't this cover almost all shooting situations?

[Alister Chapman]

As I understand it a beam splitter mirror will pass horizontally polarised light and reflect vertically polarised light (or vice-versa). So you OK with circular or randomly polarised light, but if you have a lot of H Pol then only one camera sees it.

A quarter wave retarder as it's name implies delays some of the light passing through it by a quarter wavelength and then it mixed with non-retarderd light which results in the light becoming circular polarised.

Circular polarised light contains both horizontal and vertical components so both cameras see the same thing.

Attaching the retarder to the mirror does not appear to work for a couple of reasons. One is that the angle of the retarder relative to the incoming light is critical and it doesn't work well at 45 degrees. Another is that light passing through the retarder and then back through it again (as in the case of the reflected camera) tends to reduce the effectiveness. In addition it's very difficult to laminate a retarder to a front coated mirror without damaging the reflective coating.

So the most common solution is an optical flat glass with a laminated retarder placed in front of the mirror. But this introduces problems of it's own. You can get internal refections in the mirror box that are hard to eliminate. It adds considerable weight right out at the very front of the whole system, right where you don't want it and you will need even bigger flags and sun shades than you would with just the mirror. Add to all that the cost of a very large optically flat piece of glass and the cost of the laminated retarder layer and it quickly becomes a very expensive, often less than satisfactory solution to an occasional problem.

Depending on the polarisation of the incoming light a rotating linear polariser fitted to each camera can be used to tune out the most severe problems, but this can be difficult to set up and get right, plus there is the additional light loss to take into account.