# How many quarter-wave plates needed to create circular polarized light for 3d Cinema?

1. Aug 19, 2011

### robphippen

So... 3d cinema relies on circular polarized light these days.

Creating circular polarized light requires the use of a quarter-wave plate, which exploits birefringence to;
(a) effectively create two rays of light linearly polarized at right angles
(b) retards one of them by a quarter wavelength
...and the interference between them creates circular polarized light.
But it occurs to me that a quarter-wave plate must be tuned to work for a specific frequency.

So, my question is;
For 3d cinema, where we're talking about light of multiple wavelengths, do they
(a) Have R,G,B beams, each with a narrow wavelength, pass each through a wavelength-specific wave-plate, then combine them back to form the multi-coloured image
(b) somehow or other create a multi-wavelenght quarter wave-plate.

2. Aug 19, 2011

### sophiecentaur

Re: How many quarter-wave plates needed to create circular polarized light for 3d Cin

I imagine that getting good circular polarisation it is not very critical. Failure will only give a bit of crosstalk between right and left channels. If you get it right in the green region then the eye will probably not spot much wrong.

The visible spectrum extends on either side of green by only about 50% and I think there is probably a pretty broad region over which the circularity is fair. Bearing in mind that one doesn't actually watch 'critical' material in 3D - the most gross stereo effects are where the film maker is really trying to Wow you and I'll bet you eyes are doing their very best to get the most 3D information out of what they are seeing.

Also, I wonder whether the inadequacies of the glasses may not be greater - they must be the cheapest possible design, of course, which won't use a quarter wave plate but a version of polaroid material (I'd guess).

3. Aug 19, 2011

### xts

Re: How many quarter-wave plates needed to create circular polarized light for 3d Cin

As for my understanding: in simple projectors, using RGB matrices, you utilise just one circular polariser for each of R/L, made of linear one and QWP fit to middle of spectrum (qwl = 140nm)

In proffesional Imax cinemas, you have six separate projectors for each of RL/GL/BL/RR/GR/BR, so you may take QWP dedicated for each of the wavelengths.

The 3D glasses are just qwp (screen side) combined with linear polariser (eye side)

( I just tested that for my Quantum Eraser at Home project... )

4. Aug 19, 2011

### LostConjugate

Re: How many quarter-wave plates needed to create circular polarized light for 3d Cin

sophie might be right. Check out the picture of the projector, looks like just one is used.

http://www.reald.com/content/proProducts.aspx?pageID=122

Now I wonder if this is the cause for the ghosting problems.

5. Aug 19, 2011

### sophiecentaur

Re: How many quarter-wave plates needed to create circular polarized light for 3d Cin

I suppose what you really need, for a QWP, is a highly dispersive medium in which the velocity is proportional to wavelength (or is it the other way round?) so that the width is a quarter wave at all frequencies.

6. Aug 19, 2011

### LostConjugate

Re: How many quarter-wave plates needed to create circular polarized light for 3d Cin

I also heard that since they are using a semi-conductive screen that some polarization is lost. This could be why it can get so bad at the far edges of the screen.

7. Aug 19, 2011

### sophiecentaur

Re: How many quarter-wave plates needed to create circular polarized light for 3d Cin

I don't think anyone cares too much, whilst they're packing the cinemas with the present system. They can always change to sequential switching when the price of the specs comes down and / or they come up with a better shutter method.

8. Aug 19, 2011

### K^2

Re: How many quarter-wave plates needed to create circular polarized light for 3d Cin

I haven't looked at the projectors, but I've played with polarizer glasses they give out. These things definitely consist of a single quarter wave layer and a single linear polarizer layer. Polarizing light with one pair and blocking it with the other, you can see quite a bit of red light leaking through. Blue gets through poorly. I'm guessing, these are designed for a wavelength somewhere in greens, as suggested earlier in the thread.

Given that, there is no real reason to make the projection more complicated than that. So single quarter wave plate is most likely there.

9. Aug 26, 2011

### andonrangelov

Re: How many quarter-wave plates needed to create circular polarized light for 3d Cin

Robphippen very good question I was going to ask the same question, so thank you for asking instead of me …

10. Aug 26, 2011

### Andy Resnick

Re: How many quarter-wave plates needed to create circular polarized light for 3d Cin

The RealD movies I have been to use inexpensive glasses using plastic components. There are some color artifacts (which you can easily see by looking through the glasses the wrong way), but they are good enough.

The real trick is preserving polarization information after the light scatters off the movie screen- there is a metallic scrim placed behind the screen to accomplish this.

http://www.edcf.net/edcf_docs/real-d.pdf