Circular & Linear Polarizers

[lefei]

What's the diff?

[Ian]

Quote: Originally posted by lefei What's the diff?

The difference is how the polarizer deals with light, linear polarizers polarize light in a linear fashion, while a circular polarizer consists of a linear polarizer and a 'depolarizer' that scatters the linearized light in a random manner.

There are no differences to the result on film between linear and circular polarizers.

However, Linear polarizers generally intefere with the auto focus sensors on AF cameras and therefore you should always use a circular polarizer with an Auto focus lens. In some cases linear polarizers will also intefere with modern exposure metering systems and lead to exposure errors.

[mervlam]

Quote: Originally posted by Ian The difference is how the polarizer deals with light, linear polarizers polarize light in a linear fashion, while a circular polarizer consists of a linear polarizer and a 'depolarizer' that scatters the linearized light in a random manner.

Actually, a linear polariser consists of only a layer of polarising film. But a circular polariser consists of both a layer of polarising film plus a quarter-wave plate. You have to learn some physics of light at university level to know what is a quarter-wave plate. I had forgotten......

a circular polariser actually polarise light in a same manner as a linear polariser, just that the behaviour of the polarised light is different. Therefore, circularly polarised light does not interfere with the metering system of an AF camera

[roygoh]

Quote: Originally posted by mervlam Actually, a linear polariser consists of only a layer of polarising film. But a circular polariser consists of both a layer of polarising film plus a quarter-wave plate. You have to learn some physics of light at university level to know what is a quarter-wave plate. I had forgotten...... a circular polariser actually polarise light in a same manner as a linear polariser, just that the behaviour of the polarised light is different. Therefore, circularly polarised light does not interfere with the metering system of an AF camera

I believe the quater wave plate rotates half of the linearly polarised light by 90 degrees as well as delay the propagation by a quarter wavelength. The resulting wave has 2 components of E field, orthorgonal to each other and with 90 degrees phase difference. If you plot the E field at any point against time, you actually get a vector turning in a circle with constant amplitude. That's why it is called a circular polariser. Just my guess.

[mervlam]

Quote: Originally posted by roygoh I believe the quater wave plate rotates half of the linearly polarised light by 90 degrees as well as delay the propagation by a quarter wavelength. The resulting wave has 2 components of E field, orthorgonal to each other and with 90 degrees phase difference. If you plot the E field at any point against time, you actually get a vector turning in a circle with constant amplitude. That's why it is called a circular polariser. Just my guess.

ar... yes u are correctly.... u DO remember your work.

[roygoh]

Quote: Originally posted by mervlam ar... yes u are correctly.... u DO remember your work.

Hehe...actually did not take physics in U.

Remembered something about Maxwell's equations from my electrical engineering classes.

Recently I was reading Issac Asimov's "The Left Hand of the Electron" and in one of the articles discussing Newton and his contemporaries' experiments to understand light, he mentioned about crystals with molecular structures that rotate the polarity of light. The amount of rotation is proportional to the thickness of the crystal the light passes through.

That's how I got the idea of what a circular polariser actually is. Haven't verified my idea with any physics guru yet.

[mervlam]

actually that's a first year engineering physics course here in NTU....

going off topic liaoz

[roygoh]

OK. Let's get back on topic.

To lefei,

Apart from the technical differences between a linear and circular polariser, the main concern for a user is the compatibility with the camera's optical and metering systems.

More and more cameras are using some form of light-splitting prisms in the internal optical design to cater for exposure metering, autofocusing, as well as view finder applications. Such metering systems will be messed up if the light entering the camera is linearly polarised.

That's when a circular polariser is required to maintain the compatibility.

There is not much difference in the price, I believe, and if a camera is compatible with both types, then the effect will also be the same.

As such, going for a circular polariser will be a better bet.

[roygoh]

By the way this topic has been discussed numerous times before. Do a search on "polariser" and you will get a lot of useful information.

[Ian]

Quote: Originally posted by mervlam Actually, a linear polariser consists of only a layer of polarising film. But a circular polariser consists of both a layer of polarising film plus a quarter-wave plate. You have to learn some physics of light at university level to know what is a quarter-wave plate. I had forgotten...... a circular polariser actually polarise light in a same manner as a linear polariser, just that the behaviour of the polarised light is different. Therefore, circularly polarised light does not interfere with the metering system of an AF camera

This will teach me to give a simple answer with an admittedly poor choice in wording

However, the net result is that a retarder (the correct name for a 1/4 wave plate) as Roy has pointed out gives a delay in propogation along one of the two output axis.

Oh it's almost 18 years since I completed my MSc so the old brain is starting to ossify big time.

[Ian]

Quote: Originally posted by roygoh Recently I was reading Issac Asimov's "The Left Hand of the Electron" and in one of the articles discussing Newton and his contemporaries' experiments to understand light, he mentioned about crystals with molecular structures that rotate the polarity of light. The amount of rotation is proportional to the thickness of the crystal the light passes through. That's how I got the idea of what a circular polariser actually is. Haven't verified my idea with any physics guru yet.

The material is Calcite and it's used in the Glan-Thompson polarizer design that is used in instrumentation where a wide optical bandwidth is required amongst other considerations.

[lefei]

Quote: Originally posted by roygoh By the way this topic has been discussed numerous times before. Do a search on "polariser" and you will get a lot of useful information.

not much info on the search...anyway, i've gotten a circular polarizer

thanks to everyone that answered to this thread

[mervlam]

This will teach me to give a simple answer with an admittedly poor choice in wording



Didn't know that my English is that bad..... Maybe that happens after 30 months in the SAF



However, the net result is that a retarder (the correct name for a 1/4 wave plate)

But 1/4 wave plates are stated as a mica quarter-wave plates on most American physics textbooks.

[Ian]

Quote: Originally posted by mervlam But 1/4 wave plates are stated as a mica quarter-wave plates on most American physics textbooks.

Since when were the Yanks known for using correct terminology

[mervlam]

Quote: Originally posted by Ian Since when were the Yanks known for using correct terminology

Aussies and the Brits can't be correct all the time either.

[stl]

If i have 2 lenses with different filter size eg 55mm and 62mm. what is the converter called? or there is no such thing? meaning i have to purchase 55mm and 62mm filters...

[mervlam]

You can purchase a 55 to 62mm step-up ring. You can screw the ring on a 55mm thread lens and screw-in a 62mm filter.

[ckiang]

Quote: Originally posted by stl If i have 2 lenses with different filter size eg 55mm and 62mm. what is the converter called? or there is no such thing? meaning i have to purchase 55mm and 62mm filters...

You SHOULD buy a 55mm and 6mm filters, helps to boost the economy.

Otherwise, you can get a stepping ring of the correct size. E.g. for your example, you would get a 62mm polarizer filter, and get a 55 (lens)->62mm (filter) stepping ring.

Regards
CK

[TME]

Quote: Originally posted by ckiang You SHOULD buy a 55mm and 6mm filters, helps to boost the economy. Otherwise, you can get a stepping ring of the correct size. E.g. for your example, you would get a 62mm polarizer filter, and get a 55 (lens)->62mm (filter) stepping ring. Regards CK

Am I correct to assume that generally, u put a larger filter over a smaller lens? This way, there is no danger of vignetting? How much difference in diameter can be tolerated by the camera in general? I have a 49mm and a 72mm lens, I guess that is way too far? And the next lens I am going to get is 77mm in diameter. Would it be okay for 72mm to step up to 77mm? BTW, the 77mm is an ultra wide.