[atupdate]

I just scored a couple of Hoya rotatable linear polarizing filters from Craigslist (among other items). Is there any benefit to stacking them? I would need to get a step-up adapter to stack them and I don’t want to go through the effort if someone knows that it doesn't work.

Tim

[scratchpaddy]

Interesting question. If you stack them, you get an ND filter with unlimited stops, right? Two polarizers perpendicular to each other block all light. That's how LCD screens work. You'll get an adjustable ND filter with a range of between ~3 stops and total darkness! Awesome!

[atupdate]

Interesting question. If you stack them, you get an ND filter with unlimited stops, right? Two polarizers perpendicular to each other block all light. That's how LCD screens work. You'll get an adjustable ND filter with a range of between ~3 stops and total darkness! Awesome!

That's what I was thinking but I'm sure there must be a downside somewhere.

Tim

[scratchpaddy]

I'm sure there must be a downside somewhere.

There always is. You said they're linear, so your autofocus won't work (or is it the metering? Somebody correct me if I'm wrong). I'm guessing there's going to be some color cast, too. Just try it without the step-up ring first. You want one because you're worried about vignetting, right? How wide is the lens you want to use them on?

[JimJohnson]

You said linear polarizers. As I understand it, they can muck with your exposure metering and that is why all current cameras specify circular polarizers. If you hold a circular polarizer up you will see there is a difference whether you are looking from the back side or the front side. With linear polarizers the filter acts the same regardless which side you are facing.

Does this still matter is you are stacking filters? I'd love to hear the answer to that one myself. It would be a LOT more convenient to pack a couple step rings and get double duty out of my CPL's than invest in comparable quality ND filters.

[Docrwm]

You said linear polarizers. As I understand it, they can muck with your exposure metering and that is why all current cameras specify circular polarizers. If you hold a circular polarizer up you will see there is a difference whether you are looking from the back side or the front side. With linear polarizers the filter acts the same regardless which side you are facing.

Does this still matter is you are stacking filters? I'd love to hear the answer to that one myself. It would be a LOT more convenient to pack a couple step rings and get double duty out of my CPL's than invest in comparable quality ND filters.

+1. They can really mess with metering and other aspects of the digital age. They were made for film use.

[atupdate]

You want one because you're worried about vignetting, right?

No, my kit lenses are 52mm and the LP's are 58mm and 62mm.

Tim

[Not a Number]

Have some fun with the birefringent effect. Put a piece of clear plastic (like the cover of the case for a filter) between the two polarizing filters and see what happens when you look through on of the filters.

[abmj]

Technically, unless you get them lined up exactly with each other, they should cancel everything and give you no picture. In reality, it probably won't be perfect and you will get some light transmission but I wouldn't know how to measure or meter it.

[AstroDave]

Technically, unless you get them lined up exactly with each other, they should cancel everything and give you no picture. In reality, it probably won't be perfect and you will get some light transmission but I wouldn't know how to measure or meter it.

Not true! If the light from the first polarizer is perfectly polarized (it should be close), then it is the cosine of the angle between the polarized light (or the axis of the first polarizer) and the second polarizer's axis of polarization that will determine the amount of light that gets through. If they are parallel (angle=0 ==> cosine = 1), all the light that gets through the first filter goes through the second one (this is probably the case you were thinking of), If they are at 45 degree angle (cosine = 0.707), then 0.7 of the light gets through. Only at 90 degrees (axes crossed exactly; cosine=0) will (nearly) all the light be attenuated.

Think of looking through a polarizer at something that is highly polarized (reflection from water, say). The reflected light doesn't go away for all but one angle of the filter. The variation in light CHANGES as you rotate your filter. Same effect here (and this is how variable ND filters work, in fact).

[scratchpaddy]

Technically, unless you get them lined up exactly with each other, they should cancel everything and give you no picture. In reality, it probably won't be perfect and you will get some light transmission but I wouldn't know how to measure or meter it.

Not true! If the light from the first polarizer is perfectly polarized (it should be close), then it is the cosine of the angle between the polarized light (or the axis of the first polarizer) and the second polarizer's axis of polarization that will determine the amount of light that gets through. If they are parallel (angle=0 ==> cosine = 1), all the light that gets through the first filter goes through the second one (this is probably the case you were thinking of), If they are at 45 degree angle (cosine = 0.707), then 0.7 of the light gets through. Only at 90 degrees (axes crossed exactly; cosine=0) will (nearly) all the light be attenuated.

Think of looking through a polarizer at something that is highly polarized (reflection from water, say). The reflected light doesn't go away for all but one angle of the filter. The variation in light CHANGES as you rotate your filter. Same effect here (and this is how variable ND filters work, in fact).

Okay, to simiplify, abmj has it backwards. If the polarizers are perfectly perpendicular, they will block all light. Any other angle, and they will block varying amounts of light. The computer screen you're looking at right now is proof of this.

[Nass]

When I tried it a couple of years ago I got x (cross) patterns of darker areas

[dms]

I have no problem metering with K20D and linear polarizing filter (LPL). And yes stacked they act as a variable ND filter. Although i don't use them stacked--I do use LPL sheets on my lights and LPL om my lens and the colors are true (use to photograph paintings--so off color would be an issue).

[Stone G.]

Well I don't have two linear polarizers, but I have one linear and one circular, and I have never had any real problems with metering using either on my Pentax DSLRs. And yes, you can stack them and use them as a variable ND filter. Here's a quick and dirty proof-of-concept made a few minutes ago with the stack on my DA 18-135 and K-5. Av-mode and AF used. The light meter determines the exposure time.

• First Image - Filters almost parallel: 1/13 s, AF worked
• Second image - Pretty close to perpendicular filters: 4s, AF still worked
• Third image - very close to perpendicular filters (oops, I need to dust off something): 10 s, AF didn't work because there was so little light.

You will note the color cast that occurs near total blocking of light. That will have to be balanced in post-processing.

But there is more fun with two linear filters: You can use one on your flash and one on your macro lens and reduce flash reflexions in macro and clos-up photography. I am not the inventor of that idea, but I have shown some examples here: Cross-Polarization and the - Completely - Lost Reflexions

[Lowell Goudge]

I did this to shoot the passing of Venus a year or so ago. Gave a good high FStop ND filter