peta's got a nice write up on one:
http://www.petapixel.com/2010/08/27/...ensity-filter/

although they use 2 CPs and need to flip one. Couldn't you use on CP and one LP?

I actually saw the same thing on instructables today. looks cool. Said it could get to 14 stops??!

interesting

curious....may need to try it out

I tried with 2 mis-sized CPs. I doubt if it even got me 5 -6 stops.. def no where near 10

nevermind...

in theory... yes. but in practical application and field testing... no

it all comes down to the Quality of the CP in question. They isnt Magic in there, its just a sheet of metal wires at high density, low size.
The better the alignment of the "strings" well call them that run through the polarizer, the better the ratio - take a look at my old employers website, they offer a few varieties.
http://www.edmundoptics.com/onlineca...productid=1912
The truth is in the numbers, look a contrast ratio to allow you to asses how good the polarization is. The contrast is between 2 layers, between parallel and crosses states. these state that there is about 0.04% transmission in crossed state.
I used this crap in my "cross polarization" experiments. The results were OK, maybe 3-4 stops
Now check out these. These are what are going to get you 10-14 stops crosses.
http://www.edmundoptics.com/onlineca...productID=2464
obviously, they arent feasible, but these are the type that the GOOD filters will use for vari filters.
it all comes down to how well the "strings" are aligned inside the glass. Higher tolerance = better performance = higher price.