Re: A thinner sensor stack may be possible after all!
Actually, there is a lot more to it than the thickness of the filter, although for in-situ filters (vs interference filters like the B+W 486) thicker absorbs more (of everything, including IR).
The selection of the glass type can easily have orders of magnitude more effect than reducing the thickness by a factor or two or three.
The challenge is finding a filter that both provides the correct (or close enough) visual band shaping to maintain the white balance tuning of the sensor+firmware system while still providing good enough IR and UV blockage. I believe the M8 issues were less about the thickness, although it contributed to the problen no doubt, and more the filter glass selection.
If anyone would like to study this more, the Schott glass transmission curves are all in their technical data section. You will see that the IR blockage of the various Blue-Green glasses used by virtually all manufacturers vary by several orders of magnitude. Reducing the thickness by half effects it only by a factor of 2.
As to why not use interference filters, which are surface coatings and unaffected by glass thickness, they have (as mentioned above) issues with large ray angles, so are not a good solution for this application.
Michael
Feb 24, 2015 at 01:00 AM
Previous versions of mdemeyer's message #12878405 « A thinner sensor stack may be possible after all! »
