[Leica] George: M8 IR Different filters

[Frank Verizon]

Paul.... The thickness of a piece of glass does not affect its transmissivity of wavelengths.  The coatings do.



Frank Filippone
Red735i at verizon.net

On Nov 30, 2015, at 1:10 PM, Paul Roark <roark.paul at gmail.com> wrote:

The relevant M8 data sheet seems to be this:
 
(Thank you Marty for sending the URL off list.)
At page 28 it shows the graph of the cover glass spectral transmission.
Interestingly, the graph does not look all that different than what
Kolarivision shows for various OEM cover glasses.  See
http://kolarivision.com/articles/two-spectrum-conversion/
However, the M8 cover glass appears to have, as you said, some IR getting
through it way out into the longer wave lengths.  The little chart on the
Datasheet shows this, whereas none of the graphs show enough information to
see this with either the M8 or the other OEM filters.  I just have to
assume that the thicker glasses reduced this IR "leakage" to nil.

Paul
www.PaulRoark.com

> On Mon, Nov 30, 2015 at 8:12 AM, Howard Ritter <hlritter at bex.net> wrote:
> 
> Since you mentioned it, Paul, I have looked at the graph of the M8’s
> KAF-10500-CX sensor’s spectral response with a 0.5 mm BS-7 coverglass. I
> don’t think I can attach it to this post, but I can tell you that it shows
> that both the green and the red photosites have a non-zero, and
> non-trivial, response out well into the IR. By inspection, it suggests that
> the red sites’ response even at 1100nm is maybe 5% of, or only about 4
> stops down from, its peak sensitivity at about 610nm, with only a slight
> falloff in efficiency from 800 to 1100. The green sites’ efficiency is
> maybe half of the reds’. Unfortunately, the graph extends no further. And I
> don’t know whether this coverglass is the same as is used in the M8.
> 
> Since the 093’s passband begins just to the long side of 800nm, there’s
> plenty of overlap between the M8’s sensitivity and the 093’s passband. But
> there’s a wide difference between the shortest and the longest wavelengths
> detected by the M8/093 combination, and even greater in the M8/092 combo,
> meaning that unless the lens being used is just fortuitously achromatic
> across this part of the IR spectrum, it may be impossible to get an image
> that shows well-focused fine details. (If there were a sharp peak in the IR
> part of the response curve, that would mean that when that wavelength,
> contributing most of the light to the image, is in f