What does IR Filter thckness Have to do With Effectiveness?

[barjohn]

Advertisement (gone after registration)

I got to thinking about the controversy about the camera's thin (0.5mm glass) IR cut filter and how they couldn't make it any thicker without it projecting too far, etc. It got me to thinking. The filer is 0.5mm thick. However the IR absorption coating is in the microns of thickness. Why couldn't adding additional micorns of lens (Glass) coating to improve IR aborption be done to eliminate the need for external filters? Glass is very transmissive of IR until it is trapped by a short distance between two or more pains of glass as anyone who has sat in a car with the windows up on a hot day knows. It isn't the thickness of the glass that makes the filter effective but rather the coating and the coatings are in the microns not milimeters.

[ho_co]

John--

Not all glass is the same. Some glass blocks UV, some blocks IR, some blocks yellow, some blocks green, as you know.

 

Check the Kodak specs on the M8 chip. It does not use a coating to block IR, but a particular type of glass. (Snippet attached.)

 

Check the B+W specs for the 489 filter, which blocks IR simply with glass type. (Or see Edmund Optics' site about their two types of IR-blockers, which are the same as B+W's: Edmund Optics - Mounted IR Filters.)

 

There are dichroic coated filters that actually cut transmission of IR. And there are absorptive glasses that attenuate IR transmission.

 

This topic has been handled often on the forum and by Leica. The problem with using coatings on the IR filter over the sensor is both angle of incidence and internal reflections. The problem with using thicker glass is refraction.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[rvaubel]

John--Not all glass is the same. Some glass blocks UV, some blocks IR.

 

Check the Kodak specs on the M8 chip. It does not use a coating to block IR, but a particular type of glass.

 

Yes, Kodak uses an absorbtive, not reflective technique to reduce IR. Canon and others use BOTH. However, Leica was afraid that the reflective IR coating would lead to cyan vignetting with off axis rays, which it does. And hence the mess we are in. However, the next generation of Leica sensors could use a more effective absorbitive filter and perhaps a microlensing reflective IR coating that could go a lot farther to solving the purple penquin/cyan vignetting than the current model. However, the technical challenge is great but Kodak could do it if sufficently motivated. Meanwhile, they came fairly close, for a first effort. I will live with the filters, etc until the M9 comes out.

I am not holding my breath. We are talking >5 years at Leica speed.

 

Rex

[Vivek Iyer]

Yes, absorption depends on the thickness for colored glasses. This is expressed by Beer-Lambert's law:

 

A=ebc

 

Beer's Law - Theoretical Principles

[barjohn]

It appears to me that Leica may have made a mistake in not useing both the B7 glass which gives a 90% reduction and the Dichroic coating which might have been sufficient to eliminate the last 10% (I know maybe more like 9.5%) not absorbed by the B7 glass. Especially since they have the cyan corners anyway, but the purple-black would have been eliminated. I can understand that this would have entailed sensor replacement and that would have been expensive so perhaps they chose the most expediant and least costly solution. If the sensor is a mounted module, maybe they will offer a replacement some day. Were I Leica, I might see this as a way to sell both a new M8n (new revenue) plus a path to upgrade revenue for a new and better sensor if they combined better low light sensitivity with the better filter.

[Vivek Iyer]

Leica did not put any filter on the Kodak sensor. Kodak did. Whatever spectral transmission they have put out is at best suspect.

 

Kodak also claims that this M8 sensor has no UV sensitivity with the cover glass in place.

They are wrong. I do not believe the spectral trace for the Kyocera cover glass is correct at all.

 

See: http://www.leica-camera-user.com/digital-forum/19983-ultraviolet-sensitivity-m8.html