Quote Originally Posted by gainer
I don't know, for example, what the visible channel or the UV channel are measuring.?
That's why we need to check with the manufacturer and see if they list the specifications for our particular instruments. As Sandy shows with his Gretag D-200 II, they are very specific in this information. Xrite says that his model 810 densitometer uses Status A color filters for reflection and Status M color filters for transmission. I have a MacBeth TD-903 Transmission densitometer that has both Status A and M color filters for transmission. My Noritsu DM-201 uses Status M for transmission. As you can see, each manufacturer may take a different apporach.

So what are Status Filters? The following is taken from a Tobias densitometry technical bullitin:

Spectral Response
The reflection densitometer uses similar color filters to those that produce the separations. So you might think that different densitometers would read the same. However, there are some differences in the filters of various types of densitometers causing discrepancies in readings between units. To overcome this, ANSI specifications have defined several system responses for densitometers. Units conforming to these specifications should have a reasonable agreement. Among these spectral responses are Status A and M which are used in photographic applications and Status T, which is generally accepted as the Wide Band Graphic Arts response in North America. European responses are different giving a higher reading on yellow ink; this response is called Status E. Other responses exist such a Status I which is a narrow band response. Because these varying system responses exist, it is important that the Status response of the densitometer that you are using be included in any communication between customer and vendor. (End of Tobias Quote)

Note that last sentence - that was something I mentioned in the other, long thread while I we were discussing the calibration of densitometers.

According to the Kodak Filters for Scientific and Technical Uses book I have from 1980, Status A and M filters "are carefully selected and calibrated to conform to close spectral tolerances. Their use allows more direct comparison of measured densities with other similarly equipped electronic densitometers." It also says they must be used with the #301A IR cut-off filter.

Kodak does not sell Status filters despite making them - they must be purchased from densitometer manufacturers (at least in 1980...).

From the Wratten book, it looks like the Status filters are based on the Wratten #92 for Red, #93 of Green, #94 for Blue. The book does say the 92/93/94 combo is used for densitometric measurements of color films and papers. However, it points out that like all Wratten filters, the precision of these filters is +/- 5%. Perhaps it is these #92/93/94 filters that are being "selected" for use in the Status filters. The then goes on to say that it recommends using a #301A Infrared rejection filter in combination with these filters as they all pass IR.

For less precise work, one could use the narrow band color filters could also be used - the #25 for red, #61 for green, and the # 47B for blue. These filters are often used for making color separations with B&W film, but they will not match the characteristics of the Status filters

Quote Originally Posted by gainer
I ass-u-me that the RGB channels are measuring something related to color separation positives for dye separation or similar processes. Do the specs for a commercially available densitometer provide spectral distribution plots of what these channels measure?
See my last paragraph above. The Kodak book I mentioned above has graphs of the spectral response of all the filters it covers including the Wratten #92/93/94 densitometric color filters I discussed above, but unfortunately not the Status color filters . Gretag gives enough info about the filters in the D-200 II that we could graph the spectral response.

Quote Originally Posted by gainer
If so, what does the visible channel have to do with photosensitive materials, when panchromatic films are more sensitive to red and blue than the eye?
Excellent question! I'm not really sure if a filter is used with all densitometers when they say "Visual" but I suspect there is. The Gretag D-200 II does use one for visible readings - the manual says it has a filter that is centered at 555 nm, has a bandwith of Tmax/2 of 100 nm and Tmax/10 of 180 nm. So most of the light it is seeing is from around 505 to 605 nm, and it will not see anything below 465 nm or above 645. So it is missing all of what would be considered the Violet, Indigo, and Red portion of our visual response (depending on who's wavelength ranges you choose for each color).

Now there may not be anything really wrong with that, especially since Gretag has told us that this is the property of the filter they chose to use. But is may be a concern as you say if your film is more sensitive to red than blue and your densitometer is not measuring any red but is measuring some (albeit only a little) of the blue...

But we also need to keep in mind, that we need to pick the filter we are using to try and best match the conditions we are making these measurements for. If you have a relatively neutral material for the range we are making an exposure, say a step wedge, then it will not really matter much if our Visual filter does not cover the entire range. You would get comparible results if the material is fairly neutral. It is when we start measuring things that have significant colors in them that this will come into play and then be a big issue.

If anyone has access to the old ANSI specs for these things, I would love to see them. Or the current specs either...

Quote Originally Posted by gainer
How much of the UV that is measured will get through the glass of a printing frame?
The float glass I've measured starts to drop significantly around 360 nm and is pretty much opaque by the time you hit 300 nm. These wavelengths are lower than I gave in the other thread. I'm in better agreement with Sandy on this issue now than I was in the other thread...