UV vs visible in non-pyro negatives

[Kirk Keyes]

How is the film's spectral sensitivuty linked to the densitometer's? [...] The densitometer reads those densities based on the intensity and spectrum of the light passing through the negative. There is a complete disconnect between the two. How does the densitometer know what spectrum of light exposed the film? What am I missing?

A film's spectral sensitivity is not in any way linked to the spectral sensitivity of any densitometer. You're right. But depending on how you make the exposure onto the film, your choice of densitometer and filtration could make a difference.

If you make exposures in a camera by varying the amount of light using a shutter and aperature, it doesn't matter. As you say, there is a complete disconnect between the densitometer and the exposure.

But if you make the exposure through a step wedge, then your measurements of the step wedge can vary based on the filtration used in the densitometer. The actual exposure that is given to the film is not affected by your choice of densitometer/filter, only the numbers that you generate which will be used in your calculations later on will be affected. This variablity could be significant.

I'm not talking about measuring densities of the film to which the exposure was made when making these measurements, but the step wedge that was used to control the exposure in the first place.

So what I was trying to point out in that sentence, "But this 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... " is that if you are controlling your exposure on your film (or paper as you pointed out) and the densitometer is not able to make measurements of your step wedge under conditions that simulate the actual conditions of use for your material, then you may have some concerns about the numbers that you are reading off the display of your densitometer.

Let me restate some more of what a said in the other post with a few changes to the wording that may make it more clear -

If you have a step wedge that is a relatively neutral material over the range of wavelengths that your film or paper is sensitive to, then it will not really matter much if our Visual filter does not cover the entire range. You would get comparable results if the material is fairly neutral whether the densitometer can see part of the range of wavelengths or the entire range of wavelengths. It is when we start making exposures through materials that have significant colors in them that this will come into play and then be a big issue.

Did that help?

 

[gainer]

Not me. I still don't have a way to determine if my step wedge is neutral. If it is not neutral then I have no way to compare, to within the proverbial gnat's behind, what I measure with what you measure. I know that some of my negs are not neutral, but neither of us will know by how much from either my measurements or yours, even if we use the same filter, unless we know the native spectral sensitivities of our densitometers as well as the transmission spectrum of our filters.

It is true that when we are dealing with negatives that look neutral we can get a mutually agreeable measurement of contrast index, so that if I tell you that a certain film-developer combination developed under certain conditions gives a certain contrast index, you ought to come within experimental error of the same result in your darkroom. It is also true that what difference there is will be most likely in the form of a difference in film speed rather than in slope if it is due to, say, the difference between a CdS photoresistor and a silicon phototransistor.

I think we stand a better chance of comparing certain results if we stick to narrow band filters. The color separation filters may be narrow enough for most purposes, and a UV bandpass filter for the rest. Determinations of film speed by densitometer readings are subject to validation by practical use.

I'm not trying to pontificate here, just bouncing thoughts off the backboard.

 

[Kirk Keyes]

Not me. I still don't have a way to determine if my step wedge is neutral.

I found a few more sources and found mention of "Status V" densitometry - V for Visual. Of course, no one has gone into enough detail to actually describe the spectral characteristics of the Status V filter. But they did say that they were designed to replicate human vision for densitometric purposes. I suspect that the Gretag Visual filter that was discussed above would comform to this specification, as Gretag seems to know what they are doing when it comes to constructing densitmeters. But as we've seen, the filter used in the D-200 II densitometer, doesn't completely cover the range of human vision. But for densitometric purposes, it is probably sufficient. If more detail about the transmission characteristics of a material are needed, then Gretag and others will gladly sell us a spectrophotometer, which will be able to make density measurements on a wavelength by wavelength basis (or at least every 10 nm according to some of Gretag's brochures).

But along with the type of filters used are requirements on sensor position, light sources...

If it is not neutral then I have no way to compare, to within the proverbial gnat's behind, what I measure with what you measure. I know that some of my negs are not neutral, but neither of us will know by how much from either my measurements or yours, even if we use the same filter, unless we know the native spectral sensitivities of our densitometers as well as the transmission spectrum of our filters..

But if we both have densitometers, and the are both deisgned and contructed to conform to ANSI/ISO specifications that I assume are out there, and the filters used in them conform to the specifications as well, then we should get directly comparable results.

I would suggest that if we had the same filtration, that doing that would go a long way towards getting comparable results. At least we would be looking at the same band of wavelengths. And since most light sources used in densitometers are going to be incandescent, and even with the small range of color temps that those sources may have, they should be close as well. So that leaves detectors - I would be surprised if any modern densitometers use CdS cells (too sluggish and not much linear range) - I bet Si photodiodes are much more likely. Whether they go to the effort of using a Silicon Blue diode or not, I don't know. If they intend to do UV measurements, then it is probably not used as they drop off response faster than a regular Si photodiode.

I looked around the ISO site, and they do have several specs on densitometry, but at around $50 a standard, I have to pass on getting them. I'll see if I can find a library that has some of them around here...

It is true that when we are dealing with negatives that look neutral we can get a mutually agreeable measurement of contrast index, so that if I tell you that a certain film-developer combination developed under certain conditions gives a certain contrast index, you ought to come within experimental error of the same result in your darkroom.

I agree completely. Problems arise with materials that are not fairly neutral.

I think we stand a better chance of comparing certain results if we stick to narrow band filters.

I agree with you when you say we should get better agreeable results if we are both using narrow-band filters. In fact, I'll go further and claim that if we used filters that only transmitted very narrow bands of light, say like if we used a monochromator which could give us bands of light as narrow as a couple of wavelengths, then we should get identical results (within experimental error) as the response curve of the detector and the spectral distribution of the light source will not matter. There will essentially be no slope or curve to either one of those things to come into play. As you have pointed out before, we have reduced all the sources for error down to simply the measurements of the incident vs. the transmitted light.

PS - I like bouncing thoughts too!

 

[gainer]

Considering that in the long run, the purpose of most of our densitometer measurements is to predict with some reasonable accuracy what the printing qualities will be of some arbitrary negative on some known printing paper, we should consider other approaches. One is to print the negative of the step wedge on the printing material in question and note the number of steps between black and white. A print of the original wedge directly on the printing material will show the effective number of steps in the SBR. Dividing that by the number of steps in the print of the negative will give the effective contrast index when those development conditions are used on any scene to be printed on that medium. Any kind of densitometer can probably be used to produce the H&D curve well enough to show peculiarities in the shape and give an idea about rendition of mid tones.

 

[sanking]

One is to print the negative of the step wedge on the printing material in question and note the number of steps between black and white. A print of the original wedge directly on the printing material will show the effective number of steps in the SBR. Dividing that by the number of steps in the print of the negative will give the effective contrast index when those development conditions are used on any scene to be printed on that medium.

Alternative printers have been testing their materials for decades with step wedges. We were using the wedges long before any of us had any idea that there were actually densitometers that could measure the UV stain. Or could afford them!

This method of testing is essential because the range of exposure scale that people get in practice with altternative processes is much greater than the range people get printing different brands of silver papers. The pH of the papers, sensitizing mix, method of coating, and type of developer combine to create a much more variable scenario than silver printers see.

Sandy

 

[Kirk Keyes]

...we should consider other approaches. One is to print the negative of the step wedge on the printing material in question and note the number of steps between black and white.

Patrick, it all really comes down to what kind of precision, accuracy, and predictability we want. Sure, using paper works. Afterall, we are trying to make prints. And it would be foolish not to verify that measurements made with a densitometer did not behave as expected on the printing paper.

 

[gainer]

Patrick, it all really comes down to what kind of precision, accuracy, and predictability we want. Sure, using paper works. Afterall, we are trying to make prints. And it would be foolish not to verify that measurements made with a densitometer did not behave as expected on the printing paper.

Of course. But if you use a 30 step 0.1 wedge, you get within + or - 0.1 in estimating the SBR that your developing procedure and film could put on your chosen paper. It's not likely that two people will agree with each other that closely on measured SBR at the original scene.

If you are trying to define very small differences between films or processes for scientific purposes, then it is necessary to be very specific about the experimental conditions, including the specs of all measuring instruments. I could preach about this aspect of experimentation from my experience at writing and editing NACA and NASA research papers, but I won't because I'm sure it would be like preaching to the choir.

When I have an idea for some test that I think should be done, I may do a quick and dirty trial to see if it's worth pursuing. In a group like this, I might publish that Q&D work as such because the more people you can get to try something, the more likely the final result is to be valuable for either accepting the premise or rejecting it.

So I wound up preaching anyway. Oh well, I guess I have achieved what I never thought would happen. I'm old.

 

[Kirk Keyes]

Patrick, I'm in that choir!

And Quick & Dirty testing does have a place. Even using a step wedge is more testing than probably 99+% of photographers will do.

But as I said, it's all a matter of what kind of precision, accuracy, and predictability someone wants.