Sensitometry and results from an artificial light source

[reneboehmer]

Dear community,

recently, I drove down the roads of my imaginary world of things, which I might add to my workflow. Well, I ended up with a decision; Building or buying a small contact printer to sandwich my Stouffer step wedges to film. A little background information: In a project I will start soon, I'll have to plot an absurd amount of curves for many, different B&W films.

To me, it seems to do this properly I will have to modify/filter a tungsten source in a way that simulates a standard photographic lens, as well as a standard daylight source.

Does anyone here have experience with building something like this and can tell me if it's even a feasible idea?
I saw that old units online are quite affordable. Would it be simpler to purchase one? Are there things to look out for? I would think that the filters used would go bad at some point. Are units made for X-ray ortho film even usable for daylight tests on pan material, or are they built to fit a standard light emitter in X-ray machines?

Another question, I can't really wrap my head around the exposure time that should be used. Can anyone enlighten me? I remember back when I did my 4×5 tests years ago, I did everything in camera and just metered for the medium EV of a scene I set up. How would this translate to such a unit and a step wedge? (Sorry if this is a stupid question, I just can't solve the knot.)

Have an amazing weekend,
all the best

- René

 

[reneboehmer]

Wouldn't a full spectrum LED light like this be good enough? 3D printed housing and a simple shutter.

 

[koraks]

Does anyone here have experience with building something like this and can tell me if it's even a feasible idea?

I've not built something like this (yet), but have played with the idea. I don't see anything inherently problematic about it.

I will have to modify/filter a tungsten source in a way that simulates a standard photographic lens, as well as a standard daylight source.

Frankly, I'd just assume the lens is transparent for all intents and purposes and don't worry about that part.

Wouldn't a full spectrum LED light like this be good enough?

It depends a bit on what you want/ need to get out of it. But personally, I'd definitely go with a decent quality white LED and not worry about some of the nuances. LEDs have the advantage of allowing rapid cycling, so you can adjust the exposure time to whatever you need/want it to be.
If you need to also chart spectral sensitivity (and not just HD curves for 'white' light), then things may change a little and you may need to be more selective w.r.t. the light source.

I can't really wrap my head around the exposure time that should be used. Can anyone enlighten me?

Not really, but in principle, it's evidently possible in a camera setup to correlate the light flux at the film plane to the flux on the other side of the taking lens. I hope someone's going to drop by and fill the both of us in on the optics.

 

[ic-racer]

Reads like you want to construct a sensitometer.

I reviewed a few different types here:

The Great Sensitometer Shootout....

I have a number of sensitometers to compare. Each is a different construction. EG&G: Xenon flash Wejex: incandescent lamp ESECO Speedmaster: green LED ESECO Speedmaster: blue LED I'm going to run some step wedges at different development times and see if they all produce curves that can give...

www.photrio.com

and here:

X-Rite 334 Sensitometer: Inspection and Analysis

With the demise of film medical radiography, sensitometers have hit the used market. Currently on ebay there are about 80 sensitometers and about 80 transmission densitometers. But the sensitometers are about ten times less expensive. I found a seller with six working X-Rite 334 sensitometers...

www.photrio.com

 

[ic-racer]

If you want to build one, there are a number of resources.

This one is particularly nice. Nicer than the one I built many years ago:

http://www.jbhphoto.com/wp-content/uploads/2015/01/buildingsens.pdf

 

[ic-racer]

To sum up ten thousand pages of information:

If comparing two films, LED source is fine.

If comparing a single film to a STANDARD, on needs a light source that can be traced back to a standard. In the case of ISO 6-1993 (film speed) the light source choices are: IS0 sensitometric daylight, 3400k or 3200k (see ISO 7589-1984). LED is not specified.

 

[Nicholas Lindan]

It would help if you would describe the use you plan for the sensitometric data.

 

[reneboehmer]

It would help if you would describe the use you plan for the sensitometric data.

Hello, I am sorry to have not mentioned this. I want to use this to determine film speed and plot the characteristic curves, in order to inform future exposure/development choices. I figured the light source will have to be in line with the ones used in ISO 7589 (Photography — Illuminants for sensitometry), but I don't have this document available to me. Also, its last revision is from 2002, that's why I thought modern full spectrum LED, that is 99% to a CIE D65 normal daylight, might be a good choice. Since, in my mind, the light source's only job is to somewhat represent an average daylight scene, or am I mistaken?

 

[reneboehmer]

I've not built something like this (yet), but have played with the idea. I don't see anything inherently problematic about it.



Frankly, I'd just assume the lens is transparent for all intents and purposes and don't worry about that part.



It depends a bit on what you want/ need to get out of it. But personally, I'd definitely go with a decent quality white LED and not worry about some of the nuances. LEDs have the advantage of allowing rapid cycling, so you can adjust the exposure time to whatever you need/want it to be.
If you need to also chart spectral sensitivity (and not just HD curves for 'white' light), then things may change a little and you may need to be more selective w.r.t. the light source.



Not really, but in principle, it's evidently possible in a camera setup to correlate the light flux at the film plane to the flux on the other side of the taking lens. I hope someone's going to drop by and fill the both of us in on the

Thanks for the reply, Koraks. I see it as a requirement when doing so many different films. It seems like a good thing to have in general.

 

[reneboehmer]

If you want to build one, there are a number of resources.

This one is particularly nice. Nicer than the one I built many years ago:

http://www.jbhphoto.com/wp-content/uploads/2015/01/buildingsens.pdf

Thank you IC-racer,
you guys are always very helpful

 

[reneboehmer]

To sum up ten thousand pages of information:

If comparing two films, LED source is fine.

If comparing a single film to a STANDARD, on needs a light source that can be traced back to a standard. In the case of ISO 6-1993 (film speed) the light source choices are: IS0 sensitometric daylight, 3400k or 3200k (see ISO 7589-1984). LED is not specified.

Thank you for this! I will look further into it. Is there a way to look into these ISO norms without purchase? Public library or so?

 

[MattKing]

If you want to avoid any issues with a discontinuous spectrum light source, a tungsten bulb - including a halogen bulb - filtered to to 3400K or 3200K would remove that concern.
Otherwise you need to exercise some care with your choice of LED.

 

[Milpool]

You might want to have a look at waveform lighting. They make some pretty good continuous spectrum LED sources with fairly good photometric/radiometric specs, and they have much improved flicker characteristics over some other LED options. If you choose to use LEDs va incandescent sources these will be plenty good for any B&W emulsion sensitometry.

I assume you are wanting to estimate illuminance at the film place using luminance or EV readings (which can be converted to luminance)?

 

[Nicholas Lindan]

The easiest solution for the characteristic curve may be to contact a step tablet onto the film using an enlarger or other timed light source. The tungsten enlarger light can be balanced to 'daylight' with an 80a filter, though with black & white film it likely wouldn't make that much difference.

You would need to link the HD curve contacts with the film speed as determined with your meter and camera. Photographing a grey card for a Zone V negative density and finding the corresponding density step would let you link your camera/meter to the film's HD curve.

Or, just use D-76/ID-11/Xtol, rate the film at 2/3 box speed and don't worry about any of this. When taking pictures, unless you are working under controlled studio conditions, the exposure is a bit of a crapshoot. Just expose generously and the scene will fall somewhere on the linear part of the HD curve. Then you just fix it when making the print.

 

[Romanko]

The easiest solution for the characteristic curve may be to contact a step tablet onto the film using an enlarger or other timed light source.

There is a good video on Youtube by Gregory Davis (TheNakedPhotographer) on how he does his sensitometric tests using both an enlarger and a dedicated sensitometer.

Or, just use D-76/ID-11/Xtol, rate the film at 2/3 box speed and don't worry about any of this.

Can't argue with this advice. Unless you are shooting slide film in a 8x10 camera this is all you'll ever need. If you really insist on getting the exposure perfect, use bracketing.

As for me, I would either build a calibrated and fully ISO-compliant measuring apparatus or use a "close enough" film speed and technique that consistently produce good printable negatives. Anything in between in my opinion is a waste of time if your goal is photography. If on the other hand you are interested in sensitometry and properties of photographic materials, than definitely, go for it.

 

[reneboehmer]

The easiest solution for the characteristic curve may be to contact a step tablet onto the film using an enlarger or other timed light source. The tungsten enlarger light can be balanced to 'daylight' with an 80a filter, though with black & white film it likely wouldn't make that much difference.

You would need to link the HD curve contacts with the film speed as determined with your meter and camera. Photographing a grey card for a Zone V negative density and finding the corresponding density step would let you link your camera/meter to the film's HD curve.

Or, just use D-76/ID-11/Xtol, rate the film at 2/3 box speed and don't worry about any of this. When taking pictures, unless you are working under controlled studio conditions, the exposure is a bit of a crapshoot. Just expose generously and the scene will fall somewhere on the linear part of the HD curve. Then you just fix it when making the print.

Dear Nicholas,

For my personal work, I use one developer and one film. I tinkered with times and speeds until my negative started to print well and never changed a thing. I shoot mostly roll films, so developing is not a precise matter anyway.

But, I also operate a commercial lab and I want precision, as well as perfect control to be implemented into our B&W process. I'd like clients to be able to chose from a look-up table the Subject luminance range and have them decide what speed they will need. We have people who shoot lots of 120 in a studio environment, and for them this would be amazing. Also for people who carry different cameras or different camera backs to adjust for the scene. I would like to offer a service no professional is unhappy with, no matter the background.

We develop film sorted by brand and type, by hand. We use one shot HC110, since I wasn't happy with D76 and D76-R results. Might have to give it another go though. I don't like B&W machines, we do it all by hand. We have got one employee who only does B&W. He has a good feeling for developing properly. He's been doing it for some years, but he would also like to pinpoint some numbers.

 

[reneboehmer]

Dear Nicholas,

For my personal work, I use one developer and one film. I tinkered with times and speeds until my negative started to print well and never changed a thing. I shoot mostly roll films, so developing is not a precise matter anyway.

But, I also operate a commercial lab and I want precision, as well as perfect control to be implemented into our B&W process. I'd like clients to be able to chose from a look-up table the Subject luminance range and have them decide what speed they will need. We have people who shoot lots of 120 in a studio environment, and for them this would be amazing. Also for people who carry different cameras or different camera backs to adjust for the scene. I would like to offer a service no professional is unhappy with, no matter the background.

We develop film sorted by brand and type, by hand. We use one shot HC110, since I wasn't happy with D76 and D76-R results. Might have to give it another go though. I don't like B&W machines, we do it all by hand. We have got one employee who only does B&W. He has a good feeling for developing properly. He's been doing it for some years, but he would also like to pinpoint some numbers.

On the basis of this, I think it's understandable, when going through the amount of work it is to calibrate this process for each film type there is, that I would like to use the optimal sensitometer setup.

 

[reneboehmer]

You might want to have a look at waveform lighting. They make some pretty good continuous spectrum LED sources with fairly good photometric/radiometric specs, and they have much improved flicker characteristics over some other LED options. If you choose to use LEDs va incandescent sources these will be plenty good for any B&W emulsion sensitometry.

I assume you are wanting to estimate illuminance at the film place using luminance or EV readings (which can be converted to luminance)?

I looked at some very good LEDs, but I am unsure.

 

[reneboehmer]

To sum up ten thousand pages of information:

If comparing two films, LED source is fine.

If comparing a single film to a STANDARD, on needs a light source that can be traced back to a standard. In the case of ISO 6-1993 (film speed) the light source choices are: IS0 sensitometric daylight, 3400k or 3200k (see ISO 7589-1984). LED is not specified.

Dear IC Racer,
could you go into a bit more detail on how to create an ISO sensitometric daylight source? Is it a simple process? For B&W, I am sure it's not too critical, is it?

 

[koraks]

@reneboehmer you might want to read this post: https://www.photrio.com/forum/threads/which-densitometer.211008/post-2862923 by @dkonigs. He may also respond himself to this thread if he finds the time.

 

[reneboehmer]

@reneboehmer you might want to read this post: https://www.photrio.com/forum/threads/which-densitometer.211008/post-2862923 by @dkonigs. He may also respond himself to this thread if he finds the time.

Thank you Koraks, you got your eyes on everything. Thanks for this.

I red Dereks response and I think for his case, making it a commercially viable product is a different matter, because form factor also plays a huge roll. He will probably also want his to be used for color, which introduces a bunch of complex variables. I guess he would like to go with LEDs which would also complicate it even more After reading through your blog, Koraks, I know you have struggled with this in the past.

Getting a 12,5 × 4 cm (that's approx. the Stouffer 21 step size) evenly lit slit should not be too troublesome. I feel like getting perfectly even, diffused light for such a small surface should be a solved problem already. Thinking about screen manufacturers and their diffusion films.

My initial idea was as followed:
I have a technical Nikon body that was used in microscopes. It has an electronic shutter system controlled by a separate unit. My idea was to take either a LED like mentioned above or maybe a tungsten lamp with a filter and then 3D print a housing that lets me use the Nikon as a shutter unit. Under the Nikon, a tunnel with diffusion material is placed. An antinewton glass presses against a foam board fitted into a 3d printed lower part, that gets clamped to the upper unit when in use. This way I could sandwich the Wedge and negative properly.

The lamp can be turned on before the shutter activates to allow for heat up and stable light output.

It seems like a simple enough idea. I might give it a go.

 

[koraks]

Yeah, sounds simple enough. How simple it'll turn out to be in practice, you'll find out once you start building it. Flare and reflections are a concern that you may (will) have to address in your setup. Yes, in principle/on paper, getting even illumination for your step wedge is easy enough. Getting it right in practice is a different matter.

After reading through your blog, Koraks, I know you have struggled with this in the past.

My work with color LEDs has been mostly aimed at making color enlargements using RA4 paper from color negatives. This is a totally different ballgame.

Maybe @dkonigs might chime in on the math and optics involved in correlating scene illumination to light flux on the focal plane of a camera. I think you still need to tackle this issue if you want to establish reliable ISO-compatible speed ratings for the films you'll test, especially if you don't want to resort to an additional benchmarking test as suggested by @Nicholas Lindan (although that is an entirely realistic and smart practical workaround).

 

[Rudeofus]

Before you try to establish film speed down to 3 digits, I would like to raise the "latent image decay" issue here. It has been investigated here and tested there. Your customers can follow all the procedures to the letter, but if they wait for a few days between shooting and developing, all these measurements will be in vain. Look in the first thread, how much relative humidity affects latent image decay. Look at the second thread to see, how much difference it makes, if you develop your film right away versus after one hour.

Film, like beer, appears to be a living thing, and any quest for perfect accuracy seems to me doomed from the start.

 

[reneboehmer]

My work with color LEDs has been mostly aimed at making color enlargements using RA4 paper from color negatives. This is a totally different ballgame.

Yes, I am aware. I was referring to LEDs complicating the matter.

I would appreciate @dkonigs opinion very much.

 

[reneboehmer]

Before you try to establish film speed down to 3 digits, I would like to raise the "latent image decay" issue here. It has been investigated here and tested there. Your customers can follow all the procedures to the letter, but if they wait for a few days between shooting and developing, all these measurements will be in vain. Look in the first thread, how much relative humidity affects latent image decay. Look at the second thread to see, how much difference it makes, if you develop your film right away versus after one hour.

Film, like beer, appears to be a living thing, and any quest for perfect accuracy seems to me doomed from the start.

Thanks for this. I red through the threads, and it might actually be good to store the undeveloped negatives for some time. In order to approach a more reliable measurement.

Well, it is so many factors. I'll just try my best, do the work, see if it is good in a practical sense and if it is, ill leave it as is. After all, results are the best way to measure.