Camera+Film+FilmDeveloper Process Control idea for Zone System followers

PeterB · Apr 11, 2013

Hi folks,

As a Zone System practitioner I would like a relatively quick way to ensure that the process variables for my camera, film and film developer are "in control". I would like to propose a method to do this and would appreciate your feedback on its pros and cons.

BTW I don't (yet?) own a sensitometer nor have I purchased any KODAK Black-and-White Film Process Control Strips (CAT 180 2990). They would only test out that my film development process was in control, so instead with the following suggestion you can widen the net and include your own film plus camera system too.

Aim:

To expose a roll of film with 3 frames, each frame being a shot of the same grey surface (not necessarily 18% grey) exposed to Zones III, V and VII by only varying the camera's aperture. Process the film for N development, then measure the density of each frame to see if it matches your previously determined density targets for those zones at N development. You of course will need to have previously determined these densities using your favourite ZS methodology, and thus be in possession of a densitometer.

Method:

1. Find any mid grey surface to mostly fill the frame of your favourite format.
2. Expose frame 1 for Zone III using your light meter. Choose a shutter speed on your camera that you have measured to be within 10% of its nominal value. Use either f22 or f16.
3. Expose frame 2 for Zone V using the same shutter speed as step 2 and opening up the aperture to either f11 or f8.
4. Expose frame 3 for Zone VII using the same shutter speed as step 2 and opening up the aperture to either f5.6 or f4.
5. Ensure the light level hasn't changed between steps 2 and 4.
6. Process film for N development.
7. Measure the transmission density of each frame and ensure it matches your expectations to within your acceptable margin of tolerance (say +/- half a zone). You should understand the Zone system to know where those densities should fall.

A possible variation is to take 3 frames of each zone to check for variability of your shutter timing.

The advantage of this method is that:

It is less sensitive to flare you would get by having say a test surface/card with the three different Zones on it at once.
You also don't have to create or maintain such a 3 zoned surface which takes more time and relies on the accuracy of your light meter in high flare conditions to ensure the surface was calibrated.

dpurdy · Apr 12, 2013

For my quick tests I generally only worry about Z1 and Z7... film speed and development time.
I do have a densitometer but am often too lazy to warm it up. I have a sheet of film that I know reads 1.2 that I keep on my light table for quick visual comparisons to my test Z7 and I have pretty good memory recognition of Z1.
As you are using the Zone system I assume you have a spot meter. You can use your spot meter as a quick rough density checker if you have a good size piece of film at a known density you can put on a light table. Make a snoot for your spot meter to block flare light from the light table.
Dennis

PeterB · Apr 12, 2013

Thanks Dennis. I have both a spot meter and a densitometer. The test I propose gives a relatively quick check of ALL process variables in the system up to the processed negative.

rgds
Peter

cliveh · Apr 12, 2013

PeterB said:
Hi folks,

As a Zone System practitioner I would like a relatively quick way to ensure that the process variables for my camera, film and film developer are "in control". I would like to propose a method to do this and would appreciate your feedback on its pros and cons.

BTW I don't (yet?) own a sensitometer nor have I purchased any KODAK Black-and-White Film Process Control Strips (CAT 180 2990). They would only test out that my film development process was in control, so instead with the following suggestion you can widen the net and include your own film plus camera system too.

Aim:
To expose a roll of film with 3 frames, each frame being a shot of the same grey surface (not necessarily 18% grey) exposed to Zones III, V and VII by only varying the camera's aperture. Process the film for N development, then measure the density of each frame to see if it matches your previously determined density targets for those zones at N development. You of course will need to have previously determined these densities using your favourite ZS methodology, and thus be in possession of a densitometer.

Method:
1. Find any mid grey surface to mostly fill the frame of your favourite format.
2. Expose frame 1 for Zone III using your light meter. Choose a shutter speed on your camera that you have measured to be within 10% of its nominal value. Use either f22 or f16.
3. Expose frame 2 for Zone V using the same shutter speed as step 2 and opening up the aperture to either f11 or f8.
4. Expose frame 3 for Zone VII using the same shutter speed as step 2 and opening up the aperture to either f5.6 or f4.
5. Ensure the light level hasn't changed between steps 2 and 4.
6. Process film for N development.
7. Measure the transmission density of each frame and ensure it matches your expectations to within your acceptable margin of tolerance (say +/- half a zone). You should understand the Zone system to know where those densities should fall.

A possible variation is to take 3 frames of each zone to check for variability of your shutter timing.

The advantage of this method is that:

It is less sensitive to flare you would get by having say a test surface/card with the three different Zones on it at once.

You also don't have to create or maintain such a 3 zoned surface which takes more time and relies on the accuracy of your light meter in high flare conditions to ensure the surface was calibrated.

What do you do if the light changes between exposures?

PeterB · Apr 12, 2013

cliveh said:
What do you do if the light changes between exposures?

Well if you are taking it inside that shouldn't happen. If you are taking it outside you should check to see that no passing clouds will cause the local light level to change in the 2 minutes it will take you to take the 3 frames and stop open the aperture (which isn't too hard).

Doremus Scudder · Apr 13, 2013

Some thoughts:

I shoot sheet film and practice the Zone System. I test for many different development schemes (N, N+s and N-s). However, for roll film, when you will be shooting subjects of different contrast on the same roll, it is most important to simply find an N that gives you good shadow detail and manageable highlights. Since you use a spot meter, all you would then have to do is place an important shadow. I would deal with the different contrast situations by changing paper grade. And I would tailor my N to grade 2.5 or 3 to give a bit more latitude for high-contrast subjects.

Your test could (and I think should) be done without a densitometer, but rather by making proper proofs of your negatives on the paper and grade you will mostly use. This takes your enlarging system into account as well.

Furthermore, I think you should set up your targets in a situation where you would have "normal" flare. I set up a target card in a scene and like to have it take up no more than about a third of the entire scene. I use darker cards for the shadow values. That way I take an "average" amount of flare into account when making the test. Otherwise, you can test out everything and then get significantly higher shadow values in practice due to flare.

I find keying Zone VIII as just below paper-base white to be my most important highlight benchmark, not Zone VII.

I also like having the prints of my tests to remind me of just what I'm going to get when I place a certain value in a Zone. Densitometer readings don't help me to visualize. I used to make Zone Rulers (prints of Zone 0 through the highest Zone for a particular development scheme) to help me learn to visualize tones. I no longer need them, but they were indispensable at the beginning. One also finds that Zones for a particular scheme don't always follow the classic ZS descriptions. When Zone III and Zone VIII are perfect, Zone V is often far from 18% grey. It's more important to know what you'll get when you expose than to try to match some arbitrary standard.

The amount of variables in a photographic system is huge and little errors add up fast. Getting to within one Zone of my desired placement/development is way close enough for me. Printing controls are more than adequate for that. Realize that even your tests can be significantly "off," and make future adjustments based on field notes and experience.

Best,

Doremus

www.DoremusScudder.com

PeterB · Apr 13, 2013

Doremus Scudder said:
Some thoughts:

I shoot sheet film and practice the Zone System. I test for many different development schemes (N, N+s and N-s). However, for roll film, when you will be shooting subjects of different contrast on the same roll, it is most important to simply find an N that gives you good shadow detail and manageable highlights. Since you use a spot meter, all you would then have to do is place an important shadow. I would deal with the different contrast situations by changing paper grade. And I would tailor my N to grade 2.5 or 3 to give a bit more latitude for high-contrast subjects.

Thanks Doremus. I shoot the entire roll to the same scheme (N, N+s and N-s).

Doremus Scudder said:
Your test could (and I think should) be done without a densitometer, but rather by making proper proofs of your negatives on the paper and grade you will mostly use. This takes your enlarging system into account as well.

Wouldn't I then just be using a reflection densitometer to measure the paper densities anyway ?

Doremus Scudder said:
Furthermore, I think you should set up your targets in a situation where you would have "normal" flare. I set up a target card in a scene and like to have it take up no more than about a third of the entire scene. I use darker cards for the shadow values. That way I take an "average" amount of flare into account when making the test. Otherwise, you can test out everything and then get significantly higher shadow values in practice due to flare.

I considered your suggestion, and this is certainly one way to account for real life flare. The way I prefer to do it is to slightly increase the negative density range (by about 0.1 units) for a given SBR. i.e. increase CI/gamma. I suspect this is no more or less better at compensating for flare than you method, just a different way.

Doremus Scudder said:
I also like having the prints of my tests to remind me of just what I'm going to get when I place a certain value in a Zone. Densitometer readings don't help me to visualize.

That sounds a good idea.

I used to make Zone Rulers (prints of Zone 0 through the highest Zone for a particular development scheme) to help me learn to visualize tones. I no longer need them, but they were indispensable at the beginning. One also finds that Zones for a particular scheme don't always follow the classic ZS descriptions.

Doremus Scudder said:
When Zone III and Zone VIII are perfect, Zone V is often far from 18% grey.

I've always wondered about the non linear nature of intermediate zones. I also experience that.

You also mentioned using Zone VIII instead of VII. That is easy to do with my proposed method if you have a large enough aperture range. I initially wanted to ensure most aperture ranges could be catered for.

regards
Peter

Jim Noel · Apr 13, 2013

The north side of a building works well for this. If it has texture,such as stucco, so much the better. Develop together.

Bill Burk · Apr 13, 2013

I'm a big fan of sensitometer/test strips from step wedges contacted on film. I would go that route if flare-free testing was my aim. Though you are right, the suggested test of a single-tone target would be relatively flare-free.

There is no definition correlating density to a particular Zone, but this doesn't mean you can't use three arbitrary exposures to base a custom "test strip" upon. You can pick any two, three or more steps. Heck, you could do an impromptu test anywhere you go on the first roll of the day... Meter at f/5.6... Open to 2, shoot, stop down a stop, shoot, and quickly fire off 5 or more frames a stop apart.

Then when back from the day, develop that roll and see whether it's good for that camera/lighting conditions/day. If so, develop the rest of the rolls, if not tweak up or down a minute or two according to your preferences...

Doremus Scudder · Apr 14, 2013

PeterB said:
Thanks Doremus. I shoot the entire roll to the same scheme (N, N+s and N-s).

Well then, of course you would want different development schemes. Whenever I shoot roll film (and that's a rarity these days) I usually have different contrast situations on one roll.

PeterB said:
Wouldn't I then just be using a reflection densitometer to measure the paper densities anyway ?

No, you'd use your eyes to tell you if you got the desired print values. As you can see, I prefer a visual approach to testing that takes the final print into consideration. I never measure the densities of prints I make in order to decide what value it should be; I use my eyes. Same for testing.

PeterB said:
I considered your suggestion, and this is certainly one way to account for real life flare. The way I prefer to do it is to slightly increase the negative density range (by about 0.1 units) for a given SBR. i.e. increase CI/gamma. I suspect this is no more or less better at compensating for flare than you method, just a different way.

Again, I don't want to test for just one variable, nor do I really want to collect data; I just want to know how what I meter will end up in the print, visually. Therefore the test of the entire system as a "black box," not the individual aspects.

PeterB said:
I've always wondered about the non linear nature of intermediate zones. I also experience that.

I think that for ZS practitioners, realizing this is an important step in getting to know your materials and the process, not to mention helping you to better know what you'll get when you place something in a particular Zone.

PeterB said:
You also mentioned using Zone VIII instead of VII. That is easy to do with my proposed method if you have a large enough aperture range. I initially wanted to ensure most aperture ranges could be catered for.

I find that finding my Zone VIII, which I define as the first noticeable grey with a hint of texture (i.e., a non-specular white) is a better and easier to pinpoint benchmark than a "fully-textured off white/concrete color" Zone VII. That and the Zone III (first black with real texture) give me the endpoints of my system. Then I test to see where the other Zones will end up. I often make a Zone ruler to be able to visualize this more easily for, as already mentioned, the intermediate Zones are not always linear.

Best and good luck testing.

Doremus

PeterB · Apr 16, 2013

Bill Burk said:
There is no definition correlating density to a particular Zone,

Thanks Bill. When I initially tested/calibrated my film as per WBM I established a relationship between N+/-s, my film speed, film density and zone number (etc).

I actually exposed a Process test roll last night using the steps I outlined above and found zones III to VIII were within half a zone of my target. This is moderately encouraging. I noticed that zone III was a bit high in density, zone VIII a bit low and I calculated the gamma/CI to be 0.4 rather than the target gamma of 0.5 for those conditions.

I have a number of possible explanations for this. The most likely possibility is my XTOL stock (mixed using distilled H2O) is slightly less active than it should be as it is 1 year old and this was the main reason I wanted to do this test to see if it was still "OK". I will either extend my development to compensate or mix up another batch. I hate wasting half of what I mix up !! (I know I need to shoot more films !!)

The other less likely explanation is the formula I used to adjust the development time is not sufficiently accurate (I have never validated it for use). The temp I initially calibrated my film at was 20degC, my Process roll was developed at 23.1degC. I use the formula:

to determine the new dev time. I obtained the formula from this relatively comprehensive webpage on XTOL.

regards
Peter

PeterB · Apr 22, 2013

I have redone this test with modifications and came up with even more confusing results, the exact opposite of what I was expecting !!

To summarise my initial problem: I was getting a lower gamma (0.42 compared to the target 0.5) and I though this was due to the target I photographed introducing too much flare (in spite of me intending that it wouldn't - I suspect I wasn't thinking straight). The initial target was a matt cream coloured wall, but I only framed about a small 10% of the centre of the wall and in hindsight then I think the rest of the wall might have contributed to flare both in and out of the frame thus reducing my contrast. So to rule that cause in or out I repeated this test and found something even more strange which I need help understanding.

I decided to find a target with much less opportunity for creating flare. So I draped a white towel folded in half in front of a dark green hedge. The towel in the centre filled about 30% of the frame. I exposed the towel to sit on Zones III through VII. The hedge and the area out of the frame in front of the camera were a few stops darker than the towel. It was in shade and I couldn't see the sun.

So with another roll of film I now separately shot two targets. The first Five frames had this new towel target and another 5 frames had the original cream wall target. I fully expected the gamma calculated with the towel shots would be higher than the wall shots. It was in fact the exact opposite :confused: . Towel gamma=0.46, wall gamma=0.55 :confused: . The towel gamma should be equal to and probably higher than the wall gamma. (The target gamma in both cases was 0.6)

So what can possibly explain this difference ? Experimental error ?

Here are some misc notes in case you have read this far and might be thinking about an explanation: It turned out by chance that the exposure reading of each target was within 1/4 of a stop of each other (the cream wall was in a slightly higher light level). Accordingly I used the same shutter speed for both targets and simply varied the aperture from f5.6 to f22 in steps of 1 stop. I also took the opportunity to expose and develop for N+1 rather than N (as per the first check roll I did). This at least proved my gamma increased by about the expected 0.1 . The 5 data points from each target all lay quite well on the straight line portion of my HD curve. This rules out any light level or shutter speed variations. I double checked my densitometer calibration before and after measuring the densities. Also because both targets were on the same roll of film this rules out the possibility of dev strength/temp/timing errors. I also used a second spot meter this time to double check the exposure readings.

Bill Burk · Apr 22, 2013

First thing that comes to mind is... Color separation negatives through the blue filter have to be developed longer than the other separations because blue light results in lower gamma. (That's an explanation I've read, others may corroborate or dispute). Maybe the light on the white towel had more blue than the cream wall.

Second thing that comes to mind is... 15% isn't a great difference, I occasionally have that much difference in test results that are supposed to be the same. (Of course something always turns up to be at fault, like trying to re-use a tray of developer)

PeterB · Apr 22, 2013

Bill Burk said:
First thing that comes to mind is... Color separation negatives through the blue filter have to be developed longer than the other separations because blue light results in lower gamma. (That's an explanation I've read, others may corroborate or dispute). Maybe the light on the white towel had more blue than the cream wall.

Thanks for your thoughts Bill. I am not using any colour separation filters. My densitometer is reading greyscale/"visual" density of a B&W negative.

Bill Burk said:
Second thing that comes to mind is... 15% isn't a great difference, I occasionally have that much difference in test results that are supposed to be the same. (Of course something always turns up to be at fault, like trying to re-use a tray of developer)

Possible but something must have caused this deviation and I can't figure out what. My 1st light meter is a Sekonic L328, my 2nd light meter was the spot meter in a Nikon D90.

Bill Burk · Apr 22, 2013

PeterB said:
Thanks for your thoughts Bill. I am not using any colour separation filters. My densitometer is reading greyscale/"visual" density of a B&W negative.

Not that you used a blue filter, but that the distribution of spectral energy on the towel might be predominately blue because you are in a shadow, taking light mostly from the blue sky. I only mention blue-filter exposure to support my idea.

PeterB · Apr 23, 2013

Bill Burk said:
Not that you used a blue filter, but that the distribution of spectral energy on the towel might be predominately blue because you are in a shadow, taking light mostly from the blue sky. I only mention blue-filter exposure to support my idea.

Wow Bill. Now I understand what you are saying and it is a very plausible explanation. Thanks for your insight.

This begs my next question, why aren't (some ?) other people concerned with compensating for the colour temperature of the either the incident light or the colour of the reflected light when exposing their B&W films ? In my accidental discovery I was able to show the equivalent of one non trivial N step in contrast (for want of a better term) between two different colour combinations without trying to find the largest possible deviation/worst case.

Bill Burk · Apr 23, 2013

There's meter spectral response, human eye spectral response, film spectral response and the spectral distribution of light and the "color" of the test target. Lots of variables.

Many people inadvertently obtain a "Tungsten" speed rating by their tests without realizing it.

Others put an 80b filter over the light or lens to (at least partially) simulate daylight.

More important in my mind is to stick with a test plan where you understand it's limitations, and strive for consistency. Yes it's significant in your case, but if your test-to-test result stayed within 2/3 stop (or 2/3 N step), then you can use your test to control your processes.

Stephen Benskin · Apr 24, 2013

There's a reason why there are guidelines for testing. Why try to reinvent the wheel. You're making assumptions that may or may not be correct. Bad testing can be worse than no testing at all, and other similar cliches. I recently ran across a quote by Phil Davis on in camera testing,

"Traditionalists defend this testing method some vehemently on the grounds that involving the camera in the test simulates the conditions of practical use and is, therefore, not only convenient but desirable, Similarly, they are apt to argue emphatically that, after all, the purpose of this whole thing is to produce prints, so appraising print values must therefore be the most appropriate way to judge the materials performance.

In fact, thats a technical non sequitur. These traditional testing procedures cant supply material-specific information any more than driving your car around the block can inform you about the comparative quality of your motor oil, You can obviously tell whether the car runs satisfactorily or not, but you cant know for sure what part the oil has played in that performance. There are simply too many unrecognized or uncontrolled variables in the procedure; there is no accurate way to quantify the results of such subjective tests, and you have no logical basis for assuming that the conclusions drawn are valid."

PeterB · Apr 24, 2013

Thanks Stephen.

I hold a very different position to the one you suggest. I am a professional engineer having worked in the medical device industry for 20 years now. I have had plenty of experience with (among other things) proving designs are fit for their intended purpose - a necessity required by regulatory bodies worldwide.

Attempting to test a product or a process by focusing on the individual sub systems would never cut the mustard. Along with low level testing (known as Verification testing) to prove that sub systems meet their functional/engineering requirements, it is imperative to also perform testing at the system level - this is known as Validation testing. Validation testing is performed without needing to know any internal implementation details. It is akin to black box testing. It proves the product meets the needs of a variety of stakeholders such as those of the customer and regulatory bodies (such as the FDA).

As a photographer if all I did was individually test the sub sections in my photographic process I would only ever be performing verification testing and not validation testing. Both are necessary and valuable.

regards
Peter

PeterB · Apr 24, 2013

Bill Burk said:
There's meter spectral response, human eye spectral response, film spectral response and the spectral distribution of light and the "color" of the test target. Lots of variables.

Many people inadvertently obtain a "Tungsten" speed rating by their tests without realizing it.

Others put an 80b filter over the light or lens to (at least partially) simulate daylight.

More important in my mind is to stick with a test plan where you understand it's limitations, and strive for consistency. Yes it's significant in your case, but if your test-to-test result stayed within 2/3 stop (or 2/3 N step), then you can use your test to control your processes.

Thanks again Bill. I agree with what you wrote above. I might perform one more test before deciding on what process control test I will regularly perform. That test I'll do is to repeat my testing of photographing a transmission step tablet under particular controlled conditions. That is to stick the tablet onto a piece of white translucent perspex, masking off all areas outside the tablet as seen by the camera then affixing that assembly to an outside facing window, and finally draping a black cloth between the mask and the camera to ensure no additional light can enter. This is about as controlled as I can get the conditions when including the camera in the testing process.

regards
Peter

Stephen Benskin · Apr 24, 2013

Michael R 1974 said:
I'm not defending OPs particular test method, but I think Phil Davis overstates the case slightly. After all he had to sell books too.

Michael, you've said it yourself how a lot of people get by with bad testing techniques because of the tolerances in the photographic process. In order for a black box approach to work properly, you need to know what goes in as well as what comes out. The higher the confidence with the input, the higher confidence in the results, and the more likely testing errors will be identified.

Stephen Benskin · Apr 24, 2013

Good analogies are hard.

I was fortunate enough to have my sensitometer calibrated at one time. I also used a calibrated step tablet, and my old boss paid. Whether it's still calibrated is anyone's guess. I took it to that level because once done, there was less worry, and someone else paid (I paid for the sensitometer though).

Bill Burk · Apr 25, 2013

PeterB said:
This is about as controlled as I can get the conditions when including the camera in the testing process.

I'd recommend an artificial light source. I think actual daylight is too hard to control, it may vary too much to give you consistent results (so while it is the correct light source to match your shooting planned conditions, it's not as good for testing since it is different each time you use it).

Your test sounds simple enough that you could make a "light box" to hold behind the white translucent perspex that turns it into a "light table"

Bill Burk · Apr 25, 2013

PeterB,

I don't know if you saw this thread where we explored a test strip maker. I found it remarkably easy to do.

(there was a url link here which no longer exists)

PeterB · Apr 25, 2013

Bill Burk said:
I'd recommend an artificial light source. I think actual daylight is too hard to control, it may vary too much to give you consistent results (so while it is the correct light source to match your shooting planned conditions, it's not as good for testing since it is different each time you use it).Your test sounds simple enough that you could make a "light box" to hold behind the white translucent perspex that turns it into a "light table"

Thanks Bill. After my interesting results with two considerably different colour temperatures giving noticeably different gammas, it makes sense to control for that by using an artificial source to provide a more repeatable colour temperature.

rgds
Peter

Camera+Film+FilmDeveloper Process Control idea for Zone System followers

Privacy & Transparency

Privacy & Transparency