Long Exposures: Film Reciprocity Failure
One of the things that I love about photography is the ability to capture images that you cannot see with the naked eye. In particular long exposures of sea scape, with silky smooth water and streaky clouds.
Rock Pool: Nikon D90, ISO 200 ,15mm, 67sec, f16
This image was taken with my Digital SLR in Cronulla over a year ago before I started to much around with film. If I was to go back there today to take this same shot film I would still shoot with a tripos and a cable release, but I would have to adjust the exposure because film has something that digital does not, but can easily be compensated for. But first lets revise the key concepts of exposure.
The foundation of exposure is the idea that you reduce the area of your aperture opening in half (increasing your f/stop by one stop eg f5.6 to f8) you have to double your exposure time (increase it by one stop, i.e. 1/60 to 1/30) to let in the same about of light to obtain the same photograph.
For example say we have just taken a photograph of a rock on the edge of the stream and that the correct exposure for the photograph was f5.6 at 1/8sec. Now suppose that we want to show more of the landscape in the shot to do this we would want a greater depth of field say f16. To take a shot at f16 (which is three stops away from f5.6 (f5.6, f8, f11, f16) we would then want to let in three more stops of light in which would give us a shutter time of 1sec (1/8, 1/4, 1/2, 1).
Relationships like this are said to be reciprocal. The reason for it is that if close the aperture down a stop we halve the amount of light which we compensate by doubling the stutter time. So what we have done here is 2*1/2=1, so there is no overall change in the exposure of the image.
Reciprocity failure occurs when this relationship no longer holds and is a byproduct of the way film records light. At its very core film photography is a chemical reaction where light provides the energy for the reaction to take place. This failure is inherent in all films to varying degrees and generally starts to occur at the 1 second mark. One notable exception is Fuji Across 100 does not suffer any reciprocity failure for exposures under 2:00.
I liken it to shaking up a bottle of soft drink and then opening it, the reaction starts off explosively and slowly drops away. So here you have given the bottle energy to perform the reaction, like having a lot of light hitting the film. Now if you were to open the bottle with out shaking it you will get a little bit of fizz to start with then a slow release of gas that will take a long time to release the same about of gas as when you shake up the bottle, which is similar to how reciprocity failure works.
For long exposures (longer then a second) we can adjust the exposure time to compensate for the reciprocity failure, if we do not we will end up with an underexposed photograph.
Below is a table with adjusted exposure times that compensated for the Reciprocity Failure. These time are given in the data sheets for each film, the Ilford sheets have a graph showing the adjusted times while Kodak and Fuji just have a table with how many stops to adjust you exposure by. Please note that all these times are for a meter reading with ISO 100.
|Meter at 100 ISO||Fuji Across 100||Tmax/Delta 100||Tri-X 400/ HP5/FP4||T-Max/Delta 400|
From Night Photography by Lance Keimig
Now just looking at this table one might be lead to believe that the best films to use for long exposures would be Kodak T-Max or Ilford Delta 400 as they have the shortest exposure times of any of the films given a meter reading at ISO 100. However there is another side effect that occurs with Reciprocity Failure and that is increased dynamic range.
Of course there might be a creative reason why you want more tonal range in your photograph in which case you get it automatically in a long exposure.
Lets see where this increased dynamic range comes from, to do this lets consider the following example. We have judged that the correct exposure for our shot is to be 4 seconds at ISO 100, the shadows are sitting at and exposure of 16 seconds (+2 stops) and the highlights are sitting at 1 seconds (-2 stops). Here we would have the dynamic range be 4 f stops of light. The below table shows the time for the exposures of the highlights, mid tones, shadows and then finally the dynamic range in f stops of light (rounded).
|Exposure metered at 4 seconds|
|–2 stops||On Meter||+ 2 stops||Range in Stops|
|Fuji Across 100||1||4||15||(2+2)=4|
|T-Max/ Delta 100||1||6||30||(6+2)=8|
|Tri-X 400/ HP5 /FP4||1/4||1.5||12||(2+4)=6|
|T-Max /Delta 400||1/4||1||6||(2+6)=8|
The original meter reading would have had a total range of 4 stops. Here we can see that only that only Fuji Across 100 keeps its tonal range while the others tonal range expands. The tonal range will only increase from here so it is quite easy to lose details in the shadows or blow the highlights if you are not careful. Now not all is lost as there are still two ways you can compensate for this, either just use Across 100 or adjust your development time or concentration.
So far I have only been using Fuji Across 100 for long exposure and as long as they are under 2min I have nothing to worry about as there is no failure to compensated for. What I want to start looking into is how you can control the tonal range on the negative though compensating development, there will be a blog post on this some time in the future.
Note: In this post I have only considered B&W film photography. When you start to use colour film for long exposures an extra difficulty occurs where you can experience colour shifts. The reason why this can happen is beach colour film contains three light-sensitive emulsion layers, which sometimes will have difference reciprocity properties. This colour shift can be corrected via the use of colour correction filters. Each individual films data sheet should should contain the specific filter that should be used.