The above image was shot on Fuji Neopan Acros, a standard black and white film stock. Yet, as is plainly visible, a full range of color has been reproduced. How? The oldest color-reproduction trick in the book — color separation. By recording a series of images (in this case, three), each containing a different isolated color element (which we might refer to as a channel), we get pieces of a whole that can be reconstructed to form a color image. Here, I’m using the logical combination of three primaries — red, green, and blue.
Specifically, the three filters I’m using (seen above) are a Wratten 29 (red), a Wratten 61 (green), and a Wratten 47 (blue). I don’t think this is actually a perfect set — if memory serves, a deeper red would actually fit the green and blue better. But, getting filters in the odder Wratten numbers is tricky enough, and the 29, 61, 47 set works fine for my purposes. I’m using Tiffen filters, which are not necessarily the quality of B+W or Heliopan, but are affordable and, more importantly, available in a wide range of Wratten numbers. These filters are quite deep, and do rob the film of light — 3EV for the red, and —2.5EV for the green and blue.
Since the process involves shooting three frames, and attaching/detaching three filters, care must be taken to keep the camera stable. The sturdier the tripod, the better. I suppose this is one instance where having a motor drive might help, reducing the potential for camera movement during frame advance. I’ve had decent luck, however, with just a small Novoflex tripod, and a manually-advanced body. I never screw the filters on too tightly — each one will only be on for a single frame at a time anyway.
Since the filters are quite dark, I frame the shot before any of them go on. I meter while I’m framing, just to save time and energy after. For each shot, I compensate manually with the previously listed values. I always shoot them in order — red, then green, then blue — so that there’s no confusion afterward. Reconstructing the image in Photoshop is relatively simple. Bringing in all three layers is always the first step, and then either allowing Photoshop to auto-align them, or aligning them manually by viewing the differences of two at a time and adjusting accordingly. From there, the file can be converted from greyscale to RGB, and each layer reassigned to the corresponding channel. But, the method I prefer is to compensate for exposure and set each layer to screen over the next. I have a Photoshop Action just for this. Simply name the red layer ‘r,’ the blue layer ‘b,’ and the green layer ‘g,’ and the Action does the rest using Adjustment Layers for non-destructive behavior.
I guess the only thing left to address here is the matter of why.While home processing of color film has been made far more accessible recently, black and white is still cheaper, safer, and simpler. That’s kind of a cop-out reason due to the impracticality of trichrome photography — consider, for example, getting a human subject to stay still long enough for a crisp portrait. A more reasonable reason is simply that the temporal gaps between the three color channels leave plenty of opportunity for creative effects. Subjects moving in and out of frame will leave brightly colored ‘ghosts.’ Trees blowing in the wind will have multicolored glows about them. Water will shimmer prismatically and clouds will be like those from a dream. This is my reason — experimentation and a resultant image distanced yet further from reality.
While most b/w films are these days, it’s still wise to make sure you’re using panchromatic stock. Orthochromatic films like Efke lack sensitivity on the red side of the spectrum, and won’t do well for color reproduction. Be sure to check out the trichrome group on Flickr.