Slit-Scan Photography
Andrew Davidhazy
School of Photographic Arts and Sciences
Rochester Institute of Technology
Link to main ARTICLES page
Interesting distortions can be made by photographing
reflections in irregular surfaces of glass or mylar. They can also be
produced by passing either the image forming light either at the taking
or printing stage through prisms or transparent materials having
nonhomogeneous composition. Both of these methods depend on altering
the normal, expected or standard behavior of image forming light rays
at the instant in time that the picture is made.
Another
technique for creating distortions involves the use of a fine
slit that moves past the film at the time the picture is made. In fact,
a focal plane shutter is just such a device and the distortion produced
by these shutters is fairly well understood by photographers at large
and the distortions thus produced are called focal-plane shutter
distortion.
As a matter of background, it is generally impossible to tell
whether a photograph was made with a leaf or diaphragm shutter or with
a focal plane one. Unless the focal plane shutter is not functioning
well the statement above is almost invariably true.
The one condition under which it starts to become possible to
differentiate and identify the kind of shutter that was used to make a
given photograph is when the subject is not stationary but is in
motion. If we simply concentrate on the subject then the statement must
be modified and what must happen is that the image of the subject must
move with respect to the film while the exposure is taking place in
order to be able to pick up the characteristics associated with focal
plane shutter distortion.
Below are a series of idealized results that might be
produced with leaf, diaphragm or, in the digital realm, "global
shutters" and with focal plane
shutters or "rolling shutter" type digital cameras. Notice that the
images associated with leaf
shutters simply exhibit what might be called motion blur while the
ones made with a focal plane shutter the results show not only blur but
also a distortion that is associated with the essentially "scanning"
nature of such shutters.
<>To understand just why focal plane shutters distort rapidly
moving images one needs to simply establish that a moving slit shutter
takes some time to travel across a stationary image of some subject. If
the image then moves in the same direction as the slit shutter is
moving then the shutter will take longer to traverse the image of the
subject than under the previous condition and this will produce on the
film a stretched out image of the subject. On the other hand, if the
image is moving in the opposite direction as the focal plane shutter
slit, then the shutter curtain traverses the subject's image in less
time than if the image had been standing still resulting in a
compressed version of the subject. >
Movements at right angles to the motion of the slit introduce
tilting of the subject either in the direction of motion of the subject
or in the opposite direction depending on the direction of motion of
the shutter slit with respect to that of the subject's image.
A
very famous photograph that depends on focal plane shutter distortion
for its visual impact is one made by Henri Lartigue with a large camera
equipped with a focal plane shutter. The photo shows a race car leaning
in one direction while spectators and telephone poles included in the
picture are leaning in the opposite direction. This indicates that the
camera was panned causing the background to tilt one way while the
panning speed was not fast enough to keep up with the vehicle thus
causing the race car to lean in the opposite direction. Close
examination of the degree of tilt in the image seems to also indicate
that either the car changed velocity, the camera was not panned at a
constant velocity and/or that the shutter curtain velocity across the
film gate was not quite constant.
Another
photographer who took advantage of the characteristics of focal
plane shutter distortion but in a more controlled way was Robert
Doisneau who photographed a couple spinning on a turntable as a slowed
down focal plane shutter exposed the scene sequentially from one edge
of the film to the other.
Now one might wonder why it is that there is no significant
body of work
that illustrate the effects of focal plane shutter distortion. It turns
out
that the reason for this is that most focal plane shutters traverse the
film
plane quite quickly in relationship to the speed at which the images of
most
subjects move along on the film. It is important to realize that this
is not
related to the fact that one can achieve short exposure times with
focal
plane shutters since short times can be achieved by making the shutter
slit
very narrow. The reason is that modern shutters have better springs and
other
mechanical characteristics that allow them to accelerate and decelerate
more
rapidly than in older shutters.
Focal plane shutters associated with 4x5 Speed Graphic and
Graflex cameras move much more slowly than today's 35mm camera
shutters. Some photographers purposely slowed the shutter tensioning
down to achieve startling focal plane shutter distortion effects. Some
of these cameras may even be still in use today.
Finally, there have been many amateur and also some
commercial photographers who have devised ingenuous means for
introducing focal plane shutter distortion by modifying their 35's,
their Wideluxes or related cameras, or simply building matte-boxes that
were fitted with a moving slot to produce a piecemeal but continuous
exposure of the film.
Which brings us to the illustration at the top of this
article!
It
was made with a home-built moving slit type matte box shown in the
illustrations below. The device reminds one of an oversized lens shade
and consists of a 12 inch cube made out of plywood that has an open
front and on the back wall it has a hole through which the camera lens
looks into the box.
The open front of the box is flanked by a pair of "L" shaped,
metal (could be some other material!) "lips". These are attached to the
box edges in such a manner that they reach over the edges forming a
channel. Within this channel, and covering the open front of the box
rides a 12 inch by 3 foot opaque, black, cardboard has been cut mask
into which a slot about 2 or 3 millimeters in width and extending
almost the full width of the mask. Because this leaves a very weak
connection at the edges, a piece of clear glass is glued across the
slot to strengthen the mask and prevent the slot from breaking or
varying in width once tensioned when the mask is placed in motion.
In the slit-scan box illustrated here the mask is moved by
passing it between rubber coated rollers, one of which is motorized. In
this case one of the roller's shafts as well as the motor's shaft are
fitted with sprocketed belt drive gears and a flexible sprocketed belt
connects the two. The motor is a 24 volt DC gear head, high torque,
motor. It can be operated at variable speed by simply powering it at
lower than maximum voltage and it can be made to move the mask up or
down by the simple expedient of reversing the polarity of the voltage
supplied to the motor.
As shown in the illustrations and the drawing, the camera is
firmly
attached to the box by screwing the lens onto a step up ring that has
threads onto which the lens can be screwed and whose outer flange is
kept attached to the box by four or more flat headed screws whose heads
reach over the ring's edge and that are screwed into the box just
enough so that the ring can still turn. The fit should be snug. This
will allow the camera to be turned to suit the photographer but still
prevent the camera from detaching itself from the box accidentally.
Just don't knock the camera off the ring's threads!
The
scheme for moving the mask could be something very basic and I
recommend that you strongly consider a simple "pulling" mechanism as
shown in the attached drawing.
To use this moving slit matte box you simply aim the
box/camera combination at some location where there will be some
uncontrolled or contrived action. Generally you would place the
box/camera assembly on a sturdy tripod to make sure that the camera
remains pointed at the area of interest. You then move the mask so that
the slot is either at the very top of its travel or the very bottom,
depending on where you want the exposure to start.
At this time you open the shutter of the camera and lock it
open. Since the camera is looking into the black box the film is not
being exposed at all. Now you power-up the motor and the mask starts to
move, hopefully in the right direction (if not then reverse the
polarity of the power to the motor) across the open end of the box. As
it moves, different areas of the scene in front of the camera will be
exposed sequentially on the film. Much like if the exposure had been
made with a slit moving at the film plane.
It is recommended that you use a fairly small aperture to make
the slot sharper, its edges better defined, than if you use a large
aperture on the camera lens. This makes the edges of moving objects
more distinct and the slit-scan effect more obvious and dramatic.
Determining proper exposure is a bit of a trial and error
situation since the slot has a tendency to interfere with the aperture
that you are setting your lens to and thus the amount of light at the
film plane will not be quite what one would expect for any given
aperture. In any case, one can start with determining the exposure time
by dividing the width of the slit by the rate at which it moves at the
voltage you will be operating the motor. It helps to make oneself a
chart of voltages vs. exposure times and attaching it to the side of
the box.
Because during the time that an exposure is actually made the
shutter is blocking the viewfinder it is often necessary to choreograph
the event that one will be photographing. The subjects will need to be
given directions as to when and how to move and when to do it in
relation to the position of the slot on its way across the field of
view of the camera.
One solution might be to use a camera such as the Canon RT or
the older Pellix which have fixed, semitransparent mirrors that do not
move during exposure. These would allow the photographer the
opportunity to see in the viewfinder the relationships between the
moving image of the slot and the subjects in front of the camera.
In any case, this slit-scan photography matte box is a
wonderful special effects device. It is intended to produce images that
will, to some extent, startle an audience and make them ask "how did
they do that?!".
Click here for
an overview article including further applications
Click on
this line to visit my main articles page
If you'd like to communicate with me about this topic
simply
send me e-mail.
Andrew Davidhazy - andpph@rit.edu
or by clicking HERE!.
|