A camera is essentially a light-tight box with a lens at one end to form an optical image and a fixture to hold light-sensitive material at the other to record this image. Most cameras have some means of focusing the image sharply at the photoplane for subjects at different distances. The duration of exposure to light is controlled by a shutter system, either mechanical or electronic. During the exposure the image illuminance on the film is controlled by an iris diaphragm, the aperture diameter of which can be varied. The optimum settings of the shutter and iris diaphragm are linked to an exposure determination system using light metering, usually measuring through the lens. Finally, the camera has a view-finder system by which the subject area to be included in the image may be determined. Essential features are some means of changing the exposed film in the gate for the next frame, and flash synchronization of the shutter. A data display panel may be needed for camera control. Both film and digital cameras share all or most of these features.
The format of an image as recorded in the camera is the dimensions of the film gate in the photoplane or the photo-sensor area. Originally, photographs were taken on glass plates followed by contact printing. Plate sizes up to 305 × 381 mm (12 × 15 inches) were used. Improvements in lenses, emulsions and illuminants made projection printing practicable and a decrease in format with later enlargement possible. Roll film speeded this process and currently the 24 × 36 mm (135) format is the popular size. The relative areas of various film formats are shown in Figure 11.1.
The 120 roll film size, using film of 62 mm width or ‘gauge’ and a backing paper, provides medium formats such as 60 × 60 mm. Adaptors can provide alternative smaller formats, e.g. roll-film backs for technical cameras and 35 mm film adaptors for 120 size roll-film cameras.
Sheet film was once considered an inferior alternative to plates but polymeric base materials such as polyester, with improved dimensional stability, have replaced these. Large-format work is standardized on 102 × 127 mm (4 × 5 inch) and 203 × 254 mm (8 × 10 inch) sheet film sizes.
Many types of camera have been produced both for general applications and for specialized purposes in conjunction with a range of accessories. Well-defined types include simple, compact, rangefinder, twin-lens reflex, single-lens reflex, technical, and specialized designs (see Figure 11.2).
A ‘simple camera’ is made for ease of operation, with little choice of control settings. The primitive box camera has a single meniscus or doublet lens of aperture f/14. Smaller apertures are selected by ‘weather’ symbols on an aperture control. The lens is fixed-focus, set at the hyperfocal distance to give reasonably sharp focus from 2 m to infinity. A ‘portrait’ supplementary lens reduces the sharp focus to about 1 metre. Alternatively, a three-point symbol focusing system gives sharpness zones of 1–2 m (portrait), 2–8 m (group) and 3 m to infinity (landscape). An ‘everset’ shutter offers two settings of ‘I’ for ‘instantaneous’ (about 1/40 s) and ‘B’ for time exposures. The viewfinder is either a direct optical one or a brilliant (reflex) type. Formats vary from 60 × 90 mm (eight exposures on 120 film) down to 30 × 40 mm (16 on 127 film), as well as 28 × 28 mm (126 cartridge), 11 × 17 mm (110 cartridge) and 8 × 10 mm (disc film cassette).
The single-use camera (SUC) or ‘film-with-lens’ camera is the current embodiment of the simple camera, intended to be used and then sent away complete for processing, when the camera components are largely recycled for reloading and further use. The plastic body shell with modest fixed-focus lens and a single speed shutter (1/100 s) contains pre-loaded film in cassette (135 format) or cartridge (APS format). The pre-advanced film is rewound into the cassette after each exposure. The shutter is inoperative after the last exposure. Fast film of ISO 400 or 800 speed is loaded to allow use in a range of conditions, taking advantage of the exposure latitude of colour negative material. Cameras are available with or without an integral flash, with output suited to the modest lens aperture. The popularity of these cameras is due to their small size, light weight, modest cost and disposability. A variety of alternatives are available, including underwater versions, and long-focus and wide-angle (panoramic) types.
A compact camera is intermediate to simple cameras and single-lens reflex (SLR) types, with a non-interchangeable lens and useful specifications and considerable automation, in a body of modest dimensions, using standard 135 or APS type cassettes. A coupled rangefinder, large-aperture lens, bright-line finder and choice of shutter- or aperture-priority automation may be provided. Other features are an integral electronic flash unit and autofocus using a phase detection system.
A fixed (non-zoom) lens has a typical maximum aperture of f/2.8 and a focal length of 35–40 mm. A silicon photodiode (SPD) monitors subject luminance and operates an exposure program with a between-lens shutter offering a range of some 1/20 to 1/500 s and an aperture range of f/2.8 to f/16. Flash exposure automation is by alteration of lens aperture in accordance with distance data from the autofocus system (the flashmatic system). Film speed is set by the DX coding system on the film cassette. A backlight control provides one to two stops exposure increase for high-contrast scenes. The flash can be used for fill-in purposes too, or a matrix-type metering system may combine flash with the ambient light exposure.
A zoom lens typically has a range of 35–70 mm focal length with the viewfinder frame altering to match. Other zoom ranges are available, such as 38–140 mm. To keep such lenses compact enough to telescope back into the camera body, maximum aperture decreases with increase in focal length, possibly to as little as f/11, from some f/4 at the short-focus end. Film loading is semi-automatic with advancement to the first frame on closing the camera back. Film advance is motorized, as is rewind of the exposed film. An alternative panoramic format of 13 × 36 mm is given by selecting masking blades for the film gate. The viewfinder is of complex optical design but kept small to fit within the camera body. Usually real-image types are used, based on Kepler telescopes.
The APS type (which is no longer a supported camera format by the majority of manufacturers) offered additional features, apart from very small camera bodies. There was a selectable choice of three alternative formats, labelled C (classic), H (HDTV) and P (panoramic), giving aspect ratios of 1:1.5, 1:1.77 and 1:3 respectively, obtained by selective enlargement of areas of the film frame area. The film cartridge is easy to load, with film advance and rewind completely automatic. Mid-roll film interchange was possible to allow switching of film types. A range of ‘intelligent’ features provide a range of user options, especially at the printing stage, when multiple prints or different sizes and backprinting of data may be made individually for each exposed frame. The film base was coated with an additional transparent layer of material for magnetic encoding of data.
These cameras use a coincidence-type rangefinder system, coupled to the focusing mechanism of the lens to ensure sharp focus at the subject distance. This method is essential for accurate focusing of large-aperture lenses. Combined rangefinder–viewfinders have sets of bright-line frames. Interchangeable lenses are limited as the fixed base-length of the rangefinder means that focusing accuracy decreases with increase in focal length. A rangefinder camera is compact and capable of accurately focusing large-aperture lenses in low-light-level conditions. The viewfinder is complex with alternative bright-line frames, parallax correction linked to focusing and exposure metering systems by through-the-finder (TTF) methods.
Twin-lens reflex cameras are essentially two cameras mounted one above the other, the upper for reflex viewing and focusing, and the lower for exposing the film. Both lenses have the same focal length and are mounted on the same panel, which is moved bodily to provide continuous viewing and focusing. The reflex mirror gives an upright, laterally reversed image on a ground-glass screen. The screen is shielded for focusing by a collapsible hood with a flip-up magnifier. The hood may be interchangeable with a pentaprism viewfinder for eye-level viewing and focusing. The focusing screen incorporates a Fresnel lens, split-image rangefinder or microprism array to assist focusing. The viewing lens is always used at full aperture. Film transport and shutter setting is normally by means of a folding crank device. Cameras used the 60 × 60 mm format on 120/220 film. Interchangeable lenses are uncommon with these cameras. There is parallax error in the viewfinder screen image because of the separation between viewing and taking lenses. For close-up photography a wedge-shaped prism can be fitted to the viewing lens when supplementary lenses are used. Few TLR cameras are now in use.
The principle of the SLR camera is illustrated in Figure 11.3. A plane front-surfaced mirror at 45° to the optical axis diverts the image from the camera lens to a screen for focusing and composition. For exposure the mirror is lifted out of the way before the camera shutter operates and returns to the viewing position after exposure. Design advantages are the ease of viewing and focusing, and freedom from parallax error. The depth of field at the preselected aperture can also be estimated.
Some designs use rectangular formats, and a rotating back may be provided so that upright (‘portrait’) or horizontal (‘landscape’) pictures can be composed.
SLR design has been fully developed for 135 and 120 film formats using focal-plane shutters and a springoperated mirror. An instant return mirror, giving uninterrupted viewing (except during the time the shutter is open), is now standard in most SLR cameras. For wide-angle lenses, the depth of the mirror box in the camera requires lenses with a long back focal distance, i.e. the retrofocus type.
A rectangular format is less convenient when vertical framing is required for a picture, as turning the camera sideways causes the image on the viewing screen to be inverted. A square format allows the camera to be held in the same way for all photographs. The pentaprism view-finder (Figure 11.4) permits eye-level viewing and focusing with the image erect and laterally correct for both horizontal and vertical formats. The erect but laterally reversed image on the focusing screen undergoes three reflections inside the roof prism to appear correctly oriented. The necessary precision of the angles of the prism requires it to be made of glass, so it is weighty and expensive, but hollow prism versions using mirrors reduce both. The iris diaphragm, which was normally stopped down manually just before exposure, evolved into the fully automatic diaphragm mechanism (FAD). Improved viewfinder focusing screens use passive focusing aids in the form of microprism arrays and split-image rangefinders, with alternative screens for specialized purposes. Manual focusing has been superseded by autofocus systems (see below).
Available lenses range from fish-eye to extreme long-focus types. Fully interchangeable magazine backs allow for changing of film type in mid-roll, or rapid reloading. Exposure determination uses through-the-lens (TTL) metering.
The SLR design is the basic unit for a wide range of accessories. The medium-format SLR camera can replace a technical camera for many applications not requiring a large format or extensive use of camera movements.
The terms technical camera or view camera cover two types of camera. The first is the monorail type (Figure 11.2f), based on the optical-bench principle to offer the widest possible range of camera movements. Focusing and composition are done using a ground-glass screen, and the camera needs a tripod. The second type is the folding-baseboard variety, and may have a coupled rangefinder and optical finder as well as a ground-glass screen. It can also be used hand held.
Early wood and brass studio and field cameras were adequate for large-format work but were slow in operation. Improvements in lenses and smaller formats needed greater precision in manufacture. So metal was substituted for wood, giving rigidity at the expense of weight. Some standardization was achieved in the sizes of items such as lens panels, filmholders and camera backs. Technical cameras can use medium formats such as 60 × 70 mm to 60 × 120 mm, but monorail cameras in these formats are less common. The term ‘large format’ covers film sizes of 102 × 127 mm (4 × 5 in) and 203 × 254 mm (8 × 10 in). Cameras feature reducing backs and format changing for alternative formats.
As a ‘system’ camera the basic body can be fitted with alternatives for almost every component. Monorail designs are produced on modular principles so that rails, front and rear standards, bellows, focusing screens, lenses and shutters are interchangeable to adapt for a range of formats or types of work.
The folding-baseboard camera may use a high-precision rangefinder with coupling to alternative lenses. Viewing is by means of a multiple-frame, bright-line viewfinder. Sheet film, roll film, self-developing material and even plates can be used. It has a more limited range of movements than the monorail type. The ground-glass screen must be used for close-up work or camera movements. The bellows is of ‘triple-extension’ type, allowing a magnification of ×2 with a standard lens for the format.
Innovations in technical camera design include: forms of through-the-lens exposure determination using a probe in the focal plane; preset mechanisms for shutter speeds and aperture settings; electronic shutters; extreme wide-angle lenses of large useful aperture; computer coupled and indicated camera movements; and binocular viewing and focusing aids.
Special cameras use self-developing (instant print) materials manufactured by the Polaroid Corporation and others. These are available in roll-film, sheet-film and film-pack forms and formats up to 8 × 10 inches and even 20 × 24 inches. Both colour and black-and-white materials are available. Cameras range from simple types with minimal control of functions to sophisticated models equipped with electronic shutters, TTL exposure metering, autofocusing and automatic ejection of the exposed film frame. The geometry of image formation, and the sequence of the layers in the particular film type, may require use of a mirror system in the camera body in order to obtain a laterally correct image in the print. This gives a bulky, inconvenient shape or requires considerable design ingenuity to permit folding down to a compact size for carrying. Various adaptor backs permit the use of self-developing materials in other cameras for image preview purposes. Such camera and film combinations are often very convenient to use, and still have useful applications.
Most aerial cameras are rigid, remotely controlled fixtures in an aircraft, but for hand-held oblique aerial photography specialized types are used that are simplified versions of technical cameras without movements and with rigid bodies and lenses focused permanently on infinity. A fixture for filters, a simple direct-vision metal viewfinder and ample hand-grips with incorporated shutter release complete the requirements. If roll film is used (normally 70 mm), film advance may be by lever wind or may be electrically driven. To give the short exposures necessary to offset vibration and subject movement, a sector shutter may be used. This is a form of focal-plane shutter in which an opaque wheel with a cut-out radial sector is rotated at high speed in front of the film gate to make the exposure. Data imprinting can record positional and other information at the moment of exposure. These data may be taken directly from the navigational system of the aircraft.
For underwater use, cameras can be housed in a pressure-resistant container with either an optically flat window or a spherical dome port. The housing may be an accessory for the camera. Underwater housings are available for 35 mm, digital SLRs or compact camera formats and use rods attached to the main camera controls, which pass through the casing to oversize knobs on the outside and access the controls inside. Such housings can be expensive but may be safe down to depths of around 40 metres. They also have connectors to attach external flash units. Dome ports are used for wide-angle lenses and the curvature of the dome is ideally matched to the lens’s focal length. The lens focuses on a virtual image created by the dome. Often a dioptric supplementary lens is required on the camera to facilitate this. Dome ports do not introduce problems associated with refraction, radial distortion and axial chromatic aberrations. Flat windows are used with long lenses for close-ups or shots that start or end above the water and don’t require a dioptre. However, they are unable to correct for the distortion produced by the differences between the refractive indices of air and water, and introduce refraction, distortion and chromatic aberration problems when used underwater. In some cases the housings are made with other additional optics to make the apparent angle of view wider. This is particularly useful to some digital cameras with small sensors that do not achieve wide angles of view with the conventional lenses. Both wide-angle and close-up supplementary lenses are available, often as ‘wet lenses’ that can be added or removed underwater.
A few cameras have a pressure-resistant, water-tight body and interchangeable lenses to make them directly usable to specific depths. They are also useful in adverse conditions on land where water, mud or sand would ruin an unprotected camera. These cameras usually have a simple direct-vision or optical finder and manually set focusing controls, but SLR versions have been made. The lenses are interchangeable and have suitable seals, but changing must be done out of the water. A short-focus lens may be fitted as standard to compensate for the optical magnification effect that occurs due to the change in apparent subject distance caused by the refractive index of the water in the object space. A field of view similar to that of a standard lens on a conventional camera is then obtained. Special lenses corrected for underwater use, i.e. with water actually in contact with the front element, are also available. Some compact cameras are specially sealed to allow underwater use to depths such as 3 metres and will withstand adverse weather conditions.
When a sufficiently large angle of view cannot be obtained from the usual range of lenses available for a camera, and when distortion considerations rule out the use of a fisheye lens, then an ultrawide-angle camera may be used. This may be a camera body of shallow depth with a lens (usually non-interchangeable) with a large angle of view, e.g. some 110° on the diagonal, as obtained by the use of a 65 mm lens with a 4 × 5 inch format. Focusing of such a lens is usually by scale, and the viewfinder may be a simple direct-vision or optical type, preferably incorporating a spirit level as an essential aid in aligning the camera. Limited shift movements may be provided, such as rising front. Medium-format versions find special use as precision shift cameras for architectural work.
It is possible to obtain single-frame panoramas using film cameras with rotating (or swing) lenses. The lens rotates around its rear nodal point – from which the back focal length is measured – while a vertical slit exposes a strip of film which is aligned with the optical axis of the lens. Lens and slit scan horizontally across the panoramic scene. Typically, the shots encompass a 110–140° field of view horizontally and 60° vertically; therefore the horizontal image size will commonly take up one and a half to two and half times the length of a common 35 mm frame. Unfortunately most swing lenses come with a fixed focal length and cannot focus well at a distance of less than 10 metres. Capturing subjects at a smaller distance requires use of a small aperture to provide a large depth of field. Additionally, the number of available shutter speeds is limited, causing problems for subjects at low light levels.
Rotating panoramic cameras function similarly, but in this case the camera body rotates around the front nodal point of a static lens. A mechanism rotates the camera continuously while the film, in a curved film gate with a centre of curvature at the rear nodal point of the lens, is transported in the opposite direction. The speed of movement of the film matches the speed of image movement across the image plane. The film is exposed through a thin slit producing a sharp image throughout; therefore, such cameras are also referred to as slit or scanning cameras. The horizontal panoramic coverage can be up to (sometimes more than) 360°.
The use of swing lenses or panoramic cameras does not result in the same extreme distortions of lines often seen in ultra wide-angle lenses; however, image perspective is unique. The camera must be correctly levelled, parallel to the subject plane, to prevent the introduction of other distortions, such as the curvature of horizontal lines near the bottom or top of the frame if the camera is tilted up or down. If a subject is at a close distance, horizontal lines will converge in the centre at the edge of the frame, even if the camera is parallel to the subject plane. On the other hand, a curved subject – such as a long group of people sitting around a concave row of chairs – will be reproduced as a straight line. Tilting the camera also causes convergence of vertical lines at the top or bottom of the frame.
In other cameras, known as wide-angle or wide-field camera systems, a high-aspect-ratio format (from 2:1 to 4:1) is used with a flat film gate and a very wide-angle lens, usually needing a centre spot filter to improve image illumination at the edges and eliminate optical vignetting. This reduces exposure of the film at the edges of the frame, but at the expense of approximately 2EV. These are the most common panoramic cameras, available in various image formats, and may vary significantly in quality and price. The image produced by true wide-field cameras is relatively distortion free but the panoramic coverage is restricted when compared to swing lenses or to rotating cameras. Nevertheless, they are popular for shooting architectural panoramas as they do not cause lines to curve or produce perspective distortions. Another great advantage of this type is the instantaneous exposure of the frame as opposed to the longer, sweeping exposures of other types of panoramic cameras, meaning that the use of flash is not restricted, as often happens with cameras which only expose one part of the image at a time using a slit or scanning mechanism. Typically, the maximum angle of view with a flat film camera is about 90°, although lenses with up to 150° can be used. Formats from 60 × 120 mm to 60 × 170 mm or even 60 × 240 mm are used.
A selection of types of special-purpose camera is shown in Figure 11.5.
Most cameras are ‘automatic’ in that there is some form of digital control using solid-state micro-electronics. Functions previously operated by mechanical, electrical or electromagnetic devices are controlled by electronic circuitry. This replacement of gears, rods and levers varies from primitive controls, through electronic ‘enrichment’ to fully digital operation. Functions such as shutter operation and metering are controlled by a central processing unit (CPU) that makes logical decisions based on the input of digital data. Subsequent control commands or output can be carried out by the camera to give full automation, or with the aid of the user to change settings as appropriate, to give semi-automatic operation.
The operation of all-mechanical cameras depends on the film advance mechanism to tension actuating springs and the shutter timing depends on escapement devices and gear trains. An electronically controlled shutter uses a resistor–capacitor circuit for exposure timing and automation is possible when the exposure metering photocell output is used to control the timing circuitry. The exposure-determination circuitry of the camera uses the exposure equation relating shutter speed (t), aperture (N), subject luminance (L) and the film speed (S), involving the meter constant (K) for reflected light where:
Analogue converter devices set appropriate resistance values for these variables into a primitive computing device such as a Wheatstone bridge arrangement. The shutter-speed setting control, film-speed setting and exposure-compensation controls all use forms of fixed and variable resistors. The control circuitry uses only a few components such as transistors, resistors and capacitors. A silicon photodiode with its very fast reaction time and freedom from unwanted ‘memory’ allows metering to be carried out in ‘real time’ in the short interval between depression of the release button and the mirror rising or the shutter operating, or, better, during the shutter operation itself by metering from the actual film surface. The photodiode requires additional circuitry such as operational amplifiers, and draws current from a battery which also powers the shutter and viewfinder display. A meter needle is replaced by LED displays, still as an analogue arrangement. Increasing demands are made on simple analogue circuitry, especially with multi-mode metering systems and the short time available to carry out the necessary operations for each exposure, in particular with a motor drive operating at several frames per second. The possibility of overload is real. Most of these limitations are overcome by use of digital control.
Digital control refers to the use of data and control commands in digital form. An early digitally controlled system was a viewfinder display using either alphanumeric characters formed from seven-segment LEDs or a liquid crystal display (LCD), where letters and symbols can be formed as well as numbers. Power-consuming LED displays can be replaced or complemented by large LCD panels that give a comprehensive readout of the current status of the camera (such as exposure mode, film frame number, film speed in use) (see Figure 11.6). Such displays require more complex control circuitry than a simple LED display, where the LED is either lit up or not. Activation is from the CPU via devices called decoders and multiplexers. A multiplexer is a circuit which selects with the aid of a clock circuit from a number of inputs and outputs in turn. The clock timing is provided by an oscillator circuit using quartz or lithium niobate crystals.
The CPU uses additional devices such as analogue-to-digital (A/D) converters and both random access memory (RAM) and read-only memory (ROM). A clock circuit monitors the sequence of operations and all are connected by buses for data flow.
Some systems need a 6-volt supply, others only 3 volts, determining battery needs and space for them. Devices may be sensitive to voltage changes, so a constant-voltage supply circuit is needed. Others are sensitive to temperature, so another compensating circuit is needed.
Various types of CPU are used, depending on needs. All have large numbers of circuit elements per unit area, given by large-scale integrated circuits. The high packing density is given by metal oxide semiconductor (MOS) circuitry. These are sensitive to voltage and temperature, and are not easy to interface with other devices in the camera. The alternative transistor–transistor logic (or TTL; not to be confused with ‘TTL’ metering) costs more, and uses more power, but needs less precise control. Processing speed is higher than with the simple MOS type. An intermediate choice is integrated-injection logic or I2L type, which has the packing density of the MOS type but is slower than the TTL type. Operation requires a read-only memory or ROM device to store a fixed program of operational steps in the correct sequence to guide the CPU through its calculations in the correct order. The ROM is static, and data does not disappear when the camera is switched off; the program (‘firmware’) can be added to or modified subsequently.
The clock circuit acts as a reference for the order of events. A quartz oscillator gives 32,768 beats per second or hertz (Hz). These pulses are counted by appropriate circuitry, and can also be used to control shutter speeds, self-timer delays, meter ‘on’ timing controls and intervalometers. An exposure of 1/1000 s is timed as 33 pulses, to an error of less than 1%. The counting circuit delays travel of the second blind until sufficient pulses have elapsed. An alternative ceramic oscillator device gives 4,194,000 Hz for even greater accuracy and for control of complex data displays.
Digital control means that the data and calculations are handled in discrete form and not as a variable electrical quantity in their analogue form. The circuitry is basically an array of transistor switches, each of which can be in only one of two states, either ‘off’ or ‘on’, to shunt the current flow through logic gates of the ‘if/then’ and ‘and/or’ configurations. The decimal system of 10 digits from 0 to 9 is replaced by the binary system using the two digits 0 and 1, corresponding to the two alternative states of a switch. The ‘number-crunching’ power of a digital processing system, plus its speed and error-free operation, more than offsets the additional circuitry needed compared with analogue systems.
All input data must be coded into binary form and then the output commands decoded for implementation. The analogue data from transducers such as potentiometers in the film speed setting control is changed into binary data by an analogue-to-digital (A/D) converter, and the reverse operation by a digital-to-analogue (D/A) circuit. This conversion is doubly difficult in the case of data from the silicon photodiode photocell. This gives subject luminance data in analogue form, and covers an enormous dynamic range: a 20EV metering range corresponds to a light level range of 1,048,576:1. This range is compressed by converting values to their logarithmic equivalents before conversion to digital form, using a logarithmic compression circuit. Data encoding by a transducer may be by movement of a wiper arm over a patterned resistor or by an optical device, the output being a string of pulses corresponding to the required 0 and 1 values. To guard against encoding inaccuracies, the strict binary code may be replaced by, for example, the Gray code, which permits the checking of values by ‘truth tables’, so that impossible values cannot be sent to the CPU. Other components of the camera such as flash units, power drives and multifunction backs can transduce and transform their data as necessary. By means of the DX film coding system in binary form on a 35 mm film cassette, the film speed, exposure latitude and number of exposures may be encoded directly into the CPU by means of an array of contact pins in the cassette compartment.
The CPU performs various actions on its output. Values for the viewfinder display are ‘rounded off’ to conventional values or to two decimal places. The display can be updated every half-second or faster and alter its intensity in accordance with the photocell output by the simple means of altering the pulsing rate of each segment (the intensity of an LED cannot be varied by other means). A ‘bleeper’ can operate to warn of camera shake or inadequate light, generate an error signal, and so on.
A separate CPU may be used for the complex signal processing operations required with the data from arrays of sequentially scanned charge-coupled devices (CCDs) used in autofocus systems. Accessories such as flashguns may also have LCD digital displays to supplement the main display on the camera body or in the viewfinder.
Other interfaces in the form of arrays of electrical contacts are used in the lens mount, film gate, viewfinder well and hot shoe to allow the flow of data and commands from an autofocus lens with its own ROM and transducers, with a multi-function back, an accessory automatic metering viewfinder system and various forms of dedicated flashgun respectively. A generalized picture of the flow of data and control signals in an automatic camera under digital control is shown in Figure 11.7.
The camera shutter controls the duration (t) of the total exposure (H) given to the sensor. Image illumination (Ei) is controlled principally by the aperture stop according to the law of reciprocity:
Quantity t is called the shutter speed, and both Ei and t are in a simple reciprocal relationship in that to keep H constant, as either Ei or t increases, the other must decrease (see also Chapter 8, Eqn 8.1). For exposure control, the shutter speed range is usually greater than the aperture range, e.g. a 14-stop range from 8 s to 1/2000 s is 16,000:1, but a 50 mm f/1.4 lens stopping down to f/16 has only a seven-stop range giving 128:1. A shutter speed range from 1/12,000 s to 30 s is typical. The shutter is located either in front of, within or behind the lens (a leaf or diaphragm shutter) or close to the film plane (focal plane (FP) shutter).
Shutter speed also determines the amount of subject movement (blur) discernible in the image and is used to ‘freeze’ motion by short exposures. Long exposures can be used deliberately to give blur to provide information about subject movement or for creative effect. For a subject moving with velocity V orthogonal to the optical axis (parallel to the film plane) and distance u from the lens of EFL f, the image velocity Vi is given by:
When u is large, magnification m ≈ f/u, then:
The image distance (blur) B moved in time t is given by B = Vit and must be less than or equal to the local value for the circle of confusion (C), so B = C. By substitution, the necessary exposure time (maximum) for sharp images is given by:
So for a static camera and subject 5 m away moving at 1 m s−1, for a 50 mm lens with C = 0.05 mm, te must be 1/200 s or less. If the camera is moving, the relative velocity of camera and subject is used. If the subject velocity is at an angle θ to the film plane, then Vcos θ replaces V in the above equations, i.e.
Another cause of image blur is camera shake where the camera is inadvertently moved during exposure due to user tremor, vibration of the camera platform and other environmental effects. Blur worsens with increase in lens EFL (in mm) and an empirical rule is that:
so for a 2000 mm lens te needs to be 1/2000 s or less. Some cameras give an audible or visible warning if shake is anticipated with the lens and shutter speed combination selected. A tripod or image stabilization system (see below) is needed to improve sharpness. Electronic flash can provide the short exposure needed to arrest motion, but in high ambient light levels, depending on shutter synchronization speed (usually from 1/30 to 1/250 s, but 1/1000 s is possible with leaf shutters) a secondary blurred image may be overlaid.
Shutters primarily use mechanical actuation systems to move the blades or blinds by the action of springs, levers, electromagnets or linear motors. These open/close systems require a timing arrangement to delay closure and provide a range of exposure times. Mechanical timers such as clockwork escapements were replaced by analogue electronic timers, then by digital circuitry using a quartz (32,768 Hz), ceramic (524,288 kHz) or lithium niobate (4,000,000 Hz) oscillator. Shutter speeds of 1/1000 and 1/15 s require counting of 33 and 2185 pulses respectively using a quartz oscillator.
Leaf shutters have multiple leaves or blades pivoting about one end to uncover and recover the lens aperture. Top speed is about 1/500 s. Shutter blade activation can be by spring, lever, electromagnet or linear motor. A ‘cocking’ or resetting action is necessary unless the shutter mechanism is self-cocking. The shutter may have a complex operation when used in SLR cameras, as it must normally be open for reflex viewing. There is a time lag before shutter operation of some 8–600 ms. A value of 120 ms is typical for a 35 mm SLR camera. Shutter blades use blackened steel 0.05 mm thick and the tips can reach velocities of 60 km h−1 (16.7 m s−1) in use. Between one and seven blades can be used. Advantages include reliability, ease of flash synchronization, electronic timing, a clear aperture with 100% transmission, no distortion of the image and usability with all formats. Leaf shutter characteristics are shown in Figure 11.8.
The traditional design of a focal plane shutter uses a slit of width W formed between the trailing and leading edges of two successive ‘blinds’ travelling at velocity Vs across the film gate close to the photoplane. The effective exposure duration (te) is the time required for the slit to move its own width, so te = W/Vs.
The blinds can be rubberized cloth, thin metal structures or multiple overlapping blades. Materials include patterned titanium and carbon fibre/epoxy composite materials. Usually exposure is varied by altering W for a nominal constant velocity, but W may increase during travel to compensate for acceleration during travel. Slit velocity can be as high as 6.7 m s−1. The slit can travel in four alternative directions across the gate, dependent on camera design. Shutter blinds may be patterned for use with off-the-film (OTF) automatic exposure systems. Timing mechanisms are mechanical or electronic as for leaf shutters, but shorter exposure times are possible, currently some 1/12,000 s. Shutter characteristics are shown in Figure 11.9.
Early flash photography used the camera on a tripod, an open shutter on the ‘B’ setting, a manually ignited tray of magnesium powder, then closure of the shutter. Later, flashbulb firing was synchronized with shutter opening, allowing a hand-held ‘instantaneous’ exposure. A separate synchronizer device was replaced by flash contacts incorporated in the shutter mechanism.
Flash synchronization of leaf and focal-plane shutters presents different problems. With the former, the aim is to arrange for the peak of the flash to coincide with the period over which the shutter blades are fully open. There is a short delay between the moment of release and when the blades first start to open (approximately 2–5 ms), and a further slight delay before the blades are fully open. Flashbulbs have a delay after firing before combustion occurs and light is produced. By comparison electronic flash reaches full output virtually instantaneously.
Two types of synchronization are used, called respectively X- and M-synchronization. With X-synchronization, electrical contact is made at the instant the shutter blades become fully open so this type can be used with electronic flash at all shutter speeds. Compact cameras with an integral electronic flash unit are X-synchronized only, as are hot-shoe connections. With the obsolescent M-synchronization the shutter blades are timed to be fully open approximately 17 milliseconds after electrical contact is made. This was to allow flashbulbs time to reach full luminous output when the shutter was fully open, requiring a delay mechanism in the shutter. M-synchronization allows synchronization of flashbulbs at all shutter speeds. It is not suitable for electronic flash, which would be over before the shutter even began to open, hence no image. This form of synchronization is now obsolete and only X-synchronization is usually provided. The exceptions are lenses with leaf shutters intended for large-format and some medium-format cameras. Some shutters manufactured have a lever marked ‘V-X-M’. The ‘V’ in this case is a delayed-action (‘self-timer’) setting, which is synchronized to electronic flash only.
Synchronization of focal-plane shutters presents a particular problem. Exposures with electronic flash can be made only at slower shutter speeds, where the shutter slit is the same size as the film gate, so that the whole film frame area can be exposed simultaneously by the flash. When large flashbulbs had to be synchronized with fast shutter speeds up to 1/1000 s, a special slow-burning flashbulb (class FP), now obsolete, was available. This bulb gave a near-constant output for the whole transit time of the shutter blind. The modern electronic flash counterpart has a rapidly pulsed or strobed output that permits the use of flash even at 1/12,000 s.
Unlike a fully synchronized between-lens shutter, which has only one flash connection and a two-position switch to select ‘X’ or ‘M’ synchronization, some focal-plane shutters had one, two or even three flash connections in the form of PC (coaxial) sockets, hot shoes and special-fitting sockets. If only one, unmarked, outlet is fitted to the camera, such as a simple hot-shoe or PC cable socket, the shutter is X-synchronized only. Some older cameras have a separate control linked to this sole outlet, able to select different forms of delay, but this too is now obsolete. A pair of outlets may be X and M, or perhaps X and FP. Use of the appropriate one of these with a suitable shutter speed automatically gives correct synchronization. The extension of exposure automation to the use of electronic flash has resulted in most cameras now giving automatic setting of the electronically controlled shutter to the correct X-synchronization speed, when a suitable dedicated flashgun is attached. This reverts to an alternative setting for ambient light exposure while the flash is recharging.
In X-type synchronization, electrical contact is made when the first shutter blind has fully uncovered the film gate, and before the second blind has begun to move. Depending on the design of the shutter, X-synchronization may be possible up to a shutter speed of 1/300 s. Medium-format cameras have synchronization restricted to a maximum shutter speed of typically1/30 to 1/125 s. Due to these comparatively long shutter-open times in relation to the brief duration of an electronic flash unit, the effect of the ambient light can be a problem, and an electronic flash exposure meter that also takes into account the shutter speed in use is helpful. A comparison of shutter operation and flash source characteristics is shown in Figure 11.10.
The alternatives rear curtain synchronizations is when the flash is triggered just before the second blind starts its travel. This is useful for longer shutter speeds to give a better visual effect when there is a mix of time exposure and flash exposure.
The interface between the electronic flash and camera shutter is by means of an electrical connection in the form of a PC socket, hot shoe or special connection (such as by a pulse of infrared radiation to allow cordless operation in off-camera use). A PC (for Prontor-Compur) connection is a 3 mm co-axial plug-in socket. The hot shoe is a more reliable single-contact outlet, utilizing the original accessory shoe as used for viewfinders, rangefinders and simple flashguns. The centre contact provides X-synchronization. Other contacts provide additional features with ‘dedicated’ flashguns.
Many cameras now have integral flash units or use small units which clip on, both of which place the flash close to the camera lens. The result is near-axial lighting along the optical axis of the lens, which can give ‘red-eye’ where the light illuminates the retina of the eye that is red with blood vessels (see Chapter 3). The result is undesirable and unattractive. The best remedy is to remove the flash unit from near the lens, e.g. on an extension cable or by use of a slave unit. Alternative bounce or indirect flash can be used, or a flash diffuser may help. There is a loss of effective output. For direct flash, multiple pre-flashes may help to reduce the pupil size, an intense beam of light may be directed from the camera to the eye, the subject may be able to look away from the camera, or the ambient light may be increased.
The intensity of the light transmitted by a photographic lens is controlled by the iris diaphragm or stop, which is usually an approximately circular aperture. In some simple cameras the aperture is fixed in size, in others the diaphragm consists of a rotatable disc bearing several circular apertures so that one of them may be brought in line with the lens. Such fixed apertures are known as Waterhouse stops and are used, for example, in some fish-eye lenses. This arrangement is limited in scope. Most lenses use an iris diaphragm, the leaves of which move to form an approximately circular aperture of continuously variable diameter. When the camera shutter is of the between-lens type the diaphragm is part of the shutter assembly. It is operated by a rotating ring, usually with click settings at half-stop intervals, and calibrated in the standard series of f-numbers. The interval between marked values will be constant if the diaphragm blades are designed to give such a scale; in older lenses with multi-bladed diaphragms the scale may be non-linear and the smaller apertures cramped together (Figure 11.11).
The maximum aperture of a lens may not be in the conventional f-number series, but can be an intermediate value, e.g. f/3.5. The minimum aperture of lenses for small-format cameras is seldom less than f/16 or f/22, but for lenses on large-format cameras, minimum values of f/32 to f/64 are typical. An SLR camera requires automatic stopping down to the chosen aperture immediately before exposure, in order to permit viewing and focusing at full aperture up to the moment the shutter is released.
In early SLR cameras, the lens was stopped down manually by reference to the aperture scale. The use of click-stop settings assisted this process. Next a pre-setting device was introduced which, by means of a twist on the aperture ring, stopped the lens down to the preset value and no further. This arrangement is still occasionally used, for example, on perspective control (PC) lenses. The next step was to introduce a spring mechanism into this type of diaphragm, triggered by the shutter release. The sequence was thus speeded up, but the spring needed resetting after each exposure and the diaphragm then reopened to its maximum value. This was known as the semi-automatic diaphragm. Finally, with the advent of the instant-return mirror came the fully automatic diaphragm (FAD). In this system an actuating lever or similar device in the camera, operated by the shutter release, closes the diaphragm down during the shutter operation. On completion of the exposure the diaphragm reopens and the mirror returns to permit full-aperture viewing again. A manual override may be fitted to allow depth-of-field estimation with the lens stopped down. Cameras with a shutter-priority mode have an additional setting on the lens aperture scale, usually marked ‘A’, where the necessary f-number is determined by the camera and set just before exposure, the user having selected an appropriate shutter speed. The aperture set may be in smaller increments than the usual half-stop increments, possibly to 1/10 stop if motorized and under digital control.
In most exposure modes the TTL metering requires the maximum aperture of the lens in use to be set into the metering system to allow full-aperture metering with increased sensitivity. Lenses fitted to large-format technical cameras are usually manually operated. Some lenses have fixed apertures such as mirror lenses and those used in simple cameras.
The functions of the viewfinder are: to indicate the limits of the field of view of the camera lens in use; to enable the user to select and compose the picture; to provide a data display; and to assist in focus or exposure determination. The viewfinder also acts as a ‘control centre’ and has a data display system primarily for the exposure measurement system, with a variety of indices, needles, icons, alphanumerics, lights and camera settings visible around or within the focusing screen area. There may even be either an ‘eye-start’ function where putting an eye to the view-finder switches on the camera functions, or an interactive viewfinder where the direction of gaze activates different autofocus zones.
A versatile simple finder for use at eye level is the frame finder. A metal open frame, with the same proportions as the film format, is viewed through a small peep-sight to define the subject area. It is used with technical, aerial and underwater cameras.
A direct-vision optical finder uses a reversed Galilean telescope, giving a small bright erect virtual image. An improvement, the van Albada finder, uses a white-line format mask and a partially reflecting rear surface on the front negative lens so the lines are seen superimposed on the virtual image. The view extends beyond the frame lines so that objects outside the scene can be seen. A separate mask with selectable frame lines for different lenses can be superimposed on the field of view by a beamsplitter in the optical finder. The frame lines are separately illuminated. This type of finder usually incorporates a central image from a coupled rangefinder (Figure 11.12). Other refinements may include compensation for viewfinder parallax error by movement of the frame line with the focusing mechanism, and a reduction in the frame area when focusing at closer distances. A through-the-viewfinder (TTF) exposure metering system can also be incorporated. As seen in such finders, the image size is usually around ×0.7 to ×0.9 life-size.
The reversed Galilean type of finder gives a bright, upright, unreversed and virtual image. It is not easy to make this finder very compact or to provide a zoom mode. Instead, another type of telescope is adapted, the Kepler telescope used in reverse mode. This gives a real image but it is inverted and laterally reversed, so requires additional optics in the form of a prism erector system (Figure 11.13). The light path can be folded optically to fit inside a compact camera body. A very bright image is given and an intermediate lens gives a zoom action coupled to a zoom lens. An LCD plate in the system can display format lines and data. Aspherical plastic elements reduce distortion and a beamsplitter can sample light for through-the-finder (TTF) exposure metering.
Early cameras used a plain ground-glass screen giving a real image for composing and focusing, as is still used in technical cameras. The advantages of exact assessment of the subject area covered by the lens, accurate focusing and appraisal of the effects of camera movements offset any inconvenience in viewing an image that is dim, inverted and reversed.
A reflex system, with a front-surface mirror inclined at 45° to the optical axis, gives an image on a ground-glass screen in an equivalent focal plane. The image is the same size as it will be recorded on film, and erect but laterally reversed. In a TLR camera, viewing and taking are by separate lenses; in an SLR camera, viewing and taking are by the same lens using an instant return mirror.
A plain ground-glass screen indicates correct focus, but gives a rather dim image, with a rapid fall-off in illumination towards the corners. Evenness of illumination is improved if the screen is etched on the flat base of a planoconvex lens, or if an accessory Fresnel screen is used as a field-brightening element.
A supplementary focusing aid can be incorporated in the centre of the screen. This is a ‘passive’ device, i.e. it has no moving parts. It is usually in the form of a split-image rangefinder or a microprism array. The reflex mirror may be multi-coated to improve reflectance and give a brighter viewfinder image. Some of the incident light may be transmitted to a photocell for exposure metering or to a photosensor array for autofocusing, usually via a ‘piggyback’ mirror hinged to the reverse side.
Lateral reversal of the reflex image is troublesome, especially in action photography, but addition of a pentaprism viewfinder gives an erect, laterally correct and magnified screen image for focusing. There is provision for eyepiece correction lenses or dioptric adjustment for users who would otherwise need spectacles for viewing. The view-finder prism may incorporate a through-the-lens exposure metering system for measurements from the screen image.
Lenses can be used as fixed-focus objectives, using small apertures, depth of field and distant subjects to give adequate image sharpness, but this is not feasible with lenses of large aperture or long focal length, or for close-up work. To ensure that the most important part of the subject is in sharp focus it is necessary to have some form of focusing system, as well as a visual indication of the state of focus. Current focusing systems use a variety of mechanical, optical and opto-electronic arrangements (Figure 11.14). Focusing satisfies the lens conjugate equation (see Chapter 6), where a change in subject distance (u) requires a corresponding change in image distance (v) for focal length (f).
The simplest method is by movement of the entire lens or optical unit (Figure 11.14a). This is called ‘unit focusing’, as used in technical cameras. Monorail types are focused by moving either the lens or the back of the camera. Rear focusing alters the focus only, whereas front focusing alters the size of the image as well as the focus. Other cameras have the lens unit installed in a lens barrel or focusing mount. Rotation of a ring on the lens barrel moves the lens in an axial direction. A helical focusing mount causes the lens to rotate during focusing, whereas a recti-linear mount does not. A double-helicoid arrangement provides macro lenses with the extended movement necessary for continuous focus to magnifications of 0.5 or more. The focusing action may be coupled to a rangefinder or to an autofocus system, or it may be viewed on a screen. The subject distance may also be set by a scale.
Focusing is possible by varying the focal length of the lens and not varying the lens-to-film distance. This is done by moving the front element to alter the separation of the front element from the other groups (Figure 11.14b). A slight increase in this separation causes an appreciable decrease in focal length, giving a useful focusing range. Close focusing to less than 1 metre is not satisfactory due to lens aberrations. Zoom lenses may use a (well-corrected) movable front group for focusing. Rotation of the front element is a nuisance with the use of polarizing filters and other direction-sensitive attachments such as graduated filters.
A positive supplementary lens can be fitted in front of almost any lens (see Chapter 10) to give a fixed close-up focusing range (Figure 11.14c). If the subject is positioned at the front focus of the supplementary lens, the camera lens receives parallel light, so it gives a sharp image at its infinity focus setting. By varying the focus setting on the camera lens, a limited close-up range is given. The design of the supplementary lens is important, as the curvatures of its surfaces determine its effect upon the aberration correction of the prime lens. Usually a positive meniscus shape is used, with its convex side to the subject. This simple lens may be perfectly adequate when used at small apertures, but an achromatic cemented doublet design with anti-reflection coatings is preferable for improved performance. A +1 dioptre lens has focal length 1000 mm, and if of diameter 50 mm has an aperture of f/20, which allows a doublet design to give excellent correction.
With internal focusing (Figure 11.14d), the focusing control moves an internal group of elements along the optical axis, thereby altering focal length but not the external dimensions of the lens. The front of the lens does not rotate either. An extensive focusing range is possible with only a small movement. The lens is not extended physically, and the whole unit can be sealed against the ingress of dust and water. The focusing movement can also incorporate the correction for lens aberrations that increase as the focused distance is decreased. Internal focusing is particularly suitable for autofocus lenses and the drive motor can be located either in the camera body or in the lens housing. Internal focusing is used for a variety of lenses, including zoom, macro and super-telephoto types. Close focusing is provided, with retention of image quality.
Many lenses can only be focused close enough to provide a magnification of about 0.1 or less. Most standard lenses have a minimum focusing distance of 0.3–1.0 metres. If the lens is removable, the use of extension tubes and bellows extensions between the lens and camera body provide the additional lens-to-image distance needed to give greater magnification. (Non-interchangeable lenses are limited to use of a supplementary close-up lens.) An extension tube has fittings for attaching the lens to one end and the camera body to the other. Tubes of various lengths are available for use singly or in combination. For SLR cameras, automatic extension tubes have the necessary mechanical or electrical linkages to retain the operation of the iris diaphragm and TTL metering systems.
Where magnification is greater than 1.0, the optical performance of a lens may be impaired, as corrections are normally computed for work at infinity, and macro settings reverse the usual proportions of conjugate distances (i.e. v >> u). A lens reversing ring, to mount the lens with its rear element facing the subject, reduces this problem.
Lenses are usually focused without reference to focusing scales when a rangefinder, focusing screen or autofocus system is provided. Scales are useful for use with electronic flash, so that the aperture may be set according to a flash guide number, and to check if the subject is within the operating distance range of an automatic flash exposure system. Also, the distance values can give an estimate of the depth of field by reference to the appropriate scales on the mount, or to tables. The focusing scale may also carry a separate index for infrared use usually denoted by a red dot or letter ‘R’, and the distance value first set visually or by autofocus is then transferred manually to the IR index.
The traditional ground-glass screen or focusing screen is a most adaptable and versatile focusing system, giving a positive indication of sharp focus and allowing depth of field to be estimated. The screen can focus any lens or optical system, but accuracy depends on screen image luminance, subject contrast and the visual acuity of the user. Supplementary aids such as a screen magnifier or loupe and a focusing hood or cloth are essential, especially when trying to focus systems of small effective aperture in poor light. The plain screen has evolved into a complex optical subsystem, incorporating passive focusing aids (see below) and a Fresnel lens to improve screen brightness. The screen can be a plate of fused optical fibres or have a laser-etched finish to provide a detailed, contrasty, bright image. No single type of screen is suitable for all focusing tasks and many cameras have interchangeable focusing screens with different properties. Precise visual focusing is aided by accurate adjustment of the dioptric value of the eyepiece magnifier to suit the vision of the user.
A coincidence-type rangefinder uses two windows a short distance apart, through each of which an image of the subject is seen. The two images are viewed superimposed, one directly and the other after deviation by an optical system (Figure 11.15a). For a subject at infinity the beamsplitter M1 and rotatable mirror M2 are parallel, and the two images coincide. For a subject at a finite distance u the two images coincide only when mirror M2 has been rotated through an angle x/2. The angle x is therefore a measure of u (subject distance) by the geometry of the system, and may be calibrated in terms of subject distance. By coupling the mirror rotation to the focusing mount of the lens in use, the lens is automatically in focus on the subject when the rangefinder images coincide. The accuracy of this coupled rangefinder system using triangulation is a function of subject distance u, the base length b between the two mirrors and the angle x subtended at the subject by the base length. Mechanical and optical limitations make this system of focusing unsuitable for lenses of more than about two and a half times the standard focal length for the film format, e.g. 135 mm for the 24 × 36 mm format. The method is unsurpassed in accuracy for the focusing of wide-angle lenses, particularly with a long-base rangefinder and in poor light conditions. Alternative systems use a fixed mirror and beamsplitter, obtaining the necessary deviation by a lens element which slides across the light path between them (Figure 11.15b). The rangefinder images are usually incorporated into a bright-line frame viewfinder.
A split-image rangefinder is a ‘passive’ focusing aid in that it has no moving parts, unlike the coincidence-type rangefinder. It is a small device, consisting basically of two semicircular glass prisms inserted in opposite senses in the plane of the focusing screen. Any image that is not exactly in focus on the central area of the screen appears as two displaced halves. These move together to join up as the image is brought into focus, in a similar way to the two images of the field in a coincidence-type rangefinder. The aerial image is always bright and in focus, even in poor light conditions. Because focusing accuracy by the user depends on the ability of the eye to recognize the displacement of a line, rather than on the resolving power of the eye, which is of a lower order, the device is very sensitive, especially with wide-angle lenses at larger or moderate apertures. The inherent accuracy of this range-finder depends on the diameter of the entrance pupil of the lens in use, so large apertures improve performance. Unfortunately, the geometry of the system is such that for effective apertures of less than about f/5.6, the diameter of its exit pupil is very small, and the user’s eye needs to be very accurately located. Slight movements of the eye from this critical position cause one or other half of the split field to black out. Consequently, long-focus lenses or close-up photography result in a loss of function of the rangefinder facility and produce an irritating blemish in the viewfinder image.
The same principle is used in the microprism grid array where a large number of small facets in the shape of pyramids are embossed into the focusing screen surface. An image from the camera lens that is not precisely in focus on the screen undergoes multiple refraction by these facets so that the image appears broken up into tiny fragmentary areas with a characteristic ‘shimmering’ effect. At correct focus this image snaps back into its correct appearance, giving a very positive indication of the point of focus. The pyramids are small enough (<0.1 mm diameter) to be below the resolving power of the eye. The disadvantage of the microprism is the same as that of the split-image rangefinder, the microprism array blacking out at effective apertures less than about f/5.6.
Visual focusing can be slow, inaccurate and tiring. Methods of obtaining or retaining focus automatically are helpful and such autofocus systems can operate in various modes. Autofocus can be linked to the shutter release and may be blocked until focus is achieved. The focus may then lock at this setting or it can search continuously for focus as the camera or subject move. An ‘active’ type of autofocus scans a beam of infrared radiation across the scene; the reflection is measured and triangulation is used, similar to the coupled rangefinder. A ‘passive’ type uses no moving parts and a pair of images are compared using the output from a CCD array, similar to a split-image rangefinder.
A focus indication mode or assisted focusing mode is a form of electronic rangefinder that is an analogue of an optical rangefinder or focusing screen. An LED display lights up in the viewfinder to indicate correct focus when the lens is focused manually; the direction of focusing movement required is indicated if the focus is incorrect. A predictive autofocus mode anticipates the position of a moving subject after the time delay due to the mirror rise time and will ‘drive’ the lens to this focus setting in compensation. An automatic depth-of-field mode is where the lens is focused automatically on the near and far points of the scene to be rendered in focus; the necessary intermediate focus is set and the automatic exposure determination system then chooses the appropriate aperture to give the correct depth of field, given an input of data from the lens as to its focal length. Another type is a motion-detection mode, or focus-trap mode, where the camera is autofocused at a suitable distance; when a moving subject enters this zone, the camera automatically fires the shutter. A focus maintenance mode retains sharp focus once this has been set visually. Autofocus systems should be rapid-acting, function at low light levels and give accuracy comparable to visual methods. Different systems are in use.
One autofocus method uses direct measurement of the subject distance by an active ranging method using pulses of ultrasound, as introduced by Polaroid in cameras for self-developing film. A piezo-electric ceramic vibrator (PECV) emits a ‘chirp’ of ultrasonic frequencies and elapsed time for the return journey to the subject is proportional to subject distance. This system operates even in the dark, but cannot penetrate glass.
Other ‘active’ systems involve scans across the subject area. Electronic distance measurement (EDM) uses an infrared-emitting diode moving behind an aspheric projection lens to scan a narrow beam across the scene. Reflection from a defined subject area is detected by a photodiode, and controls the synchronous focusing movement of the camera lens. This is the electronic analogue of a coupled rangefinder. Infrared beams can operate in darkness and through glass, but can be fooled by unusual reflection of the radiation from various materials such as dark cloth.
Phase detection autofocus is a passive system using one or more linear CCD arrays and no moving parts. The CCD array is located in the mirror chamber (Figure 11.16) beyond an equivalent focal plane and behind a pair of small lenses that act in a similar manner to a prismatic split-image rangefinder, as in Figure 11.17. Divergent pencils of rays beyond the correct focus position are refracted to refocus upon the array, and the separation (or ‘phase’) of these focus positions relative to a reference signal is a measure of the focus condition of the camera lens. The phase information operates a motorized focus control in the lens. Very rapid focusing is possible, even in continuous-focus mode. The system will even focus in the dark, using the emission of infrared radiation from a source in a dedicated flashgun. The autofocus zone or zones are shown in the viewfinder and can be selectively located on subject detail to be in focus. The focused distance data may be used in exposure calculations or in the operation of dedicated flash systems. Accuracy of operation is dependent upon using a lens aperture of f/5.6 or larger and also a reasonable light level. Low-light performance is given by an EV scale of values, similar to that used for exposure metering systems. A dedicated flashgun fitted to the camera may emit a small pre-flash with a striped pattern to fall on the subject and assist autofocus in dim light. Alternatively, a continuous beam of deep red light or even infrared may be used.
An SLR autofocus system involves a beamsplitter mirror, whose effect is altered by linearly polarized light, so errors may occur if a linear polarizing filter is used over the camera lens. A circular polarizing filter is needed.
Most cameras incorporate some form of exposure metering system. The meter with a calibration constant (K) measures the luminance (L) of the subject by reflected light, and given a film speed (S) the exposure equation t = KN2/LS is solved by an integral calculator to display the various combinations of shutter speed (t) and f-number (N) required. The value of film speed may be automatically input by a DX coding system. The values of N and t may be selected automatically by a selected program mode. Various forms of photocell have been used in light-metering systems, including selenium cell and cadmium sulphide cell types, now effectively replaced by the silicon photodiode. Incident light on this photovoltaic device generates a very small current. This output is amplified and converted to a voltage which is linearly proportional to the incident light. Response is good even at very low light levels, and the linearity is maintained over a wide range of illumination levels. The response time of a silicon photodiode is very short, in the order of microseconds, and cell area can be very small while retaining adequate sensitivity. Spectral sensitivity is from 300 to about 1200 nm, with a peak around 900 nm. For photometric use in a camera or light meter, filtration to remove much of this natural IR sensitivity is necessary to give a suitable response.
A light-metering system built into a camera body coupled to the camera controls can be used either for manual setting or for one of several alternative automatic exposure (AE) modes. The photocell(s) may be located in various positions, requiring different optical systems for each variant. In non-SLR cameras the metering system reads directly from the subject. The cell is placed behind an aperture located either on the camera body or near the camera lens. Metering sensitivity is high as light losses are low compared with TTL metering. A frame line in the viewfinder may indicate the measurement area. To provide allowance for film speed, a metal plate with a series of graded apertures, or a tapered slit, is located in front of the photocell and linked to the film speed setting control.
Through-the-lens (TTL) measurements have advantages in that only the imaged area is used, regardless of the lens fitted, and exposure compensation is automatic for bellows extension and optical attachments. Such compensation is important for zoom lenses, as the f-number usually changes as focal length is altered.
The best location for a photocell is in the true photo-plane of the camera, as realized in some insertable metering systems for large-format cameras, but an equivalent focal plane can be used instead. Details of photocell locations and optics are given in Figure 11.18. Such TTL systems are normally limited to simple measurements of subject luminance. Uncertain accuracy with non-average scenes using a simple integrated measurement requires alternative metering sensitivity patterns to improve the proportion of acceptable exposures. Alternative patterns use approximate full-area integration with central bias, centre-weighted measurement (possibly with directional bias) and small-area measurement. Some may use spot metering of very small areas, or zone measurements by multiple zone or segmented photocells to provide scene categorization information (matrix metering).
An accurate form of TTL metering is that of off-the-film (OTF) metering, in which image luminance measurements are made using the light reflected from patterned shutter blinds or from the film surface itself during exposure, even with rapidly fluctuating light levels. For short exposure times, the patterned shutter blind simulates average image luminance over the obscured area of the film gate during travel of the slit. Flash metering uses the open gate full frame area.
In-camera metering is normally used in semi-automatic or automatic mode. In the former, either shutter speed or aperture or both together are varied manually, until a readout indicates that correct exposure has been set. The camera can be used to scan and sample the scene in chosen areas, and a modified exposure given as judged necessary. Automatic exposure offers a choice of shutter priority, aperture priority and various program modes. Additionally, metering pattern may be altered, an exposure compensation control may be used or an exposure memory lock employed to improve the proportion of acceptable pictures from non-average scenes.
Metering sensitivity is dependent upon the effective maximum aperture of the lens in use. Measurement may have to be carried out in a stopped-down mode if a lens without automatic iris diaphragm operation or maximum-aperture indexing is used. Metering is of the light reflected from a subject, as the incident-light mode is not easily available to TTL metering systems. Metering sensitivity with an f/1.4 lens and ISO 100 film is generally of the order of 1EV, significantly less than that of a hand-held meter.
Multi-mode cameras offer a choice of user-selectable AE programs. The programmed exposure mode is the ‘snapshot’ form of setting and was the original form of automatic exposure. The principle of this ‘P’ mode is that photocell output and the set film speed will give the necessary exposure for a scene in terms of exposure value (EV), which represents a wide range of equivalent combinations of lens aperture and shutter speed. Associated circuitry is programmed to give one particular combination according to the EV data fed in. Examples of such programs are shown in Figure 11.19, in which a program is given as a straight line with a chosen slope and one or more changes in direction to span the EV measuring range of the camera.
A horizontal or vertical line shows that one variable is kept constant while the other changes continuously; for example, shutter speed may change continuously over the range 8 to 1/8 s while the aperture is held constant at f/1.4. A sloping line shows that both variables change continuously. A steeply sloping line shows that priority is given to the maintenance of short exposure durations, to minimize the risk of camera shake or to stop subject motion, while relying upon lens performance and accurate focusing at the corresponding large aperture. A required EV of (say) 12 for correct exposure is interpreted in various ways by programmed cameras. It could mean settings of 1/30 s at f/11 for one camera or 1/80 s at f/7, 1/125 s at f/5.6 or even 1/500 s at f/2.8 for other cameras. An electronically controlled shutter and fractional control of lens aperture setting allows non-standard exposure settings.
The user can decide whether shutter or aperture priority might be appropriate under different circumstances, such as moving subject matter, due to depth-of-field requirements, low light levels or long-focus lenses. So there is a choice of ‘normal’, ‘action’ and ‘depth’ AE programs. Additionally, the insertion of an alternative lens of different focal length may automatically select one of three shutter-biased program modes usually referred to as ‘wide’, ‘standard’ and ‘tele’, with the turning point of the program line on the EV graph at the point where the minimum hand-held shutter speed may be used. The slope of the program line may increase progressively from ‘wide’ to ‘tele’. A zoom lens may alter the program selected continuously as focal length is altered.
In any programmed AE mode the aperture value is selected by the camera in increments that may be as small as 0.1EV. When used with a metering-pattern system capable of some analysis of tonal distribution of the subject matter, programmed exposure modes are capable of producing a high proportion of acceptably exposed, sharp results.
Accurate determination of camera exposure, especially by in-camera TTL AE systems, is limited by variations in subject luminance ratio and non-typical subject tone distributions, which necessitate their recognition and classification by the user. Examples of such images include those containing large areas with very light tones or very dark tones; in these cases auto-exposure will often result in incorrect results, as the camera will be basing the exposure on an average of the scene, assuming that it should be a mid-grey. The use of spot metering, small-area and various weighted integrated readings can help, with intelligent use. A greater percentage of successful results is possible using a segmented photocell or matrix metering arrangement for scene classification. The viewfinder image is measured, assessed, compared to stored data on a large number of scene patterns and categorized for subject type. Adjustments are then made automatically to the basic camera exposure. For example, the screen area may be measured as five or six separate zones (with some overlap) by two symmetrically opposed segmented photodetector arrays, both with three independent measurement areas. Other arrays can provide spot, small-area and weighted readings as well (Figure 11.20). The weighted response data from the separate segments is classified by signal-processing techniques. Analysis of zonal luminance difference values and the total subject luminance value, plus the metering pattern, allows classification into a number of computer-simulated subject types and the exposure is based on stored data for several hundred subject types as determined from practical tests or evaluation of photographs (see also Chapter 12).
The tone distribution may be determined as a comparison of central foreground to background luminance, and if a chosen difference value of EV is exceeded, then an integral flash unit may be charged automatically and used for fill-in purposes. The lens aperture and shutter speed are optimized separately for subject distance (using autofocus data) and flash synchronization. Additionally, separate segments may be switched out or used in different configurations to provide various alternative forms of metering sensitivity patterns (spot, small-area, centre-weighted).
An improvement in scene classification is to provide subject distance (u) information from the autofocus module and focus setting on the lens. The latter requires that the focusing control has a suitable transducer and data link with the microprocessor in the camera. From the matrix array of photocells used to divide the scene into segments, the scene can be classified using the three pieces of data, overall scene luminance (L), luminance range of the subject (LRS) and the subject distance (u), then subdivided into a series of scenes for which optimum exposure data is stored, based on extensive practical tests, especially for non-typical scenes. The scene is matched to one of these and the optimum exposure given. The defocus data from the AF system helps determine whether the subject is central or off-centre in the frame, as this can give emphasis to different segments of the matrix. By the additional use of ‘fuzzy logic’ procedures, the scene type boundaries are not sharply defined but overlap to some extent to avoid sudden changes in exposure, e.g. when using a motor drive, when small variations in an essentially constant scene may cause a noticeable change in exposure as the scene is inadvertently reclassified and exposure altered. The same principle can be applied in a simpler way to the use of fill-in flash to give a balanced exposure to subjects, especially of excessive contrast. A matrix array of CCD photocells filtered to blue, green and red light, together with scene classification data, can also be used in-camera to assess the colour temperature of a scene (see Chapter 14).
Most cameras depend upon battery power for their functions, such as metering, autofocus, microprocessor operation, electronic shutter control, self-timer, viewfinder displays, and for motors to advance the film, rewind it and focus the lens. Accessories such as power winders, flash-guns, data backs and remote controls all use batteries. Without functioning batteries the equipment is inoperative. An exception is a ‘mechanical’ setting of a shutter providing a single speed suitable for flash synchronization, given a failure of the battery operating the shutter timing circuitry.
For camera equipment only low direct current (DC) voltages are required, seldom more than 6 volts. Small batteries are favoured given space limitations, but have reduced capacities, especially given the number of functions to be powered. Batteries should be inserted correctly regarding polarities and may be in a removable holder or a shaped grip on the camera body. Performance is related to temperature and a cold weather adaptor kit may be available to allow the battery pack to be kept in a warm pocket and a long cable terminates in an adaptor to go into the camera compartment. A battery condition test and indication is vital. To conserve battery power a timing circuit may switch off the camera after a short time, varying from 30 seconds to several minutes. Slight pressure on the shutter release will keep power on. Other means of saving battery power include having mechanical latches on camera shutters for long exposure times using the ‘B’ setting. Electronic flashguns use very effective thyristor-controlled energy-saving circuitry to discharge the capacitor only as much as necessary; also, the unit may switch off if not used for a few minutes. Battery drain depends on circumstances, type of circuit and type of battery. Some metering or shutter control circuits are largely independent of the state of the battery; others such as silicon photodiode circuits require a stabilized voltage, so batteries are favoured that have a near constant output during their operation and life. Heavy-duty applications such as motor winders and flashguns need moderate voltage and short bursts of heavy current, supplied by clusters of alkaline–manganese or nickel metal hydride (NiMH) cells. Medium-format cameras with integral motor drives also have heavy-duty needs.
A battery uses cells in series and each electrochemical cell uses dissimilar electrodes and a suitable conducting electrolyte, which can be liquid, gaseous, viscous or near solid. Leakage of electrolyte can be disastrous and damage equipment severely, so batteries should be removed when equipment is not needed for a time. Chemical reactions cause a gradual loss of voltage from its initial value. Shelf life varies for unused batteries, but can be 5 years or more in the case of lithium types. A battery tester may be incorporated to check voltage before use. Batteries should be changed in sets, not just one discarded when found to be discharged. Batteries are either of the throwaway or rechargeable type. The first is disposed of when exhausted and it is very dangerous to attempt to recharge it as the gases formed may cause it to explode. The rechargeable cell uses a trickle DC current of low voltage to reform the chemical cell by inducing reverse reactions. With such cells, too rapid a discharge or charge must be avoided as internal damage and heating may occur. Types of rapid recharging batteries are available such as lithium-ion and nickel metal hydride types. Batteries should be inspected for corrosion before insertion and the terminals wiped with a cloth to remove any oxide film and improve contact.
A number of battery types are in use. The carbon–zinc cell, the oldest type, is cheap, has a short life and cannot deliver large amounts of current. It is best used only in an emergency if other varieties are unobtainable. Zinc chloride types give improved performance. The alkaline cell, or manganese–alkaline type, is similar to the carbon–zinc type but with a different electrolyte and near-leakproof construction. It has about 10 times the life and a wide operating temperature range. The mercury button cell is no longer available for safety reasons and has been largely replaced by the silver-oxide cell, which has a very stable output, but is costly so is frequently substituted by alkaline or lithium cells. The lithium cell uses lithium instead of zinc for its positive terminal. Combined with manganese dioxide, a long shelf life is given and a useful life in a camera. Low-temperature operation is possible to – 20°C. The smallest voltage available is 3 volts so it can only replace two silver-oxide types used in series. Other types include lithium-ion and nickel metal hydride batteries, both of which are rechargeable and retain no memory of previous use.
Image degradation and loss of resolution occur if the image moves across the photosensor area during exposure duration (t). This can be due to movement of the subject, the camera or both, referred to as camera shake. The subject velocity (Vs) and image velocity (Vi) are related by magnification (m) in that for motion parallel to the film plane Vi = mVs. For motion in a direction inclined at angle θ to the optical axis:
In photography, blurring of an image point due to movement can be reduced to an acceptable level, such as to the value of C, the diameter of the circle of confusion, by a short exposure as given by shutter speed or flash duration, so:
The technique of image motion compensation (IMC) matches photosensor and image velocities during exposure to reduce blurring. For a fixed camera position and viewpoint, even the image of a static subject may move a distance exceeding C during exposure, due to camera shake. This is compounded by photography from a moving platform such as a car or helicopter. Taking m = f/u as above, then:
So t is proportional to 1/f, hence for a lens with f = 500 mm, t must be 1/500 s or less for shake-free results. Personal limits for hand-held exposures can be determined using a test subject to compare results for tripod and hand-held exposures at different shutter speeds with different focal lengths. Cameras with auto-exposure systems interface with the lens to monitor focal length and give a visual or audible warning if the shutter speed exceeds a safe value. An autofocus module using a CCD array of cruciform shape can similarly be programmed to warn of camera shake by comparing phase information in both x and y directions to obtain a measure of image velocity. Camera performance is always improved by use of a suitable tripod or other support. For hand-held use a method of offsetting the blurring effects of movement is to stabilize the image optically.
Methods used have included a gyro-stabilizer using a heavy flywheel rotated at high speed, when inertia reduces high-frequency oscillations or a body harness with balanced springs to allow movement to follow action smoothly without vibration by distribution of mass and balance. The effects of camera movement are reduced if the lens is supported at a position beneath its rear nodal point, as a lens can rotate about this point or optical centre without movement of the image. Such mounting can be difficult as the node can be in front of the lens if it is of telephoto design, or it can change position if it is a zoom lens.
Compensation is possible by additional refraction of incident light at an angle proportional and opposite to the camera motion angle using a liquid lens formed by a volume of liquid between two optical flats of the same refractive index. One flat is rotatable about two axes to produce a variable angle prism for compensation. This has little effect on lens properties. Contemporary methods use solid-state accelerometers to detect pitching and yawing movements in orthogonal axes and to operate piezo-electric actuators within 10 ms to adjust either a compensating lens group within the lens (see Figure 11.21), or a liquid-filled cell in front of the lens (used in camcorders).
Alternative non-optical methods use digital technology. The image is stored in a frame store, segmented into several areas and some 30 points in each area are compared to the next frame. The frame is cropped and reduced, rotated to align, and enlarged back again for display and recording. Alternatively, gyro sensors can adjust the timing of the CCD array to ‘move’ the image in a compensating direction.
For most photography it is usually sufficient to have the optical axis of the imaging lens orthogonal to the photo-plane, intersecting it at (or near to) the point of intersection of the format diagonals, i.e. the image centre. This point is the principal point (of autocollimation) (P), not to be confused with the principal points of the lens, and its position is the origin of a Cartesian coordinate system for image points.
As well as axial focusing movements of the lens or photoplane, some camera designs offer the possibility of a range of camera movements, where the optical axis of the lens and the camera axis of symmetry are displaced (translated) and/or rotated relative to each other (see Figure 11.22a). The movements are usually called shifts, swings and tilts, and have distinct practical uses. They are used to control focus distribution (image sharpness) and image shape (‘perspective’) by use of the lens or photo-plane movements respectively. Application of the Scheimpflug Rule (to give collinearity at the intersection of subject plane, lens rear principal plane and photoplane) gives optimum distribution of acceptable image sharpness within the volume of the subject space (Figure 11.22b). Some large-format cameras of the monorail type are equipped with transducers and sensors to measure swings and tilts, while a small on-board computer will indicate optimum settings to control sharpness as required from application of Scheimpflug.
A similar range of movements is provided in some enlargers for the restoration of image shape and scale when previously recorded by a tilted camera or without corrective movements.
A variety of movements find specific uses, including correction of verticals in architectural photography, lateral shifts by a ‘perspective control’ (PC) lens in panoramic photography and lens rotation in panoramic cameras, as well as lateral shifts and tilts for stereo photography, close focusing in photomacrography and the restoration of perspective from oblique views that include a subject reference grid.
Electronic imaging systems may offer the ability to restore or produce image shape changes by alternative means, such as the removal of keystoning (converging verticals when the camera is tilted up) in images for video projection. Digital image processing may use geometric transformations to restore image shape or distort images to conform to a particular form of image projection.
Goldberg, N., 1992. Camera Technology. Academic Press, San Diego, CA, USA.
Harris, M., 1998. Professional Architectural Photography. Focal Press, Oxford, UK.
Jacobson, R.E., Ray, S.F., Attridge, G.G., Axford, N.R., 2000. The Manual of Photography, ninth ed. Focal Press, Oxford, UK.
Ray, S., 1983. Camera Systems. Focal Press, London, UK.
Ray, S., 1999. Scientific Photography and Applied Imaging. Focal Press, Oxford, UK.
Ray, S., 2002. Applied Photographic Optics, third ed. Focal Press, Oxford, UK.
Stroebel, L., 1993. View Camera Technique, sixth ed. Focal Press, Stoneham, MA, USA.
18.191.237.131