In this document, the term flash is exclusively used to denote a short burst of light. The words flash unit are used to denote a device capable of emitting a flash. In everyday speech, flash units are very often called just flashes. Some manufacturers use the term speedlite or speedlight to denote a flash unit.
It is assumed in most of the document that the flash unit is a separate device which is attached to the camera (typically a SLR). Most cameras also have a built-in flash unit which works on the same principles. Its electronics is usually more integrated with the camera and some modes (such as full power) may be unavailable or at least require some effort to achieve.
Electric energy from a battery (or other sources, such as household current) is converted to high voltage (300 volts or more) and is used to charge a capacitor. The converter sometimes makes a high-pitch sound which you can hear when the unit is charging. The capacitor is permanently connected to two electrodes in a glass tube ("bulb") filled with xenon gas. At this stage, the gas does not conduct electricity and emits no light.
Another, small, capacitor is charged at the same time as the big one. When the flash unit needs to fire, this small capacitor is discharged through a transformer, which generates a pulse of very high voltage (several thousands of volts). This voltage is applied to a third electrode in the xenon tube. The high-voltage pulse causes the gas to ionize. Ionization makes the gas conductive, and the big capacitor starts to discharge through the xenon gas. Bright light is emitted by the xenon gas during this process. Since the resistance of the gas is very low at this stage, the discharge is rapid, with the current following an exponential curve. About 1/1000 - 1/200 seconds later the capacitor is essentially empty, and the voltage has dropped so low that the xenon stops to conduct electricity, and the light pulse dies off. At this point, the process can be started from the beginning.
This mode of operation is often called "full power" or "manual flash" and can, in theory, be achieved with any flash unit (unless 'clever' electronics in the flash unit disallow it).
In order to limit the amount of light emitted by the flash unit, the discharge needs to be cut short before all the charge available in the capacitor has made its way through the xenon. In most flash units produced today, a semiconductor device (most often some type of thyristor) is placed in the discharge path, which can switch off the discharge current when needed. This also means that any energy which is left in the capacitor is preserved and can be used for the next pulse.
The signal to turn off the flash can come from various sources. Some flash units have switches or menu selections which limit the power to 1/2, 1/4, 1/8, etc. of the maximum. In these cases a simple timing circuit can turn off the flash. Some (typically older) units have a photo cell on the front side. The light reflected from the subject back into the cell is measured, and when the resulting exposure is judged to be sufficient, the flash is turned off. Flash units built into cameras (as well as some advanced standalone units) may use distance information from the camera's focusing system to calculate the required duration of flash.
In TTL (Through The (camera) Lens) systems, the photo cell is located inside the camera, and (typically) measures light reflecting off the imaging sensor (film in older cameras) during exposure. Again, when enough light has been registered, a signal is sent from the camera body to the flash unit to turn off the discharge.
Many modern systems use a low-powered pre-flash to illuminate the scene just prior to the exposure and measure the reflected light using the camera's primary light-metering sensor (which may be the actual imaging sensor in certain designs). This infprmation is then used to calculate the amount of additional flash lightning necessary.
Some experimenatlly measured flash discharge curves are available at a separate page.
For some applications (see below), it is important that the subject would be illuminated for a longer time than the typical duration of a flash of 1/1000 seconds. One way to achieve this is to deliver a rapid sequence of flash pulses. This is done by repeatedly turning the current off by a timing circuit, and then starting the next pulse a little (e.g. 1/5000 sec) later. Since the charging circuitry does not have the time to re-charge the capacitor during these 1/5000 seconds, the total energy available in the capacitor is divided between several flashes, reducing the average brightness of the flash.
Other types of modern electronics may reduce the flash current to a lower steady level and provide essentially the same result.
The further away the subject is located from the flash unit, the more energy is needed to give sufficient illumination for proper exposure. If the power or the flash stays constant, it is necessary to open up the camera's aperture as the distance between the flash unit and the subject is increased. The distance and the aperture value for proper exposure are connected by this formula:
This constant is called the Guide Number (GN for short). Since f-stop values have no units attached, the guide numbers are expressed in units of length (feet in the U.S., meters most elsewhere). Guide numbers are valid only for a given sensitivity (ISO value) of the sensor or film, and typically are expressed for ISO 100.
In order to double the guide number, four times more energy needs to be stored in the capacitor. This puts practical limits to guide numbers, since the size, weight, and cost of the capacitor increase as its capacity is increased. In practice, most flash units built into SLR cameras are limited to about 15m/45ft or even less, while off-camera units are typically in the 24-50m/80-150ft range, with some pro models slightly higher. Studio flash units, which can be powered from AC power and need not be carried around, may have much higher guide numbers. Modern compact cameras tend to have very low (10m/30ft) guide numbers. The size of the camera limits the size of the capacitor and the latter in turn puts limits on the guide numbers.
The flash unit produces a cone of light which lights up the field of view of the lens. There is no point in lighting up parts of the scene which are not captured on the film. Each lens has a certain angle of view, and the manuals of flash units usually mention the shortest focal length (widest angle) lens which can be used with the given flash unit.
Lenses with longer focal length have narrower field of view. When using such lenses, much of flash power is wasted to illuminate areas of the scene which are outside of the field of view of the lens. If this light energy would be used to illuminate the actual picture area, the guide number could be increased without using a larger capacitor.
Some flash units have the ability to direct the power of the flash into a narrower angle when needed. These units are known as zoom flashes (zoom flash units would be proper in this document, but I have never seen such a term). When zoomed "in" for longer focal lengths, their guide number is increased. When zoomed "out" for shorter focal lengths, the guide number decreases. Manufacturers often state only the maximum guide number (at tele position) in the sales literature. For a fair comparison with a non-zoom unit, the coverage angle of the non-zoom (typically corresponding to 28mm or 35mm lenses) should be known, and the GN of the zoom flash at that focal length should be used in the comparison.
For a given flash unit, lens, and sensor (film) sensitivity, there is a limit of the unit-to-subject distance which can get sufficient illumination to produce proper exposure. This limit corresponds to the flash unit operating at full power, and the lens aperture wide open. For example, a point-and-shoot camera lens could have a maximum aperture of f/4.0, and the unit's guide number of 12 meters, which means that the working distance is limited to 12/4 = 3 meters (about 10ft) with ISO 100 film.
These matters are further complicated when zoom lenses and zoom flashes are in use. For example, a 28-80mm SLR zoom lens could have f/3.5 maximum aperture at the wide end, and f/4.5 aperture at the long end. The zoom flash would have GN 15m at 28mm, and GN 24m at 80mm. Now the maximum working distances become 15/3.5 = 4.3m at wide end, and 24/4.5=5.3m at the tele end. In the worst situations, the maximum working distance does not increase with focal length at all - a result often overlooked when evaluating flash unit and lens combinations.
One may be surprised that there is often also a minimum working distance specified in flash unit manuals. This has to do with the speed of the electronics in the flash unit: for the smallest working distances the flash durations become very short, and switching the flash off with sufficient accuracy becomes impossible at one point. Other factors which limit working distance are limits on the sensitivity of the light meter in the flash unit or in the camera, as well as parallax problems, when the lower border of the light cone gets into the field of view of the lens at shorter distances. Neutral (gray) filters over the flash head can be used to further reduce the minimum working distance, but only tilting the flash (often hard or impossible) will take care of the parallax problems.
Doubling of sensor sensitivity (film speed) will increase the guide number by a factor of square root of two (approximately 1.4) . Going from ISO 400 to ISO 1600 sensitivity, the guide number is doubled. Going from ISO 400 to ISO 100, the guide number is halved.
Always be careful about the units of length: meters or feet. For purposes of these calculations, 1ft=0.3m is good enough. Guide numbers in feet are typically in the 30-50 range for built-in flash units and between 50 and 150, possibly more, for stand-alone units. Guide numbers in meters are typically below 15 for built-ins, and 15-50 for stand-alones. As you can see, there is little (if any) overlap within a class. Always check with a reliable source when in doubt.
Guide numbers of zoom flashes typically increase by 1.3 to 1.4 times when the focal length is doubled. If the manufacturer states that a zoom flash has GN 35m at 100mm, then at 28mm it will be roughly half of that, or 17m. How did I get it? 100mm and 28mm relate approximately as 4 and 1, so we have two operations of doubling through the range. If each of them divides the GN by 1.4, we get a total decrease of 1.4*1.4=2 (all calculations very approximate).
Unfortunately, the short answer is "no". Various techniques are used to show higher Guide Numbers than would be measured by independent testing. Here is a short list of the tricks, provided by Rui Pedro Mendes Salgueiro:
There are two types of shutters in common use in todays' cameras. Leaf shutters are typically used in point-and-shoot cameras. These are located in or near the lens, and have no serious interference problems with the flash. These shutters will not be discussed here.
Focal-plane shutters are used in most SLR (Single-Lens Reflex) cameras. They are located near the film, and work by sliding two curtains across the frame. When the exposure time is relatively long, the first curtain has time to completely remove itself from the frame before the second curtain starts to cover it up from the other edge. When the exposure time is short, both curtains move at the same time, and the full frame is not completely uncovered at any given time. You can think of this as a slit between the curtains making its way across the frame.
Additionally, many modern cameras use electronic shutter which is not a mechanical device at all. Instead, the electrical signals to/from the sensor are manipulated in such a way that a short-duration reading of the image is achieved. This is often combined with a mechanical shutter of either type to cover the sensor either before or after (or both) of the exposure.
As discussed above, the flash itself has a very short duration (1/1000 seconds or less). For proper exposure, the shutter must not block any part of the frame at the instant the flash is fired. Otherwise, part of the frame will not "see" the flash and will not be exposed. Therefore, flash should be used at shutter speeds which are long enough to include a moment where the whole frame is uncovered. The highest speed at which this is possible is called the top or highest synchronization speed of the camera. Flash photography is possible with this shutter speed and with all longer ones as well.
For some applications (such as daytime fill-in flash) it is necessary to use faster shutter speeds than the highest synchronization speed available. With many flash units, there is nothing you can do here. Some newer flash units, however, offer a high-speed synchronization mode, which works by emitting a rapid series of short flash pulses (see section 1.3 above for technical details). As the slit between the shutter curtains moves across the frame, these short pulses illuminate the scene in a rapid sequence, eventually resulting in exposing the complete frame.
The flash unit distributes its light energy to the whole scene in all cases. With normal (low-speed) synchronization, all of the light which is collected from the scene by the lens, is available to expose the film. In high-speed mode, part of the frame is blocked by the shutter at any given moment, and therefore part of the flash energy is wasted. As a consequence, flash units have lower guide numbers in high-speed mode than in normal mode.
Use of electronic shutters opens another possibility of high-speed synchronization. Many modern SLRs use this method to syncronize at 1/500 seconds or higher with a normal single-pulse flash, which would require very expensive focal-plane shutters otherwise.
When one uses long exposures to blur the motion of a moving subject, it is sometimes nice to add a flash picture of the subject into the same scene, to create a better illusion of movement. With the standard construction of shutters, the flash is usually fired as soon as the shutter gets completely opened. This results in ugly pictures where the movement trails are ahead of the flash image. For nicer looking results, its is therefore good to synchronize the flash just before the shutter is about to close. Then the movement blur is behind the subject, providing a better visual clue to the motion.
While the technology to achieve this is quite simple, it requires support from the camera body, and in cases of some manufacturers, is implemented via proprietary communication between the flash unit and the camera body, being therefore available on only selected flash units.
The delay between closing the synchronization circuit and starting of the flash is extremely small and can be ignored for most applications. Until some decades ago, there existed another kind of flash units, known as "flash bulbs" or "flash cubes", which used a chemical reaction (burning) to generate the light. These units required a fraction of a second of early warning, so that the chemicals would be burning at full rate by the time the shutter opens fully. For this reason older cameras have a special synchronization circuit which closes the contacts very slightly before the shutter opens, thus giving time for the flash cube to ignite. Such operation is called the "M" synchronization. The regular timing for electronic flash units is called the "X" synchronization. While "M" has become almost extinct by now, the term "X synchronization" is still widely used to denote the regular no-advance-warning kind of operation.
Any stand-alone (i.e. not built-in) flash unit gets at least the synchronization impulse from the camera. In old days, and even today in studio situations, a standard connector, called "PC plug" was used. (This has nothing to do with Personal Computers - the term pre-dates personal, and probably all, computers). PC connectors are becoming increasingly rare on today's cameras.
The now-standard way is to use a "hot shoe" to both mount the flash unit on the camera, and communicate information to and from it. Some manufacturers (Minolta, Olympus IS series) use a proprietary construction, which will not be discussed here. All the other shoes incorporate at least one contact, which is located in the centre of the shoe. When the camera wants to trigger the flash, this contact is shorted with the metal parts of the shoe itself. The flash unit needs to apply a voltage (see "trigger voltage", below) between the shoe itself, and this contact, and start the discharge when these contacts are connected together by the camera.
Besides that, many modern hot shoes incorporate additional contacts for communicating information such as flash ready, f-stop, TTL light metering, zoom position, etc. The position and function of these contacts differ from manufacturer to manufacturer, and the protocols are kept secret in some cases. Third-party manufacturers reverse engineer these protocols, which may lead to incompatibilities.
A flash unit which has more than just the central contact on the hot shoe needs to be manufactured for a specific camera brand, and is called dedicated flash. Most manufacturers make dedicated flash units to work with their own camera bodies. Third-party manufacturers make either specific (sub)models for different camera brands, or produce dedicated modules which can be replaced when the flash unit is used with a different brand of camera.
In general, any flash unit with a hot shoe should be usable with any camera with a hot shoe (but see "trigger voltage", below). However, if there are additional contacts on both the flash unit and the camera, it is better to avoid non-dedicated combinations. Additional (non-central) contacts on only one side (camera or flash unit) should generally not affect operation. However, the flash unit will operate at full power in these cases.
If you have the money, a dedicated flash unit from the manufacturer of the camera is usually the best bet. There will be no compatibility problems, and the various special features (zoom, exposure compensation, high-speed and second-curtain synchronization, special light metering modes) available on the body and/or the flash unit will be usable and work as expected.
If the budget is limited, you may look at the offerings from third-party manufacturers such as Metz, Vivitar, Sunpak and others. You may encounter compatibility problems, and some or all special features available on the original units may be missing. Some of the higher level offerings from these companies may have higher guide numbers or interesting accessories which are not offered by the camera manufacturer at all.
In any case, it is useful to have a bounce and swivel head on the flash unit. This means that the light could be bounced off the ceiling or wall if needed. Some units include only bounce (the head can be turned upwards) but no swivel (left and right). This may limit your bounce-flash pictures to landscape orientation.
Some more primitive flash units apply a very high voltage (200-300 volts) between the hot shoe base and the central contact. These voltages may damage the sensitive electronics in modern electronic cameras. The trigger voltage, as it is called, of newer units is lower, typically less than 10 volts. If you are getting an older (used!) or otherwise primitive model, it would be a good precaution to measure the trigger voltage before using the unit. Since the voltage can be very high, proper precautions should be taken - please read the flash troubleshooting guide before you do it. You may kill yourself otherwise!
These are variations of the TTL flash developed by Canon. They are available on Canon camera bodies and dedicated flash units. The differences can be summarized as follows:
TTL measures light reflected off the film plane during flash exposure. Once enough light has reached the film for proper exposure, the flash pulse is cut short. This method was used in film-based cameras and is not present in Canon DSLRs.
A-TTL uses a pre-flash and a light sensor on the flash unit to evaluate the proper illumination needed. This info is then combined with camera settings to produce a supposedly more uniform illumination of the scene and balancing existing light with flash. This is also an outdated mode which was used in the film era.
E-TTL (and its newer variations, sich as E-TTL II) also uses a pre-flash, but the camera's normal (multi-segment) meter is used to evaluate the results and pick a proper aperture, and balance foreground and background. This is the newest technology and is available in all the modern DSLR bodies and compatible flash units. These units usually have the letters "EX" in their name. Similar technology (involving a pre-flash) is also used in many compact cameras made both by Canon and other manufacturers.
As the camera focuses, the lenses compatible with the D system provide camera-to-subject distance to the camera body. This information is combined with light measurements made with pre-flashes after the camera's mirror had moved up, but before the shutter opens. Information from these sources is combined to possibly reduce the guide number of the flash, and to pick the proper TTL sensors for use with the scene.
Many people claim that this system is more accurate, especially for off-center subjets, than Canon's, since camera-to-subject distance is directly incorporated into the calculations of flash output power. On the other hand, Canon's E-TTL II is also claimed to include lens-to-subject distance into the calculation, so the judgement may have become outdated.
Auto focus cameras have big trouble focusing in low light (where flash is typically used). Some bodies have built-in lights (white or infrared) to add light to the scene for focusing. Some flash units have similar focusing aid lights built in. These are typically more powerful than the ones available on the camera bodies. Such flash units are sometimes called "Auto Focus" (or AF) flash units.