Why is flash sync speed much slower than flash duration?

Question

According to What is flash duration?, flash duration at full power is 1/1000 s for decent flashguns. Presumably, at low power, the duration is even shorter. I understand that film era flashguns didn't do the dual flash (once for measurement, another for exposure) but simply turned the flash off when there was enough exposure.

However, my camera (EOS RP) has 1/180 s flash sync speed.

Why is the flash sync speed much slower than flash duration? I mean, if we can control the shutter precisely enough for 1/4000 - 1/8000 s shutter speed (meaning the accuracy should probably be below 10 microseconds or else we have inconsistent exposure), why can't we control flash precisely enough for 1/1000 s shutter speed?

Clearly, the issue cannot be limited speed of light, because speed of light is 300 meters per microsecond. Microsecond is accurate enough, and nobody is using a flash 300 meters away. (Ok, the signal needs to travel twice, once to the flash via radio/infrared/cable, another time from the flash to the object via flashgun output light, but even then the limit would be 150 meters.)

Is the issue related to the slowness of igniting the arc in the flashgun? So that one can control the duration of the arc very precisely, but not when the arc actually starts?

Answer 1

Cameras that have a mechanical focal-plane shutter have two curtains, a front and a rear. For longer exposures, the front curtain opens and starts the exposure, then the rear curtain closes to end the exposure.

The mechanical shutter is relatively slow, so to create a quick exposure, the rear curtain must start closing before the front curtain fully opens. So, effectively, a slit travels across the sensor exposing each section for the desired amount of time.

The sync speed (1/180 for your camera) is the fastest speed where the sensor is fully exposed. That is, a slit is not used.

Some speedlights offer a high-speed sync option. In this mode, the flash creates a quick burst of tiny flashes so that the sensor is exposed equally as the slit travels across the sensor.

Answer 2

Flash sync speed has almost nothing to do with the duration of the flash.

Flash sync speed is mostly about the shortest shutter time that allows the flash to fire while both shutter curtains are fully open.

Most cameras with mechanical shutter curtains require anywhere from 2-4 milliseconds for the shutter curtains to transit from one side of the sensor to the other.

• 2 ms is 1/500 second
• 4 ms is 1/250 second.

For the vast majority of cameras with focal plane shutters, the curtains travel across the sensor at the same speed regardless of the selected shutter duration. What changes when the exposure time is altered is the difference between the time the first curtain begins to open to uncover the sensor and the second curtain begins to close to once again cover the sensor.

The way that a camera can take a picture with an exposure time of, say, 1/4000 second (.025 ms) when it takes 1/250 second (4 ms) for the shutter curtains to cross the sensor is that the second curtain begins to close before the first curtain has finished opening.

In our example of a 1/4000 second exposure using a camera with a 4 ms shutter transit time, the second curtain begins to close when the first curtain has only revealed 1/16 of the sensor! The second curtain then chases the first curtain across the sensor only 1/16 of the height of the sensor behind the first curtain. This means that there is no time during the exposure when more than 1/16 of the sensor is uncovered at the same time. Thus, when the flash fires for 1/1000 second, only about 1/4 of the sensor will be exposed during any of the flash burst and none of it will be exposed for the entire duration of the flash burst.

For the flash to have an equal effect on the entire frame, both shutter curtains must be fully open and the entire sensor exposed for the full duration of the flash burst. This means the shortest shutter time that can be used with the flash will be the time needed for each curtain to transit the sensor plus the length of the flash burst (plus a margin of error for coordinating the firing of the flash with the instant the first curtain is fully open).

Even with electronic shutters, CMOS sensors must be read out one line at a time. For many sensors, this readout is actually slower than the transit time of many mechanical shutter curtains. So similar limitations are in effect when using electronic first curtain or even electronic first and second curtain. For the full frame to get the effect of the flash, the flash must fire when no lines on the sensor are being read out (and thus not collecting light).

Answer 3

For focal plane shutters (most common on DSLR), the shutter is two curtains. The topic for you to research is "focal plane shutter". One curtain over the sensor moves to open the exposure, and then the other one moves to close it again. The exposure time is the delay between the two curtains.

The idea is that both curtains move in the same direction, so that all areas of the sensor receive the same exposure. As opposed to in any simpler leaf shutter, one opening and then simply closing back again the other way, so that then one side of the sensor receives more light than the other side, which can be substantially more for a fast exposure time. For focal plane shutters, the sensor area is all equally exposed, but not at the same exact time (as this open slit moves down the frame).

And then the shutter speed timing does NOT depend on any mechanical movement, but only depends on the delay of the second curtain. This delay is timed by a crystal, like any digital clock.

For a slow shutter speed, one curtain opens, standing fully open for a bit, and some time later, the other curtain closes. The flash can only be fired when both curtains are fully open (called flash sync).

For a fast shutter, one curtain begins opening, and as it is still moving and still opening, the other curtain begins closing behind it. The actual exposure then is just the width of the narrow "slit" between the two curtains (a very fast shutter time in that slit). An fast flash cannot work into this narrow slit, it would Not be in sync to expose the full opening.

In your case, the speed of your curtain travel means that 1/180 second is the fastest possible shutter time that both curtains are fully open for flash. Any faster, and the second curtain has already began closing, so that then, only a narrow "slit" would be exposed at any one time by fast flash.

Incidentally, a speedlight at full power is slower than 1/1000 second. That time is conventionally expressed as a t.5 time, meaning measured as the time between half power points, but there is much more light in the other half. A t.1 measurement (when power is above 10%) is about three times longer. But flash duration is conventionally measured to t.5 half power points. It's an engineering thing, to measure vague things which are difficult to know when it decays to exactly zero. But it is not exactly what photographers would want to know.

But, speedlight means that the flash always fires at full power, and then the output is truncated or cut off abruptly to perform any lower level, like half power or 1/4 power. This means that duration of power levels less than full power are more accurate, more similar to t.1, but the one full power is t.5. This is why specs for speedlight half power and full power timings are about the same number, but with different meanings.

But regardless, for a focal plane shutter, both curtains have to be fully open when the any flash fires. Focal plane shutter are considered the best quality shutter, but have this one down side of flash sync.