High Speed Sync

[Golem]

@Golem I think I might have figured out the confusion, its in the method and technology used in HSS. On all focal plane shutters (film or digital) there is an instant when the whole sensor (or flim) is exposed to the light source. At speeds above the sync speed there is a slit that travels across the sensor, as the back plane follows the front, such that the whole sensor/film gets exposed equally. In earlier film cameras with leaf or petal shutters there was no moving slit, so sync speed wasn't an issue. However, they had a max speed limitation of around 1/500.

With focal plane shutters I haven't tried it but I'm assuming that a continuous light would expose a sensor or film at speeds above sync, since it's basically an ambient light shot. However it is NOT HSS because there is no synchronization between the light and camera that involves multiple flashes going on.

A speed light in HSS mode is "synched" only in the sense that illumination begins as the shutter slit begins to travel and ends at the finish of travel. This conserves precious energy, necessary since HSS is a relatively weak light source. All during the slits travel the flash is a high frequency pulsed light souce, similar to household LED but at an enormously higher frequency.

Theres only a limited amount of energy in the capacitor, and we dont want to waste any of it by lighting up before the slit travels. But we DO want to use up ALL that energy during slit travel, again to wring the most light out of a rather feeble source.

IF the power source were "endless" and IF the xenon tube were very very much more rugged, then "HHS" could just stay lit like a street lamp or inert gas auto headlamp, with no need to communicate with the camera ... just an off-switch like on an LED panel.

 

[smoke665]

A speed light in HSS mode is "synched" only in the sense that illumination begins as the shutter slit begins to travel and ends at the finish of travel.

You're still missing part of the process, to understand, requires understanding of how a focal plane shutter works. There are two curtains Front and Back. When you release the shutter the first curtain drops, at the predetermined time the back curtain starts it's descent to close the shutter. At speeds "up to" the camera max sync speed there is a period of time when the front curtain has reached the bottom of it's travel and the back curtain has not started to travel. The slower the shutter speed setting, the longer that period of time. When you exceed the sync speed, the back curtain starts to travel, before the front curtain has completely opened, creating a moving slit. As you increase the shutter speed, the moment that back curtain starts to fall comes faster, resulting in a progressively smaller slit (because the front curtain hasn't reached it's full travel). The time it takes that slit to move across the sensor is equal to the max sync speed of the camera. If you were to fire a single burst from a light, the back curtain would cause a shadow to be cast on the sensor. HSS gets around that by turning the flash into a strobe firing multiple times, to coincide with the travel speed of the slit.

This conserves precious energy, necessary since HSS is a relatively weak light source

Yes and no, HSS sync uses the power stored in the capacitor, the same as a single flash. In HSS the flash is still firing at it's setting same as if it were a single flash, but because above sync speed, only the open area of the slit is being exposed to the light, it sometimes requires that you increase the flash power to achieve the proper exposure, that doesn't mean it's a weaker light source, just a different exposure time.

a high frequency pulsed light souce, similar to household LED but at an enormously higher frequency.

Again a misunderstaning of how they work now. Early LEDs powered by AC current flickered because of the alternating cycle. Modern LEDs use a transformer that converts the AC to DC power effectively eliminating the flicker due to cycle.

 

[wfooshee]

He's saying now that the multiple bursts from the flash are "effectively" continuous light. While that may be true, as in "effectively," it is NOT a continuous burst of light. No flash is capable of firing a single burst that lasts 1/250th of a second, so it's NOT continuous light.

LED headlights in cars are not continuous light, either. They flicker, as you can see when watching video of cars at night, as some video frames catch the headlights off and some catch them on. The human eye can't tell, though.

Regarding the possibility that I've misunderstood the process in calling the pulses synced to the shutter travel. That actually was my understanding from reading (and maybe misunderstanding) Nikon's description of HSS in their manuals, but examining the timing requirement, to get HSS to actually sync with a shutter speed of 1/8000 of a second, you'd need 32 pulses, one every 1/8000th of a second. That does seem potentially unreasonable, but something happening at an 8kHz frequency is well within the realm of any electronic timing system that exists today, or that has existed for quite some time. I just don't know if the flash is capable of that, physically. I still fail to see how the subject is lit evenly at the speeds we're talking about if the flash pulses randomly, as opposed to synchronously with the shutter travel; there will be milliseconds without light, it seems to me. It obviously works, though, as I've used HSS on many occasions, even with multiple off-camera speedlights.

So maybe the flash isn't synched to the shutter travel as I was thinking, but I also don't see how it's evenly lit with basically random repeated pulses of light. Maybe longer pulses, still synced like 8 pulses at 1 kHz...

 

[smoke665]

I still fail to see how the subject is lit evenly at the speeds we're talking about if the flash pulses randomly, as opposed to synchronously with the shutter travel; there will be milliseconds without light, it seems to me. It obviously works, though, as I've used HSS on many occasions, even with multiple off-camera speedlights.

So maybe the flash isn't synched to the shutter travel as I was thinking, but I also don't see how it's evenly lit with basically random repeated pulses of light. Maybe longer pulses, still synced like 8 pulses at 1 kHz...

As I said earlier understanding how HSS works, requires an understanding of how a focal plane shutter works. From my post above, there are two curtains Front and Back. When you release the shutter the first curtain opens, at the predetermined time the back curtain starts it's movement to close the shutter. At speeds at or below the camera max sync speed (1/200 on my camera), there is a period of time when the front curtain has reached the bottom of it's travel and the back curtain has not started to close yet exposing the full sensor. The slower the shutter speed setting, the longer it takes for the back curtain to start which results in more exposure on the whole sensor. It's my understanding that the "time" it takes a curtain to complete it's full travel is equal to the sync speed of the camera, which results in an equal exposure across the sensor.

Conversely, when the shutter speed goes above camera sync speed, the back curtain actually starts to move, before the front curtain has completely opened, resulting in a moving slit. As you increase the shutter speed, the moment that back curtain starts to fall comes faster, resulting in a progressively smaller slit. With the shutter travel speed is equal to the sync speed, the movement of that slit is also equal to max sync speed, providing equal exposure across the sensor.

A typical speed light duration at full power is around 1/750 of second, if you only have a single burst of light, it's still faster than the movement of the slit between the front curtain and the back curtain and results in a shadow being cast on the image. HSS doesn't sync the pulses with the movement of the slit, it pulses the light rapidly much like a strobe light, for a duration equal to the max camera sync speed. This is why I said earlier, that the action stopping ability of HSS isn't as good as a single flash. It's my understanding the power setting of the flash determines the duration of an individual pulse which could be anywhere from roughly 1/750 at full power to 1/3000 at 1/64 power, while the communication between flash and camera determine the number of pulses required to equal sync speed.

You're right on Car LED's even though they use DC voltage, the amount of voltage in a DC system can vary. To counteract that they use Pulse Width Modulation, to pulse the light on/off as an efficient way to control the voltage and brightness of the light.

 

[Golem]

He's saying now that the multiple bursts from the flash are "effectively" continuous light. While that may be true, as in "effectively," it is NOT a continuous burst of light. No flash is capable of firing a single burst that lasts 1/250th of a second, so it's NOT continuous light.

LED headlights in cars are not continuous light, either. They flicker, as you can see when watching video of cars at night, as some video frames catch the headlights off and some catch them on. The human eye can't tell, though.

Regarding the possibility that I've misunderstood the process in calling the pulses synced to the shutter travel. That actually was my understanding from reading (and maybe misunderstanding) Nikon's description of HSS in their manuals, but examining the timing requirement, to get HSS to actually sync with a shutter speed of 1/8000 of a second, you'd need 32 pulses, one every 1/8000th of a second. That does seem potentially unreasonable, but something happening at an 8kHz frequency is well within the realm of any electronic timing system that exists today, or that has existed for quite some time. I just don't know if the flash is capable of that, physically. I still fail to see how the subject is lit evenly at the speeds we're talking about if the flash pulses randomly, as opposed to synchronously with the shutter travel; there will be milliseconds without light, it seems to me. It obviously works, though, as I've used HSS on many occasions, even with multiple off-camera speedlights.

So maybe the flash isn't synched to the shutter travel as I was thinking, but I also don't see how it's evenly lit with basically random repeated pulses of light. Maybe longer pulses, still synced like 8 pulses at 1 kHz...

You seem to be supposing that the output of each ignition of the xenon tube is a square wave. If that were true there would be the dark gaps that you mention. Fortunately its not a square wave.