Paul C. Buff Einsteins For High Speed Flash Applications

One of the really wonderful aspects about Einsteins is their flexibility. They have two modes: Action and Color. For most situations, the difference between these two modes is negligible; however it becomes significant in high speed flash applications.

Background:

Simplistically, an exposure is made by photons incident upon a digital sensor. The more photons, the lighter the image. An average or  ‘correct’ exposure is when enough photons have been collected by the sensor to create a balanced image. An analogy might be that a correct exposure can be represented by a half full bucket of water. If the bucket is less than half full, it is underexposed, and if it is greater than half full, it is overexposed. It does not matter how long the bucket took to get to half full. When it is half full, we get a correct exposure. We might use a straw to fill the bucket, in which case you would take a long time to get to the correct exposure, or we may use a fire hose to fill it – in which case it would take only an instant. (The corresponding images would be blurry or sharp, but if the bucket was half full, they would be correctly exposed). Keep this in mind as we move on…

The following graph represents the light pulse output from a flash over time. The area under the curve can be regarded (sort of) as the number of photons emitted by the flash tube. The vertical axis is Energy, and the horizontal axis is time.

In ‘Action’ mode, changes in output power are reflected by the flash duration. For a half power flash compared to a full power flash, the duration is approximately halved.  For an eighth power (or -3 stops from Full), the flash duration is approximately 1/8 of the full duration.

Flash Duration:

Flash duration is usually regarded as the time that a flash tube is illuminated, however it is often quoted with two different parameters: T0.5 and T0.1.. These parameters can lead to confusion and are often exploited for the purposes of marketing. (How do I know? I have done it myself, but that is the subject of a different article.) T0.5 is described as the time between a flash pulse reaching 50% of its maximum value and decaying to 50% of its value, as shown here:

The T0.1 parameter is the time between the pulse reaching 10% of its maximum and decaying down to 10% of its maximum, as shown here:

In the above examples, the T0.5 duration is much shorter than the T0.1 duration, hence this is the value most used in marketing documentation for flash duration. This is really important for most traditional studio flashes, where the flash duration doesn’t change radically with a reduction in flash output. That is because of the circuitry used to control the energy into studio flashes. Let me explain…. A flash circuit works very simply – you fill a capacitor with charge, and then that charge is ‘emptied’ into the flash tube. To get full power, you fill the capacitor with charge. To get half power, you only half fill the capacitor. When the flash is fired, the capacitor is emptied. One of the benefits of this approach is that the Color Temperature of the flash output remains relatively constant, throughout the range of output level. This mode is available in the Einsteins and is called “Color Mode” . The output pulses will look something like the red traces in the following plots:

Einstein Color Mode:

In ‘Color’ mode, you can see that the T0.1 and T0.5 durations do not scale with the level of output. For a 1/8th output level, the T0.5 is probably still half of the value that it is for the full output level.  It is the amplitude of the light pulse that is radically reduced. But the thing to note is that the Color Temperature of the flash output remains very stable, regardless of the output level setting. Note also that the Einsteins can go all the way down to a 9 stop reduction, or 1/256th output power level.

Einstein Action Mode:

Now comes the Einstein ‘Action’ mode – which is much more akin to how a Speedlite operates. (A Speedlite is a portable hotshoe flash). As explained in a previous post the Einsteins incorporate IGBT technology, enabling the flash tube to be ‘quenched’ or switched off at any time. In other words, the energy capacitor is filled with charge, and the flash tube is triggered. During the flash, an IGBT is switched at a predetermined duration, and the flash circuit switches off immediately. This results in the flash duration only being as long as the IGBT is ON, and the energy capacitor is only drained by the amount the flash actually uses. At very low power, or short duration, the flash uses very little of the charge stored in the capacitor. This means that the capacitor is refilled with charge (recycled) very quickly. We have all seen it – set a Speedlite to full power and it will take several seconds to recharge, but if you dial it down to 1/16th power or less, it will recycle as fast as you can fire it. The other benefit of this approach is that the flash duration reduces (simplistically, both T0.1 and T0.5) almost linearly with the reduction in output level. A full power flash may have a duration of 1/1000th of a second (1 millisecond or 1000 microseconds) and a 1/128th output may produce a flash duration as short as 1/30,000th of a second, or around 33 microseconds. This is totally dependent upon the style and design of the Speedlite. As mentioned previously, Speedlites incorporate SCR’s or Thyristors for their flash switches. The Einsteins use similar circuitry, but at much higher voltages and currents, hence employ IGBT’s for the same function. In Action Mode, the Einstein’s flash output pulse might look like the red traces in these plots:

At the half power setting:

At the 1/8th Power Setting:

The difference between the two different modes are simply that in Action Mode, the flash duration tends to scale with the output level. In Color Mode, the duration does not tend to scale with the output level. BUT, in Action Mode, the Color Temperature MAY vary between output levels, and in Color Mode, it will remain constant. I have not noticed any shift in Color Temperature in either mode, but I have not measured it either. I happen to use Action Mode in my studio when doing fashion and portraiture, and have seen no ill-effects as a result, but your own mileage may vary. If Color Temperature consistency is of crucial importance to you between shots, then you have the option. That brings us to round up Action Mode….

“So What?” you may well ask……

Well, both Einsteins and Speedlites can produce short flash durations, but for a given duration, the output of the Einstein is significantly higher than that of a Speedlite. This in turn means that for a given exposure, the flash-to-subject distance with the Einstein can be greater, and/or you can stop your aperture down further to achieve greater depth of field. Secondly, the full gamut of light modifiers available to Einsteins can be employed to improve the quality of the light.

“So What?” you may well ask……

There are many photographic applications that require short duration, or high speed flash (HSF). Much of my work exploits this genre – water drops, hummingbirds, ballistics and arty stuff. Again, a lot of fodder for future posts, but let’s take hummingbirds as a good example. I am pretty well known for my images of hummingbirds in flight. Without going through all the physics, in order to ‘freeze’ a hummingbird wingtip during mid-flap (when the wingtip is moving at its highest velocity), a flash duration of around 1/15,000th of a second is required. In order to get a complete hummingbird in focus, significant depth of field is required. The more light I can get at a given duration, the more depth of field I can dial in. Hence the Einsteins….

Above is a female Allen’s Hummingbird shot with Einstein flash units.

This image used 3 Einsteins fired a couple of milliseconds apart to produce this stroboscopic effect.

8 Responses to “Paul C. Buff Einsteins For High Speed Flash Applications”

  1. Hi Roy, This is a great article I have been asking myself these questions and even called the factory and they could not answer these questions about shooting Hummingbirds.
    How many of the Einsteins are you using? Currently I am using 8 speedlights to get enough light on the subject and background for a good exposure with smaller aperture usually f/16 or smaller. Do you think three einsteins would be enough…one left, right and one on the back ground?

    Also I need a faster cycle time, I have already fried a few flashed hooked up to my Quantum battery they just overhead in quick firing modes. Will the Einsteins fire in quick succession like 10 frames a second?

    Thanks, Ray

    • admin says:

      Hi Ray – Thanks for your message. Dependent upon the situation, I use between one and five Einsteins for my setups, depending upon how truly natural I want the lighting effect.
      One of the HUGE benefits of the Einsteins is that you can ‘machine-gun’ them with a motor-drive. I have a Canon 1DX that fires at 12fps, and the Einsteins will keep up for at least 4 shots (usually a lot more) when the power is dialed down to 1/16th or 1/32nd. There is plenty enough light, and that is sufficiently quick to freeze most of the wings.

      I have burned out my fair share of Speedlites when hooked to Quantum Turbos too. Fortunately, Canon have a flat-rate for repairing flashes – I have used that many times over the years!

      Cheers,
      Roy.

  2. Bk says:

    Nice write-up, Roy.

    Could you explain how to set up the staggered flash for the strobe effect? And is it possible to vary the distance between the flash pops. I would like to use it for a jumping dog which is quite a bit slower than your hummers. Thanks so much, Bill

  3. admin says:

    Hi Bill,

    The hummer strobe shot was created some time ago with 3 heads and several PocketWizard MultiMax’s. I have long since dispensed with the PocketWizards in favor of the new Cognisys-Inc StopShot Studio camera controller. This is a 12 output box-of-tricks that enables me to hook up to 12 devices (flashes) and program them independently, sequentially or even loop them. I added a post to my Google+ page with an example https://plus.google.com/111933408744001632656/posts/U8e5akrYFUz The interface software to the StopShot Studio resides on a PC or Mac, and is really simple to use. Hope this helps, and don’t hesitate to ask more questions!

  4. rob says:

    This is the first time I’ve heard of the Einstein flashes. The info above is very technical. What is the fastest speed and smallest aperture combination.
    I’m currently using an old Olsen setup with 1/20,000 with the len’s smallest aperture.

    • admin says:

      Einsteins are capable of 10uS pulses if the parameters are set correctly (1/100,000th of a second) but that is overkill for hummers. I use them usually at about 1/64th power which produces a flash duration sufficiently short to freeze a hummer wingtip at midflap, when it is traveling at its highest velocity.

      Smallest speed/aperture combination is heavily dependent upon a number of factors: ISO, flash to subject distance, number of flashes, reflector shape, area of coverage required, diffusion or light shaping.

      I am usually shooting ISO 100, f16 with a very convenient flash to subject distance.

      Hope this helps
      Roy.

  5. James Bollinger says:

    I am excited about using these flashes for high speed water drop photography, however, the shortest flash duration I have seen for the Einsteins is just over 1/13,000 of a second. You mentioned that they are capable of 1/100,000 of a second if set correctly. How would one do this? Thanks.

Leave a Reply