This post was originally published on my www.photografica.com.au site.
I started to look into the various possibilities of high speed flash, and what I thought was a relatively simple area of photography quickly expanded and my investigations seemed to find no end. So I thought I should blog my findings here, as I was a bit frustrated about the lack of resources on the subject myself.
So what is high speed flash? Well, it’s one way to make high speed photography. I make the distinction to separate the effects of shutter speed from the effects of flash. Both can freeze motion. What I’m interested in here is the role flash plays in freezing motion, and looking at the various types of equipment that might be used in high speed flash projects. This post may well be updated along the way, as it’s a pretty big topic…
(all of the shots above taken with speedlights at low power – click to view larger)
Before we get started looking at specific equipment, it is prudent to quickly review the terminology that is used to measure flash duration. It’s a bit technical, but not actually that hard to understand. There are in fact two terms used, the t.5 time and the t.1 time. These are terms that describe the time the flash takes to respectively discharge 50 and 90% of its power. Or from the other side, the time it takes for there to be 50% or 10% left (that’s where the 5 and 1 come from). But instead of re-inventing the wheel here, I will link to Strobist’s excellent post on the subject. Go and read that and come back. Go on! The link opens in a new window. There is also another excellent explanation of flash duration at the Paul C Buff website. Go and read that as well! (I’m a patient guy and my blog post isn’t going anywhere…)
OK, so now we have a basic understanding of t.1 and t.5 times. If you read thoroughly, you will know that t.1 times are a more accurate way to assess flash duration than t.5 time, as at 50% discharge it’s reasonable to expect that the remaining 50% will still have some effect on exposure. Not so with the t.1 time. One issue however is that the industry tends to use the t.5 time as its “standard”, as that will obviously make the flash duration appear much faster. Hats off to Alien Bees who quote both figures in their manuals.
So let’s start at the simple end of things. And the joy here is to discover that these are the fastest flashes of all, bar some highly specialised, highly expensive flashes that we will look at later. But for all intents and purposes, small flashes are the fastest commonly available flashes out there. Some manufacturers state the flash durations in the flash manual, and some don’t (boo Canon), however they often don’t mention whether it’d the t.5 time or the t.1 time they are using (in the absence of that, I’d always assume it’s the t.5 time they are quoting).
Looking at the Nikon SB900 manual for example, they quote the following figures:
1/880 sec. at M 1/1 (full) output
1/1100 sec. at M 1/2 output
1/2550 sec. at M 1/4 output
1/5000 sec. at M 1/8 output
1/10000 sec. at M 1/16 output
1/20000 sec. at M 1/32 output
1/35700 sec. at M 1/64 output
1/38500 sec. at M 1/128 output
There are a few interesting observations to be had here. One is that regardless of hard and fast numbers, you can see that at low power, the flash duration is very, very short. The other thing of interest is to note that the flash duration gets SHORTER as the flash power gets LOWER. So if you want to freeze some water drops, using an SB900 at 1/8th output or lower would be a great place to start.
Canon don’t seem to list the flash duration in their manuals, and from my reading the hard and fast figures official figures are simply not available. But there are many people who have tested these flashes for speed, and they are equally as fast as the SB900. A 580EXII for example at its lowest power setting (1/128th) has been measured at a staggeringly fast 1/35,000th of a second… See this post from The Rod and The Cone for all the nerdy detail.
So without hard and fast figures for every flash you use, and knowing if they are the t1 or the t5 figures, all you can really know is an estimated speed, and what settings to use to minimise the duration of the flash. What is clear is that if speed is a necessary requirement, it would be best to use 1/8th power or below. A speedlight at full power might be 1/500th of a second or even slower. And that’s not getting into any ambient light contribution to a photograph. You clearly have to control your ambient. But here, let’s keep to the flash durations.
This is where it gets interesting. You would be totally forgiven for thinking that an Alien Bee 1600 studio head at full power would be a much better light to use than a small 580EX or SB900. And you’d be right – IF you were trying to overpower the sun for example. But as far as flash duration goes, the Alien Bee 1600 at full power is a sluggish old toad compared to the smaller speedlights. At full power, you are only getting a speed comparable to a speedlight at full power. The reason for this is found in the technology of the lights. A speedlight uses a thyristor circuit to abruptly cut off the power when the flash has discharged the correct amount. This means the back slope of a speedlight’s output is steep (the light dies off very quickly).
With most studio heads however, the primary method of flash power control is by varying the voltage to which the flash capacitors are charged. But this method means the flash duration gets longer as the power is reduced. With higher power heads there might be two capacitors, which will make the duration even longer. So in short, the fastest studio heads are the lower WS heads (such as the Alien Bee 400) at their full power setting. The Alien Bee 400 for example claims to give a flash duration of 1/2000th of a second (t1) at full power. The 1600 on the other hand is a much slower 1/600th of a second.
The need for speed
So why would you need speed from a flash? To freeze movement in a studio. When you are using flash, your shutter speed is limited to the maximum sync speed your camera can use with flash. Typically, that’s 1/200th or 1.250th of a second. That’s not very fast to freeze a milk drop or a balloon bursting. Also in the studio, ambient light is controllable. In low ambient conditions, the shutter at 1/200th doesn’t let much light in at all, and so we can just use the short flash duration to freeze movement. Water drops, milk drops, dropping products into water, splashes, powder play, balloons.. there are no end of fun things you can do in a dark-ish room with a speedlight.
To whet your appetite, here’s a video that shows just what sort of stuff is possible with high speed flash. Jaroslav uses Einstein lights, which are Alien bees big brother, and capable of very short flash durations.
I’d like to hear about your suggestions for high speed flash, just using speedlights. What sort of photography can you see it enabling? What would look cool caught at 1/20,000th of a second as opposed to just 1/500th? Hit me in the comments.