Macro the 1/f rule and flash duration

Started Sep 8, 2008 | Discussions thread
OP fred vachss Senior Member • Posts: 1,292
Mea Culpa

JimH wrote:

Are you sure that you're not just seeing the pre-flash that is
emitted when shooting in ETTL mode? Avoiding that is the main reason
that I did my tests using the manual mode of the flash.

What can I say but whoops.

Looking at Jim's results I repeated some of my tests using a lower noise detector and being careful to recognize which pulse was which. Well, in ETTL mode with the MR14 and 450D I got pulse trains that looked like this:

where each of the first two pulses zoomed in are quite broad - with durations suspiciously similar to those I described in my initial post starting this thread.

Here's what the first pulse in the train looked like (right around 400-500 usec duration):

But here's the 3rd pulse (the actual flash exposure pulse):

where the duration is a much shorter 20 usec (1/50,000 sec) - much more in line with Jim's measurements. This pulse train was generated at maximum exposure (f1.4, ISO1600, a few inches distance from flash to target and represents the true minimum duration I could get from the MR-14). Pretty clearly these first two pulses in the train (and the 350-500 usec duration flash pulses I showed in my initial post) were in fact the ETTL pre-flash pulses not the true exposure flash.

I guess my initial detector was noisy enough that it was a bit too easy to miss the correct pulse and grab the more obvious (and wrong) one. Well, embarrassing though it is, I don't want to perpetuate misinformation I've generated so the following is some corrected data on the MR-14.

I took an abbreviated set of shots using the MPE at f/2.8, ISO 1600 at 1:1 magnification, f/2.8 ISO 1600 at 5:1, f/16 1SO 1600 at 5:1 and f/16 ISO 100 at 5:1. While not inclusive this does cover about the full range of exposures one is likely to encounter with the MPE. In all cases the photographs were taken with both flash tubes firing in a 1:1 ratio and the flash pointing right at the center of the detector - yielding images like:

The flash duration results are as follows:

f/2.8 ISO 1600 1:1 ( 20 usec duration)

f/2.8 ISO 1600 5:1 ( 20 usec duration - reduced effective aperture but shorter range, apparently about a wash).

f/16 ISO 1600 5:1 ( 50 usec duration, but with a broad pedestal about 100-120 usec wide)

and finally f/16 ISO 100 5:1 (about 1.2 msec by any reasonable definition, but only about 40 usec if you use the FWHM definition of peak width)

So it looks like my conclusion was wrong (at least partially). The MR-14 operated in ETTL mode apparently can deliver flash pulses short enough to freeze motion even using the modified 1/f rule at high magnification. The question really is how fast is fast enough. The 1/f rule suggests that even at 5:1 a duration of 1/2000 - 1/5000 or so ought to be fast enough and these speeds can be obtained by bumping the ISO up a bit (and refraining from using f/16 which is pretty useless at such high magnifications in any case).

What I should try next is revisit my post on diffraction limits and show what the flash duration is for 1:1 at f/16, 3:1 at f/8 and 5:1 at f/5.6 (at various ISOs). These are the values at which I observed blurring that I contended was due to diffraction, but there was some discussion as to whether some of this may have been motion blur due to overlong flash duration. When I get the numbers maybe we can put that discussion to rest at last.

Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow