I was doing some research to see how to do a water droplet shot. Well, from what I have read I will need a speedlite. I do not own a speedlite.
So my questions:
1. Do I really need a speedlite to get a good water droplet shot with my macro and tripod?!
2. Do you have any recommendations on good ones that aren't too expensive?!
Here's the first droplet shot I took, over seven years ago(!). I constructed a reflector out of aluminium foil to bounce the light from the pop-up flash onto a piece of paper above the camera:
The "flash" or speed light more importantly gives you "action stop" in low light since the flash duration is only milliseconds, sometimes even down to a fraction of a millisecond. So it provides you much more than just "illumination." The on-camera flash can do the same thing of course. Meike is another brand of Speedlight that can work with Nikon iTTL. A Meike MK-910 is a knockoff of the top of the line Nikon SB-910 with all the features except the "aperture" set. Fully compatible with Nikon's iTTL, and works perfectly as a remote or a commander, as well as just mounted on the camera hot shoe. 1/5 the price of the Nikon. They have worked well for me in all sorts of situations.
@icamera If you check the Exif for any of the shots above, all probably register something like 1/200 of a second, the typical flash synch speed of a digital camera. It's not a speed that could stop such close up motion under gravity. What else but the flash pulse at the correct "full power" setting?
@frankhymus Agreed, although technically, the type of flash determines how you get the shortest pulse duration.
Most monolights (studio flashes) modulate the voltage going to the flash bulb to alter the brightness, which means that you get the shortest flash duration when the flash is at full power, and a longer duration as the voltage (and brightness) of the flash is decreased. For example:
Flash Duration t 0.5: 1/2600 (max power),1/1000 s (min power)
A battery powered hotshoe flash (Speedlite) on the other hand behaves in exactly the opposite manner, because the brighter power output from these flashes is caused by a longer pulse from the bulb. As the flash power is decreased, the pulse length is cut, reducing the overall light output and hence also the flash duration. From the Nikon SB800 manual, page 122:
1/1050 sec. at M1/1 (full) output
1/1100 sec. at M1/2 output
1/2700 sec. at M1/4 output
1/5900 sec. at M1/8 output
1/10900 sec. at M1/16 output
1/17800 sec. at M1/32 output
1/32300 sec. at M1/64 output
1/41600 sec. at M1/128 output
The difference is (I suspect) pretty much irrelevant for water droplet photography, which at typical falling speed will be frozen by a flash duration of around 1/1000th (the slowest from either option), but the difference can be important if you are doing really high-speed flash work, such as capturing a bullet in motion. In these cases, speedlights are very popular for this type of photography because of their incredibly short pulse duration at low power settings.
Edit: Some back-of-napkin calculations, ignoring drag on the droplet.
Assuming a 15cm drop height, a water droplet would be travelling at 1.7m/s as it passes the camera lens. That's a distance of 8.5mm in 1/200th, or 1.7mm in 1/1000th.
Estimate the diameter of a droplet of water as 5mm. (That's on the large side, but most photographers will use relatively crude methods to generate the drop, so probably fairly reasonable). The droplet in my example photo above is (very roughly) 1/12th the height of the camera frame, so a total frame height of 60mm. Take a reasonable-resolution DSLR with 18ish megapixels, and that's roughly 5000 pixels from the top of the frame to the bottom (in portrait orientation). 5000 pixels in 60mm is 83 pixels per millimeter, so if the droplet travels 1.7mm in 1/1000th of a second, that actually results in around 140 pixels of motion blur!
Drag would certainly slow it down, although this document suggests a 5mm rain droplet has a terminal velocity of around 9m/s, so I guess drag won't be a huge component by the time it's reached 1.7m/s...
So although my gut feeling was that 1/200th would be too slow and 1/1000th would be fast enough, I suspect in actual fact you could do with a considerably shorter flash duration for optimal results, especially if you are getting very close to the action -- it looks like you might need at flash duration of at least 1/10000th to get a respectably sharp shot. Interesting!
Wow, I see I've been introduced to the resident experts. I've removed my responses as they didn't stand up to splitting hairs of sync speeds and lighting conditions, nor to the calculations in regards to estimated shutter speeds and flash durations. Bottom line, all I meant to do was answer the OP's original question and say "No, you do not *need* a Speedlite (or popup flash for that matter) to make a water droplet image". Which is 100% correct. Of course without Speedlite (or popup flash), then you will need appropriate lighting conditions (like a nice sunny day!) to achieve adequate shutter speed for freezing the motion of the water droplet. And to also say that flash, in and of itself without consideration toward any additional factors (lighting conditions, sync speeds, angle of motion to sensor plane, etc), is not the sole answer to freezing motion.
@abirkill I did this boe calculation about 18 months ago and came to similar conclusions, 1/8-1/16 power max (slowest flash pulse) for respectable minimum motion blur with the top of the line Nikon Speedlight and the 24mp (APS-C) D7100.
Tweets ahoy!
Join the Group
Pin With Me
Most monolights (studio flashes) modulate the voltage going to the flash bulb to alter the brightness, which means that you get the shortest flash duration when the flash is at full power, and a longer duration as the voltage (and brightness) of the flash is decreased. For example:
http://profoto.com/int/products/monolights-kits-associated-accessories/d1-monolights/item/d1-air-500#technical-specification
Flash Duration t 0.5: 1/2600 (max power),1/1000 s (min power)
A battery powered hotshoe flash (Speedlite) on the other hand behaves in exactly the opposite manner, because the brighter power output from these flashes is caused by a longer pulse from the bulb. As the flash power is decreased, the pulse length is cut, reducing the overall light output and hence also the flash duration. From the Nikon SB800 manual, page 122:
http://cdn-10.nikon-cdn.com/pdf/manuals/Speedlights/SB-800.pdf
1/1050 sec. at M1/1 (full) output
1/1100 sec. at M1/2 output
1/2700 sec. at M1/4 output
1/5900 sec. at M1/8 output
1/10900 sec. at M1/16 output
1/17800 sec. at M1/32 output
1/32300 sec. at M1/64 output
1/41600 sec. at M1/128 output
The difference is (I suspect) pretty much irrelevant for water droplet photography, which at typical falling speed will be frozen by a flash duration of around 1/1000th (the slowest from either option), but the difference can be important if you are doing really high-speed flash work, such as capturing a bullet in motion. In these cases, speedlights are very popular for this type of photography because of their incredibly short pulse duration at low power settings.
Edit: Some back-of-napkin calculations, ignoring drag on the droplet.
Assuming a 15cm drop height, a water droplet would be travelling at 1.7m/s as it passes the camera lens. That's a distance of 8.5mm in 1/200th, or 1.7mm in 1/1000th.
Estimate the diameter of a droplet of water as 5mm. (That's on the large side, but most photographers will use relatively crude methods to generate the drop, so probably fairly reasonable). The droplet in my example photo above is (very roughly) 1/12th the height of the camera frame, so a total frame height of 60mm. Take a reasonable-resolution DSLR with 18ish megapixels, and that's roughly 5000 pixels from the top of the frame to the bottom (in portrait orientation). 5000 pixels in 60mm is 83 pixels per millimeter, so if the droplet travels 1.7mm in 1/1000th of a second, that actually results in around 140 pixels of motion blur!
Drag would certainly slow it down, although this document suggests a 5mm rain droplet has a terminal velocity of around 9m/s, so I guess drag won't be a huge component by the time it's reached 1.7m/s...
So although my gut feeling was that 1/200th would be too slow and 1/1000th would be fast enough, I suspect in actual fact you could do with a considerably shorter flash duration for optimal results, especially if you are getting very close to the action -- it looks like you might need at flash duration of at least 1/10000th to get a respectably sharp shot. Interesting!
Hope this helps.