I took most of these high speed photos using a Nikon Coolpix 950 (some with a Canon D60) and a Vivitar 285 flash. The pictures are taken in a dark room. I open the shutter for a long time (1 second), then I trigger the flash when I want the picture taken. The trick, of course, is to get the flash to trigger when you want it to.

Pretty much, you need to electronically trigger the flash. For catching a balloon in mid-pop, you need to be able to control the timing of the flash to within a single millisecond, if not a tenth of a millisecond. The Vivitar 285 has a PC cord connection which I cobbled up to interface to a standard phono jack, which makes a reasonably robust and easy to handle connector (as long as you don't get confused and plug them into your stereo). All that needs to happen to trigger the flash is to short the two wires together. Almost anything will do.

(n.b., Flash units use 300V internally, quite enough to give you a memorable shock. Apparently, some flashes out there expose this voltage externally on their wires. All the ones I played with seem to only have 5-7V external, which makes the circuitry a bit easier and safer to deal with. Full info at www.botzilla.com/photo/strobeVolts.html)

Sound Trigger

After a poor attempt at using a mechanical trigger, the web provided me with some better solutions. For taking pictures of sound-producing events, like popping balloons, I built a sound trigger using a microphone and a cassette deck's headphone jack. I used a silicon controlled rectifier (SCR) as per www.hiviz.com. (I ended up using an NTE5416 SCR, as it was locally available.) The result is a trivial-looking cable.

(Disclaimer: build this schematic or others on this site at your own risk. God knows what might happen. Good luck.)

This works very well. You can adjust the timing of the flash simply by moving the microphone closer to or farther from the balloon (or the light bulb you are going to smash, or...) By placing the microphone in contact with the balloon, I got a picture of the first contact of the needle.  By moving the mike across the room, I got it after the balloon is essentially gone (10 milliseconds or so later, since 1 foot = approx. 1 millisecond for sound through air). You can also adjust the sensitivity of the trigger using the headphone and/or microphone controls on the tape deck. I ended up building myself a small battery-powered amplifier to avoid dragging the tape deck around. Afterwards I saw Radio Shack had one for $12. Silly me.

One issue that arises with water balloons is multiple exposures. One when the balloon pops, another when the water hits the bucket (or floor) below. Also, whacking a light bulb with a hammer can get you a few flashes (producing kind of a neat effect, actually). Under some circumstances, you can adjust the sound sensitivity to avoid this, but in general, you'll want a more reliable mechanism. More about this in a minute.

Photogate

Taking pictures of drops poses a different set of issues. Using a microphone is problematic since the drop isn't very loud. (I did manage this by using a tiny electret mic protected by a balloon, submerged in a bowl of milk. It was sensitive enough to be triggered a single drop hitting the surface.)

I built a photogate from a phototransistor and simple bipolar transistor amplifier. The phototransistor is illuminated by either an infrared LED or a low power HeNe laser (from a pocket laser pointer; available from Radio Shack).

Using the infrared LED limits me to an LED-phototransistor separation of about 3 inches. Plenty for many applications. but restrictive in others. Using the laser gives me a essentially unlimited separation, but the thing is so powerful that completely breaking the beam is a chore for a little old drop. A pin-hole aperture made from some thin aluminum creates a beam thin enough that it can be effectively disrupted by a single thread.

Then there's the problem of the delay. Drops are pretty slow compared to sound. And moving a microphone down the street for a delay while still having it hear a drop in your basement just won't do.

So, I built a flash control box that contained this photogate, an adjustable one-shot lock-out timer to avoid multiple exposures, and an adjustable time delay (both timers based on the venerable LM555). It triggers based on the internal photogate or an external trigger, such as a sound trigger. Here is a complete schematic.

For taking pictures of drops, the photogate is mounted a couple of feet above the target area (e.g., bowl of milk). The time delay unit is then adjusted to capture the drop at whatever stage of splashing I so desire.

I found that my time delay unit did not have the kind of repeatable accuracy that I wanted. When trying to take a sequence of pictures to create a video clip of a drop splashing, I really want the time delays to be very accurately spaced, so the video proceeds smoothly. Also, trying to capture the special moment when the drop just breaks the surface is difficult if you can't reliably control the time delay to within a tenth of a millisecond. I couldn't with my circuit.

Damn Computer

I finally decided that the way to get accurate timing was to use a crystal oscillator. But I didn't want to put the time and effort into creating another (much more complicated) custom circuit. So, I begrudgingly decided to use a computer. A PC's parallel port, to be exact. There is a bunch of information on the web about using the damn things for all kinds of data acquisition and control. I also thought that once I got used to the idea and learned some about it, I could use it to control stepper motors (which solves other, unrelated problems). I really wanted to avoid dragging a whole computer around with me, especially since I already had a tidy little, battery-powered box that almost did what I wanted. So be it.

I built a box to hook onto a parallel port. It allows me to control two stepper motors, four arbitrary outputs, and read four arbitrary inputs. A bit of programming and I recreated what I had before in about 1/10th the time it took me to built the first box. Here are the schematics (in two parts to make them each fit on a single page) stepper motor control, arbitrary outputs and inputs.

Given the stepper motor controller, I built a small pen-wielding robot driven by the steppers from salvaged 5-1/4" floppy drives and the head actuator arm from a hard disk drive. Kind of fun, and my kids like driving it around almost as much as I do.

[Top]