How do I photograph a pellet fired from a pellet gun?

Its a typical subject to utilize strobes on.

Math is not my field, but I can imagine that your FP shutter's slit might be open, but elsewhere, while it flies through your frame? - Your curtains take sync speed = 1/250? to traveland only expose 1/32 of your frame at once when set to 1/8000 sec.

Timing seems hard to do.

I'd lean towards darkening your environment putting shutter on long enough and using a contraption to trigger camera first pistol next and flash kind of variable just in time.

While BBs are no flintlocks; it will take a while for the pellet to come out. An ability to delay your flash by 1000th of a second steps or such would be handy.

I'd put very little hope in capturing a bullet with slow motion video bursts.

You can probably find your bullet's velocity somewhere and from there on figure out how narrow the time frame to capture it photogenically might be.

Since you have multiple cameras, it might be a good idea to use one of them, synchronized to the other, to capture a longer distance of the bullet's flight, to analyze how far off into which direction your flash's timing might be, once you get it close.

Open question: "How do I get my computer to fire 2 cameras, 1 shooter and a flash?"

I'm interested too and hope I wasn't too far off so far.

Air guns typically have a velocity between 300 to 600 fps and a diameter between 0.177" and 0.250". You need a shutter speed of 1/2000 or faster to capture something better than a streak. The best bet is using a strobe at low power, which can be as fast as 1/50,000 for a typical small flash unit with adjustable intensity. In a dark room, open the shutter and trigger the flash acoustically, electrically, or with a photodiode sensor.

Edgerton used high power flash units with a duration of 1/1,000,000 or faster for his famous experiments, using low impedance, oil-filled capacitors. Some of those flash units were bright enough to ignite a newspaper held in their beam.

As a reminder, don't shoot your eye out, kid ;)

Ditto on using a speedlight set to 1/32nd output duration or shorter.

Here's a storage 'scope trace of a typical hotshoe flash (blue line), and the integral of its light output versus time (orange line).

You can see that 1/2 'power' gets you about 1 millisecond duration, and below that the exposure v time trace is fairly linear. So a 1/32nd 'power' setting corresponds to about 1/16th of a millisecond flash duration - in the region of 65 microseconds.

Since the flash can be placed quite close to the gun-barrel, getting enough light should be no issue. What will be challenging is triggering the flash at the right time, and this could turn into a lengthy and quite complicated DIY electronics project. Or you can just throw money at the problem and buy an off-the-shelf acoustic or light-beam trigger.

I suspect there are suitable circuits published online for the DIY route.

Any legal BB gun is going to have a sub speed-of-sound muzzle velocity. So a sound activated trigger seems the most foolproof way to go. It'll just need some experimentation with microphone placement to capture the BB in the right place.

Good luck! And make sure a backstop for the BB is in place and up to the job of absorbing the energy of several dozen trial shots!

I have a pellet rifle with a strong spring, rather than pumped, action, which shoots a 0.177" pellet at 1200+ fps. This is definitely supersonic, and clearly sounds like it (I have a ballistic chronograph that measures velocity up to 4000 fps). Besides being fast and loud, the air rifle is a POS as far as accuracy goes. My 50 year-old Sheridan will deliver a 0.20 pellet at 600 fps with four pumps, and make a ragged hole at 25 yards.

An air rifle 0.2" (0.177" in some states) or over 600 fps is considered a firearm in half the states, and is regulated accordingly.

I have a pellet rifle with a strong spring, rather than pumped, action, which shoots a 0.177" pellet at 1200+ fps. This is definitely supersonic, and clearly sounds like it (I have a ballistic chronograph that measures velocity up to 4000 fps). Besides being fast and loud, the air rifle is a POS as far as accuracy goes. My 50 year-old Sheridan will deliver a 0.20 pellet at 600 fps with four pumps, and make a ragged hole at 25 yards.

 

An air rifle 0.2" (0.177" in some states) or over 600 fps is considered a firearm in half the states, and is regulated accordingly.

Wow!

I'm surprised it's even possible to get an air-powered projectile to supersonic speed.

Even 600 ft/s is quite fast. That's nearly 183,000 millimetres per second and would require an exposure time of less than 10 microseconds to reduce the blur on the projectile to a recognisable level.

Better have a speedlight that goes down to 128th 'power'! And that'll leave you with a GN of between 3 and 4 in metres @ 100 ISO. So f/8 (ish) @ half a metre. Still quite doable.

C’mon, fcs, it’s not rocket science, is it.

Actually, thanks to Edgerton, it led to rocket science. His strobes were used in high-speed, low level reconnaissance during WWII, including German rocket sites. We destroyed those sites, and ultimately re-badged their rocket scientists for work on our own rocket programs.

The google suggestion is a good one. A Quantum QFlash can give you about 1/10000 sec duration but I have seen plans for using some cheap LED lights to get to the millionth of a sec range. It doesn't look like a simple build but one thing teens have is time.

The harder issue is triggering the flash at the proper moment. Edgerton (your kids should read about his work and view some of his images) used a sound trigger which was moved nearer or father from the sound source in order to change the location of the photographed object.

One of my favorite Edgerton images is of a football place kicker striking the ball. The ball has moved about an inch, is very deformed by the kicker's shoe and the dirt that was on the ball hasn't moved yet.

Is that a like?

If I did "likes" it would be;)

Just found this on ebay.

It'll need an additional thyristor (SCR) to fire a flash, but the majority of the work's been done for you.... and it's dirt cheap. I doubt you could buy the individual components for the same money.

BTW, the big enemy of getting electronic stuff to respond fast, or to light up fast, is capacitance. Any line, self or parasitic capacitance needs to be charged before anything else will turn on, light up, or whatever. So I suspect getting any source of light, be it a gas discharge tube or a LED chip, to react in the microsecond domain isn't going to be an easy task. Let alone getting enough light for a reasonable exposure out of it.

BTW, the big enemy of getting electronic stuff to respond fast, or to light up fast, is capacitance

In more general terms, the enemy of speed is impedance, which includes inductive as well as capacitative effects. Edgerton was able to achieve extremely short flash durations through the use of oil-filled capacitors at very high voltages. I helped build a dye laser, which was activated by a flash tube at about 5000 vdc and an oil filled capacitor about the size of a lunch box.

Modern flash units use electrolytic capacitors of much higher capacitance to achieve high power at lower voltage (400 vdc), but with high internal impedance, hence longer discharge time (typically 1/500 to 1/2000 second at full power). This incidentally avoids reciprocity failure in film (not significant in digital capture) and keeps construction relatively inexpensive and more compact. The discharge time is shortened (reducing net power) by quenching the flash tube.

I'm tempted to try some of this high speed/short duration stuff myself. So to that end I 'scoped the output of some of my (too) many speedlights.

The resulting graphs follow:

The Nikon, Canon and YongNuo speedlights were all set to their minimum duration of 1/128th 'power' in manual mode.

None of those 3 mainstream and high output speedlights managed a duration shorter than about 40 microseconds; the Canon doing a lot worse at about 60 microseconds. This seems due to their slow rise-times, which can't be altered or controlled at all.

The outlier is clearly a little old Sunpak Auto-122, that manages to get in the region of 10 microseconds. It has a smaller flashtube and a lower energy than its bigger cousins. It also quenches the tube by 'dumping' the excess capacitor charge, rather than simply cutting off the current to the tube - inefficient, but obviously faster.

There's no manual control over its output either, and you need to put it into Auto Aperture mode and place it close to the subject to get such a short duration flash.

BTW. I did my best to eliminate sensor rise-time errors by shunting a small-area photodiode with a low resistance of only 47 ohms. The fast trace from the little Sunpak flash would seem to confirm that sensor rise/fall times aren't a significant contribution to the results shown above.

What about a speedlight with a strobe/multi flash mode? I'm thinking it might allow a little more leeway in the triggering, if, for example, you can set a series of 60 flashes at 1/128 power at 60Hz, that would give a whole second to play with. Set the camera on bulb exposure and you might just have a chance without a fancy trigger?

Certainly following this with interest, good 'techy' topic.

What about a speedlight with a strobe/multi flash mode? I'm thinking it might allow a little more leeway in the triggering, if, for example, you can set a series of 60 flashes at 1/128 power at 60Hz, that would give a whole second to play with. Set the camera on bulb exposure and you might just have a chance without a fancy trigger?

It's a possibility. A black background would then just capture something light-coloured flying past, but you're looking at a chance of (40 x 60):1 million of catching the subject in frame. That's about 1 chance in 400 of a useable shot.

I tried 'freehand' water splash photography, and that turned out to be pretty easy after a bit of practise. The regular drip of a tap is very predictable, and synchronising the shutter release didn't seem to take much skill..... just a sense of, errrm...... ti min g.

The frequency of the repeating strobe function is too slow to capture multiple images, and unlikely to capture even one image reliably. You need to devise some electrical or acoustic trigger. It's not hard to mount components on a blank, perforated circuit board and a soldering iron. Make it work, then worry about making it pretty.

A friend gave me a digital recording oscilloscope (my old one is non-recording, analog, and the size of a beer cooler). All this tech talk inspires me to build a photo sensor for it. Capturing bullets in flight is one challenge. Measuring shutter speed and lag time is of more immediate interest.

The frequency of the repeating strobe function is too slow to capture multiple images, and unlikely to capture even one image reliably. You need to devise some electrical or acoustic trigger.

Exactly . . .

The frequency of the repeating strobe function is too slow to capture multiple images, and unlikely to capture even one image reliably.

I suspected as much, muzzle velocity is just too high, was worth a shot in the 'use what you've already got' category though.

With an electrically conductive projectile, like a BB, there's a very simple way of triggering a flash. You simply hang a couple of sheets of aluminium foil in the path of the 'bullet' with the foil connected as an open switch to the flash synch socket. The pellet or BB pushes the conductive sheets together as it punctures them and completes the circuit.

This will only capture the projectile as it's passing through, or just emerging from the 'switch', unless a delay is introduced.

The delay can be a simple home-built electronic circuit, or the built-in delay of a radio flash-trigger can be used. Radio flash triggers introduce an encoding/decoding delay of a few hundred microseconds, placing the projectile a few millimetres or maybe a couple of centimetres the other side of the foil switch.