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

Two spaced sheets of aluminum foil could be shorted by the muzzle blast alone, much like a simple microphone or pressure switch. The muzzle blast actually outpaces the bullet itself for a considerable distance (the bullet starts to slow down the instant it leaves the muzzle. The gas continues to expand and is less massive).

Two spaced sheets of aluminum foil could be shorted by the muzzle blast alone

An air weapon doesn't have a lot of muzzle 'blast', and if you position the foil a few feet away from the barrel, the puff of air will have dissipated.

 

It's been done before. This isn't my original idea, I just remembered it from reading about it many years ago in a photo-mag or some such other DIY article.

 

I think it was recommended to stretch the foil across a thin wooden frame to keep it spaced until bridged by the projectile.

 

And even if the foil was triggered prior to being hit by the bullet, so what? You just position the flash and camera in front of the foil, rather than after it.

Edited August 16, 2019 by rodeo_joe|1

If for dramatic effect, bullets in flight are usually photographed as they leave the muzzle or strick some object, like an apple or balloon. The triggering mechanism should not appear in the photograph, unless it is a subject of interest. What you describe would work, but isn't a photogenic solution. The projectile won't "bridge" the conductors, but is likely to cause the foils to touch. Lead is not a good conductor, and probably disrupts the foil so that it doesn't really touch both foils simultaneously. Holes in soft metal are typically about twice the diameter of the projectile.

 

Just about any dynamic speaker or piezo noise maker will serve as a microphone if wired as such. I wouldn't risk damaging a real microphone for an experiment of this sort.

 

An air weapon doesn't have a lot of muzzle 'blast

Perhaps you are thinking of your trusty Red Ryder. My pellet guns have significant muzzle blast, and it is supersonic. A .44 Magnum revolver, on the other hand, will make a paper target wave like a flag at 10 meters, about a second after the bullet strikes it. If you catch it at the right time, there is a shock "doughnut" about 18" from the muzzle where the blast becomes subsonic.

Lead is not a good conductor,

That'll be why it's used as the terminal posts on car batteries then.

That'll be why it's used as the terminal posts on car batteries then.

It's used in car battery terminals because it resists attack by sulfuric acid as well as normal oxidation. It's also a component of a lead-acid battery. In this application the surface area is large to minimize resistance. It's. used in solder (formerly, anyway) because it has a low melting point when alloyed with tin. It binds wires together, electrically, but you are cautioned to mate conductors intimately before soldering for best results.

 

A projectile will push closely spaced layers of foil together before it penetrates the first one. Once penetration occurs, the foil will rebound away from the projectile, making poor if any contract with it, like popping a balloon.

A projectile will push closely spaced layers of foil together before it penetrates the first one.

Exactly!

Whatever the mechanism, it works.

 

Looking at the resistivity of lead - about 10 times greater than that of aluminium and still only 208 nano-ohm.metre - a rough calculation of the length and average area of a hollow and waisted 0.22 airgun pellet, puts its resistance at a fraction of an ohm from end to end. So I'm pretty sure a .22 slug would trigger any flash if shoved into its P-C co-axial plug, or somehow bridged the hotshoe contacts. Likewise a steel BB.

A .44 Magnum revolver, on the other hand, will make a paper target wave like a flag at 10 meters

I thought we'd already established that high muzzle velocity weapons were out of the picture - literally! Due to any readily available flash having a duration too long to freeze the projectile.

Edited August 17, 2019 by rodeo_joe|1

I thought we'd already established that high muzzle velocity weapons were out of the picture - literally! Due to any readily available flash having a duration too long to freeze the projectile.

 

Who was limiting the discussion to "readily available" flashes? I don't believe that any readily available flash will freeze a pellet . . .

A ,22 slug is coated with wax, an insulator, which would render it unlikely to make electrical contact. I suspect the spaced layers of aluminum foil would work even with a BB, which is round and unable to make contact once it passes. halfway through a foil.

 

The difference, in this discussion, between a pellet gun and a .44 magnum is one of degree rather kind, including the obvious safety ramifications. Being large and relatively slow (by firearm standards), a .44 magnum slug (actually 0.429") would be less of a challenge. I was merely using it as an illustration of muzzle blast, which evolves to a toroidal wave, which can be stable for several meters. Its effects are only visible because there's so much more "blast" generated than with smaller calibers or longer barrels.

Who was limiting the discussion to "readily available" flashes? I don't believe that any readily available flash will freeze a pellet . . .

So. You're up for building or obtaining oil-filled capacitors and playing with voltages of 1kV and upwards? And buying a suitable flash tube? Building an EHT supply and trigger circuit? Or maybe building a Kerr cell as a hyper-fast shutter?

 

Let's remind ourselves of the original brief:

My teenage sons are interested in photographing a bb or pellet as it leaves the barrel of a pistol,

No mention of it having to be pin-sharp, and no mention of freezing a high muzzle-velocity projectile. Certainly no mention of wanting this to turn into an expensive and time-consuming research project requiring state-of-the-art equipment.

Many of us have built equipment to accomplish a job which was either too expensive to buy commercially, or not available at all. Once upon a time science fairs were more than poster displays describing the evils of plastic straws. Perhaps a teenager (or parent) will be inspired by what can be rather than what is easy.

Maybe find a used working GR 1531 on the 'bay for a couple hundred. I believe that has a short enough duration for most purposes. Not sure about more modern LED strobes, but in theory they can be quite short.

Maybe find a used working GR 1531 on the 'bay for a couple hundred.

A stroboscope is not what's needed, and many speedlights already have a repeating flash function built in.

Many of us have built equipment to accomplish a job which was either too expensive to buy commercially, or not available at all. Once upon a time science fairs were more than poster displays describing the evils of plastic straws. Perhaps a teenager (or parent) will be inspired by what can be rather than what is easy.

You're talking to someone that built an electronic flash power pack from recovered parts as a teenager, because I couldn't afford one as powerful as I wanted. But high-speed discharge-tube flash is in another league, and could easily be lethal by accident or ineptitude. Also the complexity of the project to build from scratch would challenge students at undergraduate level, let alone school children.

 

Not sure about more modern LED strobes, but in theory they can be quite short.

Low voltage LEDs are a possibility. I haven't explored how short a light pulse they'll emit, nor how bright they can (affordably) be made at short duration. So there's a 'science fair' project in the making.

Edited August 21, 2019 by rodeo_joe|1

Any small electronic flash (not LED) can be lethal if mishandled. All it takes is as litle as 75 mA and 40 vdc (given enough wet contact area). Most flash units operate at 200 - 400 VDC, which is enough to overcome the natural resistance of dry skin, and the main capacitor will deliver many amperes of current in less than a hearbeat (literally).

 

My brother was in junior hich school (middle school) when he built a laser using a neon sign transformer and oil capacitor to buid a laser. An old refrigerator compressor served as the vacuum pump. My science project was a Van de Graff generator.

 

My destruction of a screwdriver was accomplished by discharging a capacitor in a 50 Joule flash unit while replacing the flash tube.

Edited August 21, 2019 by Ed_Ingold

So. You're up for building or obtaining oil-filled capacitors and playing with voltages of 1kV and upwards? And buying a suitable flash tube? Building an EHT supply and trigger circuit? Or maybe building a Kerr cell as a hyper-fast shutter?

 

Nope . . . We also aren't limited to the same gear that Edgerton had available when he was building. Look around on the web. You can easily find plans for this sort of thing using relatively cheap LED bulbs and controllers to get you into the millisecond range.

 

No . . . He didn't suggest that they needed to be pin sharp OR that he wanted to stop high-velocity projectiles. Earlier in this discussion I did the math for someone who claimed that their commercial flash had a duration of 1/80,000/sec. I don't believe that number but even that isn't going to stop a pellet traveling at 300fps even reasonably well.

The GR 1531 is triggerable for single shot events. I think the duration is around 0.8 microseconds on the best setting. Commonly used for this sort of thing-

https://www.ietlabs.com/pdf/application_notes/Handbook_Stroboscopy.pdf

I've tested most of the rest of my many hotshoe speedlights and hammerhead flashes. Not one of them shows any sign of breaking the ~10 microsecond barrier - except for the little Sunpak 122.

 

Big Mecablitzes at a supposed 1/256th 'power'? Forget 'em! The flash duration is still many tens of microseconds long.

 

Osram/Cullmann flashes? They have a very strange flash curve, with a two-stage risetime... peculiar, but useless for high-speed use.

 

Macro ringlights? You'd think the low intensity would lend itself to a short flash duration, but no dice.

 

Little built-in camera popup flashes showed some promise, but my D7200 popup at the 1/128th manual setting 'only' got down to about 20 microseconds. Plus there's no way to trigger the flash independently of the camera shutter.

 

So I turned to LEDs. Eureka!

I stumbled on a little keyring LED 'torch' that uses Pulse-Width modulation to dim its output. The pulses are extremely short at 15 microseconds - see scope trace below, with comparison to shortest duration from D7200 popup.

The keyring LED is shown below the 'scope graphs.

 

Unfortunately the little LED strip only delivers about 250 Lux at 30 cm (1 foot) that's an EV of between 6.5 and 7 depending on which lightmeter I believe. About 1/60th at f/1.4 with ISO 100. So not really viable as it stands, but encouraging to see rise and fall speeds fast enough to apparently allow pulse widths in the region of a few microseconds.

 

LEDs definitely look like the way to go.

Edited August 22, 2019 by rodeo_joe|1

Shutter speeds slower than the flash duration have no effect on exposure. A guide number of 6 or 7 may be optimistic, but since you would be working inside of one foot, the aperture would be reasonable.

 

A duration of 15 microseconds (per your chart) is 1/67,000 seconds. A Nikon D910 would provide far more power, in as little as 1/38,000 second, and at a predictable point in time.

 

An Airsoft pistol has a muzzle velocity of 300 to 450 fps, and a highly visible projectile ( 5 mm), which can be stopped safely by hanging a shag throw rug. They're not toys, will penetrate a soda can, and leave a nasty welt on skin. Eye protection is mandatory.

A guide number of 6 or 7 may be optimistic...

An EV is not a Guide Number. You'd have to divide that 1/60th second of continuous light by 15 microseconds to get an exposure divisor factor/ ISO multiplication factor. Which is about 1200 x, or an ISO of 128,000 and f/1.4. Doable, but with pretty low image quality. Plus you'd have to dismantle the LED torch and figure out how to trigger it into its PWM mode.

 

The measurement of that LED keyring light was simply to demonstrate the speed that LEDs can be pulsed at.

A Nikon D910 would provide far more power, in as little as 1/38,000 second....

Of course an off-the-shelf flash will provide a lot more light at the cost of extended duration. Although the SB-910 is a poor choice compared to other speedlights that can deliver a shorter minimum flash duration. (It's not anywhere near 1/38,000th of a second - 26 microseconds. By my actual measurement it's more like 40 microseconds - 1/25,000th.) A cheap YongNuo 560 will beat it, and a Sunpak 122 almost equals the 15 microsecond LED pulse. However, if you want to get anywhere near to single microsecond figures - with electrical safety - then LEDs definitely are the way to go. See this link.

 

There's also this

that's very informative.

Single microsecond duration? By your graph, the duration is 15 microseconds. We know how to do it right (thanks to Edgerton). One inadequate solution is no better than another. Anything worth doing is worth doing well - a good lesson for teenagers (and adults). Effort is only rewarded in T-ball.

 

If all you want to see is a streak, use a cell phone in slo-mo video mode. Borrow a GoPro. The Hero7 will shoot at 240 fps. A Sony RX0 nearly 1000 fps for a few seconds.

FWIW, we have a high speed camera at work that can do up to 1 million frames per second. We never run it that high because you'd have to live on the surface of the sun to have enough light!

Single microsecond duration? By your graph, the duration is 15 microseconds.

Yes, with that un-modified little Pulse-Width-Modulated LED keyring it is. But look at the fast rise and fall times - practically a perfect square wave! With a faster drive pulse the 'flash' duration could easily be taken into single microsecond figures.

 

Also read the 'Edgerton Project' link I gave, where the author has made a sub-microsecond LED 'flash'.

 

Personally, I think the project is over-engineered in parts, and under-considered in other aspects. For example: Why house the LED modules in a black baffle? That's just throwing light away. Also the use of a microprocessor and mosfet switches seems overkill, and the electrical layout is poor. It needs RF wiring techniques (e.g. impedance matched co-ax cable) to reduce ringing and line losses for sure.

 

Anyhow. I now have a cannibalised domestic LED light bulb to check out. It has, or rather had, 10 LED chips mounted on an aluminium substrate. One chip burned out in normal use - hence it's destiny as a test item. So I have 9 more chips to characterise and if necessary test to destruction.

 

I've already discovered that each chip needs 18 volts to drive it to a reasonably bright output, and that the chips will handle a current of 100 milliamps without any apparent harm. Using the rule of thumb that you can stress a component by tenfold for millisecond durations, I suspect that each chip can be 'blasted' with nearly an amp of current for a few microseconds to boost the light output to... who knows what level?

 

With the expensive COB LEDs used in the Edgerton Project, it pays to be cautious with the components, but who cares about annihilating a £1.99 domestic LED bulb?

 

So, what has anyone else done to practically improve our knowledge of the subject?

Edited August 24, 2019 by rodeo_joe|1