It can be hard enough to take a good photograph of a running kid or pet, and if we’re being honest, sometimes even stationary objects manage to elude our focus. Now imagine trying to take a picture of something moving really fast, like a bullet. Trying to capture the moment a fast moving projectile hits an object is simply not possible with a human behind the shutter button.
Enter the ballistic chronometer: a device that uses a set of sensor gates and a highly accurate timer to determine how fast an object is flying through it. Chronometers that operate up to a couple hundred meters per second are relatively common, but [td0g] had something a little faster in mind. He’s come up with an optical setup that he claims can capture objects moving as fast as Mach 2. With this chronometer tied into a high-speed flash rig, [td0g] is able to capture incredible shots such as the precise instant a bullet shatters a glass of water.
Because he couldn’t find any phototransistors with the sub-microsecond response time necessary to detect a small object moving at 1,000 m/s, [td0g] ended up using LEDs in a photoconductive configuration, where 27 VDC is applied backwards against the diode. Careful monitoring of voltage fluctuations across the diode allows for detection of changes in the received light level. To cut down on interference, [td0g] used IR LEDs as his light sources, reasoning there would be less ambient IR than if he used something in the visual range.
What really impresses with this build is the attention to detail and amount of polish [td0g] put into the design. From the slick angled bracket that holds the Arduino and LCD to the 3D printed covers over the optical gates, the final device looks like a professional piece of equipment with a price tag to rival that of a used car.
For the future, [td0g] plans on upgrading to faster comparators than he LM339’s he has installed currently, and springing for professionally done PCBs instead of protoboard. In its current state this is already a very impressive piece of kit, so we’d love to see what it looks like when it’s “finished”.
If you don’t need something quite this high end but still would like to see how fast something is going, we have covered chronometer builds to fit every budget.
*twitch* elude.
What can you do? Typos on Hackaday these days are a diamond dozen.
You misspelled dozing!
And grammatical errors by those who post comments are a dime a dozen.
The joke.
Your head.
+1 !!!!
Roughly 2200 fps if memory serves. Not bad, but for higher velocities, hacking one of the “Shooting Chrony” boxes might be easier.
I borrowed a Chrony (F1?) to test the results of this unit. The Chrony is fine, but is sensitive to the ambient lighting. Originally I had thought of just hacking a Chrony, but that’s considerably more expensive. Somebody should do it and share their experience!
Years ago, I bought two of the sense heads for a PACT chronograph from Dillon Precision. They used a fairly ordinary photo-transistor in each one. I built a pair of I/V converters as a front end and then used a high speed comparator after that to get a digital signal out. My testing was limited to using an air rifle because I did not have the ability to run a scope at the shooting range for testing with something faster. In any event, the circuit reliably detected .177 pellets at around 750 fps. I can’t speak to the delays, which would be un-measurable with two detectors. I remember being happy with the results, but don’t have a lab notebook form that far back to be able to provide real data. I do have the detector heads still.
I just dug out the PACT screens, they use a Motorola MRD3056 phototransistor for the detector. About 1 – 2 uS response time.
A quick look at Digikey shows a photodiode with a response time of 2 nanoseconds, for about $1: https://www.digikey.com/product-detail/en/osram-opto-semiconductors-inc/SFH-2701/475-2967-1-ND/2794398
If you want to pay more, there are ones as fast as 60 pico seconds.
That’s a good point. While it’s an interesting fact that you can use an LED for a photodiode, you can actually buy devices that come with specs, app notes, and IR filters.
When the zombie apocalypse comes, we will have to make do with what’s in the parts drawer, but today, there’s Digi-Key.
Consider this a guide for building a device after the zombies take over Digikey and Mouser!
Really, I learned about PIN photodiodes AFTER finishing this project. Nevertheless, the ~500ns response time of the LED is sufficient for measuring bullet speeds.
*50 ns
I just thought, that on the way from the (slow) phototransistor to the LED “misused” as photodiode with a way higher reverse voltage than specified as safe, there are real photodiodes. And they can be blazingly fast. They need to be for multi-gigabit/s (fiber)optic communication.
I was hoping for more that one picture of the results.
Here you go! https://gerritsendesign.wordpress.com/2016/07/24/shaken-not-stirred/
Nice write up. I like all the details about timing issues and how the accuracy was required to get the shot. Great pictures too.
What about placing the cup on the piezo element and triggering at the first sensed movement? An idea…. Find some old TV picture tubes… those things collapse spectacularly!
Good thought… That actually may work. You would get the piezo wet, but with the minimal response time of a flash this may work!
The author says “the receivers (sensors) are LTR-323DB LED’s”. But LTR-323DB is a photo-diode! I think th eauthor is a bit confused…
And with a rise/fall time of 50ns it is not even that fast, but it is cheap (Mouser: €0,18@1pc) and it LOOKS LIKE AN LED: :-)
The author is generally in a state of confusion ;) 50ns, while not the fastest, is definitely sufficient for measuring bullet speeds.