High speed video capture

June 24, 2005 10:58 PM

I waited to see if anyone else might post this. I saw it on Future Feeder. Photron's model ultima APX-RS is a high speed video camera - 250,000 fps. Here's a quick link to the gallery of video (flash interface). apparently that's not the fastest. That appears to be Shimadzu’s HyperVision HPV-1 at 1,000,000 fps. They also have a gallery - but with only 3 mpg clips each a little more than 2.5 MB (1, 2, 3).
posted by tvjunkie (19 comments total)

D'oh! Apparently
posted by tvjunkie at 11:33 PM on June 24, 2005

Those little mpgs are very, very cool, especially the first one, which is just gorgeous.
posted by Dr. Wu at 11:57 PM on June 24, 2005

Interesting that on the last mpg, there's a crack that forms on the lower right fast enough that it's nearly instantaneous, supposing I wanted to see its progress, 1,000,000 doesn't seem like enough. It's startling how fast the world around us (and within us) moves. As a thought experiment, I would think that

So if they set up a beam of light to project 300 miles, and then filmed the light turning on, we would see the light project over that distance in the course of 1 second, right?
posted by Jack Karaoke at 12:34 AM on June 25, 2005

Sorry, evil numbers are confusing me. Someone help me out here.
posted by Jack Karaoke at 12:37 AM on June 25, 2005

Thanks! That water balloon on the Photron page - when the balloon skin popped it just laid there, like an egg of water...
posted by Termite at 12:50 AM on June 25, 2005

This is great stuff.
posted by catachresoid at 12:59 AM on June 25, 2005

[i]So if they set up a beam of light to project 300 miles, and then filmed the light turning on, we would see the light project over that distance in the course of 1 second, right?[/i]

Well.. you can only [i]see[/i] the light when it arrives at the camera. So it will still be instantaneous.
posted by Harry at 2:44 AM on June 25, 2005

It would take way less than 1 second for light to travel 300 miles. I seem to remember it's about 1 second for a round trip to the moon and back.
posted by Vulpyne at 2:48 AM on June 25, 2005

I saw TWiN at Semi-Permanent London and they had access to a camera like this and showed some amazing clips. Well worth a look if you're around NYC on 9/10 Sept.
posted by quiet at 3:03 AM on June 25, 2005

I was thinking of framing a shot from a great distance, so that from edge to edge, the scene included a distance of 300 miles, from a plane perhaps. I have a vague notion that the 1,000,000 frames/ second + 300 miles, given the speed of light at around 300,000,000 miles/second, some nifty parlor trick should be possible..
posted by Jack Karaoke at 5:29 AM on June 25, 2005

Stuff like this is amazing - there is actually a guide out there on how to do this yourself, too!
posted by SharQ at 5:36 AM on June 25, 2005

It would take way less than 1 second for light to travel 300 miles.
Yes, but Jack Karaoke is suggesting you use a high speed camera which would allow you to view that much smaller fraction of a second in the space of one second.

Well.. you can only see the light when it arrives at the camera. So it will still be instantaneous.
That's true if the light is pointed in your direction, but if you are looking at the course of the light beam from 300 miles perpendicular to the light beam, then light reflected from the half way point would only have to travel 150+300=450 mi to reach the camera, and light reflected from the "end" of the light beam would have to travel 300+sqrt(150^2+300^2)=635 mi total. So light from a midpoint reflector would have a shorter distance to travel, and would reach your camera first.
posted by teg at 5:48 AM on June 25, 2005

Perhaps what you'd be able to do is photograph a laser beam as it sweeps out a circle. You'd photograph looking down on it, like a radar sweeper, but with the high speed and an appropriately large frame, you'd see it behave like a garden sprinkler, slowly bending as it curves away from the source.
posted by odinsdream at 6:06 AM on June 25, 2005

Nope. Physics doesn't work that way. You're stuck with the reference frame you're in.

Now, if you had tens of thousands of cameras arranged in a grid...
posted by dmd at 8:29 AM on June 25, 2005

odinsdream: we do "see" this at very large scales and long distances: neutron stars.
posted by warbaby at 11:06 AM on June 25, 2005

and pulsars... (too fast on the click)
posted by warbaby at 11:07 AM on June 25, 2005

I'm imagining a really, really strong focused beam of light shining through dust, illuminating the particles as it goes.

The speed of light 186,282 miles/second. That's .00000536820519 seconds/mile *.
For a beam of light to travel 300 miles would take 0.00161046156 seconds. With a 1,000,000 fps camera, you're looking at 26 minutes of video.
You could have a much more riveting 5-second video of light traveling one mile...which is much more feasable.

* Some significant figures were harmed in the making of this posting
posted by jewzilla at 11:37 AM on June 25, 2005

jewzilla...this is a beautiful, beautiful idea.
posted by effugas at 1:22 PM on June 25, 2005

I get 1610 frames, jewzilla, 54 seconds at the standard 30 fps, not 26 minutes. Light traveling 1 mile would produce 5 frames, less than 0.2 seconds at 30 fps or 1s at 5 fps.

Jack Karaoke: it's 300 million meters per second.
posted by Codebender at 1:40 PM on June 25, 2005