SpudFiles

Hey Guys,

What would be a good and reasonable cost high speed camera? I want something for decent balistic analysis that does not cost a fortune.

Matt

Sure, Fastcam and Phantom both have several cameras suitable for ballistic analysis. I believe the majority of them are in the $50k - $500k price range, although they are extremely secretive about pricing. That's not a fortune in any sort of modern terms.

If you're working on a more typical hobbyist budget, you may want to look into high speed still photography, which can be done in more like the $1k - $2k range.

Unless your projectiles are the size of small cars, the currently available "high speed cameras" produced by Casio will be woefully inaccurate. They're typically hardly even able to see projectiles, much less analyze their performance.

DYI wrote:If you're working on a more typical hobbyist budget, you may want to look into high speed still photography, which can be done in more like the $1k - $2k range.

If you already have a camera that can manual focus, manual expose (maybe) and has a spot for a remote, I think you can do it for under ~$200.

Find an old flash like a Vivitar 283 and google for a circuit. You could buy a complete multi channel circuit for about $100 and still come in under $200.

I have a casio ex-fs10 and DYI is kinda right. The highest speed, 1000 fps, is far too blurry. I dont even know why I got that camera. It's a piece of crap Plan to shell out some bank if you want a decent camera.

What kind of analysis do you want to do? For muzzle velocity you can use sound or light detectors and the MIC input of a laptop. Ignoring the laptop, figure a total cost of about $10 with time resolution 1/20,000 second and distance resolution of perhaps as little as 0.1". The same setup can be used for Cd measurements and the like by just increasing the separation between the gates.

If you want to detect tumbling then a camera based system is probably required.

jimmy101 wrote:What kind of analysis do you want to do? For muzzle velocity you can use sound or light detectors and the MIC input of a laptop. Ignoring the laptop, figure a total cost of about $10 with time resolution 1/20,000 second and distance resolution of perhaps as little as 0.1". The same setup can be used for Cd measurements and the like by just increasing the separation between the gates.

You could expand further on this and use an oscillator and counter IC for higher resolution as well... don't ask me how, though, as I still can't get my little PWM circuit to even work. A microcontroller will also have a resolution roughly 333 times higher than a basic soundcard.

High speed still photography can be done as well on the (relatively) cheap to detect tumbling and other things. Maybe somebody a little more experienced here can pipe up but I think multiple exposures can be done with more than one flash unit...

mark.f wrote:High speed still photography can be done as well on the (relatively) cheap to detect tumbling and other things. Maybe somebody a little more experienced here can pipe up but I think multiple exposures can be done with more than one flash unit...

Correct. Some really expensive flashes also have a strobe function. It is VERY weak. The main issues with still photography are:

1. Ambient lighting
shoot in darkness, leave the shutter open for about 1 second

Manual exposure for ambient lighting, triggger camera shutter before
flash
Even DSLRs have at least a few tens of ms of shutter lag. That is
the minimum delay you need after you detect the projectile, so at
800 fps and 20 ms shutter lag (shorter than typical), you detect a
minimum of 16 feet in advance of where your camera is pointed.
You then wait a minimum of 20ms, then trip the flash. CHDK
enabled cameras using the falling voltage trigger should have a
much shorter shutter lag

Kerr cell shutter in front of lens. Extremely fast, faster than probably
all flashes. You will need a way to drive it with fairly high voltage. You
can pulse it for multiple exposures, even from the same flash.


2. Flash Duration
Most commercial flash guns bottom out at 1/40,000 of a second. In that time, an 800fps projectile will travel ~0.25". You really need something faster, probably homemade

If you're exposing for ambient lighting with a long shutter speed, you need a REALLY bright flash. With a 1s shutter speed and a 1/40,000 flash duration, you're looking at 40,000x brighter than ambient. If that's direct sunlight, I'm so sorry...

3. Triggering
Sound is easy, but variation in muzzle velocity will throw your
composition

Laser is great, but it will show up in your picture, unless you turn it off
before you open the camera shutter (or if its out of frame). The main
reason to use a laser is to ensure consistency

Contact is like laser, but easier to make initially and more tedious to
maintain.

Regarding homebrew high-speed photography...
I wonder how plausible it would be to make a trigger circuit from an-off-the-shelf motion detector, like in a porch light. One could presumably either control the delay with a small variable capacitor, or limit the detector's field of view severely so it doesn't fire as soon as the projectile moves into the camera frame.
It would simplify setup somewhat. Aim the camera, aim the detector, open shutter and fire.

Fnord wrote:Regarding homebrew high-speed photography...
I wonder how plausible it would be to make a trigger circuit from an-off-the-shelf motion detector, like in a porch light. One could presumably either control the delay with a small variable capacitor, or limit the detector's field of view severely so it doesn't fire as soon as the projectile moves into the camera frame.
It would simplify setup somewhat. Aim the camera, aim the detector, open shutter and fire.

Audio is actually easier. Aim camera, aim flash, start wide angle and work with the delay until its where you want it for a close up shot.

Eh, I take that back...

I would be concerned with the motion detector being ultrasonic, and therefore "blinded" by the muzzle report (not an issue if supersonic )

with delay (2 * distance from projectile )/1100 feet absolute minmum (big issue if supersonic )

and with sensing something that small.

You need a hidden camera detector
or a general high speed camera detector?