[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Sparks - Bright in the middle, how to verify it.



Original poster: "Jim Lux by way of Terry Fritz <twftesla-at-uswest-dot-net>" <jimlux-at-earthlink-dot-net>

The webcams/quickcams may be cheap, but they're VERY tough to synchronize to
an external source: it's not important for their intended application, and
there isn't much documentation available on the newer ones. (if anyone knows
of a source for this information, let me know (off list)).

Very few consumer (film or digital) cameras are synchronizable to any
degree.  While developing triggering logic for a system composed of 100
cameras to do nifty visual effects
(http://www.reelefx-dot-com/Frames/multicamframe.htm , think "the Matrix",
although we did it more than a year before that came out ), I discovered
that modern cameras use a polling scheme to look at the shutter button and
to operate the curtains in the camera. The random jitter in the "press
button to exposure" is on the order of tens of milliseconds (this is easy to
test with a spinning disk with a line painted on it, and some means of
repetitively triggering the camera).  An uncertainty of a few tenths of a
millisecond isn't a real problem in a focal plane shutter, the exposure
might vary a bit, but the spec is only 1/2 stop, and a change of 100
microseconds isn't going change the exposure that much, and any variations
will be compensated when the pictures are printed. Very few camera owners do
any sort of calibrated sensitometry.  You should see the color balance
differences between cameras! and the effects of birefringence and polarized
light with modern plastic lenses, which are otherwise of very high optical
quality.  Unless you do camera to camera comparisons, you'll never know.

Throw in a variable delay for autofocus, and it gets worse.

For digital cameras, it is much worse.  Most don't provide any external
trigger mechanism: some can generate a flash sync, but not receive one.  The
IR remote on some is a bit better, but still has uncertainies greater than
10-20 mSec.

Your best bet is going to be a video camera that has fast shutter speeds and
that can accept a "gen lock" signal (i.e. an external frame sync).

As far as exposure goes, sparks are pretty bright.

I'd love to set up an array of cameras and do 3D imaging, and if I could
find a source for inexpensive (<$100 each, including any needed
modifications) synchronizable, fast digital cameras with at least 640x480
resolution, I'd set up an array in a flash just to study things like spark
propagation.

One idea that I did have involved using a fairly slow camera, and an
external shutter (i.e. a Kerr Cell or magnetooptic).  Then, all you need is
a long enough exposure to cover all the variation in timing, and possibly to
cool the camera so that the dark current doesn't get you.



----- Original Message -----
From: "Tesla list" <tesla-at-pupman-dot-com>
To: <tesla-at-pupman-dot-com>
Sent: Sunday, January 21, 2001 8:15 AM
Subject: Re: Sparks - Bright in the middle, how to verify it.


> Original poster: "Mike Harrison by way of Terry Fritz
<twftesla-at-uswest-dot-net>" <mike-at-whitewing.co.uk>
>
> On Sat, 20 Jan 2001 11:33:06 -0700, you wrote:
>
> >> I thought of doing this a while ago as I have a camera with an
> external sync input, which could be locked from a mains derived source
> with an adjustable delay. The problem, however, with both film and
> video methods is sensitivity, or lack of it. When you take a video
> camera down to 1/10,000 sec, you need an awful lot of light, and would
> need to look at pretty big sparks to get a useable picture.  This is
> not helped by the fact that as sparks are a 3-d phenomenon, you need a
> reasonable depth-of-field, and so you may also need to reduce the
> aperture to get a sharp picture.
>
> One possibility I did think of, is that webcam type devices are
> getting very cheap nowadays, so it might be feasible to use several of
> these to take a small sequence of pics in rapid succession with
> appropriate syncing. 3-d spark pics would also be possible with such
> an array. A bunch  of disposable film cameras could also be a
> possibility.....
>
> I've also frequently noticed the 'bright centre' phenomenon on several
>