Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: videotape data on free fall



In some cameras the way in which fast shutter speeds are attained is by
sweeping a slit across the face of the film. The film is only exposed
when the slit is in front of it. Thin, fast moving slits make for short
exposures. Is this how a cam corder works? Is this relevent to producing
"clean" images of fast moving objects? Does the orientation of the camera
matter?

On Sat, 25 Oct 1997, Mark Sylvester wrote:

Ludwik wrote:

How can this help? I would have to be far away; the number of pixels (and
lines) is fixed. I do not think that errots are in dt=1/30 second. My data
were already presented (two months ago) and I have not been able to do
anything better. Can somebody else share the free fall data? By how much
do individual accelerations fluctuate in consecutive time intervals?

We did this measurement 2 or 3 years ago, so I don't have the data to hand...

Sharpness of the image did not seem to be a problem. Our little camcorder
claims a fastest shutter speed of 10 000, giving a very dark but sharp (in
the sense of eliminating blurring due to motion) image. The crudeness of the
measurements results from the fact that there is not enough definition to
read a scale to better than +-5cm. We compromised on the shutter speed and
used, I think, 250 in order to read the scale.

The v-t graphs that the students produced showed consistently points
"jumping out of line" in a systematic way, and I concluded that a timing
problem existed. (over here btw it's 50 fields and 25 frames per second, so
delta t is a nicer decimal number, for what it's worth. Also, we have 625
lines in a frame).

I have a new hypothesis: when the vcr is on frame-by-frame advance, there
seems to be an alternation between the two fields that make up the frame.
(Do all vcr's do this?). At least, that's what I think it is - there is an
oscillation in the vertical position. Perhaps the experimenter was
consistently looking at, say, the upper image, but every now and then
slipped up and measured the other one...

At any rate, the problem we were looking at was not an "error bar
situation". We'll investigate it in the next round of practicals and see
what we come up with. It will be interesting to get some results to compare
with the digital method, when we get the software.

Mark.

*************************************************************
Mark Sylvester
United World College of the Adriatic, Duino, Trieste, Italy.
msylvest@spin.it
tel: +39 40 3739 255
*************************************************************