Re: [Phys-l] Strange fall data

[John Denker <jsd@av8n.com>]

On 05/28/2010 01:16 PM, Josh Gates wrote:

> I've had this problem before, demonstrating free fall with high speed 
> video analysis, picket fences, and sonic rangers. The acceleration 
> always turns out to be a bit bigger than g.

Various things to check:
 *) What g are you using?  
   -- Textbook author's guess?  
   -- NIST "standard" value?
       http://physics.nist.gov/cgi-bin/cuu/Value?gn
   -- actual local measured value?
       http://irpsrvgis00.utep.edu/repositorywebsite/Default.aspx

The actual g value does vary from place to place to a 
quite measurable degree.  Since I haven't heard how big
the observed discrepancy is, I have no way of knowing
whether the variation is big enough to explain the
discrepancy.

  *) How about geometry, such as parallax?  If you have a
reference grid appreciably behind the falling object, 
simple geometry will give exaggerated distances, velocities,
and accelerations.

  *) Similarly, if you have a reference grid off to the
side of the object, you are at the mercy of imperfections
in the camera lens, e.g. fisheye distortion, barrel 
distortion, et cetera.

Hint:  Moving the camera far away and using a long focal
length lens may be an easy way to alleviate some of the
problems with geometry, parallax, and/or distortion.

  *) As others have mentioned, maybe the 300 fps timebase
of the camera is off.  Aim the camera at the seconds digits
of a digital clock.  Or aim it at a color TV.  TVs are
very closely synchronized to 60/1.001 Hz, approximately
59.9401 Hz.

  *) Maybe there's a huge downdraft blowing on the 
apparatus.

A more detailed description of the apparatus, maybe even
with a picture, would help folks figure out what's going
on.  At the very least, it would help to have a report
on the size of the discrepancy, so we know whether we
are hunting a mouse or an elephant.

==============

As for sonic rangers, there's a whole nother set of 
things that can go wrong with those.  For starters, 
they fundamentally measure the frequency of arriving 
wavefronts ... and temperature will mess with the 
frequency-to-wavelength conversion factor.