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]

MBL, was Air resistance, now experimental error



It seems to me that Richard is saying something we all are aware of: most
student equipment isn't good enough to get really great results (say less
than 1% error). Sure one or two experiments in a semester might approach
this but not even the majority do. This IS disconcerting to students
(particularly if they are also taking something like analytical chemistry
where the resutls really ARE that good). And it is important for them to
realize that no experimental measurement is 100% accurate, there is always
SOME error.

So I try to make a virtue out of this 'problem' of working with low grade
equipment by asking the students to esimate the BEST possible results they
COULD get with the given equipment. For example, if the best thermometer
you have is only marked to 1 degree C accuracy, what is the best error you
can expect for a measurement of say, specific heat? If a quantity is
measured by two different techniques in a series of measurments, does the
error spread justify saying that the two measurements give the same result
(ie. do they overlap).

So I would say that following up an experiment which has 20% error by
asking the question, 'well, how close did we expect to get with this
equipment?' is a viable exercise pedagogically....If we expect 2% error but
got 20% then somethings wrong. If we expect 30% but got 20% then we should
be satisfied.

kyle

At 12:18 PM 12/11/97, Richard W. Tarara wrote:
-----Original Message-----
From: John Gastineau <gastineau@mindspring.com>

One experiment I try every
year is using a tilted air track to study accelerated motion and in part
to
measure the acceleration due to gravity. This never works very well (the
technique requires a photo-gate to be triggered AS SOON AS the cart starts
to move down the incline and that is difficult to set up properly).

If the technique has an obvious flaw in requiring a difficult-to-achieve
setup, why use it? There are many other ways to measure g using a
photogate, and they work very well.


Well I continue to use this because IT IS an air track experiment where
we're looking at basic definitions of motion and basic graphing skills.
Half of the experiment is to launch the carts on level tracks measuring
distance as a function of time, graphing, and finding the speed (velocity)
from the slope. Using a tilted track with the first photogate right in
front of the cart WILL show that this motion produces a non-linear distance
versus time graph. However, using the assumption (with this geometry) that
the initial velocity is zero and that the acceleration is constant, the
instantaneous velocity at the second photo-gate is just twice the average
velocity and a velocity versus time graph IS linear and an acceleration can
be obtained. Knowing the angle of the track--obtained from knowing the
height of the raising block and the distance between the feet of the air
track, one can use the experimentally determined acceleration to calculate
'g'. Agreement is normally +/- 20% which is OK with me but disconcerting to
many beginning students. Later we measure 'g' with a pendulum to much
greater accuracy (but only as a side-bar to studying the motion of the
pendulum itself).

Rick

-----------------------------------------------------
kyle forinash 812-941-2390
forinas@indiana.edu
Natural Science Division
Indiana University Southeast
New Albany, IN 47150
http://Physics.ius.indiana.edu/
-----------------------------------------------------