Determining the value of acceleration of free fall due to gravity in the laboratory using pendulums

160150914
Department of Physics, University of Sheeld (Dated: October 22, 2013) The aim of these experiments were to compare and contrast the accuracy of measuring the acceleration of free fall due to gravity, g, in the laboratory through dierent methods. A simple pendulum, compound pendulum and Katers pendulum were each used to deduce a value of g with values, 9.855 0.043ms2 , 9.425 0.072ms2 , 9.75204 0.00170ms2 respectively. These are not in agreement with the accepted value of 9.80665ms2 [1] and may be attributed to errors in measurement and other factors. I. INTRODUCTION

Of the four fundamental forces that are present in our universe, the one we are most familiar with is gravity due to its macroscopic eects. Gravity is an attractive force that attracts masses towards each other. No matter where we are in the world we feel the eects of gravity, the same being true if we were to be in space; the whole solar system orbits the Sun due to the force of gravity. As the Earth is not a perfect sphere and there are places signicantly above sea level, the magnitude of gravitational acceleration, g, changes. This has been modelled based on the longitude and height above sea level [1]. The accurate modelling of the eects of gravity is useful in many applications from measuring the trajectory of a thrown American football to ensuring a satellite will stay in orbit. Gravitational acceleration is represented by the units ms2 . To determine the value of g varying pendulums are to be used; the simple pendulum, the compound pendulum and Katers pendulum. By using these methods and comparing them we may evaluate the accuracy of them and discuss the advantages of using a specic method.
II. A. METHODS

To calculate the value of g, dierent lengths of light, inextensible string were measured from 0.25 to 1m and the period for each when released from a small angle of 5 recorded. To reduce the error from the human reaction time when using a stop watch and minimise the error of pressing the button at the wrong time, the period measured was for 50 complete oscillations and then the period for 1 calculated. As there may be oscillations in the perpendicular axis, the pendulum was allowed to settle before timing started. These measurements were then plotted against each other as T 2 = g 4 2 l (2)

B.

Compound Pendulum

Simple Pendulum

FIG. 2: Diagram [3] showing the compound pendulum. The centre of mass is located at the centre of the black bar which represents the metal bar. The hole at the pivot is one of many that are equally spaced either side of the centre of mass.

FIG. 1: Diagram [3] showing the basic principle of the simple pendulum. The black circle represents the attached mass and connecting line is the string attached to the pivot.

The compound pendulum is an extension of this theory to improve the accuracy of an experiment involving measuring a period. A compound pendulum is a rigid body that swings in a vertical plane about a horizontal axis. The resultant force acts through the centre of mass thus the period of a compound pendulum is related to its moment of inertia. The period may be expressed as T = 2
2 (l2 + k0 ) gl 1/2

The theory behind a pendulum is that if the length, l, from the point of the pivot to the centre of mass of the pendulum remains constant and there are no other external factors, the period of swing will remain constant. This is modelled in the simple pendulum by T = 2 l/g , (1)

(3)

By varying the distance of the pivot to the centre of mass and recording the period of oscillation about a small angle of 5, g and k0 were found. They may be plotted

2 by T 2l = 4 2 4 2 2 l+ k g g 0 (4)
2 2 From this, when T1 T2 = 0 we had the value of the constant term which allowed us to calculate g

III.

RESULTS

T, Period (s)

The gradient was used to calculate the value of g and the y-intercept may then be divided by the gradient to calculate a value for k0 . Care was taken to ensure the pendulum only oscillated in the vertical plane, but there was diculty doing this for the shorter lengths from pivot to centre of mass. Once again 50 oscillations were recorded to fraction any errors in timing.
C. Katers Pendulum

By plotting (2) we obtain The value of g is obtained

10

15

20

25

30

35

40

45

4piL, Length * constant (m)

FIG. 4: T 2 plotted against 4 2 l, Eq. (2), giving 1/g using the method of least tting squares for the simple pendulum.

FIG. 3: Diagram [3] showing Katers pendulum. The triangles represent the knife edges used to balance the pendulum on the pivot. The black bar represents metal cylinder that comprises the pendulum. The adjustable weight can be screwed to and from Pivot 1.

as 9.855 0.043ms2 . The method of least tting squares was used to obtain the value of g and the value of R was found to be 0.9996 indicating a very strong correlation of results and little deviation from the trend. This implies that the measurements were very precise. By plotting (4) we obtain
1.6 1.4
TL, Period* Length (ms) 1.2 1 0.8

Katers pendulum is a further extension of the pendulum theory that negates the need to measure the distance to the centre of mass (a dicult calculation). The pendulum is a compound pendulum that is in term suspended from either end, giving two distances l1 and l2 to the centre of mass and two times, T1 and T2 for the period from each suspension point. This equation was derived
2 2 8 2 T 2 + T2 T 2 T2 = 1 + 1 g (l1 + l2 ) (l1 l2 )

0.6 0.4 0.2 0

L, Length (m)

(5)

FIG. 5: T 2 L plotted against L2 , Eq. (4), giving gradient calculated using the 2 tting method.

4 2 g

as the

Thus when T1 = T2 the l1 l2 measurement was not important. The distance l1 + l2 + was measured using a travelling microscope from the knife edge at each end of the pendulum. The period of the pendulum was measured over 16 swings using an electronic timing gate to eliminate human error in timing. The point in the period that the pendulum passes through the gate was irrelevant to the timing. The pendulum was suspended from a pivot on the knife edge present at both ends, this thin edge minimises any interference from the pivot and also reduces any oscillations in the perpendicular plane as the knife edge extends in this plane. The pendulum was released from a small angle of 5 and its period measured. The moveable mass was then altered in position for subsequent measurements to produce enough data to plot through T1 T2 = 0. The plot was given by
2 2 T1 + T2 = l1 + l2 2 2 T1 T2 8 2 (l1 + l2 ) + l1 l2 g

The value of g is obtained as 9.424 0.072 which does not meet the accepted value. The value of 2 is 31.48232799 By plotting (6) we obtain
IV. DISCUSSION

(6)

Going into the experiment the expected results were to be an increase of accuracy going from the simple pendulum to the compound pendulum to the Katers pendulum, but this was not found to necessarily be the case. The simple pendulum was shown to obtain closest value of g to the accepted value being 0.05% away at its minimum limit, but the error was large INDICATING. The inaccuracies of human time measurement were expected to have a large impact on the calculated value

3 were observable oscillations in planes other than the vertical that were dicult to minimise, these can be seen clearly in g 2. The hole in the pendulum that balanced on the pivot was seen to rotate, losing energy to friction during the movement and also increasing the surface area of the hole in contact with the pivot increasing the friction during its oscillation. This would have caused a slower oscillation, increasing T which would increase make the value of g smaller than expected, which is what we see with the results. The eects of friction were also a factor in the simple pendulum and Katers pendulum. In the simple pendulum the string was ensured to be tight enough to allow no movement along the bar-like pivot, but the string still rubbed against the knot on the pivot causing an amount of friction which would give a smaller value for g . The obtained value of g was too large however, so the major error is thought to be in the measuring of the l. In Katers pendulum the knife edge often did not rest properly in the pivot which would cause friction as the knife edge would itself move with each oscillation as it would not sit perfectly. Due to this there were observable oscillations in the perpendicular plane. This could attribute to the smaller value of g obtained as the period length would increase. In each of the above experiments the small angle approximation is used, but is this valid? An approximate expression for the time period, T varying with angle is given by, 1 T () = T0 (1 + sin2 ) 4 2 (7)

12

T1+T2, Periods added (s)

10 8 6 4 2 0

-0.2

0.2

0.4

0.6

0.8

1.2

1.4

1.6

1.8

(T1-T2)*L, Period subtracted times total length (ms^2)

FIG. 6: T 2 plotted against 4 2 l, Eq. (2), giving 1/g using the method of least tting squares.

of g , however when the associated errors are averaged out over 50 swings they become small in comparison. The average human reaction time is around 215ms [2]. Over 1 measured period this would have resulted in an error as high as 20%. However over 50 swings this becomes 0.4%. There was a very small error, T = 0.0005s , when measuring the time for each period of Katers pendulum as the measurement was performed by an electronic instrument that used a laser gate. Recordings were always found to be within 0.001 of each other which is complient with the instrument error. The error in the timing was found to be negligible and the dening error was that of measuring the pendulum. The centre of mass had to be estimated in the simple pendulum experiment, whereas it was assumed in the compound pendulum and given by l1 l2 in Katers pendulum. It was assumed that the string was light, so the length could be measured from pivot to centre of the attached mass. This proves to be a fair assumption, as the string weighs much less than the mass meaning the centre of mass moves by a negligible amount. The actual measurement of the string has a large error that was not calculated however, which is due to the mass and strings dimensions causing the alignment of a ruler to be dicult (giving a longer length and hence a larger value of g ); in comparison Katers pendulum used a travelling microscope to accurately measure the length from pivot to pivot to an accuracy of 0.00005m. The compound pendulum however also suered issues from measurement as the pendulum was in a bad state, the holes not being perfectly round and the centres of which dicult to decipher. This led to an error greater than that given by the instrument (ruler). The error of the ruler was 0.0005m. The compound pendulum was shown not to be very accurate as when the length was made to be small, there

Using this equation with the used angle 5 we may deduce any errors from this assumption. The maximum new error that is found in T is 0.000942s, 0.00171s and 0.000941s with respect to simple, compound and Katers pendulum. SEE SOMEONE ABOUT THIS TOMORROWOROWRORWOROW.

V.

CONCLUSIONS

We have shown that there are dierent ways in which to determine g using a pendulum and that none are infallible and are especially suspect to human error. The value of g can be determined to be 9.75204 0.00170ms2 using Katers pendulum as the most accurate measurement. This does not comply with the accepted value of 9.80665ms2 [1] being 0.54% o at the uppermost error. To improve the experiment if repeated, care should be taken to ensure that equipment is not worn and is t for purpose and also should be carried out in a vacuum to eliminate external sources of error due to the surrounding air.

[1] http://www.kayelaby.npl.co.uk/general_physics/2_ 7/2_7_5.html, 21/10/2013 [2] http://www.humanbenchmark.com/tests/reactiontime/ stats.php, 21/10/2013

[3] http://www.sheffield.ac.uk/polopoly_fs/1.14262! /file/G1_Katers_script_V2.1.pdf, 22/10/2013