Revista Brasileira de Ensino de Fsica, v. 35, n.

4, 4305 (2013)
www.sbfisica.org.br

Experimental study of simple harmonic motion of a spring-mass system

as a function of spring diameter
(Estudo experimental do movimento harm
onico simples de um sistema massa-mola em funca
o do di
ametro da mola)

C.A. Triana1 , F. Fajardo

Departamento de Fsica, Universidad Nacional de Colombia, Bogot
a, Colombia
Recebido em 23/1/2013; Aceito em 14/3/2013; Publicado em 30/10/2013
The simple harmonic motion of a spring-mass system generally exhibits a behavior strongly inuenced by
the geometric parameters of the spring. In this paper, we study the oscillatory behavior of a spring-mass system, considering the inuence of varying the average spring diameter on the elastic constant k, the angular
frequency , the damping factor , and the dynamics of the oscillations. It was found that the elastic constant k
is proportional to 3 , while the natural frequency 0 is proportional to 3/2 , and decreases as increases.
We also show the dierences obtained in the value of the angular frequency when the springs are considered as
ideal (massless), taking into account the eective mass of the spring, and considering the inuence of the damping of the oscillations. This experiment provides students with the possibility of understanding the dierences
between theoretical models that include well-known corrections to determine the frequency of oscillations of a
spring-mass system.
Keywords: spring-mass system, Hookes law, elastic constant, simple harmonic motion, damping.
O movimento harm
onico simples de um sistema massa-mola geralmente mostra um comportamento que e
fortemente determinado pelos par
ametros geometricos da mola. Neste trabalho estuda-se o comportamento oscilat
orio de um sistema massa-mola considerando o efeito da variaca
o do valor medio do di
ametro da mola
na constante el
astica k, a frequencia angular , o fator de amortecimento e a din
amica das oscilaco
es. N
os
encontramos que a constante el
astica k e proporcional a 3 , enquanto a frequencia 0 e proporcional a 3/2 , e
que diminui quando aumenta. Tambem se mostram as diferencas encontradas no valor da frequencia angular
quando as molas s
ao consideradas ideais (sem massa), levando em conta a massa efetiva da mola, e levando em
conta o amortecimento das oscilaco
es. Esta experiencia oferece aos estudantes a possibilidade de compreender
as diferencas nos modelos te
oricos que levam em conta correco
es que s
ao bem conhecidas para determinar a
frequencia das oscilaco
es de um sistema massa-mola.
Palavras-chave: sistema massa-mola, lei de Hooke, constante el
astica, movimento harm
onico simples, amortecimento.

1. Introduction
The study of the movement experienced by a mass suspended from the free end of a spring is a topic discussed
in most introductory physics courses, from both theoretical and experimental outlooks. The physics of the
spring-mass system oscillations have been widely studied in a great variety of texts, in which the relationship between the period and the oscillation frequency
is shown in detail [1]. Similarly experiments allow observing the dependence of the oscillatory systems on
the mechanical forces as established by Hookes law [2].
Some studies included corrections in order to take into
account the inuence of the spring mass on the oscillations of the spring-mass system [3, 4]. Likewise, stud1 E-mail:

[email protected].

Copyright by the Sociedade Brasileira de Fsica. Printed in Brazil.

ies of the behavior of the oscillations of systems constructed of plastic and non-helical springs have been
carried out [5, 6]. Previous studies have dealt with the
inuence of changing the natural length l0 of the spring
(for a xed diameter) on the behavior of the elastic
constant k, the angular frequency and the damping
factor of the oscillations [7].
In this work we study the inuence of varying the
average spring diameter , for a xed length, on the behavior of the elastic constant k, the angular frequency
, and the damping of the oscillatory motion, which
are the principal variables that determine the simple
harmonic motion of the spring-mass system. A point
that should be emphasized is that this kind of experi-

4305-2

Triana y Fajardo

ment allows showing students that some variables that

characterize a real physical system depend on the size
of the system under consideration. It is important that
students ask, for example, how the behavior of the oscillatory dynamics depends on the size of the spring used
in the experimental arrangement. In other words, it is
very important that students take into account that the
change of size or geometry of one part of the physical
system could have a large inuence on the behavior of
other physical variables that characterize its dynamic
behavior. For the foregoing reasons, this experimental
study is an excellent one for general physics courses.
The paper is organized as follows: in section 2, a fast
review of the theoretical concepts on which the work is
based is performed; section 3 describes the device and
the experimental procedure; in section 4, the discussion of results is addressed; and nally the conclusions
of this work are presented.

2.

Theoretical framework

Previous studies have shown that the longitudinal elastic constant k of a helical spring is determined by the
diameter d of the wire, the average spring diameter
(which hereafter will be referred simply as the diameter
), the number of windings N , and the shear modulus
G, which involves specic characteristics of the material
of which the spring is manufactured [8]. In agreement
with that result, it is possible to calculate the springs
elastic constant k through the ratio
Gd4
,
(1)
83 N
Eq. (1) allows one to determine the behavior of the
elastic constant k as a function of the spring diameter
, and will be used to compare with our experimental results. Taking into account that we are studying
the case of small oscillations, we have that the natural
frequency 0 of oscillation of the spring-mass system is
given by

k
0 =
,
(2)
m
where m is the mass suspended from its free end. Equation (2) describes the behavior of an isolated system in
which there are no dissipative forces and the spring has
no mass (massless spring). In the case of a real physical
system, one must take into account the equivalent mass
M /3 of the spring [9, 10], then the angular frequency
e of the system oscillations taking into account this
correction can be calculated by

k
e =
.
(3)
((M /3) + m)
k=

The principle of conservation of mechanical energy

allows us to relate the angular frequency to the oscil-

lation period T of the system, which in its fundamental

mode [11] is given by = 2/T . The amplitude of
the oscillations decreases as energy is dissipated due
to the presence of non-conservative forces of friction in
the medium in which the system is immersed. Therefore, the oscillatory behavior of a real physical system
disappears after a certain time interval. Thus, a more
realistic description of the behavior of the spring-mass
oscillating system is determined by a damped oscillation, dened by an equation of motion of the form [12]
x = A0 e

t
2

cos(t + ),

(4)

where A0 is the initial amplitude of the oscillations and

is the initial phase. The damping factor responsible
for the decrease in the amplitude of the oscillation as
a function of time is dened by = b/m, b being the
damping coecient and m the mass suspended from
the free end of the spring. Taking into consideration
the damping, which makes the oscillation frequency of
the system slightly smaller than the value dened by
Eq. (2), the angular frequency is determined by

2
k
b2
2
= 0
=

,
(5)
4
m 4m2
where 02 = k/m corresponds to the frequency in the
absence of dissipative forces. Equation (5) suggests that
the higher the damping factor , the greater will be the
attenuation of the amplitude A of the oscillations of
the spring-mass system. In this paper we will show
the dierence obtained in determining the frequency of
oscillations according to theoretical models dened by
Eqs. (2), (3) and (5) as a function of the spring diameter .

3.

Experimental procedure

Figure 1 shows a photograph of the experimental setup

used to measure the k, and variables as a function
of the spring diameter . The experimental measurements were carried out using a series of eight springs
of dierent diameters, made of a steel wire with a diameter of d = 8.1 104 m. All springs had a natural
length of l0 = 10.1 102 m and had no separation h
between their coils (step h = 0). The number of coils
for the springs was chosen to be N = 124. Table 1 summarizes the main characteristics of the springs used in
this experiment. The cost of the springs was less than
10 dollars, so this is an experiment that requires lowcost materials. Using a tape measure, the measurement of the elongation x experienced by the spring
was carried out when masses m were suspended at its
free end (see Fig. 1a). Performing a graphical analysis of the applied force F = mg as a function of the
elongation x experienced by the spring, the elastic
constant k was determined. The oscillation period T
was recorded using a Vernier VPG-BTD photogate. To

Experimental study of simple harmonic motion of a spring-mass system as a function of spring diameter

measure T , a mass m = 0.182 kg was suspended at the

free end of the spring, and it was elongated by a length
x = 1.5 102 m from the equilibrium position, and
then it was allowed to oscillate freely. In order to obtain a more accurate value, the period T was obtained
by means of averaging ten measurements. By replacing this value in the expression = 2/T , the value
of the angular frequency was found. Using a digital
force sensor Vernier Dual Force DFS-BTA, the force F
of the spring-mass system was measured (see Fig. 1b).
Adjusting the envelope of the oscillation by means of a
decreasing exponential, the damping factor was determined.

4.

4305-3

Results and data analysis

Figure 2 shows the elongation x experienced by the

springs when the masses that exerted the external
forces F are suspended at their free end. We highlights
that in all gures presented in this paper the error obtained in the calculation or in the measurement process
are smaller that the point size. A close look shows that
the initial value of the applied force is not the same
for all springs, but increases as a function of the diameter . This is because to the mass m suspended
from the free end of the springs the fraction M /3 of
the springs mass M was added; springs having higher
diameter have greater mass M (see Table 1). This
issue was done to graphically highlight the correction
to the spring mass M /3, dened in Eq. (3), which becomes more noticeable as the spring diameter increases.
The above procedure does not alter the slopes of the
lines shown in Fig. 2, and allows seeing the eect of
the correction in order to take into account the springs
mass. It can clearly be observed in Fig. 2 that the
slopes of the straight lines that determine the elongation x produced by the applied force F increase as the
spring diameter decreases. This result implies that
the springs elastic constant, dened as k = F /x, is
greatest for springs with smaller diameters .

Figure 1 - (a) Photograph of the experimental setup used to measure the k and variables as a function of the diameter of the
spring. Note that the springs upper coil is set between two plates
in order to minimize external vibrations. (b) Experimental setup
used to measure the force F in the spring-mass system with a
Vernier sensor.
Table 1 - Diameters and mass M of the springsa,b used to study
the inuence of on the oscillations of a spring-mass system.
Spring

Diameter
Mass M
(0.01 102 m)
(0.01 103 kg)
1
0.82
12.87
2
0.96
15.31
3
1.03
16.17
4
1.20
19.31
5
1.28
20.83
6
1.36
21.87
7
1.44
23.36
8
1.54
25.09
a All springs have a natural length l = 10.1 102 m
0
(N = 124 coils).
b The spring coils are in contact, without separation h
between their coils (step h = 0).

Figure 2 - Applied force F as a function of the spring elongation

x taking as a parameter the spring diameter .

Figure 3a shows the variation of the elastic constant

k (full circles) as a function of the spring diameter .
In this gure, it can be seen that the value of elastic
constant k decreases as increases. The variation of
the eective spring mass M /3 (empty circles) as a function of the spring diameter is also shown in Fig. 3a.
Note that the eective mass M /3 of the spring increases
linearly with , as expected because the increases in
rise linearly with the length of the wires that make up
the spring, and then these linear increases in the length
of the wires produce a linear increase in the eective
mass M /3 of the springs. The linear t between the
springs elastic constant k and 3 , shown in Fig. 3b

4305-4

Triana y Fajardo

(which corresponds to a linear correlation coecient of

adjustment of 0.99), indicates that the results obtained
t perfectly with that established by Eq. (1).

Figure 3 - Variation of the springs elastic constant k (lled circles) and the eective spring mass M/3 (empty circles) as a function of the spring diameter (a), and linear t between the
springs elastic constant k and 3 (b).

The functional relationship between the springs

elastic constant k and the spring diameter is given
by
k=

3.16 105
,
3

(6)

5
where the quantity
/ 3.16 10 corresponds to the con4
stant value Gd 8N dened in Eq. (1), which involves the diameter d of the wire with which the spring
was manufactured, the number of coils N , and the
shear modulus
G. Taking into account that 3.16
/
105 = Gd4 8N and substituting the values of d and
N , a value of G = 72 109 N/m2 is found for the
shear modulus. This result is slightly lower than the
shear modulus reported in the literature for steel [13],
where G = 79 109 N/m2 . We believe that this difference is due to the fact that the value of G found
in our paper was obtained from a series of springs in

which the diameter varied, whereas the value of G

reported in the literature is for a straight wire. In the
springs design, a relationship known as spring index C
is dened, which is given by C = /d. Depending on
this value, a series of second-order approximations in
the calculation of the elastic constant is generated, as
shown in previous theoretical studies with regard to the
shear modulus G [14]. In our case, the spring indexes
are C = 10.1 and C = 18.9 for the spring of lowest and
highest diameter respectively. These values of the
spring index C suggest that as the diameter becomes
smaller, the higher-order terms in the calculation of the
elastic constant may have more inuence on the oscillatory behavior of the system.
Figure 4a shows a comparison in the behavior of the
angular frequencies , 0 and e as a function of the
spring diameter . The value was calculated by replacing the period T in the expression = 2/T , while
0 was calculated by means of Eq. (2). The e value
was determined through Eq. (3), in which the correction to the spring mass M /3 was performed. In all cases
the oscillations were induced by suspending a constant
mass m = 0.182 kg and applying an initial amplitude
of A0 = 1.5102 m. The results clearly show that the
frequencies , 0 and e are greatest for small values
of the spring diameter and decrease as increases.
This behavior is in complete agreement with the analysis performed in Fig. 3, since if the spring diameter
decreases, then the value of its elastic constant k will
be greater. As established by Eqs. (2), (3) and (5) the
frequencies , 0 and e are highest for the greatest k,
i.e. for springs with small diameter .
Another important issue that can be seen in Fig. 4a
is that the values of the frequencies are not equal, since
0 > e > . This behavior is due to the fact that
the 0 value corresponds to the frequency determined
in the absence of retarding forces, while the frequency
e includes the spring mass correction M /3. On the
other hand, corresponds to a real physical system
that experiences damped oscillations and compressive
forces that tend to keep the spring coils together, an
eect that is greater as the springs diameter becomes
smaller [15]. The t shown in Fig. 4b shows the linear relationship between the angular frequency 0 and
3/2 , with a linear correlation coecient of adjustment of 0.99. The functional relationship between the
variables in Fig. 4b is given by
0 =

0.013
B
= 3/2 ,
3/2

(7)

where by the substitution

of Eq. (1) into Eq. (2) it is

found that B = Gd4 /8 Nm. Taking the value of the

shear modulus G = 72109 N/m2 previously calculated
and replacing the values
d, N and m in Eq. (7), it is
obtained that B = Gd4 /8 Nm = 0.0132, giving us a
dierence of 1.5% as compared with the result obtained
in Eq. (7).

Experimental study of simple harmonic motion of a spring-mass system as a function of spring diameter

4305-5

the highest value of , i.e. for the spring with the smallest diameter . Also, the envelope of the oscillatory
curve is shown, which corresponds to the exponential
factor dened in Eq. (4), from which the value 1 is
obtained. For the springs with other diameters, only
the envelope of the oscillatory curve is shown. In all
cases, the experimental points, obtained by measuring
the force F as a function of the time t, were t according to the theoretical prediction given by Eq. (4),
relating the values i (i = 1, ..., 8) to the values organized from smallest to largest diameter, according to
the data summarized in Table 1. The results show that
for the system built with the spring of the smallest diameter, which has the highest damping factor 1 , the
amplitude of the oscillations decreases rapidly in a time
of t 300 s, while the systems built with springs with
lower damping factors i require a much longer time for
the oscillatory motion to vanish.

Figure 4 - Comparison of the behavior of the angular frequencies

, 0 and e as function of the spring diameter (a). Note
the linear t between the natural frequency 0 and 3/2 and
an apparent linear relationship between the angular frequency
and 3/2 (empty triangles), which is not correct according to
Eq. (5) (b).

In Fig. 4b, the angular frequency as function

of the spring diameter 3/2 is also shown, and it is
possible to think that an apparent linear relationship
between the variables and 3/2 exists. However,
we recall that is a value determined experimentally
from measurements of period T of the oscillations, and
therefore implicitly has the eect of damping. Consequently, in agreement with that established by Eq. (5),
cannot be in a linear dependence with 3/2 .
The oscillatory dynamic of the spring-mass system
addressed in this paper corresponds to a real physical system, where the amplitude of the oscillations decreases gradually and disappears after a certain interval of time. Consequently, there is a damped harmonic
motion around an equilibrium position with decreasing amplitude, in which the action of non-conservative
forces leads to the dissipation of energy into the medium
in which the system is immersed. Figure 5a shows the
damping experienced by the spring-mass system with

Figure 5 - Amplitude A of the oscillations as a function of time

t for the spring with smallest diameter and the highest damping factor 1 . For the other springs, only the envelope of the
oscillations is shown and the damping factor i (i = 1, ..., 8) is
indicated, which corresponds to the springs arranged from the
smallest to largest diameter (a). Amplication of the amplitude
A as a function of time t for the spring-mass system with damping
factor 1 (smaller diameter ). The black dots in the oscillation
correspond to the experimental values and the continuous curve
to the theoretical t (b).

4305-6

Triana y Fajardo

Figure 5b shows an amplication of the oscillations

shown in Fig. 5a for the spring with the smallest diameter (i.e. the system with the largest ). The black
points in the oscillation correspond to the experimental
values and the solid curve to the theoretical t obtained
by means of Eq. (4). An analysis of the results shows
a correspondence of 99.9% between the experimental
data t with respect to the values obtained theoretically by using Eq. (4). It is found that the amplitude
A of the oscillations (in meters) decreases in time according to the relationship

A = (1.5 102 )e

i t
2

(8)

where the quantity A0 = 1.5 102 m corresponds to

the initial amplitude provided to the spring-mass system to induce the oscillations.

Figure 6 - Variation of the damping factor of the oscillations

as a function of the springs diameter .

Figure 6 shows the variation of the damping factor

as a function of the spring diameter . Note that the
damping factor given by Eq. (8) is greater for small
values of the springs diameter and decreases as increases. This can be explained by taking into account
that the springs with smaller diameter have a larger
elastic constant k, and therefore have higher damping
factors . It is important to clarify that the spring
constant is independent of the damping factor, because
if, for example, we take the same spring-mass system
and make the mass oscillate in two dierent media, two
dierent values for the damping constant would be
obtained, but the spring constant would be the same.
The results obtained show that the springs diameter
considerably inuences the behavior of the k, , and
variables of the simple harmonic motion. Note that
these variables decrease as the springs diameter increases, as shown in Figs. 3, 4 and 5, in accordance
with predictions that can be made from Eqs. (1), (2),
(3), (4) and (5). An issue to note in Fig. 6 is that at
rst glance there is an apparent linear relationship between the damping factor and the spring diameter .
However we believe that not is appropriate to make this
generalization, because the obtained correlation coecient of the linear adjustment is 0.86 which is far form
the ideal value.
Figure 7a shows the behavior of the angular frequencies , 0 , and e as a function of the springs elastic
constant k. Observe that the frequencies increase as
the k value increases (or likewise as decreases), as is
established by Eqs. (2), (3) and (5). Again we emphasize that the dierences in the values of the frequencies
(0 > e > ) are because 0 corresponds to the natural frequency, while e includes the correction to the
spring mass M /3 and corresponds to a damped oscillation under the eect of dissipative forces.

Figure 7 - Behavior of the angular frequencies , 0 and e as

a function of the springs elastic constant k (a). The linear t
between the natural frequency 0 and k1/2 (full squares) and
an apparent
linear relationship between the angular frequency

and k (empty triangles), which is not correct, according to

Eq. (5) (b).

Experimental study of simple harmonic motion of a spring-mass system as a function of spring diameter

The linear t shown in Fig. 7b shows that

the natural frequency 0 is directly proportional to k (full
squares), as is established by Eq. (2). The linear correlation coecient of adjustment is 0.99, and allows one
to determine that the functional relationship between
the variables is given by

0 = 2.34 k,
(9)

where the value 2.34 = 1/ m. Replacing the mass

m = 0.182 kg, suspended from
the free end of the
spring, we obtain a value of 1/ m = 2.344, giving us
a dierence of 0.2% as compared with the result obtained in Eq (9). It is important to note, in Fig. 7b,
that it gives the impression that there exists a linear

relationship between the angular frequency and k

(empty triangles); nevertheless this interpretation is not
correct, since in agreement with Eq./ (5) the angular
frequency must involve the term 2 4, which is associated with the damping of the system.
Figure 8a shows the graphical representation of
2
Eq. (5), i.e.
/ the behavior of as a function of
2
(k/m) ( 4), which is plotted taking into consideration the experimental values of , k and . Note
that the functional relationship between the variables
is linear, with slope 0.99 and intercept 0.9. These results are very close to those established by Eq. (5),
where the slope and intercept are equal to 1 and 0, respectively. The small dierences between experimental
and theoretical values may be due to measurement uncertainties or to the inuence of small eects that have
not been taken into account in Eq. (5), such as the
compression eects on the springs coils [16]. It is important to note that the , k, and variables shown in
Fig. 8a have all been determined experimentally, and
show a linear correlation coecient of adjustment between them of 0.99. Figure 8b shows that the value of
the angular frequency increases as the damping factor increases. This behavior is due to the fact that
the springs that have a higher damping factor have a
higher elastic constant k, and therefore oscillate with
higher angular frequency. This result can be corroborated taking into account that the values of the springs
elastic constant k correspond to powers of 101 , while the
values of the damping factor correspond to powers of
102 . This proves
/ that in our experiment the quantity;
(k/m) >> ( 2 4), and therefore Eq. (5) is always fullled taking positive values of the angular frequency .

5.

4305-7

tic constant k, the angular frequency and the damping factor . The experiment is also very instructive
for comparing the results of dierent models for nding
the angular frequencies of the spring-mass oscillator,
which involve the massless spring ideal approximation,
a spring mass correction M /3, and the damping of the
oscillations. This experiment, which uses low-cost materials, can be utilized so that through graphical analysis students can nd a great variety of functional relationships between the variables that characterize the
simple harmonic motion of a spring-mass system, making it an excellent practice or project for physics laboratory courses at the undergraduate level. One suggestion for carrying out this experiment is to use plastic
springs such as are used in book binding, since they are
available in a wide variety of diameters. Also, similar
studies for other oscillating systems can be carried out.

Conclusions

The experimental study of simple harmonic motion of

a spring-mass system shows that the principal physical
variables that characterize the oscillations, such as k, ,
0 , e , and , are strongly inuenced by the springs diameter . The results obtained indicate that decreases
in the springs diameter lead to increases in the elas-

Figure 8 - Graphical representation of Eq. (5), behavior of 2

as a function of (k/m) ( 2 /4) (a). Angular frequency as a
function of the damping factor (b).

4305-8

References
[1] A.P. French, Vibrations and Waves (Norton, New
York, 1964).

Triana y Fajardo

[9] T.W. Edwards and R.A. Hultsch, Am. J. Phys. 40, 445
(1972).

[2] S.Y. Mak, Am. J. Phys. 55, 994 (1987).

[10] R. Resnick, D. Halliday and K.S. Krane, Physics for

Students in Science and Engineering (New York:Wiley,
1992).

[3] F.W. Sears, Am. J. Phys. 37, 645 (1969).

[11] M. Leclerc, Am. J. Phys. 55, 178 (1987).

[4] J.M. Bowen, Am. J. Phys. 50, 1145 (1982).

[5] F.A. McDonald, Am. J. Phys. 48, 189 (1980).

[12] S.F. Crawford, Waves, Berkeley Physics Series (New

York: McGraw-Hill, 1968), v. 3.

[6] P. Ferreira e J.P. Rino, Revista Brasileira de Ensino de

Fsica 28, 195 (2006).

[13] H. Eskandari, Mater. Des. 24, 143 (2003).

[7] C.A. Triana and F. Fajardo, Eur. J. Phys. 33, 219

(2012).
[8] J.E. Shigley and C.R. Mischke, Mechanical Engineering Design (McGraw-Hill, New York, 1989), 5th ed.

[14] P. Mohazzabi and J.P. McCrickard, Am. J. Phys. 57,

639 (1989).
[15] P. Froehle, Phys. Teach. 37, 369 (1999).
[16] J.T. Cushing, Am. J. Phys. 52, 925 (1984).