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

2, 2504 (2013)
www.sbfisica.org.br

An experimental verification of Newton’s second law

(Uma verificação experimental da segunda lei de Newton)

Roberto Hessel1 , Saulo Ricardo Canola, Dimas Roberto Vollet

Departamento de Fı́sica, IGCE, UNESP, Caixa Postal 178, 13506-900, Rio Claro, SP, Brazil
Recebido em 6/7/2012; Aceito em 11/2/2013; Publicado em 24/4/2013

Descrevemos nesse trabalho um procedimento experimental para investigar a validade da segunda lei de
Newton. A montagem experimental utilizada permite acelerar um carrinho sobre um trilho de ar por meio de
forças constantes e conhecidas. Mostramos também como determinar a aceleração a partir de velocidades médias
calculadas para intervalos de tempo sucessivos do movimento usando vários contadores eletrônicos conectados
a um único circuito oscilador a cristal. Dentro dos erros experimentais, os experimentos realizados mostram
claramente a proporcionalidade entre aceleração e força para uma massa constante e entre aceleração e o inverso
da massa para uma força constante.
Palavras-chave: segunda lei de Newton, medida de intervalo de tempo, medida de aceleração, velocidade média.

We describe an experimental procedure to probe the validity of Newton’s second law. The experimental
arrangement allows us to accelerate a glider on an air track by means of forces that are both steady and known.
We also show how to determine acceleration from average speeds calculated for successive time intervals of the
motion measured by using several electronic counters connected to a single-crystal oscillator circuit. Within
experimental errors, the experiments clearly show the proportionality between acceleration and force for a fixed
mass and between acceleration and inverse of mass for a fixed force.
Keywords: Newton’s second law, measurement of time interval, measurement of acceleration, average speed.

1. Introduction cond law in a quick and convincing way.

To introduce Newton’s second law or concepts such as

force, inertial and gravitational mass and weight, it is a 2. Experimental set-up
common practice to use the approach offered by PSSC
Physics that, according to Arons [1], is quite reasonable The experimental set-up consists of a glider on an air
for introductory levels. In the PSSC context, Newton’s track connected by a string passing over a small pulley
second law of motion is investigated in the laboratory, to a hanging load of mass m and weight mg. We consi-
with carts, times, and a rubber loop stretched a cons- der the glider and the load as a single object, subject to
tant amount as the unit of force [2]. The choice of stret- the accelerating force mg. To show that the acceleration
ched elastics to accelerate a cart on a level table is quite of the system is proportional to the acceleration force
suitable as the starting point because it takes into ac- when the total mass is kept constant, we begin with a
count the intuitive notion of force related to a sensation hanging load of mass m and add four identical metallic
of muscular effort but, due to the difficulty of keeping discs of mass m to the glider of mass M (Fig. 1) There-
the rubber loop stretched a constant amount as the cart fore, the accelerating force mg acts on a system of total
accelerates, the quantitative results are not always con- mass M + 5m. To double the accelerating force, one
vincing. Indeed, only the more attentive pupils obtain disc is transferred from the glider to the hanging load.
satisfactory results [3]. For this reason, when we follow To triple the force, two discs are transferred from the
the sequence outlined in PSSC, we complement the la- glider to the hanging load, and so on [4]. To show that
boratory activity with an experimental demonstration the acceleration of the system is inversely proportional
that allows confirmation of the validity of Newton’s se- to its mass when the accelerating force is kept constant,
1 E-mail: fisica@rc.unesp.br.

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

2504-2 Hessel et al.

we change the mass of the system by loading the glider the S1 , S2 , . . . S5 outputs are LOW (L) and the state
with mass of different sizes or connecting another glider at each CL EN input is H due to the presence of the
to the original. NOT gate. Consequently, all counters are blocked be-
cause an H level on the CL EN input inhibits the clock’s
operation. When the clock’s input of the 4017 receives
the first pulse, the high state is transferred from S0 to
S1 and the first counter starts the timing. When the
second pulse arrives, the state at S1 changes from H to
L and S2 goes to H. Then, the first counter stops the
timing and the second starts. Finally, when the sixth
Figura 1 - A simplified drawing of the air track showing the han- pulse arrives, S5 goes from H to L and the fifth coun-
ging load and the glider loaded with metallic discs. ter stops the timing, while S0 goes to H again (shining
The acceleration can be determined from the ave- LED) because the S6 output is connected directly to
rage speeds calculated for successive time intervals of reset input.
the motion. For a question of availability and cost, we The pulses that arrive at the clock’s input of the
have measured time intervals by using electronic coun- 4017 are generated during the passage of posts trans-
ters in conjunction with a single-crystal oscillator cir- ported by the glider through a photogate. The glider
cuit operating at 1 kHz and a photogate [5]. The timing carries six posts, evenly spaced on a wooden ruler fixed
circuit is shown in Fig. 2. to it (Fig. 3).

Figura 3 - A simplified drawing of the glider carrying six evenly

spaced posts.

So, after the posts have passed through the photo-

gate, the counters record, in ms, the time intervals τ 1 ,
τ 2 , . . . τ 5 indicated in Fig 3. If the distance between
Figura 2 - The timing circuit.
successive posts is d, the average speed of the glider in
the time interval τ 1 is v̄1 = d/ τ1 , in τ 2 is v̄2 = d/τ2 ,
The counters have two inputs: the clock (CK) and etc. The set of these values may provide information
the clock enable (CL EN). The first receives the rectan- regarding the motion of the glider.
gular pulses sent by the oscillator and the other enables
the counting process when it is held at ground state.
When the logic state at this input is high, the counting
stops. The heart of the circuit is the 4017. The 4017 is 3. Analysis of the data
a decade counter with ten outputs that go to HIGH (H)
in sequence when a source of pulses is connected to the 1
∫t2
Consider the expression v̄ = ∆t vdt, where ∆t =
clock input and when suitable logic levels are applied t1
to the reset and enable inputs [6,7]. t2 − t1 . Expanding the integrand into a Taylor series
Briefly, the electronic circuit (Fig. 2) works as fol- about t = t̄ [8], i.e. about the midpoint of the interval,
lows. If the logic state at the S0 output is initially H, we obtain
An experimental verification of Newton’s second law 2504-3

same thing with the remaining discs. The time inter-

vals recorded on the counters after each run are shown
∫ /2[
t̄+ ∆t
1 I in Table 1.
v̄ = v (t̄) +
∆t I Tabela 1 - Time intervals recorded on the counters for five han-
t̄−∆t/2 ging loads.
]
1 2
a (t̄) (t − t̄) + ȧ (t̄) (t − t̄) + . . . dt, (1) Hanging load (g) ∆t (ms)
2! τ1 τ2 τ3 τ4 τ5
  50 406 297 245 214 193
 d2 v  100 292 212 175 152 136
where a (t̄) = dvdt t=t̄ , ȧ ( t̄) = dt 
2 and so on.
150 242 173 142 124 111
t=t̄
Setting τ = t − t̄ and integrating the series term by 200 220 154 126 109 97
term, the Eq. (1) can be written as 250 159 122 103 92 82

Knowing that the distance between two consecutive

∆t/ ∆t/ posts is 12.00 cm, we can determine the average speed
a (t̄) τ 2  2 1 ȧ (t̄) τ 3  2 of the glider at each time interval.
v̄ = v (t̄) + · + · +... .
∆t 2 −∆t/2 2 ∆t 3 −∆t/2 For instance, Table 2 shows the average speed cal-
(2) culated for the first run.
The terms in τ 2 , τ 4 , . . . are all zero, so that the
Tabela 2 - Average speeds calculated in five successive time in-
Eq. (2) reduces to tervals for a hanging load of 50 g.

2 4 ∆t (ms) t = t̄(ms) v̄(cm/s)

1 (∆t) 1 ... (∆t) 406 203 29.6
v̄ = v (t̄) + ȧ (t̄) + a (t̄) + ... (3)
2 12 24 80 297 554.5 40.4
245 825.5 49.0
By inspection of Eq. (3), we conclude immedia- 214 1055.5 56.1
tely that, for constant acceleration a, v̄ = v (t̄); a well- 193 1258.5 62.2
known result. But the converse is also true: if for any In Fig. 4 we have drawn v̄ versus t (dashed line) and
∆t, v̄ = v (t̄), then the acceleration a must be cons- v (t̄) versus t (solid line), assuming that v (t̄) = v̄. As
tant [8]. In fact, if v̄ = v (t̄) for any ∆t, the sum of the all points corresponding to v (t̄) = v̄ lie on the straight
2 4
terms in (∆t) , (∆t) , . . . at the right side of the Eq. (3) line, the acceleration of the glider is constant and is
must be zero for any ∆t. This condition is fulfilled if given by the slope of this line.
...
and only if ȧ (t̄) = a (t̄) = ... = 0, which only happens
when a is constant. So, to analyze the motion of the
glider, we assume initially that v (t̄) = v̄ for any ∆t. If
this is true, then the graph of v (t̄) versus t is linear and
the slope of the straight line is the acceleration of the
glider.
Evidently, the acceleration can also be obtained by
analyzing the distance traveled by the glider as a func-
tion of time [9], but, in view of the definition of acce-
leration, we prefer to use the method described above
because it involves change in speed and time interval.

4. Applications
4.1. Experiment 1 - Relation between accele-
ration and accelerating force for constant Figura 4 - Speed-versus-time curves for a system (glider + han-
total mass ging load), subject to a constant force. The dashed line refers to
the average speed v̄ whereas the solid line refers to the instanta-
The glider shown in Fig. 3 was loaded with four metal- neous speed v (t̄) at the midpoint of each time interval. Linear
fit: v = 23.28 + 30.99 × 10−3 t with t in ms.
lic discs having a mass of 50 g each and was connected
to a hanging load, also weighing 50 g, by a string pas- Proceeding in the same way with the remaining rows
sing over a small pulley. The system (glider + hanging in Table 1, we can determine the acceleration for the
load) with mass 1502 g was then released. At the end other runs. Figure 5 shows the acceleration a as a func-
of the run, the counters recorded the time intervals τ 1 , tion of the hanging load mass mhl . As expected, the
τ 2 , . . . τ 5 . Afterwards, one of the discs transported graph tells us that for a fixed mass, a is proportional
by the glider was transferred to the hanging load and to mhl or the acceleration force since the weight (acce-
the procedure above was repeated. We carried out the leration force) acting on the load is proportional to its
2504-4 Hessel et al.

mass. The fact that the straight line crosses the force mT when the acceleration force is kept constant. The
axis slightly to the right of the origin can be attributed slope of the straight line, equal to 9.53 × 104 dynes,
to the presence of friction forces. Using g = 976 cm/s2 , corresponds to the acceleration force. Within experi-
the mass of the system, (slope of straight line/g)−1 , is mental error, the agreement between this value and the
equal to 1518 g. This value is in reasonable agreement one calculated (100 g × 976 cm/s2 = 9.76 × 104 dynes)
with the value 1502 g measured. is very satisfactory.

Figura 5 - Acceleration of the system (glider + hanging load) as Figura 7 - Acceleration versus the inverses of the system’s mass
a function of the hanging load for a system of constant mass. for a constant acceleration force. Linear fit: a = −1.26 + 9.53 ×
Linear fit: a = −1.594 + 0.643 mhl . 104 m−1T .

4.2. Experiment 2 - Relation between accelera- 5. Conclusions

tion and mass for a constant accelerating
force We have described a way of showing effectively and
quickly the proportionality between acceleration and
This experiment was done using two gliders; one of
force for a fixed mass and between acceleration and in-
them having a length of 24.7 cm and the other 37.5 cm.
verse of mass for a fixed force, as predicted by Newton’s
They were used either coupled together or separately.
second law. In addition, the technique that we have
To increase the mass of the system, we also fastened
used to determine the acceleration of the moving ob-
metal bars to both sides of the gliders (Fig. 6).
ject affords a good opportunity to discuss certain ques-
tions concerning the average speed and instantaneous
speed, which have little chance of being treated in a la-
boratory class. The experiments proposed are easy to
perform and are appropriate for both undergraduate la-
boratories and demonstration in class lecture, since the
students have already acquired some level of familiarity
Figura 6 - Air track glider with metal bars fastened to both sides with basic concepts such as mass and weight.
of the glider.

The mass of the hanging load was fixed at 100 g. Acknowledgement

Table 3 shows the total mass mT used for each run and
the corresponding acceleration a. The authors thank to Rosana A. Gonçalves Pesce for
valuable discussions during the preparation of the ma-
Tabela 3 - Mass of the system (glider + hanging load) and the nuscript.
corresponding acceleration for a hanging load of 100 g.

mT (g) a (cm/s2 ) Referências

593.3 160.9
791.7 117.3 [1] A.B. Arons, Teaching Introductory Physics (John Wi-
941.9 98.9 ley & Sons, Inc, New York, 1997), p. 65-66.
1502 62.6
1878 50.3 [2] PSSC, Physical Science Study Committee, Teacher’s
Resource Book and Guide Part III (Heath, Boston,
As expected, the graph of acceleration a versus 1967), 2nd ed., Experiments III-1 and III-2.
1/mT (Fig. 7) shows that a is inversely proportional to [3] H.S. Fricker, Phys. Educ. 29, 386 (1994).
An experimental verification of Newton’s second law 2504-5

[4] E.M. Rogers, Physics For The Inquiring Mind (Prin- [7] R. Hessel, C.S. de Oliveira, G.A. Santarine, D.R. Vol-
ceton University Press, New Jersey, 1973), p. 111. let e A.C. Perinotto, Revista Brasileira de Ensino de
[5] R. Hessel, C.S. de Oliveira, G.A. Santarine e D.R. Vol- Fı́sica 30, 1502 (2008).
let, Revista Brasileira de Ensino de Fı́sica 30, 1501
[8] B. Denardo, S. Wong and A. Lo, Am. J. Phys. 57, 528
(2008). Erratum: Revista Brasileira de Ensino de Fı́sica
(1989).
31, 4901 (2009).
[6] See http://www.doctronics.co.uk/4017.htm [9] H. Cohen and D. Horvath, Am. J. Phys. 56, 950 (1988).