See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/259234790

Circular motion

Article  in  American Journal of Physics · July 2000

DOI: 10.1119/1.19501

CITATIONS READS
10 21,698

2 authors, including:

Richard Conn Henry

Johns Hopkins University
307 PUBLICATIONS   4,051 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Richard Conn Henry on 05 June 2015.

The user has requested enhancement of the downloaded file.

American Journal of Physics 68, 637, 2000 July

Circular Motion
Isaac Newton1 and Richard Conn Henry2

Abstract: An extraordinarily simple and transparent derivation of the formula for the
acceleration that occurs in uniform circular motion is presented, and is advocated for use
in High School and College Freshman physics textbooks.

PACS Numbers: 01.65, 02.90

The familiar formula for the acceleration that occurs in uniform circular motion,

v2
a= (1)
r

may be, today, a staple of freshman physics, but just 350 years ago it represented the
cutting edge of physics research. Its introduction marked the real beginning of the
mathematization of physics. How did it arise? And most importantly, how should it be
taught to students today?

The recognition that uniform circular motion does involve acceleration began with
Galileo and Descartes; however, “their handling of the problem was remarkably vague,
and their qualitative, sometimes fuzzy, explanations of circular motion allowed them to
ignore basic flaws in their respective systems of the world.” 1

The person who solved the problem was Christiaan Huygens in his book “Horologium
Oscillatorium” (1658; second edition in 1673), the actual proof only appearing in
Huygens’ posthumous book “De Vi Centrifuga” (1703). The effect of his discovery was
immense: Edmund Halley, Christopher Wren, and Robert Hooke were all able
immediately to substitute the result into Kepler’s third law and deduce that the
gravitational force must vary inversely as the square of the distance from the sun. Their
method produces the greater part of the law of gravitation, as I now show:

Consider two planets, masses m and M, that are in circular orbits and are distant from the
sun r and R, and have years of length t and T. According to Kepler’s Third Law,

T 2 R3
= 3 (2)
t2 r

1
Deceased
2
Henry A. Rowland Department of Physics and Astronomy, The Johns Hopkins University
([email protected])
 Isaac Newton and Richard Conn Henry

while Huygens provided

f  mv 2   R  m R v 2 m R 4π 2 r 2 T 2 mrT 2
=  = = = (3)
F  r  MV 2  M r V 2 M r t 2 4π 2 R2 MRt 2

(where I have also used the definition of the period, t = 2πr/v). Combining these two
results gives
 m
f mrR 3
 r2 
= = (4)
F MRr 3  M 
 R2 

m m
which yields f ∝ 2
or f = GΜ 2 using Newton’s later value for the constant of
r r
proportionality. Given this, is it any wonder that Hooke felt, to the end of his days, that
he had discovered the law of gravitation? But while none of the three could deduce
Kepler’s laws from their result, Newton could and did. Halley and Wren, properly
recognizing the genius of Newton, never made any claim of discovery.

Which, indeed, brings us to Newton. Did Newton know of Huygens’ formula? Hooke,
Halley, and Wren did, but Newton was intellectually more isolated in Cambridge, which
in those days was a backwater compared with London. Perhaps it still is!2 Newton did
obtain a copy of Horologium, from Henry Oldenburg, shortly after its publication. In any
case, Newton independently obtained the formula, and in a way that is much superior to
that employed by Huygens.

Huygens’ method will be discussed briefly below, as a matter of historical interest. The
main aim of my article, however, is to present Newton’s method so that it can be used in
effectively teaching the origin of Eq. 1 to the modern young student of physics.

The essential problem that all the early physicists faced, was the need to work a vectorial
problem, not having a real theory of vectors available. What was available to them was
the parallelogram law for combining vectors in two-dimensional problems; that result is
as old as Aristotle.3

If one does have vector methods, there are modern methods for obtaining Eq. 1 that are
delightful.4 But Eq. 1 is so crucial, that one does not want the beginning physics student
to think that it cannot be obtained except by sophisticated methods, with which the
student is only beginning to become familiar. That is one reason why a clear exposition
of Newton’s wonderful method is pedagogically valuable.

American Journal of Physics 68, 637, 2000 July Page 2

 Circular Motion

Newton’s Method

The method used by Newton, about 1665, is described by Westfall,5 and has been
analyzed in detail, from an historical perspective, by Erlichson.6 My aim, here, however,
is not historical, it is to present the method as it can and should be presented to the
modern student.

(a)
v
v cos θ θ θ

(b)

v cos θ
θ v

Fig. 1. A particle moving with velocity v strikes a hard surface at A, rebounding at the same angle θ with
which it struck. The component of its velocity in the horizontal direction is always v sin θ, and the particle
always moves with speed v.

The starting point is the resolution of forces into components (parallelogram law). As we
shall see, that is really all that is needed; and that is nice, because teaching students to
resolve forces into components is of course basic and necessary in any case, and the
present application gives the students a tremendous reward, quickly, for mastering the
notion.

In Figure 1(a), I show a particle that is moving in a straight line with constant speed v and
which will soon hit a hard surface at A. The component of the velocity v that is
perpendicular to the surface is v cos θ. In Figure 1(b), the particle has undergone the

American Journal of Physics 68, 637, 2000 July Page 3

 Isaac Newton and Richard Conn Henry

impact, and the component of the velocity that is perpendicular to the surface is again
v cos θ, but it is now directed oppositely, so that the total change in velocity is

∆v = 2v cosθ (5)

(the component of the velocity that is parallel to the surface, v sin θ, not having changed
in the impact). It is in the change of direction, that the change of velocity enters. And the
change in velocity, we see, is in a direction that is normal to the surface struck.

And that is it! All we need to do now, is to apply this result, as Newton did, to the case
of a particle that is crashing around a circle (Fig. 2). The figure shows the case “circle
traversed in n = 12 impacts,” but of course Newton and I are going to let n go to infinity
and ϕ go to zero very soon.

Fig. 2. A particle bounces around a circle, hitting the circle a total of 12 times, and each time bouncing at
an angle θ to the normal to the circle.

It should be emphasized that what is important is the change in velocity, not what
produces that change. Whether the change is produced by a crash with a wall, or by
gravitation, the change in the velocity is the change in the velocity.

The first point we note is that the change in velocity is always directed toward the center
of the circle (i.e., is always normal to the surface struck) at each impact.

We have seen that the change in velocity due to each impact is ∆v = 2 v cos θ, so the sum
of the fractional changes that result from the n impacts is

American Journal of Physics 68, 637, 2000 July Page 4

 Circular Motion

Σ ∆v = 2 nv cosθ (6)

But n ϕ = 2π and ϕ + 2θ = π so

2π π ϕ 4π ϕ 4π ϕ
Σ ∆v = 2 v cos  −  = v sin   ≅ v = 2π v (7)
ϕ  2 2 ϕ  2 ϕ 2

where in the penultimate step I have recognized that for small angles the sine is
approximately the angle; when n is infinity, and ϕ is zero, our result is exact.

So, we conclude that Σ ∆v = 2π v in going around a complete circle, whatever may be the
cause of the circular motion.

It should be pointed out again to the students that although the speed of the particle never
changes, always being simply v, because of the change of direction in going around the
circle, the velocity does change, every time the particle bounces.

We are now in a position to compute the acceleration, which is what we want. Because
all the bounces are the same, we can calculate the acceleration either from the velocity
change in any one bounce, or (as Newton did) from the sum of the fractional changes.
Newton’s approach gives

∆v Σ ∆v 2π v v2
a= = = (8)
∆t T  2π r  r
 v 

where T is the time to go around the circle. So we have our desired result!

We have already noted that the change in velocity is always in the direction perpendicular
to the surface struck, that is, in the direction of the center of the circle. So we have found
both the magnitude and the direction of the acceleration for circular motion.

Notice how unsophisticated all of the steps have been! That is the great virtue of the
method.

It is clear that Newton recognized from the symmetry of the problem that the average
magnitude of the acceleration (as we just calculated it) would be the same as the
instantaneous value of the same quantity. There is some virtue, however, in directly
calculating the instantaneous valuein fact, some instructors may find this approach
pedagogically preferable. With our same trigonometric substitution, the “single bounce”
change in velocity yields ∆v = 2v cosθ ≈ vϕ for the case of many bounces (i.e., for the

American Journal of Physics 68, 637, 2000 July Page 5

 Isaac Newton and Richard Conn Henry

ϕr
case that ϕ is small). If ∆t is the time to go through (small) ϕ, then ∆t ≈ and the
v
instantaneous acceleration

∆v vϕ v2
a= = = (9)
∆t  ϕ r  r
 v 

which is equal to our previous value.

Huygens’ method

Huygens considered a particle that is thrown sideways, which he knew would fall in a
parabolic curve. He constructed the largest circle that passes through the original position
of the particle, and yet does not cut the parabola. The method is not elegant. Newton, in
his second approach to the same problem, like Huygens considered a comparison with
motion under gravitation.

This alternate approach is presented elegantly and concisely by Tipler,7 Tipler’s

exposition requiring no vector concepts. However, these proofs by Huygens; by Newton;
and by Tipler, do mix the general concept of circular motion with the particular question
of motion under gravity, and they also do require the student to know that for constant
acceleration a, the distance traveled in time t, is at2/2. In contrast, Newton’s first method
(which I have presented above) has neither of these complications, and is therefore
strongly to be preferred for the student’s first introduction to the acceleration that is
associated with circular motion.

Acknowledgements

Both authors have published once previously in the American Journal of Physics.8 The
junior author in 1997 was Keeley Visiting Fellow at Wadham College, Oxford, site of
Christopher Wren’s “astronomy chamber,” where the Royal Society was conceived; he
thanks Wadham College for its kind hospitality. This work was supported by Maryland
Space Grant Consortium.

References

1. J. G. Yoder, Unrolling Time: Christiaan Huygens and the mathematization of nature

(Cambridge University Press, Cambridge, 1988), p. 33.

2. D. Clery, “London, Cambridge Lead Europe in Output,” Science 281, 1127 (1998).

3. M. Kline, Mathematical Thought from Ancient to Modern Times (Oxford University

Press, Oxford, 1972), p. 776.

American Journal of Physics 68, 637, 2000 July Page 6

 Circular Motion

4. F. Ninio, “Acceleration in uniform circular motion,” Am. J. Phys. 61, 1052 (1993);
K. R. Brownstein, “A simple derivation of centripetal acceleration,” Am. J. Phys. 62,
946 (1994); U. Haber-Schaim, J. H. Dodge, and J. A. Walker, 1981, PSSC Physics,
5th edition (D. C. Heath & Co., Lexington, MA, 1981, pp. 52-54; D. W. Kraft, and
L. Motz, “Geometric derivation of radial acceleration magnitude,” Phys. Teach. 33,
565 (1995).

5. R. S. Westfall, Never at Rest: A Biography of Isaac Newton (Cambridge University

Press, Cambridge, 1980), p. 149.

6. H. Erlichson, “Motive force and centripetal force in Newton’s mechanics,” Amer. J.

Phys. 59, 842-49 (1991).

7. P. A. Tipler, Physics for Scientists and Engineers, 3rd edition (Worth, New York,
1991), p. 66.

8. I. Newton, “A new theory about light and colors,” Amer. J. Phys. 61, 108-112 (1993);
R. C. Henry, “Quantum mechanics made transparent,” Amer. J. Phys. 58, 1087-1100
(1990).

American Journal of Physics 68, 637, 2000 July Page 7

View publication stats