betaFeedback

WorksPhysics and Astronomy

Elementary derivations of centripetal acceleration and works of Huygens and Newton: a revisit with a simple physical derivation

Abstract

This work shows how two simple derivations of the centripetal acceleration are fundamentally related to seminal works of Huygens and Newton. A different, more physically motivated derivation is then given.

The continued interest in elementary (no calculus, no vector, no trig function) derivations[3, 4, 5, 6, 7, 8, 9, 10] [1, 2] of the centripetal acceleration, seems not just an effort to teach a difficult concept to beginning physics students, but actually reflects how circular motion was originally understood by founders[11, 12, 13] of classical mechanics. It seems pedagogically enriching that this historical connection should be made more widely known among students and teachers of centripetal acceleration, especially when these elementary derivations are used.

The works of Huygens and Newton on centripetal acceleration are of course well known to historians.[11, 12, 13] However, they are not aware or interested in modern derivations. For beginning students of physics, Ref.8 has given an excellent historical introduction on centripetal acceleration. This work goes beyond Ref.8’s discussion by providing precise details on Newton’s[11, 12, 13] and Huygens’[12, 13, 14] contributions.

The simplest derivation of the centripetal acceleration

a=v2r(1)a = \frac{v^2}{r} \tag*{(1)}

is undoubtedly the two circle method,[5] which has been independently derived numerous times.[3, 6, 7, 8] The above acceleration follows when the angular velocity 2π/T2\pi/T in the circle traced out by the position vector

v=2πrT(2)v = \frac{2\pi r}{T} \tag*{(2)}

is substituted into the circle traced out by velocity vector

a=2πvT.(3)a = \frac{2\pi v}{T}. \tag*{(3)}

However, despite its simplicity, (3) is usually not invoked in textbooks because a circle in “velocity space”[5] was deemed too abstract for most students. Below we will first show that Newton already knew about the velocity circle (3) from his earliest study of circular motion.

Since constant angular velocity is the key, acceleration (1) can also be derived from any arc length of the circle in time Δt\Delta t via

v=θrΔt,a=θvΔt.(4)v = \frac{\theta r}{\Delta t}, \qquad a = \frac{\theta v}{\Delta t}. \tag*{(4)}

This is similar to methods used in most textbooks, by taking the limit Δt→0\Delta t \to0. However, the generalization of Newton’s Waste Book construction[12] shows that the acceleration (1) would remain the same if one replaces the arc length by its corresponding chord. This is shown in Figure 1.

Figure 1. Newton’s polygonal orbit with triangle ABOABO similar to C′CBC'CB because θ+2ϕ=180∘\theta+ 2\phi= 180^\circ.

In Figure 1, one seeks to find the acceleration a\mathbf{a} such that the change in velocity Δv=aΔt\Delta\mathbf{v} = \mathbf{a}\Delta t will produce a polygonal orbit of uniform speed vv on a circle of radius rr. Starting at AA, in time Δt\Delta t and uniform speed vv, the particle moves to BB along the chord AB=vΔtAB = v\Delta t with center angle θ\theta. If there were no force acting on it, it would move to C′C' in the next time interval unimpeded. However, if it were to land on CC, also on the circle, of equal chord length vΔtv\Delta t from BB, then there must be a change of velocity aΔt\mathbf{a}\Delta t at BB, toward the center, resulting in a change of displacement from C′C' to CC of length (aΔt)Δt(a\Delta t)\Delta t. Since AB=vΔtAB = v\Delta t is the base of an isosceles triangle AOBAOB with two side lengths rr, and C′C=(aΔt)ΔtC'C = (a\Delta t)\Delta t is the base of a similar isosceles triangle C′BCC'BC with two side lengths vΔtv\Delta t, both must have the same base to side ratio

a(Δt)2vΔt=vΔtr,(5)\frac{a(\Delta t)^2}{v\Delta t} = \frac{v\Delta t}{r}, \tag*{(5)}

thereby yielding (1) independent of Δt\Delta t. Thus (1) is true for any regular NN-side polygon on the circle. Since the length of the polygon is NvΔtNv\Delta t, in the limit of N→∞N \to\infty and Δt→0\Delta t \to0 such that NΔt=TN\Delta t = T, the polygon becomes a circle with circumference 2πr2\pi r and therefore vT=2πrvT = 2\pi r. It would seem that this should be the end of Newton’s derivation.

Surprisingly, according to historian Richard Westfall[11], Newton’s did not stop at (5). Instead, Newton multiplies it by NN, getting aNΔt/v=vNΔt/raN\Delta t/v = vN\Delta t/r and arrives at the velocity circle aT/v=vT/r=2πaT/v = vT/r = 2\pi! Westfall then concludes that when this last result is divided by T=2πr/vT = 2\pi r/v, one would get a=v2/ra = v^2/r!! Erlichson[15] disputed whether Newton actually did such a division, but it is indisputable that Newton had known about the velocity circle aT/v=2πaT/v = 2\pi since the Waste Book time around/after 1664.

The other simple derivation, also independently found by many[10] [1, 2] (including Tipler’s text[4]), was first given geometrically by Huygens[12, 13, 14], also more than 350 years ago. Imagine a stone tied by a tiny string to the rim of a merry-go-around of radius rr rotating at constant rim speed vv, as shown in Figure 2. If the string were cut, the stone would fly off the tangent at constant speed vv to BB, a distance of vtvt in a short time tt. For such a short time, the merry-go-around would have rotated from AA to CC and would view the stone as “fleeing” away from the center to a distance xx from the rim given by

Figure 2. Huygens’ key diagram with triangle ABCABC similar to ABDABD because both have the same two angles γ\gamma and α\alpha.

(r+x)2=(vt)2+r2(r+x)^2=(vt)^2+r^2
(x+2r)x=(vt)2.(6)(x+2r)x=(vt)^2. \tag*{(6)}

Neglecting xx relative to 2r2r, this center-fleeing distance is

x=12(v2r)t2.(7)x=\frac{1}{2}\left(\frac{v^2}{r}\right)t^2. \tag*{(7)}

Instead of algebra, Huygens cleverly observes that ABCABC and ABDABD are similar triangles and therefore immediately arrive at (neglecting xx in BD=2r+xBD=2r+x)

BCAB=ABBD → x=(vt)22r.(8)\frac{BC}{AB}=\frac{AB}{BD}\ \rightarrow\ x=\frac{(vt)^2}{2r}. \tag*{(8)}

Huygens view this center-fleeing distance xx as due to a constant centrifugal acceleration v2/rv^2/r. This explanation from the rotating frame is clear and obvious. (It is not clear why Ref.8 would state that Huygens’ method is “not elegant” yet praises the identical algebraic expression in Tipler’s text as “presented elegantly”.) On the other hand, if one stays in the inertial frame, then xx is the distance prevented from happening by a centripetal acceleration so that the stone would stay put on the rim. This is a less obvious point-of-view and is the fundamental reason why students find centripetal acceleration more difficult to understand.

After reviewing the above two historical derivations, this work proposes the following more physical picture. From projectile motion, one learns that a baseball thrown horizontally with velocity vv would travel a horizontal distance dd in time t=d/vt=d/v. During this time, the ball would free fall a height of

h=12gt2=12gd2v2.(9)h=\frac{1}{2}gt^2=\frac{1}{2}g\frac{d^2}{v^2}. \tag*{(9)}

If hh equals to the the initial height of the thrown ball, then the ball hits the ground at a distance dd from its original location. This is elementary to all students. However, the earth is not flat. As shown in Figure 3, over a horizontal distance dd earth’s “ground” drops below the horizon by a distance HH. This HH can be solved from the obvious right triangle BCOBCO. However, it is also easy to follow Huygens, by noting that triangle ABCABC is similar to triangle CBDCBD, and therefore

Figure 3. The dropping distance HH due to earth’s curvature. Triangle ABCABC is similar to CBDCBD because both are right triangles and α+β=90∘\alpha+ \beta= 90^\circ.

ABBC=BCBD → H=d22R.(10)\frac{AB}{BC}=\frac{BC}{BD}\ \rightarrow\ H=\frac{d^2}{2R}. \tag*{(10)}

If a baseball were thrown with velocity vv such that the dropping of the ball hh exactly matches the dropping of the earth HH, then the ball will never hit the ground. This velocity is

v=gR.(11)v = \sqrt{gR}. \tag*{(11)}

At this velocity, the baseball becomes airborne and orbits the earth in a circular orbit with radius RR, always free falling toward the center of earth with acceleration gg. Generalizing gg to any center-seeking acceleration aa and earth’s radius RR to any radius rr, then (11) reproduces the needed centripetal acceleration (1). In this derivation, the orbit is circular because the object is constantly falling toward the center, but just keep on missing the “ground”.

References

  1. [3]U. Haber-Schaim, J. H. Dodge, and J. A. Walker, PSSC Physics, D. C. Heath, Lexington, MA, 1981, 5th ed., pp. 52–54.
  2. [4]P. A. Tipler, Physics for Scientists and Engineers, Worth Publishers, New York, 1991, 3rd ed., P.66.
  3. [5]B. Wedemeyer, “Centripetal acceleration a simpler derivation”, Phys. Teach. 31, 238 (1993).DOI
  4. [6]K. R. Brownstein, “A simple derivation of centripetal acceleration,” Am. J. Phys. 62, 946 (1994).DOI
  5. [7]D. W. Kraft and L. Motz, “Geometric derivation of radial acceleration magnitude,” Phys. Teach. 33, 565 (1995)DOI
  6. [8]I. Newton and R. C. Henry, “Circular motion”, Am. J. Phys. 68, 637 (2000).DOI
  7. [9]Harvey S. Leff “Acceleration for circular motion”, Am. J. Phys. 70, 490 (2002)DOI
  8. [10]C. Corrao, “Gamow on Newton: Another Look at Centripetal Acceleration”, Phys. Teach. 50, 176 (2012)DOI
  9. [11]R. S. Westfall, Force in Newton’s Physics, MacDonald, London; American Elsevier, New York, 1971.
  10. [12]H. Erlichson, “Huygens and Newton on the problem of circular motion”, Centaurus 37, 210 (1994).DOI
  11. [13]A. Stinner, “Linking ‘The Book of Nature’ and ‘The Book of Science’: Using Circular Motion as an Exemplar beyond the Textbook”, Science & Education 10: 323–344, (2001).
  12. [14]C. Huygens, “De vi centrifuga”, in Oeuvres complètes, Vol. XVI, pp. 255-301, Translated by M.S. Mahoney, https://www.princeton.edu/ hos/mike/texts/huygens/centriforce/huyforce.htm
  13. [15]H. Erlichson, “Motive force and centripetal force in Newton’s mechanics”, Am. J. Phys. 59, 842 (1991)DOI

Paper details

vlnc:W_z7b5qj4m
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents