Hooke's Law and A Few Simple Lessons From Physics

Download as pdf or txt
Download as pdf or txt
You are on page 1of 5

Hooke’s Law and a Few Simple Lessons from Physics

Statement of Hooke’s Law. The physical setup in which a mass is connected


to a spring is a rather ubiquitous one; for example molecular structures are often so
modelled. We don’t consider gravity in this situation, so there are only two constants
of relevance to us. The first is the so-called spring constant, which measures the
“stiffness” of the spring (the larger the constant, the greater the stiffness). The second
constant is the (inertial) mass of the object connected to the spring. Hooke’s Law
says that the force exerted on the object by the spring is proportial to the displacement
away from equilibrium (unstretched spring). Since Newton’s Second Law says that
the net force on an object is the object’s mass times the accelleration resulting from
the force(s), and since acceleration is represented by the second derivative of position
with respect to time, we conclude that Hooke’s Law takes the form of a differential
equation:

d2 x
m 2 = −kx .
dt

Force on object
k










←−
m (spring is stretched)






x=0 x>0
(equilibrium)

Force on object
k











−→
m (spring is compressed)








x<0 x=0
(equilibrium)
Solution of Corresponding ODE and IVP. As seen above, Hooke’s Law is
expressed as a second order ordinary differential equation (ODE). Any student hav-
ing been exposed to the basics of differential calculus can cite two easily-arrived at
solutions:
r ! r !
k k
x(t) = sin t and x(t) = cos t .
m m
However, a moment’s thought reveals that not only is any constant multiple of either
of the above solutions again a solution, so is the sum of such constant multiples. This
affords the general solution of the above differential equation:
r ! r !
k k
x(t) = C1 sin t + C2 cos t . (1)
m m

Typically, initial values of the above second-order equation would be specified by


giving an initial displacement and an initial velocity:

x(0) = a and x0 (0) = b,


where, as usual, x0 (t) = dx
dt
. Since cos(0) = 1, the first condition says simply that

C2 = a. (2)
Next, we differentiate Equation (1) with respect to t and obtain
r r ! r r !
k k k k
x0 (t) = C1 cos t − C2 sin t ;
m m m m
set t = 0 and conclude that
r
m
C1 = b (3)
k

Of particular interest is the situation in which we displace the mass an amount a


and just “let it go.” This means that the initial
q velocity
 is zero (b = 0) and so the
k
solution of this particular IVP is x(t) = a cos m
t .

Hooke’s Law with Damping. Suppose that we modify the above physical situation
by adding what is usually called a dashpot (or “shock absorber”) with damping
coefficient c. This is depicted by the drawing:

Forces on object
@
k (−kx) (−cx0 ) @@










←− −→ @@
m c @@




 @@
(dashpot) @@
@@
x=0 x>0 @@
@@
(equilibrium) @
In this case, the differential equation governing the motion of the object is

d2 x dx
= −kx − cm ;
dt2 dt
equivalently, this can be expressed somewhat more symmetrically as

d2 x dx
m + c + kx = 0 . (4)
dt2 dt

There’s a commonly-applied trick we can use in the above. Namely, we define the
second-order differential operator by setting

d2 d
D=m 2
+ c + kI,
dt dt

where I is just the “identity operator:” Ix = x. The trick is just to treat the above
as one might treat any quadratic polynomial and factor it in terms of its zeros. Using
the quadratic formula, these zeros are

√ √
−c + c2 − 4mk −c − c2 − 4mk
r1 = , r2 = (5)
2m 2m

and so we “factor” the differential operator as


  
d d
D = − r1 − r2 .
dt dt

Therefore, the differential equation (4) now becomes

  
d d
− r1 − r2 x = 0. (6)
dt dt

Note now that if x = x(t) is a solution of either of the differential equations


   
d d
− r1 x = 0 or − r2 x = 0,
dt dt

then x is a solution of (6).1 Thus, we set r = either of r1 or r2 and solve the first-order
ODE:

dx
= rx (7)
dt
   
1 This d d
is because the two “operators” dt
− r1 and dt
− r2 commute with each other.
It is clear that the general solution of the first-order ODE above is

x(t) = Cert ,

where C is an arbitrary constant. We now turn to the three cases that arise, according
to the discriminant of the above differential operator.

Case I: c2 − 4mk > 0 (the “overdamped case.”) Here the quantities r1 and r2 defined
in Equation (5) are distinct real numbers. From this we conclude that the general
solution of the ODE (6) is given by

x(t) = C1 er1 t + C2 er2 t ,

where r1 , r2 are given in (5).


Case II: c2 − 4mk = 0 (the “critically damped case.”) Here, we have that r = r1 =
−c
r2 = and x(t) = C1 ert is a general solution of (4). However, there is a second
2m
solution of (6), namely x(t) = C2 tert , giving the general solution of (4) in the
form

x(t) = C1 ert + C2 tert .

Case III: c2 − 4mk < 0 (the “underdamped case.”) What makes this case subtle is
that the zeros r1 and r2 are now (non-real) complex numbers. What we have to
remember that if r = a + bi, where a and b are real constants, then the complex
exponential function ert = e(a+bi)t is defined by setting

e(a+bi)t = eat eibt = eat (cos(bt) + i sin(bt)) .

Furthermore, if a complex function of t is a solution of a differential equation


with real coefficients, then both the real and imaginary parts of this function
are again solutions. Therefore, we infer that the ODE (4) has the general solution

x(t) = C1 eat cos(bt) + C2 ebt sin(bt),

where

−c 4mk − c2
a= and b = .
2m 2m

Again, we recall that the specific values of C1 and C2 are determined by specifying
initial values in the IVP.
Higher-Dimensional Models. Rather than considering a single mass on a spring,
one might consider more than one mass, constrained to move on one dimension. For
the sake of simplicity, we shall stick to very symmetric configurations. The symmetric
two- and three-dimensional configurations are depicted below.

@
k k k @@



























@
@
m m @@






 





 





 @
@
@@
@@
@@
@@
@

@
k k k k @@





































@
@
m m m @@






 





 





 





 @
@
@@
@@
@@
@@
@

We conclude this brief discussion to ask a question that really belongs to the
study of linear algebra. Namely, find initial displacements of the masses so that
the resulting motion will be purely sinusoidal. For example, if in the two-dimensional
system we displace the right weight a certain distance to the right of its equilibrium
and displace the left weight the same distance—but to the left of its equilibrium
and release the weights, then the resulting motion for each weight will be purely
sinusoidal. Likewise, if we displace the right weight a certain distance to the right
of its equilibrium and displace the left weight the same distance to the right of its
equilibrium, then again the resulting motion for each weight will be purely sinusoidal.

Question. In the above situation, which initial displacement will provide the faster
oscillations? (Hint: argue as would a physicist and determine which initial displace-
ment “packs more energy into the system.”)

Exercise. Repeat the above analysis for the three-body problem. In this case there
are three distinct initial displacements2 that would result in purely sinusoidal motion
for the masses. Also, determine which initial displacement would result in the fastest
oscillations.

2 In the language of linear algebra, these distinct displacement would be “linearly independent eigenvectors” for a

suitably-chosen linear transformation.

You might also like