Scribd
Upload
43 views9 pages

Lecture 5

1. Hooke's law describes a linear restoring force proportional to displacement. 2. Newton and Hooke disputed credit for discovering the inverse-square law of gravity and despised each other. 3. Applying Hooke's law to Newton's second law yields a second order differential equation describing simple harmonic motion.

Uploaded by

ShootingStarPhotons
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
43 views9 pages

Lecture 5

1. Hooke's law describes a linear restoring force proportional to displacement. 2. Newton and Hooke disputed credit for discovering the inverse-square law of gravity and despised each other. 3. Applying Hooke's law to Newton's second law yields a second order differential equation describing simple harmonic motion.

Uploaded by

ShootingStarPhotons
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

Hooke's law

If we Taylor expand the force about a point, we find:


A restoring force linearly proportional to the displacement is named after Robert Hooke (!"# $ %&"'(
)ewton and Hooke disputed the credit for disco*ering the in*erse+s,uare law of gra*ity and despised
each other(
In -:
. / +kx
setting the e,uilibrium position to be x
&
( Inserting this into )ewton0s 1
nd
law yields:
m d
1
x2dt
1
/ + kx d
1
x2dt
1
3 4
1
x / &
This is a 1
nd
order differential e,uation, which implies the general solution has 1 constant determined
by 1 I5s( 6e*eral e,ui*alent choices:
x(t' / A sin (4t 3 7' / A cos(4t 3 8' / 9 sin 4t 3 5 cos 4t
5hoose a form of the solution suited to the I5s of the problem(
6how that : / ; k A
1
/ const(
<eriod: / 1=24 .re,uency: > / 2/ 421=
In 1- and the special case:
we ha*e
F / +kr / ma
This is 1 uncoupled ?-:s in 5artesian coordinates:
d
1
x2dt
1
3 4
1
x / &
d
1
y2dt
1
3 4
1
y / &
The solutions are thus independent:
x(t' / A
x
cos(4t + 8
x
'
y(t' / A
y
cos(4t + 8
y
'
-efining @ / 8
x
$ 8
y
, we can show these e,uations imply:
y(t' / A
y
cos(4t $ 8
x
3 @' / A
y
Acos(4t $ 8
x
' cos @ $ sin(4t $ 8
x
' sin @B
/ A
y
Ax2A
x
cos @ + CA $ (x2A
x
'
1
B sin @B
Isolating s,uare+root term and s,uaring, we find:
@ / =21 (left circular polariDed' or "=21 (right circular polariDed'
@ / & (linearly polariDed y2A
y
/ x2A
x
' or = (linearly polariDed y2A
y
/ +x2A
x
'
Eenerally we ha*e elliptical motion(
Fhen tensor of
st
deri*ati*es is not proportional to the identity matrix, we ha*e:
x(t' / A
x
cos(4
x
t $ 8
x
'
y(t' / A
y
cos(4
y
t $ 8
y
'
6uch a cur*e will be closed only if 4
x
24
y
is a rational number (fre,uencies are commensurable'( If
fre,uencies are incommensurable, cur*e will e*entually fill the entire rectangle( Important for stellar
orbits in galactic potentials(
Phase diagrams:
)ewton0s 1
nd
law is a 1
nd
order differential e,uation, re,uiring 1 I5s per degree of freedom to fully
specify the solution( It is sometimes useful to consider solutions not in the ) dimensional
configuration space, but the 1) dimensional phase space( TraGectories in phase space cannot cross(
6imple harmonic oscillators trace out ellipses in phase space:
Damped oscillators:
If the 6H? has a damping force that linear in the *elocity, the :?H becomes:
which implies:
where 8 / b21m is the damping parameter( The solutions to linear, homogenous ?-:s such as this one
are x(t' / 5e
rt
( Inserting into the abo*e, we find:
implying the general solution:
This solution has ,ualitati*ely different beha*ior in " cases:
( underdamping: 4 I 8
1( critical damping: 4 / 8
"( o*erdamping: 4 J 8
Underdamping:
In case abo*e, the general solution becomes:
where
In general, A

and A
1
are complex, gi*ing K unknowns, but the re,uirement that x and dx2dt be real
pro*ides 1 constraints, lea*ing the desired 1 constants for our 1
nd
order ?-:( Fe can either recogniDe
that A
1
is the complex conGugate of A

, or choose alternati*e linear combinations that we know to be


real, sin 4t and cos 4t:
Important points:
1. :n*elope of traGectory e
+8t
(
2. .re,uency is lower than undamped oscillator 4

J 4(
3. d:2dt / .
damp
x dx2dt / +b (dx2dt'
1
other energy is going into <(:( of oscillator(
Critical damping:
The solution is:
5heck this(
Ley feature of critical damping is that the solution approaches e,uilibrium faster than either of the
abo*e cases(
!erdamping:
In this case, the general solution is:
where:
.irst term dominates at late times(
"in#soidal dri!ing $orces:
The solution to this e,uation can be expressed as x(t' / x
c
(t' 3 x
p
(t' where the complementary solution
*anishes when inserted into the RH6 as in the absence of dri*ing( Fe guess that the particular solution
is:
and find that:
At late times t II 28, the complementary solution dies away and the system approaches the particular
solution( The phase lag @ increases from & at 4 / & to =21 at 4 / 4
&
to = at 4 II 4
&
as seen below:
The amplitude - is maximiDed where the denominator is minimiDed:
Resonance is only possible for 4
&
I 8C1, in which case the amplitude will be:
This amplification is often described in terms of the quality factor:
In the limit 4
&
II 8, the ratio -24
&
1
A goes to M( Amplification for different *alues of M is shown
below:
)otice that as M increases (8 decreases', the peak of - increases and 4
R
approaches 4
&
( The .FHH of
the resonance is:
The range of M in physical systems *aries greatly( 6tandard mechanical systems can ha*e M of a few,
the best mechanical oscillators can ha*e M N &
K
, and atomic transitions can ha*e M I &
%
(
%lectrical oscillations:
An electrical circuit with a capacitor and inductor is also a simple harmonic oscillator( Apply
Lirchoff0s law to this circuit yields (I / +d,2dt':
This oscillator has a fre,uency:
1 I5s correspond to charge and current at time t / &( Inductance corresponds to mass in the mechanical
systemO in*erse of capacitance to spring constant( :lectric potential energy of capacitor corresponds to
mechanical potential energy of springO energy in magnetic field of inductor corresponds to kinetic
energy(
A constant force (such as gra*ity' displaces the e,uilibrium position of 6H?, but otherwise lea*es
solution unchanged:
The same is true of a constant emf in an electrical circuit:
yielding a solution:
Adding a resistor to the circuit corresponds to damping:
Fe see that 8 / R21P and the solution to this e,uation are as for the damped 6H?:
Complex &mpedance
The complex impedance Q of the element of a circuit is the ratio of the *oltage across that element to
the current(
R(t' / R
&
e
i4t
Resistor: R / IR Q / R2I / R
5apacitor: , / 5R I / 5 dR2dt / i45R Q / +i245
Inductor: R / PdI2dt I / S R2P dt / R2i4P Q / i4P
Fe add the impedance of series elements and the in*erse impedances of parallel elements( .or the
circuit as a whole I(t' / R(t'2Q implying that for a sinusoidal dri*ing *oltage, the magnitude of the
current will be R
&
2Q
&
and the phase lag of the current will be ArgAQB(
"eries and Parallel
If we had springs in series:
implying that the effecti*e spring constant is:
If we add capacitors in series, the charge on the capacitors must be the same implying:
It appears at first that there is a direct analogy between springs and capacitors in series( Howe*er, 5
enters into 4
&
as the in*erse of the spring constant k( If we want to create an electrical circuit with the
same 4
&
as a mechanical circuit, we need instead
This is pro*ided by capacitors in parallel:
5on*ersely, to create the electrical analogue of springs in parallel:
re,uires
i(e( capacitors in series( Applying a time+*arying emf to an electrical oscillator is entirely analogous to
applying a time+*arying dri*ing force to a mechanical oscillator(
Principle o$ "#perposition
The 6H? operator:
is linear, implying that we can add the particular solutions for different dri*ing functions to find the
solution for the total dri*ing function(
.ourier0s theorem tells us that any periodic function (with period and fre,uency 4n / 1=2' can be
expressed as a series of sin and cos terms:
where
Fe can therefore express the particular solution to any periodic dri*ing function as:
'reen's $#nction
How do we deal with an arbitrary (non+periodic' dri*ing functionT An acceleration a applied at a time
t
&
o*er a short time inter*al JJ 1=24
&
will lead to the particular solution:
Any acceleration a(t' can be considered as a succession of these impulsesO by the principle of
superposition any particular solution can therefore be expressed as a summation of these impulse
responses:
where E(t, t0' is an example of a Green's function( Ereen0s function are incredibly useful for sol*ing
linear, inhomogeneous differential e,uations(

You might also like