Q value of an underdamped oscillator (following text
treatment of T. L.Chow)
- E = 1/2 m x-dot^2 + 1/2 k x^2
 
- underdamped oscillator: E(t) = C exp(-beta t) [function
f(t) periodic in T = 2 pi/omega]
 
- easy to show that energy is E(t) = exp(-2 beta t) [ function
g(t) periodic in T]
 
- if we evaluate at intervals of nT where n is an integer
we get
 
- E(nT) = exp(- 2 beta t) E(0)
 
- if we use the notation <E(t)> for E evaluated at
intervals of T we get
 
- <E(t)> = exp(- 2 beta t) <E(0)> 
 
- this means d<E(t)>/dt = - 2 beta <E(t)>
 
- T. L. Chow' text defines Q of an underdamped oscillator
as
 
- Q = energy/(magnitude of energy dissipated per radian)
 
- energy dissipated per radian is energy dissipated in
the time for the function to oscillate through one radian
 
- omega =radians/sec, so 1/omega = seconds/radian
 
- thus, energy lost per radian = d<E(t)>/dt * 1/
omega, or
 
- energy lost per radian = -2 beta/omega <E(t)>,
and finally that
 
- Q = <E(t)>/[2 beta/omega <E(t)>] = omega/(2
beta).
 
- Q = omega/(2 beta). [ T. L. Chow]
 
- omega^2 = sqrt(omega-0^2 + beta^2)
 
- Marion and thornton have Q = [omega at resonance]/(2
beta), slightly different from Chow
 
- Only for heavy damping will there be any difference between
the two statements.
 
Q by measuring the frequency difference between the 0.707
points 
- The response function to a sinusoidal driving function
is
 
- A/sqrt( a^2 + b^2), where a = omegao^2-omega^2 and b
= 2 beta omega
 
- at the max of the response a is nearly zero and omega=omegao
 
- so max response = A/b
 
- at the 0.707 response points a = b, so
 
- (omegao+omega)(omegao-omega) = plus or minus 2 beta omega
 
- since omega is nearly equal to omegao we have that
 
- omegao-omega = beta at the 0.707 points
 
- the frequency difference between the 0.707 points is
then
 
- delta-omega = 2 beta
 
- The Q value is given by omega/(2 beta), so we may also
write
 
- Q = omega/delta-omega where delta-omega is the frequency
difference between 0.707 points