next up previous
Next: About this document

  1. The oscillator has the particular solution

    displaymath294

    with r and tex2html_wrap_inline298 as given in section 1.9:

    eqnarray23

    The full solution will also have the general, underdamped homogeneous oscillator solution added to tex2html_wrap_inline300 , with coefficients fixed by the initial conditions. (This general homogeneous solution is also called the transient solution, because it decays exponentially in time.) Here, however, we're asked to find initial conditions such that steady state motion begins instantly; that is, we don't have to wait for the transient to decay, because its coefficients are zero. In that case the full solution is tex2html_wrap_inline302 , with velocity

    displaymath304

    This obeys initial conditions

    displaymath306

    using the oddness of sin. Note that our derivation of the response factor in section 1.9 (especially equations 1.120 - 1.123), lets us write out tex2html_wrap_inline308 and tex2html_wrap_inline310 explicitly, since

    displaymath312

    giving

    displaymath314

    Thus our initial conditions are

    displaymath316

  2. You are asked to find the particular (steady state) solution to

      equation277

    In section 1.9 we derived the solution when the right hand side of the equation (the source) was tex2html_wrap_inline318 :

      equation279

    Note, for complex solutions z = w + ix, that the imaginary part of equation (2) is exactly equation (1), since tex2html_wrap_inline322 . Thus, knowing our solution z to equation (2) gives us the solution x for a sinusoidal source, since tex2html_wrap_inline328 .

    Recall that equation (2) gave the particular solution

    displaymath330

    where

    displaymath332

    Thus

    displaymath334

    1. We're asked to find a particular solution to the ODE

      displaymath336

      for a critically damped oscillator with tex2html_wrap_inline338 . Our ODE is thus

      displaymath340

      We seek a particular solution proportional to the source, tex2html_wrap_inline342 . Noting that tex2html_wrap_inline344 , plugging into our ODE gives

      displaymath346

      Thus we have particular solution

      displaymath348

    2. Now we seek a particular solution to the ODE

      displaymath350

      and are encouraged to try tex2html_wrap_inline352 . This guess has derivatives

      displaymath354

      Plugging into our ODE gives

      displaymath356

      or, noting that tex2html_wrap_inline358 and tex2html_wrap_inline360 terms cancel,

      displaymath362

      This works if it's a time-independent equation, that is, if n=2, in which case it gives solution 2A = f. Thus we have particular solution

      displaymath368

    3. We wish to use the above particular solutions to solve

      displaymath370

      with initial conditions tex2html_wrap_inline372 . Note that

      displaymath374

      so that, by superposition, it gives a particular solution which is half the sum of particular solutions in (a) and (b):

      displaymath376

      Our most general solution is given by this particular solution plus the general solution tex2html_wrap_inline378 to the homogeneous problem (a critically damped oscillator with tex2html_wrap_inline338 ). This is

      displaymath382

      Thus we have solution

      displaymath384

      with derivative

      eqnarray130

      At t= 0 this gives

      displaymath388

      Imposing our initial conditions tex2html_wrap_inline372 gives tex2html_wrap_inline392 . Thus our solution obeying these initial conditions is

      displaymath394

    1. The RC circuit obeys

      displaymath396

      as the net voltage drop along the full circuit loop must be zero. This is an ODE with constant coefficients, whose solution tex2html_wrap_inline398 must obey

      displaymath400

      Thus the general solution is the exponentially decaying

      displaymath402

      (There is only one independent solution as this is a first order ODE.)

    2. If the initial charge on the capacitor is tex2html_wrap_inline404 , then tex2html_wrap_inline406 and the solution is

      displaymath408

    3. For the RC circuit with voltage source tex2html_wrap_inline410 , Kirchoff's law for voltage drops along the circuit loop gives

      displaymath412

      (Traversing the circuit clockwise from the switch, we have a voltage gain tex2html_wrap_inline410 followed by voltage drops tex2html_wrap_inline416 and q/C, with the total gain balancing the total drop.) Looking for a constant particular solution tex2html_wrap_inline420 , the ODE requires

      displaymath422

    4. Our full solution for the inhomogeneous ODE is tex2html_wrap_inline420 plus the general solution to the homogeneous equation,

      displaymath426

      Applying the initial condition q(0) = 0 gives tex2html_wrap_inline430 . Thus we have solution

      displaymath432

      with current

      displaymath434

    5. At t=0, we have q(0) = 0 and tex2html_wrap_inline440 . This corresponds to no charge on the capacitor (hence no voltage drop across it), and a voltage drop tex2html_wrap_inline442 across the resistor; that is, the full source voltage tex2html_wrap_inline410 is dropped across the resistor. As time goes on charge builds up on the capacitor, increasing the voltage dropped across the capacitor. This means the voltage across the resistor must drop, which it does by reducing the current I. Asymptotically, as tex2html_wrap_inline448 , tex2html_wrap_inline450 and tex2html_wrap_inline452 . In this asymptotic state, no current flows, so no voltage is dropped across the resistor and the full source voltage tex2html_wrap_inline410 is dropped across the capacitor with charge tex2html_wrap_inline456 .
    6. In this case, the particular solution is constant and does indeed persist. The general solution tex2html_wrap_inline458 to the homogeneous equation does indeed decay exponentially; it is ``transient'' in that its effects become negligible when tex2html_wrap_inline460 . So, for the RC circuit, eventually everything dies out but the particular solution tex2html_wrap_inline462 , which does indeed describe an asymptotic ``steady state.''
    1. For the LC circuit with voltage source tex2html_wrap_inline464 , Kirchoff's law for voltage drops along the circuit loop gives

      displaymath466

      (Traversing the circuit clockwise from the switch, we have a voltage gain tex2html_wrap_inline464 followed by voltage drops tex2html_wrap_inline470 and q/C, with the total gain balancing the total drop.)

      We seek a particular solution to the inhomogeneous ODE

      displaymath474

      We try a solution proportional to the source,

      displaymath476

      Plugging into the ODE gives

      displaymath478

      solved by

      displaymath480

      defining tex2html_wrap_inline482 .

    2. The full solution is the particular solution above plus the general solution to the homogeneous problem

      displaymath484

      Taking our independent homogeneous solutions as tex2html_wrap_inline486 we obtain

      displaymath488

      with derivative

      displaymath490

      Setting our initial conditions tex2html_wrap_inline492 gives

      displaymath494

      giving solution

      displaymath496

    3. Note that we may Taylor expand tex2html_wrap_inline498 near tex2html_wrap_inline500 as follows:

      displaymath502

      displaymath504

      Taking tex2html_wrap_inline506 ,

      displaymath508

      Thus

      eqnarray241

      As tex2html_wrap_inline510 , all terms with positive powers of tex2html_wrap_inline512 vanish, leaving only the tex2html_wrap_inline514 term:

      displaymath516




next up previous
Next: About this document

Katherine Benson
Fri Feb 15 15:41:47 EST 2002