MathGroup Archive 2004

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Fourier Transform with Differential Equation

  • To: mathgroup at smc.vnet.net
  • Subject: [mg48978] Re: Fourier Transform with Differential Equation
  • From: Paul Abbott <paul at physics.uwa.edu.au>
  • Date: Fri, 25 Jun 2004 17:52:20 -0400 (EDT)
  • Organization: The University of Western Australia
  • References: <cbgkes$cls$1@smc.vnet.net>
  • Sender: owner-wri-mathgroup at wolfram.com

In article <cbgkes$cls$1 at smc.vnet.net>, Lee Fisher <lfis at helix.nih.gov> 
wrote:

> I'm attempting to find the frequency response of the output of an RC
> circuit (a low-pass filter).  For some reason, the output of the Fourier
> transform in mathematica does not seem to correspond to anything I would
> expect from this differential equation.  It seems that the placement of
> the initial conditions has an enormous effect on the shape of the
> frequency response, so much so that the phase portion looks nothing like
> i would expect it to. I'm using the following code
> 
> w=1
> result=DSolve[{10 V'[t]+V[t]==E^(I w t), V[0]==0},V,t}]
> time=Table[i,{i,0,100}];
> out=Re[Evaluate[V[time]/.result[[1]]]];
> fout=Fourier[out];
> ListPlot[Abs[fout],PlotJoined->True,PlotRange->All]
> ListPlot[Arg[fout],PlotJoined->True,PlotRange->All]
> 
> and adjusting w to observe the response at different frequencies. While
> the magnitude graph is somewhat reminiscent of what I would expect, the
> phase graph looks nothing like I'd expect (it should be at Pi highly
> negative frequencies, negative Pi at highly positive frequencies, and at
> 0 at the zero frequency).  Could someone please explain to me either
> where my code is wrong, or where my expectation of the magnitude and
> phase output is wrong.

With w=1, what is the period, and what impact does that have on the 
sampling rate? Consider the following modification of your code:

For the angular frequency,

 w = 2 Pi 100

compute the period:

 period = 2 Pi/w

Solve the differential equation as before:

 result=DSolve[{10 V'[t]+V[t]==E^(I w t), V[0]==0},V,t}]

Sample the voltage over one period (not repeating the end-point):

 time = Range[period, 1, period]; 
 out = Re[V[time] /. First[result]]; 

Compute the Fourier transform:

 fout = Fourier[out]; 

Plot the magnitude,

 ListPlot[Abs[fout], PlotJoined -> True, PlotRange -> All]

and the phase, rotating the list so that zero frequency appears in the 
middle:

 ListPlot[RotateRight[Arg[fout], Floor[Length[fout]/2]], 
   PlotJoined -> True, PlotRange -> All]

Hope this helps.

Cheers,
Paul

-- 
Paul Abbott                                   Phone: +61 8 9380 2734
School of Physics, M013                         Fax: +61 8 9380 1014
The University of Western Australia      (CRICOS Provider No 00126G)         
35 Stirling Highway
Crawley WA 6009                      mailto:paul at physics.uwa.edu.au 
AUSTRALIA                            http://physics.uwa.edu.au/~paul


  • Prev by Date: Re: Simplifying with assumptions
  • Next by Date: Postscript Camera-Ready Copy (from Nicholas Georgakopoulos)
  • Previous by thread: Re: Fourier Transform with Differential Equation
  • Next by thread: [Final Call] ACA-2004 Conference on Applications of Computer Algebra