1
$\begingroup$

The oscillation of light waves and sound waves (and many other waves) can be described with sinusoidal functions in time (frequency) and in space (wavenumber). Moreover, oscillation of monochromatic waves is perfectly described by a single sinusoidal (single frequency and wavenumber) function.

But why sine function and not any other function/wavelet? Or, why pressure waves, for example, have anything to do with the Cartesian projections of circular motion?

$\endgroup$
  • $\begingroup$ Partial duplicate of Using sinusoids to represent sound waves $\endgroup$ – Alfred Centauri Sep 5 '16 at 0:58
  • $\begingroup$ But, for example when you have to solve the Maxwell Equations in cylindric coordinates the solution to are given in terms of Bessel Functions, so there you have an example of a wave that is not a sine $\endgroup$ – Jasimud Sep 5 '16 at 2:23
  • $\begingroup$ @Jasimud Bessel function may be a solution but that's not how a light/sound perturbation propagates in nature. Gaussian beams on the other hand... Hmm $\endgroup$ – Sparkler Sep 5 '16 at 15:15
2
$\begingroup$

The solutions to wave equations are sinusoids, or seem to be, because sinusoids always represent solutions to a wave equation. But in reality the general solutions are linear sums of sinusoids, with different frequencies.

When you sum a lot of sinusoids you can can get any wave packet you want. So the solution then depends on what other conditions you impose.

For one pure sinusoid the condition is a fixed constant energy. Constant energy fixes the frequency. For electromagnetism (EM) a single frequency sinusoid it is a single color (if light, a very finely defined freq. channel if radio, etc. If you have a sum of many of those with frequents spanning the spectrum (of light) you get white light. If you changes frequencies Ina radio signals to do frequency modulation you have a bunch of freqs within that band. Something similar is true for sound, where freq represents pitch.

So, yes, you can form packets or wavelets or whatever you want, by summing with different amplitudes a bunch of those.

So, it is not that sinusoids are the only solutions, it is that any solution can be represented (and calculated) as the sum of sinusoids. If you look at the signal then so you can see all its frequencies, you are looking at its spectrum. Mathematically it is simply a Fourier transform to go from the time waveform to the spectrum.. A sinusoid spectrum is simply one line in the spectrum (two, one a positive freq and one at negative freq, if you want to be exact).

It is also said that sinusoids, or complex exponentially, are eigenvectors of the wave equation represented as an operator on the eigenvectors, with fixed eigenvalues, which are the energy.

Bets to understand this about waves well before doing any real physics.

$\endgroup$
  • $\begingroup$ On one hand infinite series of an orthogonal basis is a solution to the PDE but on the other hand a single sine wave perfectly represents a certain perturbation. The question is why? Why a primitive perturbation can be represented with a single sine but never with a single other wavelet function. $\endgroup$ – Sparkler Sep 5 '16 at 1:58
  • $\begingroup$ @Sparkler: This is the case because sine and cosine functions (complex exponentials, in the general case) are the eigenfunctions of the differential operator in question. Intuitively speaking this means that sines and cosines are the natural modes of the harmonic operators that give rise to these. And, yes, for non-Cartesian geometries and boundary conditions, other sets of eigenmodes may arise, such as Bessel functions in the case of problems with cylindrical symmetries. $\endgroup$ – Pirx Nov 24 '16 at 17:57
0
$\begingroup$
  1. When I push a door at 90 degree with door surface, effect of force is largest. If I do same thing at 0 degree with door surface, effect is zero. If I do between two extremes it is represented by $F*\cos A$, where $A$ is angle.

  2. Electricity produced is directly proportional to rate of change in flux. Whatever large magnet you place across wire loop, electricity won't be produced. But, electricity increases if loop is exposed with zero to maximum magnetic flux that too in very small time. Flux is directly proportional to area. Area of loop exposed to flux is zero at 0 degree and maximum at 90 degree.( make a loop of wire and see at various angles). Electricity is then rate of change in $\cos A$ function. Derivative of $\cos A$ is $\sin A$. So electricity produced is sin function.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.