General Question

Mariah's avatar

Physics homework question about waves.

Asked by Mariah (25883points) January 20th, 2015

I don’t want to give all the details for the problem out of academic honesty concerns, but some guidance would be nice.

I have the following graph. The blue and pink waves show the relationship between time and y displacement for two different points along the x axis of the wave.

I need to estimate the wavelength of this wave. Note that it cannot just be directly estimated from the graph because the x axis is time, not x displacement.

I fixed t at various values and drew a number of y versus x graphs, but there are multiple interpretations for how the wave looks as it passes between the two points, as shown in this sketch of mine: http://i.imgur.com/GvzQdPy.jpg

I don’t believe I know any formulas that would relate wavelength to any of the information that I already know (amplitude and period/frequency).

A hint, please?

Observing members: 0 Composing members: 0

11 Answers

kritiper's avatar

“Frequency times wavelength equals speed.” So frequency divided by speed equals wavelength.
(Don’t know if this helps but I thought I’d take a crack at it. Info here is from “The Wonders of Physics: An introduction to the physical world.” A deluxe Golden Book, text by Irving Adler, copyright 1966 by Western Publishing Company, New York.)

RocketGuy's avatar

Use @kritiper ‘s info. Then assume speed = speed of light, since this could be an electromagnetic wave. You should be able to calculate your answer.

Mariah's avatar

I don’t know the velocity. It cannot be assumed to be an electromagnetic wave.

RocketGuy's avatar

You are right. Look at the graph. At t=0.01s, x=0m At t=0.035s, x=0.090m Now you can calculate m/s.

Mariah's avatar

No, unfortunately. The graph is really ambiguous, I apologize. The entire pink curve is the y position of the point x=0 for various times. The entire blue curve is the y position of the point x=.09 for various times. Unless I misunderstand you.

Mariah's avatar

Thanks for trying to help me solve it, by the way, guys.

LuckyGuy's avatar

I agree there is not enough info, but practically, that sure looks like an audio speaker.

You know the diaphragm is moving +/- 4 mm. And you know the frequency, since the period is 0.04 sec 1/0.04 = 25 Hz. It looks like it is working in air but it might also be in helium or another gas.

Mariah's avatar

I’m a lot closer to a solution now.

I have the formula y = amplitude*cos(2*pi*x/wavelength – 2*pi*t/period). I know the amplitude and period. So I can substitute in any (x,y,t) point from the graph and supposedly solve for the wavelength.

I have done this, but I’m confused. Shouldn’t I get the same wavelength for any point along the wave that I substitute in? But I’m getting lots of different answers for various points.

gasman's avatar

It’s been decades since I was a physics major, but maybe this link will help:
Traveling Sine Wave

y = A sin ( kx – wt + p) where w=omega (angular frequency) and p = phi (phase offset)

Looks like you’re already on to this kind of solution.

Mariah's avatar

Thanks. Do you have any clue why plugging in different points along the wave would lead to a different value k????

gasman's avatar

I don’t know—the given info is presented in a weird way. It looks from the diagram that period T is 0.04 seconds, amplitude A=4, and an x-offset of .09 m translates to a time offset of around .023 sec, but I’m afraid I can’t help much coming up with the desired function y(x,t).

Answer this question

Login

or

Join

to answer.

This question is in the General Section. Responses must be helpful and on-topic.

Your answer will be saved while you login or join.

Have a question? Ask Fluther!

What do you know more about?
or
Knowledge Networking @ Fluther