Visualising electromagnetic (radio) waves.

[mark g]

Hi.

I am having trouble visualising radio waves.

Assuming a perfect isotropic antenna (I know they don't exist) an electromagnetic is just a spherical pulse that radiates outward, this is quiet easy to visualise. However all texts refer to the magnetic and electric fields being 90 degrees apart, which leads one to believe that one field is parallel with the Earth and the other is perpendicular (if antenna was placed perpendicular).

With the isotropic antenna, electric and magnetic fields would surely be experienced in every possible direction and not at 90 degrees to each other.

Can someone please help me see this. It is driving me crazy and cannot get an answer anywhere.

Thanks,

Mark.

 

[UltrafastPED]

See this discussion, and the spherical wave:
http://physics.stackexchange.com/questions/20331/understanding-the-diagrams-of-electromagnetic-waves

 

[jtbell]

A Google search for "dipole radiation animation" turns up a number of animations that show the electric field lines "spreading out" from an oscillating dipole. I haven't seen any yet that also show the magnetic field lines.

Imagine a sphere with longitude and latitude lines like the ones we use on a globe, with the dipole at the center. The electric field generally runs more or less parallel to the longitude lines, looping towards or away from the center of the sphere as they approach the poles. The magnetic field lines (which are perpendicular to the electric field lines) are circles parallel to the latitude lines.

 

[Drakkith]

Let's see if I can make this work...

Let's start with something easier to visualize. How about a wave on the surface of a pond.

So we throw a rock into this pond and create a circular wavefront that propagates outward in all directions. Notice that the water itself is NOT moving outwards, it is moving up and down. The wavefront, which is the disturbance of the water, is what is moving outwards, similar to how a wave of people standing up in a crowded stadium propagates without any person actually moving around.

Okay, so we have this wavefront moving on our 2d surface. How can we relate that to 3d space? Well, imagine that instead of having a 2d circle we have a 3d sphere. From any angle, the part of the wavefront that is propagating in a direction perpendicular to your line of sight has oscillations which are moving "towards and away" from your point of view, just like the oscillation of the water when you are looking straight down at the surface of the pond. In contrast, the part of the wavefront moving towards or away from you, aka parallel to your line of sight, has its oscillations moving to the left and right. As you can see, the oscillations are always perpendicular to the direction of propagation.

But wait, an EM wave has 2 separate oscillating parts. Let's imagine that the one we've been talking about is the electric field vector (the vector is the direction of the force you would feel from the electric field). The magnetic field vector would oscillate perpendicular to both the direction of propagation AND the oscillation of the electric field vector. This means that the part of the wavefront moving parallel to your line of sight has its electric field vector oscillating left and right, while the magnetic field vector oscillates up and down.

So BOTH vectors oscillation perpendicular to each other and the direction of propagation. Make sense?

Remember that the wavefront is 360 degrees and consists of both the electric and magnetic field vectors.

 

[mark g]

Hi Guys.

Thanks for the responses. I can visualise it much better now. I still have a bit of an issue with the received signal.

I have attached an image which has two parts.

I the first part (top half) the receive antenna will receive the transmitted signal (-free space path loss).

In the second part (bottom half) the antenna is rotated 90degrees. Will it receive the signal due to the electric field now being perpendicular to the antenna.

Thanks,

Mark.

 

[davenn]

In the second part (bottom half) the antenna is rotated 90degrees. Will it receive the signal due to the electric field now being perpendicular to the antenna.

no, because the transmit antenna and the receive antenna are cross-polarised and the receive antenna will pick up little to no signal
And if that receive antenna is end on to the transmit antenna, as hinted by your diagram, the transmit signal will be totally nulled.
This is the method of how we direction find a transmitter

the top image shows the 2 antennas with the same polarity and so maximum signal will be induced into the receive antenna

Dave

 

[mark g]

Hi Dave.

That's what I thought. But say for example you are lying down with your phone or the radio aerial is at a funny angle the signal is ok. How is this? Would it be that the signal is in fact weaker and that the device just up's the power?

Mark.

 

[sophiecentaur]

It's worth noting that there is no such thing as an isotropic radiator of EM (transverse) waves. There are always nulls in the radiation pattern. Easy to show this is true if you ask yourself what happens when you follow the field patterns around a simple dipole. At the ends, the tangential (i.e. the radiated) fields go to zero and no energy is radiated. You can't 'add in' another dipole to fill in for these fields without producing a null in another direction. The best radiator to consider, when trying to visualise things is the dipole, which it's the one dealt with in all textbooks. That is no cop-out. It's just being realistic.
In real life, there are antennae that will appear to be 'almost' omnidirectional but that will be due to the wide dynamic range of the receiver you are using and the effect of nearby objects which can easily produce radiation patterns which are different over the band (due to wave interference) and they can easily appear to be filling in the obvious nulls and also changing the plane of polarisation. In the end, though, it's analogous to a partially filled air mattress - you can push down on one bit but another bit will always pop up.

OTOH, for a longitudinal wave, it is easy to produce and visualise an omnidirectional radiator - a small, radially oscillating sphere will radiated equal pressure (sound) waves in all directions.