polarization of light and power factor

[Jamie Morken]

Hi,

Does circularly polarized light have a rotating electric field, or is it
light that has a phase shift between the electric and magnetic fields?
I thought it was a phase shift between the electric and magnetic fields,
with a +90 or -90 degree phase shift being circularly polarized light
(left or right circularly polarized), and anything in between 0 and 90
degrees is elliptically polarized. However when I read about it, the
electric field rotating is always mentioned to described circularly
polarized light instead of the electric and magnetic field being phase
shifted.

If circular/elliptical polarized light is caused by a phase shift
between the electric and magnetic fields, this is similar to the concept
of power factor in electronics, and I think they are related, as power
factor describes the phase relationship between the voltage (electric
field) and current ((magnetic field) waveforms. So light that is linear
polarized, with the magnetic field and electric field in phase, would
have a power factor of one I guess!

cheers,
Jamie

 

[Tim Williams]

The phase shift between E and B is 90 degrees in a propagating wave. If
it weren't, the Poynting vector would point somewhere else, indicating
decaying power or a "turn" or something like that. None of which makes
any sense for a propagating wave (doesn't satisfy the wave equations)
under ordinary conditions.

Polarization has to do with the apparent axis of the fields. The fields
are always 90 degrees to each other, but relative to, say, the emission
point, they can rotate to the left or to the right. This quantity is
spin, which arises from quantum mechanics in 3 dimensions.

The linear combination of left and right produces the equivalent
orthogonal polarizations of vertical and horizontal. Any nonorthogonal
combination (from either basis) is "eliptical".

QM seems to suggest that circular is physically fundamental, but this has
no practical significance for antennas or anything.

Tim

 

[Martin Brown]

The phase shift between E and B is 90 degrees in a propagating wave. If
it weren't, the Poynting vector would point somewhere else, indicating
decaying power or a "turn" or something like that. None of which makes
any sense for a propagating wave (doesn't satisfy the wave equations)
under ordinary conditions.

Polarization has to do with the apparent axis of the fields. The fields
are always 90 degrees to each other, but relative to, say, the emission
point, they can rotate to the left or to the right. This quantity is
spin, which arises from quantum mechanics in 3 dimensions.

The linear combination of left and right produces the equivalent
orthogonal polarizations of vertical and horizontal. Any nonorthogonal
combination (from either basis) is "eliptical".

Wiki has a rather cute animation of it online:

http://en.wikipedia.org/wiki/Circular_polarization

QM seems to suggest that circular is physically fundamental, but this has
no practical significance for antennas or anything.

Actually it does - weakly at least in radio astronomy. Powerful natural
radio sources frequently have a significant linearly polarised component
from synchrotron radiation and so the Stokes parameter I-V (using
circular polarised feeds) is a much better representation of total
brightness than I-Q (using linear polarised feeds). The latter almost
always having to be corrected by a second set of observations of U+iV.
Circular polarised feeds were harder to make though in the early days
and ISTR the VLA was the first to exploit it fully.

http://www.atnf.csiro.au/computing/software/atca_aips/node11.html

I don't think any natural astronomical sources with strong circular
polarisation have ever been observed, but I could be out of date here.
There is a tiny amount of it from dust grain scattering in the galaxy.

Regards,
Martin Brown

 

[Tim Williams]

I'm not sure if you're implying that there's no practical significance
in whether or not circular polarization is fundamental...?

I mean in terms of circular vs. rectangular.

Polarization is indeed handy... in E&M lab, we were determining the
radiation pattern of a horn antenna. After the measurements, I suggested
turning the whole setup 90 degrees to observe polarization. Not a damn
thing showed up on the HP reciever (at least -24dB, as you say).

Tim

 

[Clifford Heath]

QM seems to suggest that circular is physically fundamental, but this has
no practical significance for antennas or anything.

I'm not sure if you're implying that there's no practical significance
in whether or not circular polarization is fundamental...?

Circular polarization itself *is* significant, of course. In the
same way that a vertically-polarized antenna will tend to reject
a horizontally-polarized signal, a clockwise-spiral antenna will
reject an anti-clockwise signal... by 24dB in many cases.

Clifford Heath.

 

[Jamie Morken]

The phase shift between E and B is 90 degrees in a propagating wave. If
it weren't, the Poynting vector would point somewhere else, indicating
decaying power or a "turn" or something like that. None of which makes
any sense for a propagating wave (doesn't satisfy the wave equations)
under ordinary conditions.

Hi,

Are you referring to the axis of the electric and magnetic fields being
orthogonal or are you referring to a 90 degree phase shift between the
electric and magnetic fields? I understand that the electric and
magnetic fields are always orthogonal apparently, but is it also true
that the electric and magnetic fields are always in phase? This picture
from wikipedia seems to show the electric and magnetic fields (the two
sine waves) orthogonal to each other, and on the travelling axis there
is also a 90 degree phase shift between the electric and magnetic fields.

"http://upload.wikimedia.org/wikiped...ed.Light_With.Components_Right.Handed.svg.png"

My main question is whether circular polarized light is caused by the
electric and magnetic field rotating around the travelling axis of the
light while maintaining their orthogonal alignment, or if circular
polarized light is caused by a phase shift between the electric and
magnetic fields, causing the phase relationship to change, similar to
power factor.

cheers,
Jamie

 

[Jamie Morken]

The phase shift between E and B is 90 degrees in a propagating wave. If
it weren't, the Poynting vector would point somewhere else, indicating
decaying power or a "turn" or something like that. None of which makes
any sense for a propagating wave (doesn't satisfy the wave equations)
under ordinary conditions.

The phase shift between E and B is 90 degrees in a propagating wave. If
it weren't, the Poynting vector would point somewhere else, indicating
decaying power or a "turn" or something like that. None of which makes
any sense for a propagating wave (doesn't satisfy the wave equations)
under ordinary conditions.

Hi,

Are you referring to the axis of the electric and magnetic fields being
orthogonal or are you referring to a 90 degree phase shift between the
electric and magnetic fields? I understand that the electric and
magnetic fields are always orthogonal apparently, but is it also true
that the electric and magnetic fields are always in phase? This picture
from wikipedia seems to show the electric and magnetic fields (the two
sine waves) orthogonal to each other, and on the travelling axis there
is also a 90 degree phase shift between the electric and magnetic fields.

"http://upload.wikimedia.org/wikiped...ed.Light_With.Components_Right.Handed.svg.png"

My main question is whether circular polarized light is caused by the
electric and magnetic field rotating around the travelling axis of the
light while maintaining their orthogonal alignment, or if circular
polarized light is caused by a phase shift between the electric and
magnetic fields, causing the phase relationship to change, similar to
power factor.

cheers,
Jamie

 

[Tim Williams]

Hi Tim, I used to think the same thing. Turns out in a traveling E-M
wave in free space the E and B are in phase. Weird huh? How does it
'know' which way it is going?

-- Oops, that should be 90 degrees angle, not phase!

Mixing my angles again...

Direction, of course, is the cross product of E and B, which is true at
any point. If they were out of phase, the wave would have 0 power factor,
in other words, it wouldn't carry any power, which is just silly!

Tim

 

[Jon Kirwan]

On 23/08/2010 05:20, Tim Williams wrote:

elliptical

Wiki has a rather cute animation of it online:

http://en.wikipedia.org/wiki/Circular_polarization
<snip>

Reminds me:

http://arxiv.org/abs/quant-ph/0012141

Given linear, circular, and elliptical polarization.... why
not ... hyperbolic polarization?

Jon

 

[Jamie Morken]

Hi Tim, I used to think the same thing. Turns out in a traveling E-M
wave in free space the E and B are in phase. Weird huh? How does it
'know' which way it is going?

I don't think it has an actual direction, rather it is constantly expanding.

cheers,
Jamie

 

[Jamie Morken]

Yes that's it.



No, there is no phase shift between E and B in free space (far field)
propigation of light.

http://en.wikipedia.org/wiki/Electromagnetic_wave

Thanks for the clarification(s)! So the electric and magnetic fields of
light are crossing zero amplitude at the same time, and are displayed on
orthogonal axis'. And for circular polarized light, both the electric
and magnetic fields are rotating for some reason while maintaining their
orthogonal alignment and phase synchronization.

So "circular" polarized light, is really two separate light waves, ie
one with a horizontal and one with a vertical electric field (vertical
and horizontal magnetic field respectively) that are out of phase, and
the amount they are out of phase determines if they are elliptically or
circular polarized and left or right polarized? That would seem to make
sense as it is hard to understand why a wave would have any inherent
rotating action while travelling. This image showing two light waves
electric fields seems to support that idea:

"http://upload.wikimedia.org/wikiped...ed.Light_With.Components_Right.Handed.svg.png"

cheers,
Jamie

 

[Jon Kirwan]

I still find EM waves confusing at some point. What does it
mean to say that they experience no time? Sorry for the
anthropomorphizing.

This is a question I sometimes consider at night, when
walking through my woods.

I'll suggest another question that arrived, many years ago,
when I first asked it. "What does distance mean, if there is
no time? And if light experiences no time and therefore no
distance, what does this imply about the universe?"

Add to this, a question that arrives from the special theory
of relativity, which flows entirely from two very simple
expressions (four, after a step): x=c*t and x'=c*t'. These
are two completely different frames of reference, entirely
separate universes, if you will. The only thing they share
is 'c'. Might particles reside in entirely isolated and
completely empty separate universes, with perhaps the only
thing shared between them being photons, which create
simultaneously the concept of (x,t) for those particles?

What does distance mean?

Jon

 

[Tim Williams]

In the near field there is a phase difference between the E and B
fields.

A dipole has current at a maximum when voltage is at a minimum, and vice
versa, and the position of these is off by lambda/4 (strong magnetic field
towards the center, electric field towards the tips). Wouldn't seem to
make sense. If anything, the magnetic field propagating should
superimpose on the 1/4-wave-later electric field and propagate off into
space (if not for the overall antenna being 1/2 wavelength, disallowing
axial radiation).

So does the magnetic field part matter at all? I suppose in this type of
antenna, it doesn't. It's like an electric field waveguide-to-coax
antenna, expanded to free space. You can replace the magnetic part
altogether with a loading coil (at the expense of physical size = gain),
whose field usually points in completely the wrong direction anyway (an
axial solenoid having an axial, rather than radial, field).

A loop antenna, OTOH, is all magnetic field. Little voltage appears on
it.

Seems to suggest that an antenna is tailored to one or the other, and you
let space sort it out as far as turning E into B into E. In the near
field, the proximity of the antenna gives off E or B fields that
apparently confuse the situation (electrostatic and induced E and B
fields), giving rise to all sorts of voltages, currents and phases.

Also makes me wonder if one were to make an antenna which does both.
Contrive a structure which launches E and M waves, in the correct phase,
angle and proportion, so there is no near field. I wonder if it would
look something like those metamaterials -- coaxial loops and sticks, with
connections somewhere.

Tim

 

[Tim Williams]

You can use either basis (is that the right physics term?)

Yes, though not from physics, linear algebra actually. An intimidatingly
simple subject, quite useful on occasion.

Tim

 

[[email protected]]

-- Oops, that should be 90 degrees angle, not phase!

Mixing my angles again...

Direction, of course, is the cross product of E and B, which is true at
any point. If they were out of phase, the wave would have 0 power factor,
in other words, it wouldn't carry any power, which is just silly!

I just think of the impedance of free space as being 377ohms, so E and B
pretty much have to be in phase. ...or space would be reactive. ;-)

 

[Jon Kirwan]

" What does distance mean? "

Yup, exactly the same problem. At the speed of light distances go to
zero. All very strange. We sell this muon lifetime apparatus that
touches on the time dilation issue. The muon life time is about 2
micro seconds. All the muons come from high energy cosmic ray
collisions in the upper atmosphere. They have speeds near c. Now if
their own clocks didn’t run a lot slower than ours only a tiny
fraction of them would make it down to the ground. (2us X c = 600 m)
But we see lots!

Yes, I remember reading about this a long time ago -- can't
remember where. It was one of the confirmations performed
about relativity. I remember some of the details of the
experiment, too, in order to actually test the idea.

By the way, there is this:

http://www.physics.smu.edu/~yejb/MuonLifeTime.pdf

Do you know any of the authors of that?

Jon

 

[Tim Williams]

As far as transmitting a ‘far field’ pattern, I wonder if a microwave
horn does a good job of that?

It would seem like a good candidate if anything. Goes from a confined
wave somewhat tighter than c (remember, phase velocity > c in waveguide,
of course group velocity < c so it's okay), to something expanded, closer
to free space dimensions, then the horn ends and it's like... what? oh
well carry on...

Reminds me of a quote from my professor. Spoken with a Serbian accent.
"You kind of fool the wave, open it up gradually so it doesn't reflect
until it's too late, and by then it's already gone... kind of like how I
got my wife!"

Tim

 

[Jamie Morken]

Hmmm Well to say it is 'really' two linear waves is the same as saying
that linear polarization is 'really' the combination of two circularly
polarized waves.. added with the correct amplitude and phase. You can
use either basis (is that the right physics term?) linear or circular
to describe a photon, both are equally true.

George H.

Hi,

You can have a linear polarized wave without a circular polarized wave,
but can't have a circular polarized wave without it being composed of
linear polarized waves, so the linear polarized wave is the fundamental
type of wave.

I received this email from an antenna expert showing this geometrically:

A circularly-polarized EM wave can be generated (at radio frequencies,
at least) by any of several methods:

(A) Feed a signal to two orthoganal antenna elements (e.g. crossed
dipoles), introducing a 90-degree phase shift
into one of the feed
arrangements (e.g. using a 1/4-wavelength-
long coaxial delay line).

(B) Feed a signal (with the same phasing) to two orthoganal elements
which are separated by 1/4 wavelength in the
direction of travel.

(C) Use an helical antenna in end-fire mode - either feed the helix
directly (direct generation of a circularly-
polarized signal), or
generate a directional unipolar waveform
(e.g. via a dipole) and
use this to excite a helically-wound director.

examples A and B generate circularly polarized waves using two linear
polarized waves, and example C can be thought of as emitting a very high
number of linear polarized waves each offset by a small angle depending
on the helical antenna diameter and turn pitch. So you can take a
circular polarized wave and see how it is made of linear polarized
waves, but can you take a linear polarized wave and see how it is made
of circular polarized waves? Espeically since linear polarized waves
are much more common in nature I think they are the fundamental type of
wave, and all other types are superpositions of linear waves.

cheers,
Jamie

 

[JosephKK]

The phase shift between E and B is 90 degrees in a propagating wave. If
it weren't, the Poynting vector would point somewhere else, indicating
decaying power or a "turn" or something like that. None of which makes
any sense for a propagating wave (doesn't satisfy the wave equations)
under ordinary conditions.

Polarization has to do with the apparent axis of the fields. The fields
are always 90 degrees to each other, but relative to, say, the emission
point, they can rotate to the left or to the right. This quantity is
spin, which arises from quantum mechanics in 3 dimensions.

The linear combination of left and right produces the equivalent
orthogonal polarizations of vertical and horizontal. Any nonorthogonal
combination (from either basis) is "eliptical".

QM seems to suggest that circular is physically fundamental, but this has
no practical significance for antennas or anything.

Ahem. For earth to satellite and satellite to earth communications;
adjacent transponder channels have opposite circular polarizations,
this increases effective bandwith as the sidebands of the two adjacent
channels can be significantly overlapped without interference. The
two polarizations can be sorted out at the antenna with a little help
from the receiver.

 

[Tim Williams]

You can have a linear polarized wave without a circular polarized wave,
but can't have a circular polarized wave without it being composed of
linear polarized waves, so the linear polarized wave is the fundamental
type of wave.

I received this email from an antenna expert showing this geometrically:

<snip>

Or you can perform the exact same procedure using the endfire helical, or
phase shift / space shift orthogonals, and generate linear polarization
back again. So by applying your argument again (almost sounds....circular
), you've just proven that circular is also the fundamental type of
wave. ;-)

Tim