WHAT IS THE SPEED OF ELECTRICITY?
TBoudr5803
What is the speed of electricity?
Hanspeter Schmid
On 22 May 1996 02:23:53 -0400, TBoudr5803 (tboud...@aol.com)
wrote (<4nubtp$9...@newsbf02.news.aol.com>):
: What is the speed of electricity?
I'd say about 180000 mph.
It all depends on what you mean by ``the speed of electricity''.
Electrical energy travels at about that speed (which is the speed of
light), the precise speed depends on the material it travels in.
The electrons themselves are much slower, in normal copper, for direct
current, they are slower than 1 mph. But if you turn a switch on, a
shockwave will run through the copper wire which makes the electrons
start to move. This shockwave has about the speed of light, and it is
this shockwave which decides on how fast energy can be transmitted.
Cheerio, Hanspeter
--
-===-=-====-=-===== Hanspeter == Schmid =====-=== hobby-musician classical-
e-mail: sch...@isi.ee.ethz.ch | You can do what you like
Signal and Information | and you think you are free?
Processing Laboratory, Switzerland | I can like what I do,
http://www.isi.ee.ethz.ch/~schmid/ | and that's freedom for me.
>ETH-Z<====-=-===== orienteering runner ===== pan-flute trombone =========-
Roy McCammon
This is going to be a fun thread.
I think what you are asking (and what I am going to answer) is :
"how fast does a change in the electric field travel." This includes
the question of how soon after a the throw of a switch does load feel
the voltage.
Well, "electricity" travels as an e-m (electro-magnetic) wave which
is just another name for light.
The speed of light in a vacuum is 1 (foot per nano-second, or one
billion feet per second). Light and e-m waves never (almost never*)
travel faster than the speed of light in a vacuum.
But, the speed of light in a good conductor like copper is real hard to
measure. Through careful experiments and measurements of indirect
effects it has been determined that the speed of an e-m wave in
copper is real slow. I'll have to consult some references that I don't
have access to right now, but as I recall, the speed of an e-m wave in
copper is slower than a person can run. It turns out that this has
something to do with skin effect. At high frequency, the e-m wave
doesn't get very far into the copper before the direction is reversed
and so only the outer skin is available to support conduction.
If thats so, why can't you throw a switch and then run to the end of
an extension cord before the light turns on. Well, the answer is
that the wave travels partly in the copper and partly in the insulation
(where the speed is fast). The part of the insulation that the wave
travels through is very close to the outer skin of the copper, so
the portion in the copper sort of tugs on and slows the portion in the
insulation and the portion in the insulation tugs on and speeds up
the portion in the copper. Just how much is in each media depends
on a lot of things, but, the upshot is that typically with the things we
usually use, the speed of "electricity" ranges from 1 ft/ns to one
third or so of that value.
* I once read in a physicts text book that it is possible for rapidly
moving media like water to pull the light along with it to a speed
greater than the speed of light in a vacuum. I don't know if it is
true or not. Relativistic physicists please comment.
Now, since the speed of light is one (foot per nano-second) in American
units (miles, feet, hours, pounds), why in the world would anybody
want to use the metric system? No replies requested. (Yeah, I know
that volts and amps are already metric).
Opinions expressed herein are my own and may not represent those of my employer.
Douglas W. Jones,201H MLH,3193350740,3193382879
From article <4nubtp$9...@newsbf02.news.aol.com>,
by tboud...@aol.com (TBoudr5803):
> What is the speed of electricity?
The question is ambiguous. Electrons moving through a wire travel
fairly slowly, but changes in the electric field travel at the speed
of light. For practical purposes, you generally want to know when
the motion of some electrons at one end of the wire will cause
electrons at the other end to move -- by way of analogy, you can
think of a domino effect, where the electric field of each electron
that moves pushes on its successor, causing it to move.
If you have a wire and apply a voltage to one end, the voltage change
will be seen at the other end pretty quickly -- but the apparent speed
with which the voltage travels down the wire isn't going to be the speed
of light! The reason is that the wire has something called a
characteristic impedence (a property that combines inductance, resistance
and capacatence) that delays the signal. Impedence is, incidentally, a
complex number, involving the square root of -1.
So, what does it come down to? In the real world, with real wires
in typical arrangements found in real wiring, electrical signals
travel up to fairly close to the speed of light (in free-hanging wires
that aren't close to anything else), or they travel as slow as perhaps
a third of the speed of light (in some coaxial cable or shielded twisted
pair cable). Even slower propagation speeds are found inside integrated
circuits (seems paradoxical, doesn't it? Some of the slowest speeds
are found in some of the fastest circuits!)
Doug Jones
jo...@cs.uiowa.edu
Sam Goldwasser
> On 22 May 1996 02:23:53 -0400, TBoudr5803 (tboud...@aol.com)
> wrote (<4nubtp$9...@newsbf02.news.aol.com>):
> : What is the speed of electricity?
> I'd say about 180000 mph.
> It all depends on what you mean by ``the speed of electricity''.
> Electrical energy travels at about that speed (which is the speed of
> light), the precise speed depends on the material it travels in.
No, the speed of light in a vacuum is roughly 186,400 miles PER SECOND.
> The electrons themselves are much slower, in normal copper, for direct
> current, they are slower than 1 mph. But if you turn a switch on, a
> shockwave will run through the copper wire which makes the electrons
> start to move. This shockwave has about the speed of light, and it is
> this shockwave which decides on how fast energy can be transmitted.
Actually, somewhat less than the speed of light depending on the
characteristics of the wire.
--- sam
> Cheerio, Hanspeter
Wayne Shanks
Well Roy.... Dead Physic Professors every where are turning over in
there graves. You were righ on one point though. your's was a fun
thread:). See other posts
Wayne S
W. John Guineau
s...@stdavids.picker.com (Sam Goldwasser) wrote:
I remember from basic electronics classes that the instructor told us
the "voltage" travels at the speed of light while electons only drift
at a few hundred feet per second (rough numbers given variance of the
conducting medium)
john
--
W. John Guineau gui...@ultranet.com
Brookline, NH
http://www.ultranet.com/~guineau
Robert B. Kennedy
Depends on what it is traveling down.
In theory it is as fast as light.
--
Robert Kennedy: Sysop: Birds On A Wire AVIAN BBS!!! (206)557-0318,14.4,8n1
http://www.seattleu.edu/~rknndy/robert.html
"Karma can only be aportioned by the cosmos!" -Homer Simpson-
Boston Technicians
You know that executive toy where there are steel balls hanging from
string and you take the first ball and knock it into the next one and the
ball at the end moves but all the other balls remain stationary?
Therefore, in order to accurately answer your question, we need to know
specifically which electron you are refering to.
Also, do not confuse this issue with RF or EM propogation since they are,
in fact, different physical phenomenon.
John Whitmore
In article <31A41C...@wam.umd.edu> Wayne Shanks <Al...@wam.umd.edu> writes:
>Roy McCammon wrote:
>> I think what you are asking (and what I am going to answer) is :
>> "how fast does a change in the electric field travel." This includes
>> the question of how soon after a the throw of a switch does load feel
>> the voltage.
>>
>> Well, "electricity" travels as an e-m (electro-magnetic) wave which
>> is just another name for light.
Not the same thing, at all.
But, the electric disturbance of one end of a wire DOES travel
over the wire at roughly the velocity of light in the medium (which
is usually going to be the wire's insulation).
>> The speed of light in a vacuum is 1 (foot per nano-second, or one
>> billion feet per second).
true.
>> ... the speed of an EM wave in
>> copper is slower than a person can run.
No, not the speed of an EM wave; it's the speed of the
individual moving charges (electrons, of course) that's in the cm/sec
velocity range. The first electron to go into the lamp cord might
take quite a few seconds to reach the light bulb. But, the light
starts glowing a lot faster than that (because that first electron
pushed a few billion others up the wire...).
>> If thats so, why can't you throw a switch and then run to the end of
>> an extension cord before the light turns on.
Because the POWER propogates in fields outside the wire,
and moves at nearly the speed of light. By analogy, if you connected
your
garden hose to a hot water tap, then turned it on, the water would
come
out of the hose promptly, BUT it'd take it a few seconds to come out
warm (because the hose was previously filled with cold water, it's the
cold water that actually sprays out the business end.
>> * I once read in a physicts text book that it is possible for rapidly
>> moving media like water to pull the light along with it to a speed
>> greater than the speed of light in a vacuum. I don't know if it is
>> true or not. Relativistic physicists please comment.f my employer.
>Well Roy.... Dead Physic Professors every where are turning over in
>there graves.
And one Professor Fizeau is rather pleased that you remember
his little experiment (yes, moving media DO have some peculiar effects
on light... not enough to get the light moving faster than it would
in a vacuum, but it DOES change the velocity. This experiment
was performed in 1859...
John Whitmore
Tom Sgouros
On Thu, 23 May 1996, Wayne Shanks wrote:
> > * I once read in a physicts text book that it is possible for rapidly
> > moving media like water to pull the light along with it to a speed
> > greater than the speed of light in a vacuum. I don't know if it is
> > true or not. Relativistic physicists please comment.f my employer.
It is possible for speeds of "things" to exceed the speed of light. There
is no violation of relativity when the things that are moving are more
perceptual phenomena than obects.
For example, imagine a rotating light source shining on a nearby straight
line. When the angle of incidence of the light is close to 90 degrees, the
patch of light on the line is moving slowest. As the light source rotates,
the patch of light on the line accelerates until it has to reach an
infinite speed before it leaves the line altogether.
|
|
O<--patch of light moving up
Light source---> O |
rotating ccw |
|
So the speed of the patch of light eventually exceeds the speed of light.
But so what? You can't harness the effect to carry any information, and it
isn't really an object at all, just some kind of perceptual phenomenon.
These are the kinds of things that can easily be found that appear to
violate relativity. The question about information is the key one to ask
about objects that are said to move faster than the speed of light.
-Tom
---------------------------------------------------------------------
to...@ids.net - 401-861-2831 - 42 Forge Road, Potowomut, RI 02818 USA
Joe Seymour
any other medium, the speed is reduced. For practical purposes, the speed
thru air or conductors is close enough to be considered the same. Don't
confuse the speed of electricity (the propagation of an emf) with electron
movement or so called "current flow."
It is a common mis-perception that current flow is similar to the flow of a
material like water. In an unbiased conductor, there is random motion of
electrons at a high velocity (temperature dependent). Because this motion
is random, if you looked at a cross-section of the conductor, there would
be no net flow in a particular direction. If a voltage is applied across
the conductor, there would be a slight bias in the net direction (over a
period of time the motion would be more in one direction) of electron
movement. This is known as electron drift and occurs at approximately
.1 cm/sec. This means that if you hooked up a DC source to a 1 meter wire,
it would take about 17 minutes for an individual electron to move thru the
length of the wire. If you use an AC source(sinusoidal), the polarity is
periodically reversing and the net drift is again zero.
Joe S.
Richard Rasker
In article <4odegk$n...@canyon.sr.hp.com> jo...@sr.hp.com (Joe Seymour) writes:
>From: jo...@sr.hp.com (Joe Seymour)
>Subject: Re: WHAT IS THE SPEED OF ELECTRICITY?
>Date: 27 May 1996 23:40:04 GMT
>TBoudr5803 (tboud...@aol.com) wrote:
>: What is the speed of electricity?
>In a vacuum, the speed of electricity is equal to the speed of light. In
>any other medium, the speed is reduced. For practical purposes, the speed
>thru air or conductors is close enough to be considered the same. Don't
>confuse the speed of electricity (the propagation of an emf) with electron
>movement or so called "current flow."
> It is a common mis-perception that current flow is similar to the flow of a
>material like water. In an unbiased conductor, there is random motion of
>electrons at a high velocity (temperature dependent). Because this motion
>is random, if you looked at a cross-section of the conductor, there would
>be no net flow in a particular direction. If a voltage is applied across
>the conductor, there would be a slight bias in the net direction (over a
>period of time the motion would be more in one direction) of electron
>movement. This is known as electron drift and occurs at approximately
>.1 cm/sec.
In my humble opinion this is not really correct: The average drift speed of
individual electrons with an electrical field applied is dependent on the
current flowing, the wire diameter and the material the wire is made of. This
shows clearly when you consider that current can be seen as the amount of
charge (= electrons) flowing through a cross-section of a conductor. Ergo when
the current is zero, tyhe net drift speed is also zero (as you correctly
poited out), but when the current increases, so does the drift speed. The
type of conductor material determines the number of free electrons per volume,
and the less free electrons are available, the faster they have to drift to
result in the same current flowing. I've posted some extensive calculations on
exactly this same question some months ago, if anyone's interested I could
mail them; furthermore, please correct me when I'm wrong here.
>This means that if you hooked up a DC source to a 1 meter wire,
>it would take about 17 minutes for an individual electron to move thru the
>length of the wire. If you use an AC source(sinusoidal), the polarity is
>periodically reversing and the net drift is again zero.
> Joe S.
Regards,
Richard
Richard Rasker
Calslaan 54-11
7522 MG Enschede
Holland
tel. +31-(0)53-4350834
Roy McCammon
The practical answer, is that with ordinary conductors, the electric
signal takes typically 1 to 3 nano-seconds per foot to go from one end
of a wire to another. That is an observed fact.
As to how it gets there . . .
First let me say that no one really knows how electricity flows. All we
really know are our mathematical models. For many applications, the model
of water flowing through a hose is accurate. That's the model I use most
of the time. I even make my drafting technicians draw schematics so
that the most positive voltage is at the top. It helps me visualize
the current tumbling down from a high potential to a low potential.
Lets call this the first level model.
Now let me describe a second level model.
1. First, everything you thought you new about the propagation of
electricity down a wire is wrong (not really, but I always wanted to
say that) except maybe for DC , which doesn't really exist (you had to
turn it on). This model is based on the notion that electrical energy
always travels as an em (electromagnetic wave). It takes no account
of electrons (well, not explicitly). It simple considers a conductor
as something that will support a current flow when there is a driving
voltage. It does not care whether that current is a lot of charged
particles moving slow or a few moving fast.
2. The speed of an e-m wave in vacuum is fast, about one foot per
nano-second. The speed in other dielectrics (air, water, plastic,
glass) is somewhat slower, but still real fast.
3. The speed of an em wave in a conductor is real slow. One reference
I have ('Engineering Electromagnetics', by Hayt, McGraw-Hill, p345)
puts it at about 7 mph (miles per hour). Some athletes can run a
mile at faster that 15 mph. Notice that I said the speed in a conductor,
not along the outside. This is probably very contrary to some peoples'
notions. I expect to recieve a number of posts saying it ain't so.
Cite your references please. I can't say mine is a classic, but its
what they used to teach e-m to electrical engineering students at U of
Texas. (A pretty good school, if you ask me).
4. Do you remember Schnell's law of refraction? When a light beam
is incident on the interface between a low n (index of refraction) media
and a higher n (and therefore lower speed of light) media from the
low n side, at an angle, it enters the higher n media with its direction
of propagation bent toward the direction that is perpendicular to the
interface. The slow speed of em waves in a conductor gives rise to the
same bending. No matter what angle an em wave strikes a conductor, the
extreme ratio in speeds bends the direction so close to perpendicular
that for all practical purposes em fields in good conductors travel
perpendicularly to the surface. They go straight into the wire. They
do not go in the direction of the wire. So how, you ask, does the
energy get to the end of the wire. Glad you asked.
5. The energy is delivered to the end of the wire by the wire acting
as a wave guide. The energy is carried along the wire in the dielectric
next to the conductor. Visualize an em wave traveling along the wire with
its e field oriented radially away from the center of the wire and it m field
oriented along the circumference of the wire. (Don't take this as the exact
configuration, just use it as a visualization model.) Perhaps you thought
dielectrics like plastic and glass and air and vacuum were insulators that
blocked the flow of electricity. Well, that's true for dc, but AC currents
and energy pass through dielectrics. You can easily get a fatal electric
current through a plastic film capacitor. The current goes right through
the plastic film.
6. The wave guided by the wire, partly enters the wire and due to the
previously mentioned Schell's law effect propagates at a right angle into
the wire, where it wastes power. This waste is the cost of having the
wire as a wave guide. Is this ridiculous. well , not really. Remember,
the e and m fields are perpendicular to the direction of travel of the wave.
So, if the em wave is traveling into the wire with the e field directed
along the length of the wire, it will cause the currents (and electrons)
to go one way along the wire during the positive half cycle and the other
way in the negative half cycle. That's pretty much what the water flow
model says. Dang, maybe this em wave theory works after all.
7. How does all this collapse down to the waterflow model in the dc case.
Well, I don't know. I suspect it has something to do with the fact that
when the em wave enters the copper, it is almost but not quite perpendicular.
The very small component remaining along the wire may give rise to dc
currents in the wire.
Matt Jones
In article <4o6bl1$r...@nntp5.u.washington.edu> John Whitmore,
wh...@hipress.phys.washington.edu writes:
> Because the POWER propogates in fields outside the wire,
>and moves at nearly the speed of light. By analogy, if you connected
Could you please elaborate on this idea of things propagating *outside*
the wire?
I thought power was V*A, both of which are only measurable within the
wire. Well, maybe voltage can be measured outside the wire, but current
is the flux of charge within the wire, isn't it?
Thanks,
Matt
Michael Bruss
Roy McCammon (rbmcc...@mmm.com) wrote:
: But, the speed of light in a good conductor like copper is real hard to
: measure. Through careful experiments and measurements of indirect
: copper is real slow. I'll have to consult some references that I don't
: have access to right now, but as I recall, the speed of an e-m wave in
: copper is slower than a person can run. It turns out that this has
: something to do with skin effect. At high frequency, the e-m wave
: and so only the outer skin is available to support conduction.
: If thats so, why can't you throw a switch and then run to the end of
: an extension cord before the light turns on. Well, the answer is
: (where the speed is fast). The part of the insulation that the wave
: travels through is very close to the outer skin of the copper, so
: the portion in the copper sort of tugs on and slows the portion in the
: insulation and the portion in the insulation tugs on and speeds up
: the portion in the copper. Just how much is in each media depends
: on a lot of things, but, the upshot is that typically with the things we
: usually use, the speed of "electricity" ranges from 1 ft/ns to one
: third or so of that value.
OK, let's do a little experiment. We'll get in the space shuttle and go
into orbit. Then we'll attach a 1-km-long copper wire (no insulation) to
each of your arms. We'll even clean the surface of the wires so there is
not even an oxide layer. Then we'll apply a couple thousand volts to the
other end of the wires and flash a light at the same time. You are not
allowed to pull off the wires until 5 sec after the light flashes. If your
theory about the role of insulation in the propagation of electricity is
correct, you won't be harmed. Ready to try it?
Mike
Bob Myers
Matt Jones (jone...@ohsu.edu) wrote:
> Could you please elaborate on this idea of things propagating *outside*
> the wire?
> I thought power was V*A, both of which are only measurable within the
> wire. Well, maybe voltage can be measured outside the wire, but current
> is the flux of charge within the wire, isn't it?
That's how power is measured, but what John's pointing out is that the
actual transport mechanism (or at least a model which is usually applicable
over a wider range of problems) is that the fields are "really" where the
power is carried. These fields exist around/outside the conductor. This
isn't to say that the conductor and its properties aren't important - it's
just a different, and often more useful, way to look at the situation.
Bob Myers KC0EW Hewlett-Packard Co. |Opinions expressed here are not
Workstations Systems Div.|those of my employer or any other
my...@fc.hp.com Fort Collins, Colorado |sentient life-form on this planet.
Roy McCammon
fzb...@dale.ucdavis.edu (Michael Bruss) wrote:
>Roy McCammon (rbmcc...@mmm.com) wrote:
>: If thats so, why can't you throw a switch and then run to the end of
>: an extension cord before the light turns on. Well, the answer is
>: that the wave travels partly in the copper and partly in the insulation
>OK, let's do a little experiment. We'll get in the space shuttle and go
>into orbit. Then we'll attach a 1-km-long copper wire (no insulation) to
>each of your arms. We'll even clean the surface of the wires so there is
>not even an oxide layer. Then we'll apply a couple thousand volts to the
>other end of the wires and flash a light at the same time. You are not
>allowed to pull off the wires until 5 sec after the light flashes. If your
>theory about the role of insulation in the propagation of electricity is
>correct, you won't be harmed. Ready to try it?
>
>Mike
I used the word "insulation" instead of dielectric, because I thought
it would be a more generally understood. To be more acurate, the energy
travel through the dielectric, along the conductor. Vaccuum is a dilectric,
so the wires in space would guide the energy just like wires on the ground.
Electric energy does readily flow through insulation, by the way. Thats
how electricity flows through a capacitor.
Here's a better thought experiment. You are standing at the center of
a huge copper disk. The disk is 7 miles thick and the radius is a light
year. Directly on the other side, someone shoots a really big electron
gun at the disk. An em wave begins to propagate both through and along
the copper. The part that is guided along the copper takes about two
yeas to reach you, but the part that is coming through the copper
gets there in about an hour.
By the way, my refernce is 'Engineering Electomagnetics', William Hayt,
McGraw-Hill, 1967. Page 345 gives the speed for 60Hz of 7.2 mph. This
is inferred from the formula v=(f)(l) where f = 60 Hz, l=wavelength
in copper of 5.36 cm. v=velocity.
I'll see if I dig out a couple of more references over the weekend.
Roy McCammon
The power (actually the power density) in an e-m wave is the vector
cross product of the e field strength and the magnetic feild strength.
If the feilds are at right angles, as they are in a traveling wave,
this reduces to e field times m field. Now, the m field around a wire
is proportional to the current and the e field directed radially away
from the wire is proportional to the voltage at that point. So by
choosing units carefully, the total power guided by the wire is just voltage
time current. So you see, that the wave propagation model reduces to
the simpler model with reguard to total power. The wave model just
says the the enery flow is in the field around (and mostly next to) the
surface of the wire rather than in the wire.
Please remember that the question deals with the "speed" of electricity,
which I have defined as "how soon after I throw the switch does the other
end feel the effect?" Others have posted answers to the question "how fast
do the electrons move?" During the time in which the disturbence is
traveling to the of the wire, we are dealing with high frequency AC
signals. The disturbence is propagated to the end along the surface
of the wire long before anything significant happens inside the wire.
Kevin M. Jones
Is this why when you look closly at power substations, you see hollow
pipes instead of large wires connecting some of the larger transformers?
Michael Bruss
Kevin M. Jones (kjo...@mail.premier.net) wrote:
: Is this why when you look closly at power substations, you see hollow
This is an interesting point. If electricity only travels at the
dielectric interface, as some have claimed, then hollow conductors and
stranded cable should be a lot more common. It would save material and
weight.
Along the same lines, how does the "electricity only travels at the
dielectric interface" theory cope with conductance through salt water?
Mike
Julian Rendell
As a complete aside to the discussion so far, I'd just like to say
thanks to all of those who've added their 2c's worth so far- I've
recently been 'catching up' on EM fields... which are foreign to my
comp sci background.
It's been instructive to see other representations/visualisations of
these ideas- many of which are foreign to me (I've got a comp sci
background...)
Many thanks for your efforts- especially those backed up with
references.
Regards to all
Julian Rendell
Roy McCammon
This is a consequence of the "skin effect". The skin depth of copper at
at 60 Hz is about a third of an inch, so once the necesary diameter exceeds
about two thirds of an inch, the copper in the middle isn't doing any thing
except adding cost and weight.
A simple model that works well in this case is that the current is confined to
the skin of the conductor. The depth of this skin, in copper, is given
by the formula 66mm/sqrt(f) , f is the frequency in Hz.
The motion of the charge carriers does not stop abruptly at the skin
depth, so that there are some losses deeper in the conductor, but you will
get the same answer as if the current motion were uniform in the entire
skin and zero deeper than that.
Skin effect is a consequence of the difficulty an e-m wave has penetrating
a good conductor like copper.
Ron Schmitt
I feel that I need to clear up some of this confusion and misinformation
that has been spread. Here's my 2 cents: (And for those of you who need a reference, look in "Fields and
Waves in Communications electronics" by Ramo, Whinnery, Van Duzer which does a good job of
showing the EM field background for circuit elements and teh basic laws used in electronics.)
1) There is *big* difference between electricity (the flow of charged, massed particles) and
electromagnetic field/waves (the flow of virtual and
real massless, chargeless photons, which are the carriers of the EM force).
They are not equivalent.
2) In electric circuits the electric field exists inside AND outside
the conductor. This is not some myth, it is theory and observed fact.
(See above reference.) EM fields do NOT exist inside *perfect*
conductors, which are often what is assumed in certain textbook examples.
There are no perfect conductors in the real world.
For DC, there is a constant net drift of electrons through a circuit
(how else would a battery become depleted). For AC, the electrons slosh
back and forth.
Even with transmission lines, antennas, and waveguides, where the behaviour is dominated by EM waves, there
are
E fields inside the conductor, and real electrons are moving in relation to
these fields. This is also observed fact.
3) Introductory EM textbooks give examples that illustrate EM
phenmonen and make many assuptions so that closed form solutions
can be found and so you don't need to invoke a computer to solve them.
Solutions for uniform EM plane waves in semi-infinite (infinite
in 1 or more directions) conductors, and infinite length transmission lines with perfect conductors are NOT
*real world*. We use similar
models in physics all the time, even though we know that they're not real-world.
For instance, in many introductory mechanical problems, we assume that
there is no friction.
4) Electricity (charge particles) does NOT flow through air other dielectrics. Air is not a conductor
(except when exposed to very large voltages and the air gap then becomes ionized as in lightning and static
electricity). As everyone learns in intro electronics, charge does not jump across the gap of a
capacitor; electrons accumulate on one side and deplete on the other. For AC electricity, the electrons
accumulate and deplete alternately allowing us to use the myth that AC current flows through the capacitor.
On the other hand, EM *fields/waves* do flow through air very easily without much loss.
5) Ok, Here's what actually happens inside electrical circuits.
In typical circuits at DC and audio frequency, the transfer of energy is dominated by fields *INSIDE* the
conductor, and by the resulting electrical currents. Whenever there is a current inside a conductor, there
must be a proportional E-field in the same direction. The reverse is also true. This is stated in ALL EM
texts by the following relation:
J=siqma * E
where J (a vector) is current per unit area, sigma is conductivity, and E (a vector) is the electric field.
J and E point in the same direction.
I beleive that you can find a derivation of this formula
in the advanced text "Classical Electrodynimaics" by Jackson. This formula is not derived in most intro
texts.
This is the EM way of stating Ohms law (V=IR).
In such circuits, there are also external fields, but they are not the dominant phenomenon.
The current does travel on the surface of the conductor. This is known
as the skin effect. How deep the surface is depends on the conductor and the frequency of the current. For
DC and a typical wire, the "skin" is actually deeper than the wire itself, and the current denisity is
virtually constant throughout the wire. As frequencies get higher, the skin gets thinner and the E fields
and current only exist very close to the surface, but they are still *inside* the condutor.
In the referecned text, there are diagrams of the electric field and current denisty that occurs *inside* a
wire at various frequencies.
AS wires get longer (close to the size of the wavelength of the transmitted
energy), the EM fields betwen wires and between wires and ground start to dominate the circuit. Most of the
energy is transferred as a wave, external to the conductors. The wires now act as waveguides and the
characteristic impedence of the dielectric between conductors is used
for resistance calculations, instead of the ohmic resistance. There are still currents and ohmic losses in
the wires, but they are not as important because the fields that travle outside carry most of the power.
I don't know why these effects occur at higher frequencies (shorter wavelengths). Anyone know?
6) The speed of electricity (net drift of the electrons) is quite slow. I can't quote any numbers,
but someone did post this recently on this group. I think it was something like a centimeter per minute.
The speed of EM waves *into* a large (in x-y direction) block of a conductor is very slow, as quoted
by the originator of this stream. In addition, it varies with the conductor and the freqeuncy and can be as
slow
as a few miles an hour. BUT because of the skin effect, a wave will also not travel very far
into a conductor. So it's a mute point. (This effect is why a thin sheet of conductor can serve as
suitable EM shield for electrical equipment). I don't know what the speed of a wave travelling *along* the
skin the conductor, which is the phenonenon that occurs in electric circuits. Common sense tells me that
since the external wave is travelling at the speed of light within the given dialectric, then the wave
inside travels at the same speed. Elsewise there would be a discontinuity at the surface of the wire, and
nature does not like discontinuities. Commonsense also tells me that since the
EM fields must co-exist inside and outside the conductor for transmission to take place, and since it
doesn't take 8 minutes for my phone to ring when someone calls me (and trust me there is current moving in
that wire), then the surface wave probably travels at the same speed as the external wave.
I welcome any comments.
Ron Schmitt
rsch...@nortel.ca
Nortel / Northern Telecom
Philip Van Houtte
Michael Bruss (fzb...@bullwinkle.ucdavis.edu) wrote:
: Mike
To my understanding, it's the electric field that carries the power (or
information), the motiion of the electrons and ions (Na+, Cl-) simply
react to field.
Philip
Roy McCammon
fzb...@bullwinkle.ucdavis.edu (Michael Bruss) wrote:
>Along the same lines, how does the "electricity only travels at the
>dielectric interface" theory cope with conductance through salt water?
Oh dear. This will tke a lot of words, and I will probably do a poor job.
But first, let me quote a portion of my previous posting that was so kindly
quoted previously my Michael Bruss :
"that the wave travels partly in the copper and partly in the insulation"
Did that create impression that "electricity only travels at the
dielectric interface"? If so, I must apologize for not being clearer
on that point. I live firmly in the 'never say never' camp.
I must say, that I don't know how electricity really propagates, but I am
familiar with three theories about it. They are QED (quantum
electro-dynamics), Maxwells equations, and circuit theory (lumped parameter).
Maxwell is regarded as an approximation of QED, and circuit theory is an
approximation of Maxwell. QED is the 6000 pound gorilla of all physical
theories. It wins all arguments against of theories. It even drove
Einstein into silence when he disagreed. Maxwell wins all arguments
where it disagrees with circuit theory.
I've never measured the field around a wire, but antennas do work. The
energy must get into the dielectric somehow. Likewise I have observed
delays in coaxial cables that are just what Maxwell says they should be.
I know a lot about circuit theory, some of Maxwell and very little about
QED. The most useful thing I know about QED is that, according to
Richard Feynman (in his book QED, which is for the general reader and
should be available in most libraries) is that QED is unexplainable
and cannot be intuitively understood even by Feynman. Thats a relief,
because I don't understand it either and it looks like I am in very
good company.
Circuit theory covers most of ordinary electronics, including lighting,
house wiring, audio circuits, computers and most things up to video
frequencies. One failing of circuit theory is that it does not handle
the delay in transmission lines, the importance of the dielectric, crosstalk,
reflections, antennas, capacitors, or inductors. For example, at very high
frequencies, you may find that wire with a certain kind of insulation
does not work well. Circuit theory does not provide a clue. In fact, if you
believe the "electricity" is entirely confined to the wire, you may not even be
willing to try another type of insulation. Likewise, according to circuit
theory, two coaxial cables that are identicle except for the dielectric
should have the same delay and loss. After all, if the electricity stays in the
copper, how could the plastic effect its propagation? But, you need look
no further than the Belden catalog. Types 8221 and 8254 have identicle dc
specs, but the latter is faster and lower loss at 1GHz. Maxwell takes over
and handles these thing as well also handling ordinary optics, fiber optics,
refraction, and radio propagation. Maxwell doesn't handle semiconductors,
ionic transport and super conductivity. These things require QED or one of
its other subsidiary theories (Chemistry).
So, what follows is the interpretation according to Maxwell.
1. All electric energy transmission is by em waves.
2. When we are using a transmission line which is made up of a good conductor
(like copper at 60Hz) and a good dielectric (like vacuum, air, or
polypropelene), the feild outside the conductor and in the dielectric is
responsible for carrying almost all the power to the load. The field
in the conductor is responsible for loss (heats up the conductor). The
direction of power propagation in the conductor is almost
directly into the conductor (and is absorbed there as heat). The
approximate depth of penetration of this power is known as the skin depth.
The very small component in the conductor that is along the direction of the
conductor is a very small fraction of the power disapated in the conductor.
This model covers the case of house wiring, coaxial cables, and most
intentional circuits.
3. Finally we get to salt water.
Case 1, A transmission line. We'll fill a plastic pipe up with salt water
and use it to transmit electricity. Everything I said above about
transmission lines applies.
Case 2, you want to talk to a submarine in the ocean. Salt water is not
nearly as conductive as copper, so the skin depth is much deaper for
salt water at the same frequency. If you choose a low frequency, and
the submarine is not too many multiples of the skin depth deep, you
can transmit to it. In this case, the major portion of the power is still
guided (reflected) by the interface to keep it in the dielectric (air), but
just like the copper wire, some penetrates into the conductor, where it
is absorbed as heat and a little is absorbed by the submarine receiver
as useful signal.
Case 3. You want to electrocute a felon in the electric chair. You make
the connection to the felon using sponges loaded with salt water and plan
to use 60Hz electricity. The skin depth at 60Hz is much greater than the
thickness of the salt water so the em wave could progate right through it.
There is a good chance that there is ionic transport. That would require
QED or chemistry. Anyway, the felon recieves a lethal current. Circuit
theory works just fine on this example. You simply postualte that there
is a certain series resistence associated with the salt water.
I promised that I would find some more references about the speed of
em waves in good conductors. But, this post is getting long, so I think
I'll post that as a reponse to this.
Roy McCammon
As promised, here are references on the speed of light through conductors.
It turned out that every book I picked up about the subject of em waves said
explicitly or implicitly (just combine a few formulas) that the speed of light
in copper or silver was on the order of 7 miles per hour at 60 Hz.
Ref 1 says explicitly that the speed is 7.2 mph. Ref 2 requires you to
combine two equations. Ref 2 is by the Nobel prize winner Richard Feynman.
He wouldn't lead us astray. Refs 3 and 5 refer to silver rather than
copper, but get about the same answer. Refs 3 and 4 give examples with
higher frequencies that have to scaled. All say the formuli apply for
sub optical frequencies (say less than 1 Terra Hz)
Most libraries will have a copy of ref 2.
References:
1. 'Engineering Electromagnetics', Hayt, McGraw-Hill,1967, p. 345.
(Third year engineering text)
2.'Lectures on Physics, Volume 2',Richard Feynman et al, Addison-Wesley,
1964, chap 32 pp 6,11.
3. 'Foundations of Electromagnetic Theory', John Reitz and Frederick
Milford, Addison-Wesley, 1967, pp 303-305. (Third year physics text)
4. 'Time Harmonic Electromagnetic Fields',Roger Harrington, McGraw-Hill,
1961, reprinted 1987, pp 5,52-53. (Graduate engineering text)
5.'Waves, Berkely Physics Course-Volume 3', Frank S. Crawford Jr., McGraw-Hill,
1968, pp 56, 58, 579, 583. (Second year physics text)
Ref 2
Chapter 32 page 11 gives Nr = sqrt[ s/(2ew) ]
for frequencies less than 1 THz
where s = 57,600,000 mho/meter (copper)
w = 377 rad/sec for 60Hz
e = 8.85 pico farads/meter
Nr = real part of index of refraction
which gives Nr = 92,909,000 (for 60 Hz)
Chapter 32, page 6 gives the speed of an em wave in a metal as C/Nr
which gives v = 3.2 meters/sec = 7.22 mph for 60Hz
= 41,300 meters/sec for 10GHz
Ron Schmitt
Roy McCammon wrote:
>
> As promised, here are references on the speed of light through conductors.
> It turned out that every book I picked up about the subject of em waves said
> explicitly or implicitly (just combine a few formulas) that the speed of light
> in copper or silver was on the order of 7 miles per hour at 60 Hz.
>
> Ref 1 says explicitly that the speed is 7.2 mph. Ref 2 requires you to
> combine two equations. Ref 2 is by the Nobel prize winner Richard Feynman.
> He wouldn't lead us astray. Refs 3 and 5 refer to silver rather than
> copper, but get about the same answer. Refs 3 and 4 give examples with
> higher frequencies that have to scaled. All say the formuli apply for
> sub optical frequencies (say less than 1 Terra Hz)
>
I definitely don't argue about these calculations for the speed of a TEM (E
field is perpendicular direction of travel) waves into a conductor. But the
waves inside a circuit wire, whether a short wire or a 1/4 mile long
transmission line, consist of an E field directed 100% in the direction of
propagation (i.e. along the wire not into the wire), along with an (almost)
TEM wave travelling outside the wire. My guess is that the wave front in the
surrounding air travels at the speed of light (for air) and locally causes an
E field inside the surface of the wire as it propagates. This would allow for
the E field to propagate inside the wire at essentially the same speed but gets
around this issue of the slow speed of EM waves in a conductor. For example,
in the case of a 1/4 mile long telephone line, the EM waves would reach the
load (your phone) at the same time inside and outside the wire, even though
the speed through air and the conductor are different. A good analogy would be
a speed boat travelling at a hundred miles an hour through the water. The
waves from the boat don't travel at a hundred miles an hour, but waves do
immediately appear all around the boat no matter how fast it goes.
--
Michael Bruss
Let's take the salt water problem one step futher. Suppose a couple of 1
cm diam. round platinum electrodes are immersed 1 meter into salt water.
They are 10 cm apart, the salt water pool is very large,
and they have insulated leads which can be connected to a power supply.
Then several hundred volts (we'll say DC, but if anyone wants to address
the AC situation, go ahead) are applied. Current will flow without there
being an obvious dielectric interface. Is this simply the case of a
distributed submicroscopic dielectric (the water) surrounding a
distributed submicroscopic conductor (the ions)? Is the propagation of
electrical force due to movement of the ions or a disturbance of their
electron clouds?
Mike
Roy McCammon
Very Good analogy.
The apparent speed (frequency times wavelength) of the field just inside
the conductor in the direction along the conductor is called the phase
velocity. Phase velocity can exceed the speed of light in the media.
John Whitmore
In article <4ou0dr$h...@mark.ucdavis.edu> fzb...@bullwinkle.ucdavis.edu (Michael Bruss) writes:
>Kevin M. Jones (kjo...@mail.premier.net) wrote:
>: Is this why when you look closly at power substations, you see hollow
>: pipes instead of large wires connecting some of the larger transformers?
No, it's because of skin effect. Even at 60 Hz, it is
possible
to measure the exclusion of current from the center region of a
conductor (but it's a small effect until the conductors get an inch
or so in diameter). That current exclusion is due to the lag
in moving a magnetic field line through the bulk of the wire...
>This is an interesting point. If electricity only travels at the
>dielectric interface, as some have claimed, then hollow conductors and
>stranded cable should be a lot more common. It would save material and
>weight.
No one said 'electricity only travels at the dielectric
interface'. The claim was that the ENERGY FLOW in current-carrying
wires was mainly by electric and magnetic fields surrounding
the wires (hence, fields that penetrate the insulation of the wire).
Do the math; the kinetic energy of a few lightweight electrons
isn't nearly enough to make a light bulb glow.
Hollow conductors only work out to be efficient when the
slow magnetic field penetration of the conductor inhibits
current in the core of the wire, which happens at 1" size wires
for 60 Hz, and at smaller wire sizes for higher frequency.
The phenomenon is called 'skin effect'. Radio frequency wiring
frequently uses thin silver plate on copper, or copper plate over
steel, or even gold plate on printed wiring traces, for this reason.
John Whitmore
Roy McCammon
1. First, let me reiterate a general principal of using
conductors to guide electric energy: The fields at the
dielectric/conductor interface are responsible for the
electric power carried to the load, the fields inside the
conductor are responsible for the electric losses in the
conductor.
2. The fields inside the conductor are thought of as arising
from the fields outside the conductor by a Snell's law type
mechanism. In a transient event, the fields inside appear
to lag the fields outside.
3. Saltwater is very complicated, involving ion transport,
mass flow, mixing, convection, eddies, gas generation and so
forth. It is outside the realm of a purely Maxwell based
analysis. I will ignore that and assume that salt water is
just a linear conductor and that the movement of the water
does not effect the movement of the charge.
4. Maxwells equations are hard to solve. One approach is to
decompose the driving source into sine waves, assume a
constant fixed frequency and solve the wave equation for
each frequency, and then recombine the results using Fourier
techniques. It is so hard, that its only done when circuit
theory doesn't work, or doesn't address the issues (like how
much energy is radiated?) In what follows, I certainly
haven't solved the wave equation, even for dc. I'm simply
going to give a qualitative description of what I think
would happen according to Maxwell. Here goes:
5. Since we are at dc, the skin depth is infinite.
6. There is a current flow and hence an e field pointing
generally from the positive electrode to the negative
electrode. There is power loss in the conductor (RI^2) and
hence there must be energy flow into the conductor.
7. The field and the energy flow are created and maintained
by em waves. This field pointing from one electrode to the
other cannot be set up by an em wave traveling from one
electrode directly to the other, because the efield of a
wave is always at a right angle to the direction of travel.
8. Thee field is at a right angle to the em waves
direction. These waves extend the entire width, length,
breath of the saltwater, and appear to be sourced and
reflected at the salt water to dielectric interfaces.
9. The saltwater is large, but small with respect to the
infinite skin depth, so where ever the saltwater stops, be
it at air, or rock or plastic insulation, there is a
water/dielectric interface.
10. The interface between the water and the wire insulation
is particularly important.
11. This is how I see the waves propagating: the are guided
by the copper wire and travel in plastic insulation until
the insulation stops. At that point there is a triple
junction of copper, water and plastic. Some of the wave
enters the water and continues to be guided by the copper
but traveling in the water. The rest of the energy is
guided by the conductive water to make a turn and go the
other way in the plastic insulation, this time at the
water/plastic interface. These fields may partially cancel
the fields going the other way. Once the waves reach the
water/plastic/air junction, almost all the wave turns and
travels along the water/air interface to the other wire
where it is guided in by the water and back out by the
copper and onward to the load. The waves also goes all over
the surface of the water, including the underside. All
though it goes everywhere, it is strong only near and in
between the wires. For example if the wires are a meter
apart, the voltage may be undetectable a mile away. As
these waves travel, they partly enter the water from the
surface at almost a right angle due to the Snell phenomenon.
These waves refracted into the water are oriented at just
the right direction to set up the internal field that pushes
the charge carriers and they are traveling the right
direction to supply the power that is lost to the water.
12. Final conclusion: fields in the dielectric interfaces
carry energy to the load (if there is a load), fields in the
conductive media carry the power to the media that is lost
to the load.
13. You may note that the salt water has played the role of
both dielectric and conductor. As you may expect, these are
relative terms.
Roy McCammon
wh...@hipress.phys.washington.edu (John Whitmore) wrote:
>The phenomenon is called 'skin effect'. Radio frequency wiring
>frequently uses thin silver plate on copper, or copper plate over
>steel, or even gold plate on printed wiring traces, for this reason.
*** And just to elaborate a little. The skin depth at microwave freqs is
very shallow, hence the gold over copper. Gold is not a better conductor
than copper, but it is a heck of a lot better than copper oxide.
> No one said 'electricity only travels at the dielectric
>interface'. The claim was that the ENERGY FLOW in current-carrying
>wires was mainly by electric and magnetic fields surrounding
>the wires (hence, fields that penetrate the insulation of the wire).
>Do the math; the kinetic energy of a few lightweight electrons
>isn't nearly enough to make a light bulb glow.
*** Here is a supportive little thought experiment. This is safe to
try at home, so go ahead (I couldn't resist saying that. I've had to
sit through too many episodes of the science guy on PBS)
Let there be three loads, and four sources. The sources are
approximately 1Kv, -999V, 1V, and 1V.
The loads are 1Mohm, 1Mohm, and one K ohm. The first load (1M) is
connected from the 1KV source to ground. The second load (1M) is from a 1V
source to the -999V source. The third load (1Kv) is from a 1V source
to ground. The wires connected to the positive side of each load are
identicle. The actual source voltages are adjusted slightly to make
the current in each load equal to exactly 1 milli amp.
Each wire is identicle. Each wire sees the same current, the
same internal field, and has the same electron drift rate. Each wire
receives the same amount of power that heats the wire.
If you could examine the conditions inside each wire, you would find them
indistingishable (identicle). Yet, the wires connected to the first two
loads are delivering 1000 times the power to the load as is the other
wire (1 Watt vs 1 milliwatt). If the conditions inside the wires are
the same, then at least 99.9% of the power in the first two instances
must be carried somewhere else (ie, in the external fields).
You can readily extend this to 1Mv, 1Gv and 1Tv sources and see that all the
power delivered to the load must be external to the wire. This same analysis
would apply to salt water.
Shaun Reemeyer
For those of us that don't want to get into the nitty gritty of EM etc, i have read several
sources (mail me) that say the speed of electrons through copper wire is about 2% the speed
of light, and that superconductor technology aims to improve this. Which is why superconductors
are used for high-speed applications.
Reemus
Electronics Research Engineer
"kindness in another's trouble, courage in your own"
-Adam Lindsay Gordon , an Australian
Massoud Ajami
In article <31BFDAB8...@sedal.usyd.edu.au> Shaun Reemeyer <sha...@sedal.usyd.edu.au> writes:
>From: Shaun Reemeyer <sha...@sedal.usyd.edu.au>
>Subject: Re: WHAT IS THE SPEED OF ELECTRICITY?
>For those of us that don't want to get into the nitty gritty of EM etc, i have
>read several
>sources (mail me) that say the speed of electrons through copper wire is about
>2% the speed
>of light, and that superconductor technology aims to improve this. Which is
>why superconductors
>are used for high-speed applications.
Well, find better book to read. :) :)
Two things to notice. Since electrons can not be distinguished from each
other, if you insert one electron into a wire, how fast the _effect_ of that
action would be detected at the other end? That is the speed of the
traveling signal, and about 2/3 of c.
Now if you insert one [colored] electron (for hypothetical distinction) into
a wire, how fast you can catch the same electron at the other end? Well, you
can walk to the other end, _wait_ until the electron arrives.
The reason is that the path of electron is not directed toward the direction
of moving charge (goes in zig-zag). Based on these paths, and also electron
losing the energy due to collision, if you can make the non-collision path
and straight line (almost), then it is a superconducting material.
--
Peace and Prosperity!
---==< 110 >==---
Joss Buttler
In most common electrical circuits, the speed of the electric current is quite slow, typically on the order of millimeters per second. This slow movement of individual electrons is due to the fact that electrons drift through the conductor in response to an electric field, colliding with atoms and other electrons along the way.
It's important to distinguish between the speed of individual electrons (drift speed) and the speed at which electrical signals or energy propagate through the circuit, which is essentially the speed of electromagnetic waves (such as light). In electrical circuits, signals or changes in voltage propagate very quickly, typically close to the speed of light, which is approximately 299,792,458 meters per second (or about 186,282 miles per second) in a vacuum.
So, while electrons themselves move relatively slowly within a wire, the electrical energy, information, or signals carried by the electrons can travel through the wire and a circuit at nearly the speed of light, depending on the properties of the materials and the geometry of the circuit. Source https://lescobillonline.net/