# Speed of electricity

when a light switch is flicked, does the light have to wait for electrons to get from the switch to the bulb before it lights up, or would electrons already lined up along the length of the wire simply move towards the bulb and so light up the bulb in a much shorter time?

Meir Achuz
Homework Helper
Gold Member
Yes, an individual electron does not travel very far.

russ_watters
Mentor
That's yes to the second part....

If I remember correctly, I think electric flow can be modelled as a diffusion of electrons, similar to heat flow.

Because of resistance it would take a percievable period of time to turn a light on with a distant switch (say 10 ft away) if the electrons were not already "lined up" as you say.
(someone correct me if I'm wrong)
The drift speed of electrons in a copper wire is something around 0.0043m/s.
See http://hyperphysics.phy-astr.gsu.edu/hbase/electric/ohmmic.html for more details.

when a light switch is flicked, does the light have to wait for electrons to get from the switch to the bulb before it lights up, or would electrons already lined up along the length of the wire simply move towards the bulb and so light up the bulb in a much shorter time?

Yes, that is more or less how it works. There are a lot more theoretical details that go into this, but what you've said reflects the basic idea. If you want numbers, I remember that my junior year experimental physics book mentioned that electric signals travel down a wire at about 2/3 the speed of light (we were using wires of different length to get two signals to reach an apparatus at different times). But if you want an explanation as to why this is the case, I'll need to think a bit before I respond.

Last edited:
An electric wire is a bit like a pipe crammed full of peas. Push a new pea in one end and one falls, almost instantly, out the other.

An electric wire is a bit like a pipe crammed full of peas. Push a new pea in one end and one falls, almost instantly, out the other.
No. It takes time for the process to reach the other end. I'm fairly sure it's a diffusion process.

No. It takes time for the process to reach the other end. I'm fairly sure it's a diffusion process.

Yes, it takes time, otherwise it would exceed the speed of light, which is a gigantic no no. But that's why I said "almost".

I would call it more of a propagation.

Maxwells Demon
Like DaveFaz says, a pipe crammed full of peas...

The drift speed of electrons in a copper wire is something around 0.0043m/s.

and another one said

electric signals travel down a wire at about 2/3 the speed of light

which one is right.. must be the one with the 2/3 right?

ZapperZ
Staff Emeritus
They are all correct. This is because in the semi-classical treatment of conduction electron in metals, these electrons are considered to be an "electron gas" within the material. And if you have learned anything about the ideal gas law, you will know that these particles continue to make "random" motion with collisions with each other and the lattice ions. The average speed (not velocity) of each of these electrons is LARGE, while the whole group speed, which is the drift velocity, can be rather low. In fact, without any external field, the drift velocity is zero.

Zz.

Maxwells Demon
velocity is speed in a straight line (or same direction) right ?

all right so the speed the electrons bang into each other with is great but the "forward going" speed is quite low..

velocity is speed in a straight line (or same direction) right ?

all right so the speed the electrons bang into each other with is great but the "forward going" speed is quite low..

The drift velocity figure given was absolutely correct. The (2/3)c value is the approximate speed of the signal itself. In practical terms: if you connect a wire to an apparatus that can acquire the signal (a DAQ card, for instance), and then send a signal to it, the computer will not receive the signal instantaneously. The information is transmitted at about (2/3)c, but any given individual electron still travels at a very slow speed. Make sense?

Maxwells Demon
yeah it does.. Could you define what a signal is?

yeah it does.. Could you define what a signal is?

Here I'm using the word to describe the information transmitted by the electric current. When an electric current flows, it's because an electric field is created in the wire, and this field causes the migration of charge. Though the speed of the charge carriers (drift speed) is very low, the net flow rate of current is high.

russ_watters
Mentor
The "signal" is essentially the electromagnetic version of a pressure wave. If you were to open a water valve quickly, the water near the valve would start flowing right away, but water at some distance away would not. This creates a shock wave that travels through the water (starting at the valve) at the speed of sound. Once the shock wave passes any point in the pipe, water at that point will flow. Effect isn't noticeable when opening a valve, but it can be very noticeable when closing one: it is the cause of the "water hammer" effect.

Last edited:
Maxwells Demon
Very good Russ.. I like these very essential explanations.. "Water hammer" effect is where the valve kind a shocks or how to say it...right? (I'm not from an English speaking country)

russ_watters
Mentor
In the water hammer effect, the shockwave will travel back and forth until it dampens itself out, banging against valves, elbows, everything that is a little bit of an obstruction.

Maxwells Demon
if you stop the water at the valve by closing it then all the water some distance away will come smacking in to the valve and then going back and forth ?

HallsofIvy
Homework Helper
I might point out that your wires probably are AC (alternating current) rather than DC (direct current) so NO electrons ever travel from the lightswitch to the lamp- they just move back and forth a very short distance.

russ_watters
Mentor
if you stop the water at the valve by closing it then all the water some distance away will come smacking in to the valve and then going back and forth ?
The wave goes back and forth the entire length of the pipe. Individual water molecules only move a tiny distance.

Maxwells Demon
That's also what I meant, but thank you for the notification mate :)

Hi all:
Have been searching for the exact distance of this as well. I can remember that a nano second move the distance of a foot or so. Vague at best. That was the analogy used to define what a nano second was. The time it takes for electricity to travel from one end to the other.
I am toying with the tesla thing using dc pulses and am trying to figure out how short the pulse timing needs to be so that I can turn it off just before the end of the wire. This reminds me of using echo testing of cables. If you got a large return signal, then it was bouncing off of something that should'nt be there as in a bad connector or such. A small return was good. If I had a good scope to use, that would tell the story.
So now if the person here has a fast scope, the wire length will determine how long it takes the signal to bounce back. Devide that by two and you should have your answer.

Hope that is of some help.

Thaelin