Speed of light demo using pulsed electromagnet

[taylaron]

Greetings PF.

I've got an idea for a demonstration of the speed of light utilizing a pulsed electromagnetic coil and ferrofluid.

I'm concerned with the feasibility of this project as it may push the limits of semiconductor technology. I would appreciate any insight on the difficulties of this project since I'm not an electrical engineer. However, if it is feasible, i'll try to recruit some EEs to help me out.
Here's how I expect it to work:

When ferrofluid is placed on an electromagnet, the fluid is drawn to the magnetic field lines. With that in mind, I want to create a pulsed electromagnet where segments of the coil are energized and other aren't. This will result in the ferrofluid being attracted to the energized segments. The pulse duration will need to be about 1.2 nanoseconds with a coil diameter of 0.3m. calculation below:

At 0.66c the speed of light (signal propagation in copper) is 197863022 m/s. With a 0.3m diameter coil and a 0.23m pulse wavelength, the oscillation rate comes down to ~1.2 ns pulse rate.

The standard spiral coil which may make things difficult as more wire is required to make a revolution with the increasing radius. I wonder if there is a better winding geometry for this purpose. The wire on the coil will also need to be extremely small in diameter to allow for low current switching. I'm also concerned about rise/fall time and timing accuracy (I assume we'd use a crystal oscillator for a trigger).

I'm sure there's a lot of issues I haven't thought of which is why I'm posting here. I appreciate any help you can provide.

 

[Baluncore]

Given those ingredients, how do you operate the apparatus to get a result?
How do you interpret the result to get the speed of light?

The speed of signal propagation on bare copper is close to 98% of the speed of light.
It is about 66% in a coaxial cable with a sold polyethylene dielectric.

Ferrofluid will not respond to high frequency RF currents.
How fast do you expect the ferrofluid to move?

Do you need to know the frequency or the wavelength of the signal in the coil?

 

[taylaron]

Given those ingredients, how do you operate the apparatus to get a result?

The device would simply be turned on and the ferrofluid would be attracted to the energized field segments. Tuning the pulse widths to match the coil will be the trick.

How do you interpret the result to get the speed of light?

I admit the experiment is a bit of a stretch, but you could say that in about a bilionth of a second, electricity has traveled 'this' far. I suppose I should re-title post to speed of signal propagation, not speed of light.

The speed of signal propagation on bare copper is close to 98% of the speed of light.
It is about 66% in a coaxial cable with a sold polyethylene dielectric.

I got the figure from http://en.wikipedia.org/wiki/Propagation_delay which states "In copper wire, the speed s generally ranges from .59c to .77c.[1][2]"

Ferrofluid will not respond to high frequency RF currents.
How fast do you expect the ferrofluid to move?

I wanted to use a transistor that would simply turn on and off and not do a full wave from +5v to -5v.

Do you need to know the frequency or the wavelength of the signal in the coil?

I'm beginning to wonder if I've got my understanding backwards. I thought the signal would erect a standing wave in the coil, but you can't do that with electricity... The signal would propagate around the coil like anything else. The pulse wouldn't stay in the same location for no reason... Well, that really screws up this idea.

 

[Baluncore]

I got the figure from http://en.wikipedia.org/wiki/Propagation_delay which states "In copper wire, the speed s generally ranges from .59c to .77c.[1][2]"

Ah, but that is wikipedia. A signal actually travels in the insulation, on the copper, not in the copper. The velocity factor on a wire is simply determined by the dielectric constant of the insulation.
http://en.wikipedia.org/wiki/Velocity_factor

To measure the speed of propagation, (phase velocity), you will need to know two of three parameters. Only velocity, wavelength or frequency can be computed, given the other two.

Wavelength was used to specify very high frequencies because wavelength can be measured by setting up standing waves on an open ended transmission line.
http://en.wikipedia.org/wiki/Lecher_lines
http://en.wikipedia.org/wiki/Lecher_lines#Measuring_the_speed_of_light

Ferrofluid has variable dielectric properties due to the oil and colloidal content that will change the phase velocity.
Ferrofluid responds to fixed or very slowly changing magnetic fields, not to high frequency RF fields or standing waves.

I wanted to use a transistor that would simply turn on and off and not do a full wave from +5v to -5v.

That would generate an RF signal, the AC component, plus a DC offset. The DC component would control the ferrofluid, the AC component would not effect the ferrofluid.

It is hard to tell if a spiral pancake coil would couple between windings more or less than be like a length of transmission line in a spiral. The 1 nsec pulses would not fill the space about the spiral in any simple way. The pattern could be calculated if you knew the speed of light, but ferrofluid and your eyes would be too slow to show a pattern.

 

[davenn]

I got the figure from http://en.wikipedia.org/wiki/Propagation_delay which states "In copper wire, the speed s generally ranges from .59c to .77c.[1][2]"

Ah, but that is wikipedia. A signal actually travels in the insulation, on the copper, not in the copper. The velocity factor on a wire is simply determined by the dielectric constant of the insulation.
http://en.wikipedia.org/wiki/Velocity_factor

yes exactly

I was going to post the same answer till I saw Baluncore's response
The velocity factor on a bare copper wire is ~ 0.98 c

as soon as you put plastic insulation on it, make it a track on a PCB etc you have introduced a dielectric into the equation and it significantly lowers the VF

and just in case you didn't understand this part of Baluncore's response ...

A signal actually travels in the insulation, on the copper, not in the copper

and question how a signal travels in the insulation

The signal travels as an EM wave outside the copper(other metal) wire (PCB track)
The wire primarily only acts as a "waveguide" guiding that EM wave from one part of the circuit to another.


Dave

 

[taylaron]

Unless I'm wrong, there isn't a way to make this demonstration work. Correct?

-Tay.

 

[Baluncore]

That would appear to be the case.
I see many obstructions, but no solutions. I can see no way to do it.