Electric field , sphere , transformer.

1. Jul 19, 2013

Crazymechanic

Hi , one of my electric field and how it works questions.

So we have a sphere made from wire or metal beams etc doesn't matter.Now we wrap wire around it and leave the ends separate, we attach a circuit of a resistor and an ammeter , for the sake of it a voltmeter in parallel.
Now we put a smaller sphere inside this bigger one , using electronics we switch the spheres on , one at a time , then we switch the other one on while the first one off and so on at a rate of "x" hz.
Now is there current induced in the windings and does this act like a transformer , only the difference that instead of primary windings which create magnetic field in a iron core which in turn creates current in the secondary, here we have a sphere made of metal and windings around it. Because the field strength inside the sphere is zero , all the field resides outside of the sphere so of every winding one side becomes polarized , the field is pushing on the electrons in the wire , now when the outer sphere is switched off and the inner turned on the same thing happens only in the opposite manner.
I will add information if necessary tomorrow as it's late and im tired.

2. Jul 19, 2013

Simon Bridge

Where is it?

You mean a spherical framework made of a conducting material? The geometry of the framework maters.

Just randomly or according to some pattern? Is the wire insulated?

So this will be an L-R circuit.

- similarly constructed?

How? Do we suppose that the inner inductor has a couple of wires coming from it that pass through the outer inductor?

So we want to drive the two inductors from voltage sources which are 180deg out of phase?

The geometry of the framework and the windings and how the two spheres are oriented wrt each other will be important to figuring the operation of the device described. The inductors (wire spheres) could, in principle, be modelled as two cylindrical coils. You then get to think about what happens when you put one coil inside another.

Only for a uniform conducting sphere. But then you wouldn't have an inductor. Make up your mind.
It also only applies at electrostatic equilibrium - constraining the charges to move along wires means that it could take a while for them to reach equilibrium, by which time, you may have changed the applied electric field.

3. Jul 20, 2013

Crazymechanic

Oh thanks for the reply and please excuse me for not attaching the drawing as promised , again I was working late had a lot of pain putting together some old carburetors etc.

Ok so now with a better mindset here we go.
The two spheres for our thought experiment are not the same , the outer one which has the wire wound on it looks like the one in the drawing the inner one is just a closed metal ball with an empty middle.To make the charge more uniform on it.
The winding ofcourse is and insulated one , it is not electrically connected to the framework sphere it sits on.Just like a transformer winding is insulated from the core.
Yes you could say the spheres pulsate with a chopped dc 180 dg out of phase, something similar like a full bridge dc dc converter.

Now let me explain why am I bothering with something like this at all , I have attempted and finished some smps in my life and the problem that i recognize with them always or almoust always is that the current that goes through the coil creates a magnetic field in the core and when this field goes to it's highest point we say the core is saturated , the problem with this is and we all know it that the magnetic field in the core lags behind a little but that little is enough in high frequency to make undesirable effects on the switching transistors etc.

So here the current is induced from high voltage electric fields and polarization not magnetic fields.Now as I said the inner sphere is smaller now you switch a high positive potential to the outer framework sphere , you get the field inside the sphere is zero it's only outside so the negative charge should be accelerated in the wires to the sides that are located outside , now you switch the inner one to negative potential and you get the positive charge in the wire is being dragged to the inner side of the coil.
We could even leave the negative potential out of this and switch positive potentials from one sphere to the other (one at a time) which I believe would make the negative charge either flow on the outside of the sphere (the bigger on switched on) or to the inside of the sphere (the smaller one switched on )
The thing is I believe switching high voltages to create electric fields which would do work on charges would be easier for transistors than to switch high currents that tend to lag under loads , and also create more heat.
Well that is if this kind of thing works the way i think it should.

And yes by the way there should be some insulation between the spheres like a good enough vacuum in the case for them not to make a dielectric breakdown and a short circuit.
Ok here is my reasoning give it a try. :)

Attached Files:

• doc23 001.jpg
File size:
27.6 KB
Views:
211
Last edited: Jul 20, 2013
4. Jul 20, 2013

Simon Bridge

OK - a toroidal coil (a flattened toroid in your case) about a conducting sphere.
You charge and discharge the sphere out of sync with the coil.

What was the question?

5. Jul 20, 2013

Crazymechanic

the question was , I think I mentioned it , maybe not, anyways how would this concept work , would it work at all (i believe it should) and can you say something about the efficiency of such a device?

6. Jul 20, 2013

Simon Bridge

Whether or not it "works", and how efficiently, depends on what you want to it to do.
Like I said, you can model it as a toroid about a spherical conductor to see what would happen.

You want to drive it with with square-waves, then you get an impulse response at the leading and trailing edges of the pulses. You have to decide how long the pulses will be and stuff like that.There will be all kinds of feedback in there.

I suspect there will probably be better ways to achieve your goal.

Properly analyzing the configuration would be non-trivial so passes out of the scope of these forums and into the realm of paying an electrical engineer to do the math. Since there are so few specifics, it's a big job.

7. Jul 26, 2013

Crazymechanic

Now the problem I realize with what I described here earlier is that even though I would have a both time and amplitude varying electric field on the inner sphere , there is no flow of current because while the inner sphere is switched on at + potential the - is off and then when the + comes off the - is switched on to the inner sphere , now that makes the field to be a non static.
It induces polarization in the outer coil but I guess that is not enough to induce current flow.
So here goes the question to which I kinda think I know the answer already but anyway.

Can a varying (time and magnitude or just one or the other) electric field induce current in a winding if there is no magnetic field just the electric one like in my drawings where a sphere is switched on at a given potential ?
I guess without some kind of current flow you cannot induce a current in secondary windings , but where there is current there is a magnetic field which induces the current as we know from inductors and transformers etc.

So can electric field transfer energy or only a magnetic field?I know they come together but in my case where I have a sphere switched on to only + or - at a time there is only a electric field.

8. Jul 26, 2013

Simon Bridge

The fact that you are switching voltages on and off already tells you that you have a non-static field. However, there is nothing wrong with non-zero currents in a non-static field.

Use ideal components when modelling electric circuits.

If you have a time varying electric field, you also have a magnetic field - by Maxwell: $$\nabla \times \vec{B} = \mu_0\left ( \vec{J}+\epsilon_0 \frac{\partial}{\partial t}\vec{E} \right )$$ ... like I said, when you switch an electric field on and off, you get an impulse-responce at the leading and trailing edges of the square pulse where dE/dt=∞.

In your setup - you are supplying charge to the central sphere right? So there must be some current flow to that sphere sometime right?

Both the electric and magnetic fields carry energy - remember, they are both manifestations of the same underlying physics.

But the charges don't get to the inner sphere by teleporting!
If you were to switch the sphere between +Q and -Q (if I take your meaning) in some time dt, then the average current is 2Q/dt in that time. The shorter the time, the higher the current - you appear to have been imagining a situation with extremely high currents.

To get a grip on the situation you want to model, 1st go look up toroidal solenoids.

9. Jul 27, 2013

Crazymechanic

Thanks Simon for the answer , well I got some info on toroidal solenoids and it seems that they are toroidal shape coil which has a magnetic field going in a circular way inside it.

Now the thing that in my case is different that the sphere is not so much toroidal as it is round only with slight openings at the ends for the wire to go through.Also i don't know from where the magnetic field could come from.
I made another drawing to illustrate more clearly the thought behind it.
So as I said a sphere windings around it and two possibilities , first either the windings go around the spherical armature which is then switched on and off with high frequency "chopped dc"
or there is the spherical armature on which the windings sit (50% of them on the inner side and other 50% outside) and then there is another inner sphere which is a perfect conducting sphere, now this sphere just as the armature in the first case is switched on and off with + and -.
Now in both cases there is an electric field that is changing due to the fact that the sphere or armature is switched between + and -.
Now where does the magnetic field arise in this situation if there is no where for the current to flow as there is a sphere which is isolated electrically from the windings and the armature and is switched on at either positive or negative potential (one at a time)
As I said earlier this will induce polarization but will it induce current flow in the windings that is what I am not sure about.

Excuse me that I have been going back and forth about this issue it just is kinda hard to grasp , as my mind and logic tells me there can't be current flow in the windings in the setup the way I describe it because there is no path for the primary current to go no current flow , only electric field which arises dues to the fact that the sphere is charged for a given amount if time , but then again the law states that a changing electric field induced a magnetic one.But the changing electric field in our case is a field which has a dead end.Now I'm kinda confused.

Attached Files:

• dirsaa 001.jpg
File size:
30.2 KB
Views:
139
10. Jul 27, 2013

Simon Bridge

It's not a torus fersure, but it is toroidal.
It shares a lot of properties with the simpler geometry - like it goes in a circle and there is a space inside it.
The math for yours will be similar to the torus math - a bit harder - so you should be able to duplicate the calculations - see how the external field behaves with different currents.
Note: tesla coils use torus solenoids...

In your initial picture, you have a toroidal solenoid (flattened to a spherical shell with holes top and bottom), with a conducting sphere in the center.

There is only one coil, so there is no "primary" and "secondary" so that part of your description does not make sense.

There should be a wire going to the inner sphere so you can bias it wrt ground.
There should be two wires - being the ends of the solenoid winding - which allow you to put a current through the solenoid.

I think you need to be clear about what you think is preventing a current from existing and where.

11. Jul 28, 2013

Crazymechanic

Well I;m actually not perfectly sure you are following my crazy mindset but I don't blame you for that.:)
Ok I'll try as simple as I can.
So a spherical armature , a winding on it , a sphere (smaller one) inside it.
Now the primary in my case would be the sphere , the secondary the winding on the armature.
There is a wire attached to the sphere and the sphere is switched on and off with + and - respectively so it creates a time and magnitude varying electric field on the sphere and also on the inside of the larger sphere where the windings sit on , it polarizes the outer sphere into + and - one in the inner side the other on the outside.
So according to what I know from theory a changing electric field like this should induce current in the windings on the outer armature although because the inner sphere which creates the electric field is only switched on to a single polarity at a given time I'm not sure can it induce a current as there is no path for the primary current to flow it just sits there in the sphere and the voltage creates a electric field.

I guess you understood it the other way around , that I am passing current through the windings on the outer sphere and they induce a em field inside and on the armature?
But I am doing it the other way around.I am trying to induce current in those outer windings with a changing electric field that originates from a closely put inner sphere which makes a uniform electric field around it in all directions at a given time.
In other words I am trying to induce a current flow by jerking around with electric field but with no path for the current to actually flow in the primary section like it would normally be in a transformer both impulse and ordinary ac where you pass current through the primary to make a magnetic field in the core.
This is why I have doubts about it.

12. Jul 28, 2013

Simon Bridge

... OK so you are giving the names "primary" and "secondary" arbitrarily - that's allowed but you have to actually say.

The "armature" would, more conventionally, refer to the moving part of an electric motor, the primary and secondary would normally refer to the roles of two coils where the primary is intended to be driven to induce emf in the secondary. If you are not using the terms conventionally, you have to make your usage clear: don't assume others know what you mean.

I think you need to go through your description more pedantically and ask yourself where you have used a personal shorthand in your description that may differ from the conventional jargon.

Like - what does this mean? What does "+" and "-" refer to? Do you mean the positive and negative terminals of a battery?

When you connect the sphere to the positive terminal of a battery, you are draining electrons from it, leaving behind a net positive charge on the outer surface and a neutral inner surface (if the sphere is hollow). So there is no "polarizing" going on, in the conventional sense. Is this another personal jargon?

This does not happen instantly. While the electrons are being drained off - there is an electron-current flowing about the sphere and down the wire (the conventional current flows the other way).

The electrons are pumped through the battery to accumulate against the negative terminal (details depend on the battery)... the setup is called a capacitor: the role of the -ve "plate" of the cap is played by the (-) terminal of the battery, and the role of the +ve "plate" is played by the sphere. Keep these "plates" well apart and you can probably get away with modelling the sphere as a time-varying charge-distribution with a spherically symmetric field: which I gather is what you want. If you switch the wire between positive and negative terminals of the battery sufficiently quickly, you can make square pulses of emf or spikes etc. and thus a changing current and alternating electric field.

the battery+wire+sphere will have currents flowing in it from all this transfer of charge.
the setup is an EM transmitter ... like a radio transmitter aerial with a knob on the end.

You are mixing static and dynamic models.

Nope - I figured you were changing the charge on the inner sphere, and thus varying the electric field.
Your initial post seemed to suggest you also wanted to drive the coil with the opposite phase but I focussed on the central sphere.

But you just told me there was a wire to the central sphere, your "primary section" ... why can't there be a current along that?

Try drawing the PSU in your picture - how is it hooked up?

It looks like the setup is expected to provide radial EM wavefronts ... changing the magnetic field through a coil induces a current in the coil - however, the angle of the coils to the field is important. All your coils will be 90deg on to the field won't they?

See why I keep directing you to the physics of a toroid?

13. Jul 29, 2013

Crazymechanic

I don't have my drawing friend online so can you please refer to the very first drawing I attached in this thread.

Yes there is a wire attached to the inner sphere, that same wire goes to my (until now imaginery) PSU where it would be attached to some semiconductor (probably mosfet) switches that would switch the sphere between (in battery terms just like you started) + and .
In other words the electrons and the current would travel to the sphere and back.
That would create a time and amplitude changing uniform electric field around the symmetrical sphere.
The wires would go vertically just like shown in the first drawing or like they go normally on a toroid.
Now I hope so far I am right and clear about what should be on the setup.
Now this time and amplitude varying electric field from the sphere would as I guess you confirmed create a induced current in those wires around the sphere due to induction.

If so then I have a question , could an induction made this way be as efficient as the one we get from a typical transformer where there is a primary winding creating a magnetic field which induces current in the secondary?
I am asking this as I feel that the way I am describing this setup is not the most efficient way as there is more electrical field than magnetic one due to the fact that instead of multiple wires and iron cores a spherical conductor is used.Which in this case puts more emphasis on the electric field part rather than the magnetic one?

I think I see why you are directing me to look on toroids and I have actually multiple times also before etc.
Now my setup is kinda like a toroidal transformer the difference only is that the primary in my case is not a winding but rather the core and shaped in the form of a spherical conducting sphere , the secondary is the winding into which the current induction should happen as it also normally happens in transformers and there is no core, which maybe changes if I use a metal grid in the form of a sphere on which the secondary windings would sit but that would alter the electric field from the primary , I'm just not exactly sure what would change.
Maybe it would be wiser to use the spherical grid on which the windings would sit as the primary sphere and abandon the inner sphere , so that the sphere on which the windings sit would become the one which would create the field?

I guess it would be much more efficient and better to use a sphere of twisted wires , where one would be the primary and the other the secondary.(metal wouldn't be good for high frequencies due to eddy currents and ferrite doesn't conduct electricity)
Now pass the current through the primary wire and the end of the primary wire or coil is connected to the inner sphere , then the inner sphere is connected to the switch (semiconductors) which would discharge the sphere fast (high frequency) That would result in current drawn through the primary wire creating a magnetic field around it , that magnetic field would induce a current in the secondary coil right next to it + the inner sphere making the same thing too.Now I'm not perfectly sure how this idea would work out , it's rather an intellectual guess.Well I hope I haven't lost the last patience nerve in you. :D

14. Jul 29, 2013

Simon Bridge

Without doing the math - I would not expect so since the loops are 90deg to the field-vectors. i.e. there is no changing flux going through the loops.
You may get some effect from the wire.

In a standard transformer, the coils share an iron core ... the flux of one is being directed through the center of the other. This would maximally couple the coils so I would not expect another configuration to be better ... maybe one coil inside another?

15. Jul 30, 2013

Crazymechanic

Ok I see , I guess the coils would have to go the same way the electric field lines go from the conducting sphere uniformly to all directions?
Just like the windings sit on the transformer core in the way the magnetic field lines go so that current could be induced.

Well, one wire into the other would be good I guess.
Well the question is like that , if I would manage to put the coils so that they would be in the direction of the electric field lines and close very close to one another could I then get the best of both of them , electric field and the magnetic one , atleast in theory I guess I should be able.

16. Jul 30, 2013

Simon Bridge

You need a changing magnetic field through a loop to induce a current in the loop.
An electric field along a wire will produce a current in the wire. It's how normal circuits work, and how aerials work.