Determination of direction in AC Power Flow

--------- Sorry, I gave my compass answer before checking other responses. Honestly, I didn't cheat. I have done this to determine whether my alternator was actually charging the battery in my car. Of course, as Daestrom has noted, the voltmeter has to be a DC voltmeter (or a dynamometer meter).

You assume that AC power has a specific direction in each line - It does not - rather, it has direction in TWO lines (from the generator to the load).

Power is always positive into the load, and in AC it's magnitude varies in time and magnitude.

I.e., there is a voltage difference across the load for half the cycle, and then there is the same voltage difference but of different polarity across the load for the other half cycle. Both portions of the waveform deliver power to the load (equal if the waveform is symmetrical. (dP=dE^2/R at any given instant)

Neutral is a reference point for voltage only (usually the same as earth ground in residential systems)

(continued below)

You could measure the field from the generator through the walls with a directional field meter, since only the generator will put out a magnetic field.

Other than that, I don't think it can be done using only the buss bars, for the previous reasons and because.

1) You have E-field, H-field (B), which vary with frequency (i.e., they reverse according to frequency along the time scale), and TTBOMK give no indication of direction. 2) Kirchoff requires current in=current out at any given instant in your loop. Thus, current on either bar is the same 3) Gauss requires any point along the no-resistance bussbars to be the same voltage with respect to any point - i.e., you have two perfect conducting busbars, thus voltage on a bar on one side of the room with reference to the other bar is the same

Resistance is the same, as long as the lines are intact.

Given the constraints and looking at it from s physics, engineering and a technical standpoint, I don't think it can be done as ststaed.

But, as Einstein said - it only takes one....

Again, Kirchoff says current in=current out.

Gauss says voltage along the line is the same -

Electron drift in one bar is to the left, and in the other bar is to the right, so that tells nothing.

>

that would better read "varies instantaneously with time"

Well, electron drift is related to current flow. The important concept for this puzzle is current flow. Long ago in the 19th century before anyone even knew about electrons, it was decided that current flows from the positive terminal to the negative terminal.

That convention has stayed with us today, even though we know that electrons are moving from negative to positive and the actual speed of any one individual electron in this flow is very slow, the current is effectively traveling at/near the speed of light.

In the case of this puzzle, you need a dc voltmeter to determine the positve buss bar conductor.

In the positive conductor, power will flow in the same direction of the dc current, relative to this conductor. That is, current will flow from the (more positive) terminal at the battery side to the less positive terminal (at the resistor side). This DC current will produce a steady-state magnetic field. A compass can determine the direction of this magnetic field, either by a derivation from the theory or, has been suggested, doing a reference experiment under controlled conditions.

Thus, the most electrical engineeringly elegant and simple answer to the problem is: A DC voltmeter and a compass. These are two very common items.

Of course, the suggested Hall effect measuring devices and temp gradient probes (and super-sensitive voltmeters) might also work, but I should have added the real world condition that, the company you work for expects you to solve this problem with existing equipment and a spending budget of no more than $50, if necessary. (Hmmm...Seems like I've worked for cheap outfits like that before...)

Thus, most of us already have dc voltmeters and we could buy a decent compass for under $50 if we had to.

Beachcomber

How do you distinguish a load from a source? What is if your load is a negative resistance? Is circular reasoning showing up here?

Bill

-- Fermez le Bush

Tell that to a lead-acid battery that is being charged.

What happens if you happen to use bismuth conductors rather than copper? Or p doped silicon, for example? What if you use a conductor made of acidified water?

ill

-- Fermez le Bush

Gernerally speaking, current DOES flow from the positive to the negative terminal. You are bringing up a special case. Charging was not mentioned in the original puzzle.

But, even so, in order for the battery to charge, the charger + terminal needs to be even more positive than the battery + terminal. for charging current to flow.

Don't know. Unless you are a research scientist, your boss and co-workers might think you are nuts.

I really jumped from AC to DC without thinking on this one - After reading your comments on DC, it occurred to me I have an old field-driven DC meter that does exactly that in my lab.

If it is DC, I just put that DC current meter in the line and read off the direction.

Sorry bout that...

Ok, change that to "current flows from the *most* positive terminal to the "least" positive terminal.

What changes, other than the dielectric constant? The speed of an individual electron is still slow compared to the SoL.

True life is full of special cases.

I bring such items up because others tend to over generalize explanations. In acidified water, the primary carrier will be hydronium ions. In a proton beam, there are no electrons at all. Current still flows.

Bill

-- Fermez le Bush

The drift velocity is very slow. The actual speed of an electron is often on the order of 10^6 m/s as it pinballs its way through the lattice.

j

The problem in my physics book (example problem solving for electron drift) has it in inches per hour.

|>

|> >

|> > That convention has stayed with us today, even though we know that |> > electrons are moving from negative to positive and the actual speed of |> > any one individual electron in this flow is very slow, the current is |> > effectively traveling at/near the speed of light. |> >

|>

|> The drift velocity is very slow. The actual speed of an electron is often |> on the order of 10^6 m/s as it pinballs its way through the lattice. | | The problem in my physics book (example problem solving for electron drift) | has it in inches per hour.

Coulombs is a specific number of electrons. Amps is coulombs per second. Figure in the electron density (figured from mass, atomic number, etc) and conductor cross-section and you should be able to get into the right ballpark.

Coulombs is a unit of charge, not a specfic number of electrons. While a given number of electrons have a specific charge, a certain charge does not have a specific number of electrons.

the formula for drift speed is

v= j/ne, where n = dN/M , where

j=current density, e = electron charge, d =density of atoms, N= Avagadros number, M =atomic weight

e.g., the drift speed in a copper wire 1.6 mm dia which is carrying 10 amps is 3.6 x 10^-2 cm/sec.

per Halliday and Resnick.

|---------------------------------------/----------------------------------|

|------------------------------------/-------------------------------------|

Think about this again.

Really? Have you thought through what you just said?

Ok. so what don't you get about 'e'? Hint: You've just hoisted your own petard.

They knew what they were talking about, anyway.

This agrees with what I said but maybe I didn't write very clearly. I am differentiating between 'drift velocity' and 'actual speed'. Many people think of the 'two speeds' of electricity ... the effects that travel at the speed of light, and the electrons that move slowly (maybe inches per hour of drift velocity). There is a third speed. The instanataneous speed of the electrons which is very high (IIRC it calculates around 10^5 m/s using basic calcs, but there are some relativistic effects or something which led to the figure of 10^6 being used as the 'actual' typical kind of number). However, these electrons take 1000000 steps forward and 999999 steps back as they pinball around in every direction. Overall there is a tendency to move in the direction the E field pushes (trajectories are slightly curved in the direction dictated by the E field), and this results in the slow net migration of electrons, the average speed in the direction of flow over time, the drift velocity.

j

I understood your point, but the electron doesn't go banging

*through* the lattice at 1E6M/S.

Ok, from one atom to the next, perhaps. How do you tag an electron to measure its individual velocity?