Induction and circuits
The voltmeter paradox
Two identical voltmeters are wired to the same two points of a circuit, and they read different voltages. Neither is broken. A changing magnetic flux runs through the middle of the circuit, and the voltage between two points then depends on which way round you go.
Before you look
Predict first: Both voltmeters are joined to A and B. The left one reads +0.1 V. What does the right one read?
What you are looking at
A ring of two resistors, 100 Ω on the left and 900 Ω on the right, round a long solenoid seen end-on. Its current ramps up and down, so the flux through it changes steadily, then the other way.
Amber arrows are the induced electric field, curling round the solenoid. Gold dots are the current round the ring, 1 mA at the peak EMF of 1 V.
Voltmeter 1 (coral leads) sits left of the solenoid, voltmeter 2 (blue leads) right. Each has its + lead on A. Drag a meter across, under or over the solenoid and watch its reading jump.
The physics
Outside a long solenoid there is no magnetic field, but there is an electric field. With the vector potential ,
round any loop that encircles the solenoid once. So ∫E·dl from A to B is not unique: along a wire that sweeps an angle Δθ round the solenoid it is . A voltmeter reads the ∫E·dl through itself, which equals ∫E·dl along its leads (inside the wires E = 0).
Round the ring, . The left meter and make a loop with no flux inside, so the meter reads V; the right meter likewise reads V. Their difference is the EMF.
Going deeper
Vector calculus and Maxwell’s equations
Kirchhoff’s voltage law, that voltages round a loop add to zero, is : true only when no changing flux threads the loop. Here every loop through both meters contains the solenoid. The lab gives each wire the EMF for the angle it sweeps, then solves the circuit by Kirchhoff’s current law, which still holds.
Walter Lewin made this demonstration famous in his MIT lectures; Romer’s 1982 paper is the careful account. The meters “disagree” only if you think voltage between two points is a property of the points. With a changing B it is a property of the path.
What to measure
Readings IR₁ and −IR₂; their difference is the EMF
- Read both meters during a ramp: +0.1 V and −0.9 V.
- Subtract: the difference is the EMF round the ring, 1 V.
- Drag meter 2 to the left of the solenoid: it now reads +0.1 V, the same as meter 1.
What the model leaves out
- The solenoid is infinitely long and thin: no field outside it at all.
- Wires have no resistance and no self-inductance; the meters draw 10⁻⁷ of the ring’s current.
- The ramps are slow enough that the circuit is quasi-static.
Things to try
- Make the resistors equal: the meters read +0.5 V and −0.5 V.
- Switch to the sine drive: the readings swing in step, always differing by the EMF.
- Drag meter 1 to the right of the solenoid: now both meters read −0.9 V.
References
- R. H. Romer (1982), What do “voltmeters” measure? Faraday’s law in a multiply connected region, American Journal of Physics 50, 1089.
- R. P. Feynman, R. B. Leighton and M. Sands, The Feynman Lectures on Physics, Volume II.
- D. J. Griffiths (2017), Introduction to Electrodynamics, 4th edition, Cambridge University Press.
More experiments, and how the lab works, are in the lab’s overview.
