Ludwik asked: . . .
|But Faynman argues that two ways of
|changing the flux result in two different phenomena.
| Right or wrong? I wish I could refer to an experimental
| verification of this theoretical claim.
| . . .
It may help to compare the cyclotron vs. the betatron:
The cyclotron employs only the qVxB effect to force an injected charged
particle into a circular "current". The B field is static in time. The
orbit is governed by |qVxB| = mV^2/R . There are no conductors to
constrain the motion, so that no static charges accumulate. The magnetic
field does no work - the energy of the "current" is still what it had at
injection.
The betatron does the same thing but also adds a time variation to the B
field, creating a non-conservative E field of circular, unending "lines of
force" . Here both effects are operating. The instantaneous value of B
determines the orbit through the qVxB force (which does zero work). The E
field generated by Curl(E) = -dB/dt directly does work on the particles,
increasing the "current". (In real devices the spatial and time
variations of B are typically controlled to maintain an orbit of fixed
radius, taking due account of the relativistic "mass" variation of the
particle.)