A construction accident that may involve $2 million in damage has disabled one of two electrical generators expected to supply power for the nation's first "break-even" experiments in nuclear fusion energy.

However, Anthony R. DeMeo, a spokesman for Princeton University's Plasma Physics Laboratory, said the experiments are still expected to proceed on schedule. He said the accident, which occurred shortly after noon Thursday, was caused by some as yet undertermined "failure" in an overhead crane that was holding a 350-ton portion of a generator being installed to power the experiments.

When the crane gave way, the $2 million part, called a stator, which is 25 feet in diameter, fell 15 feet onto a concrete floor.

The stator was being fitted over the generator's $4 million rotor, DeMeo explained. "Damage to the rotor, if any, was slight," he said. "As for the stator, we don't know yet whether it was totalled."

Yet the accident, while expensive, was not catastrophic, he said, noting that scientists are certain the other generator, already assembled, will be sufficient to provide power for the $284 million Tokamak Fusion Test Reactor.

The giant reactor, financed by the U.S. Department of Energy, is itself under construction about 100 yards from the building housing the two generators.

Tokamak is an acronym for Russian words that mean "current in a doughnut-shaped device," and the words describe what it is: a giant doughnut in which electrical energy will be used to heat a "soup" of magnetically contained ionized gases.

Sources at Princeton said one generator will be sufficient to heat the soup to 100 million degrees Celsius. When that happens, probably sometime in 1983, the thermonuclear fusion of the gases will produce about as much energy as went into the effort to heat the soup.

That "break-even" point is considered an essential step toward the achievement of a practical means of turning fusion energy into an unlimited soure of electrical energy.

Sources at Princeton said that the second generator will be needed later, when scientists attempt to achieve even higher temperatures. They said they expect repairs can be made in time.

Dr. Robert W. Conn of UCLA, a member of the energy department's fusion review panel, argreed with Princeton scientists that the experiments can proceed on schedule. "Motor generators in general are pretty rugged and reliable," he said, adding that he thinks the intact generator is unlikely to have problems that would create major delays.