He stood alone at the master control board. This was located in the engine room, aft of the compartment that held the reactor and the heat exchanger/steam generator, the latter placed right at the submarine's center of gravity. The reactor was pressurized to twenty kilograms per square centimeter, about twenty-eight hundred pounds per square inch. Only a fraction of this pressure came from the pump. The higher pressure caused a higher boiling point for the coolant. In this case, the water was heated to over 900° Celsius, a temperature sufficient to generate steam, which gathered at the top of the reactor vessel; the steam bubble applied pressure to the water beneath, preventing the generation of more steam. The steam and water regulated one another in a delicate balance. The water was dangerously radioactive as a result of the fission reaction taking place within the uranium fuel rods. The function of the control rods was to regulate the reaction. Again, the control was delicate. At most the rods could absorb just less than one percent of the neutron flux, but this was enough either to permit the reaction or to prevent it.

Petchukocov could recite all this data in his sleep. He could draw a wholly accurate schematic diagram of the entire engine plant from memory and could instantly grasp the significance of the slightest change in his instrument readings. He stood perfectly straight over the control board, his eyes tracing the myriad dials and gauges in a regular pattern, one hand poised over the SCRAM switch, the other over the emergency cooling controls.

He could hear the vibration. It had to be a bad bearing getting worse as it wore more and more unevenly. If the crankshaft bearings went bad, the pump would seize, and they'd have to stop. This would be an emergency, though not really a dangerous one. It would mean that repairing the pump—if they could repair it at all—would take days instead of hours, eating up valuable time and spare parts. That was bad enough. What was worse, and what Petchukocov did not know, was that the vibration was generating pressure waves in the coolant.

To make use of the newly developed heat exchanger, the Alfa plant had to move water rapidly through its many loops and baffles. This required a high-pressure pump which accounted for one hundred fifty pounds of the total system pressure—almost ten times what was considered safe in Western reactors. With the pump so powerful, the whole engine room complex, normally very noisy at high speed, was like a boiler factory, and the pump's vibration was disturbing the performance of the monitor instruments. It was making the needles on his gauges waver, Petchukocov noted. He was right, and wrong. The pressure gauges were really wavering because of the thirty-pound overpressure waves pulsing through the system. The chief engineer did not recognize this for what it was. He had been on duty too many hours.