S5G Natural Circulation
The S5G pressurized water reactor prototype first operated in 1965. The S5G was the direct precursor to the eighth generation S8G reactor used on the Ohio class ballistic missile submarines; a quiet submarine design. S5G was a prototype of the USS NARWHAL (SSN 671) submarine and had the capability to operate in either a forced circulation or a natural circulation mode. In the natural circulation mode, cooling water flowed through the reactor by thermal circulation; pumps were not needed. This reduced plant noise level, which is vital to submarine stealth.
S5G was used operationally only on the USS Narwhal (SSN-671). The propulsion system consisted of 1 x S5G reactor rated at 90 MWt; secondary system supplying steam to one large-diameter, directly-coupled main turbine (no reduction gear), driving a single shaft. The reactor core life was 10,000 equivalent full power hours. The single main turbine measured 12 feet in diameter, and about 30 feet long. This powerplant required a larger submarine hull diameter than the previous Sturgeon-class with an S5W reactor plant (33 ft. vs. 31.6 ft.)
As the Cold War intensified, the United States and the USSR poured their latest technology into the theory and practice of undersea warfare. They wired the ocean floors with sound detectors. These called forth technology to quiet the submarines. One source of noise came from the pumps that circulated the coolant through the reactor and kept it under pressure. The art of sound detection became so refined that skilled listeners could identify the unique sound patterns of individual boats. So the mission of S5G was to eliminate noise.
When the Department of Defense asked the commission on 3 September 1958 to develop and test a natural-circulation reactor for submarine propulsion, Rickover began pressing the commission for funds to undertake research, procure long lead-time items, and begin construction of a prototype at the Commission's National Reactor Testing Station in Idaho. In all his prototypes Rickover insisted that each consist of an actual reactor and propulsion-plant components arranged as if they were in a ship. The new facility would follow the same pattern, but added one important innovation. To make sure that the concept would work at sea, the prototype would simulate the motion of a ship in operation.
The commissioners demurred. They were willing to order components that took a long time to manufacture, but they thought the technical data were too uncertain to commit themselves to prototype construction. To get an independent assessment, they called for a review panel. In two meetings in September, Naval Reactors and Bettis won agreement that even if detailed answers were not available for all the technical and engineering problems, enough was known to warrant going ahead. Unable to resist tweaking the commission, Rickover observed in February 1959 that in the past he had been criticized for not developing new reactors: now that he was doing so the commission was holding him up.
The commission placed the item in its budget, and Congress duly authorized and appropriated the funds." With the project gathering momentum, Rickover called a meeting between Naval Reactors and Bettis. On 11 May 1959 he made sure that everyone understood the ground rules. Design had to be kept simple. He would approve no development work unless it was absolutely necessary. He wanted the principles of personal responsibility followed. In practice this meant he would not tolerate anyone hiding behind a title or organizational chart. That applied to his own engineers as well: technical recommendations had to be accompanied by the names of the people proposing them.
Late that summer the work at Bettis received an unexpected blow Without informing Rickover, Westinghouse took several key personnel out of Bettis and assigned them to a new laboratory to work on projects for the space program. Rickover was furious. He argued that the company was siphoning off individuals who had been carefully trained at unique government facilities. Furthermore, the company was casually breaking the ties of confidence that had been painstakingly erected between Naval Reactors and the laboratory. Loudly and vehemently he protested to the commissioners and the joint committee. Even though Westinghouse rescinded the action, Rickover took the natural-circulation project away from Bettis and assigned it to Knolls.
As Rickover saw it, Westinghouse had attempted a sharp and deadly thrust at a principle upon which the program was based. Although Naval Reactors made the technical decisions, it depended upon the laboratories for proposals, recommendations, advice, and work. He frankly considered the laboratory personnel as "his" people. He admitted — reluctantly — the right of the company to shift its personnel, but not unilaterally and not without offering qualified replacements for approval who, if they performed satisfactorily, agreed to serve for a number of years. Almost overlooked in the dispute was the swiftness with which General Electric and Knolls picked up the work—an indication of the high competence of both laboratories and their ability to work together.
May 1961 saw the beginning of construction of the S5G (the 5th submarine design, made by General Electric) prototype with completion planned for 1963. The schedule had already slipped a year when Rickover briefed senior naval officers on 2 July 1962, urging them to include a natural circulation reactor submarine in the fiscal year 1964 shipbuilding program.
He told them that, in some respects, he considered the project's drive for simplicity a return to earlier engineering concepts. If the machinery was less efficient then, it had the compensating virtues of ruggedness, reliability, simplicity, and easy maintenance. These qualities were vanishing as the navy was installing complicated high speed machinery, often beyond the abilities of officers and men to maintain, in order to squeeze the most energy from every ounce of fuel oil.
Rickover's strategy for reducing noise was to get rid of equipment; if that was impossible, to turn it off during quiet operation; to slow down the component; or to redesign the particular equipment to get rid of rotating components. Naval Reactors was analyzing the design of the fluid systems in the propulsion plant and scrutinizing every valve to see if it could be eliminated.
Rickover used full-scale wooden mock-ups, a device he had begun with the Nautilus and continued to use with the S5G. They were fascinating. Built largely of cardboard and wood, they made it possible to trace every pipe in its actual size, see the location of every valve, and observe the overall arrangements of components.
Rickover took a great deal of time in his frequent inspections of a mock-up, often remaining transfixed while he visualized the motions that men would have to make to maintain or repair equipment. The mock-up even showed whether lighting was sufficient to read instruments. Each component received a tag; one if its position was approved, another if it was still under consideration. The full-scale mock-up exposed problems that would not have been apparent from blueprints or a model. It allowed shipyard personnel—such as welders—to be sure they could perform their job in the ship wearing full working gear. One curious phenomenon that Naval Reactors field representatives had to watch for: frequently carpenters and woodworkers fell into the error of making their work more finished than it need be.
After a mock-up had served its initial purpose, it still remained useful. If a plant had to be modified later, its mock-up could be used in training people, making sure procedures were correct, and ensuring that operations could be carried out as planned.
The schedule for the prototype continued to slip. Some of the problems were inherent in any large construction project with rigid specifications. In early 1963 Electric Boat was completing construction of the hull, beginning to install the piping for the reactor system, and building some of the off-hull facilities. The work showed the need for improved communications between the Idaho site and Groton, for speeding up the training of engineers, and for establishing a new and more vigorous system of quality control. The unique characteristics of some major components were leading to a host of manufacturing problems, and it took time to work out and bring into effect new fabrication techniques. Stringent testing of development items, if time consuming, was proving valuable and forcing some rethinking and many design changes. And there were other factors, among them a fire in a vendor's plant that destroyed a number of components.
Toward the end of 1964 Knolls reported that the core was in the final stages of assembly. On 22 January 1965 a strike stopped the testing of the main engine at the General Electric plant at Lynn, Massachusetts, adding delay to a schedule already in trouble. March saw the reactor fueled, and May the arrival of the main engine at the site. Finally, on 12 September 1965 the reactor reached criticality and, to the satisfaction of Naval Reactors and Knolls, within close agreement with prediction. Months of testing followed, not only to determine the actual characteristics of the plant in physics, fluids, and hydrodynamics, but also for noise reduction. Power range testing began on 13 November, and once again operation verified theory.
The S5G reactor had two coolant loops and two steam generators. It had to be designed with the reactor vessel situated low in the ship hull and the steam generators high in order for natural circulation of the coolant to be developed and maintained using the chimney effect. It was largely a success, although the design never became the basis for any more fast attack submarines beside the Narwhal.
Reactor primary coolant pumps are one of the primary sources of noise from submarines in addition to the speed reduction gearbox and cavitation forming collapsing bubbles from the propeller. The elimination of the coolant pumps and associated equipment also reduced mechanical complexity and the space required by the propulsion equipment.
The S5G was also equipped with coolant pumps that were only needed in emergencies to attain high power and speed. The reactor core was designed with very smooth paths for the coolant. Accordingly, the coolant pumps were smaller and quieter than the ones used by the competing S5W core, a Westinghouse design, and were also fewer in number. In most situations, the submarine could be operated without using the coolant pumps, useful for stealth operation. The reduction in the electrical requirements enabled this design to use only a single electrical turbine generator plant. The S8G prototype used natural circulation allowing operation at a significant fraction of full power without using the reactor pumps, providing a silent stealth operation mode.
To further reduce engine plant noise, the normal propulsion setup of two steam turbines driving the propeller screw through a reduction gear unit was changed instead to one large propulsion turbine without reduction gears. This eliminated the noise from the main reduction gears, but at the expense of a large main propulsion turbine. The turbine was cylindrical, about 12 feet in diameter and 30 feet in length. This large size was necessary to allow it to rotate slowly enough to directly drive the propulsion screw and be fairly efficient in the process.
The USS Narwhal had the quietest reactor plant in the USA naval fleet. Its 90 MWth reactor plant was slightly more powerful than the other fast attack USA nuclear submarines of that era such as the third generation S3G and the fifth generation S5W.
The prototype testing included the simulation of the engine room of an attack submarine. By floating the plant in a large pool of water, the whole prototype could be rotated along its long axis to simulate a hard turn. This was necessary to determine whether natural circulation would continue even during hard maneuvers, since natural circulation is dependent on gravity.
S5G, like S1W, was built in a huge box of a building, its roof presenting an inscrutable flat surface to satellite cameras passing overhead. The reactor went critical for the first time in September 1965. The hull section floated in a “basin” of water. Operators inside the hull used equipment to make the hull rock back and forth, adding more realism to the simulation. In this setting, the Navy developed a method of circulating the reactor’s cooling water without using a pump and exploiting the principle that warm water rises.
USS Narwhal was the first boat equipped with the system. At high speeds, pumps were still needed, so the controller could move the coolant either by so- called “forced” or “natural” methods. The S5G nuclear propulsion plant, installed only on the SSN-671 Narwhal, was a natural-circulation reactor intended to test the potential contribution of this technology to submarine quieting. Reactor primary coolant pumps are one of the primary sources of noise from submarines, second only to steam turbines, and the elimination of coolant pumps and associated equipment also results in the reduction of mechanical complexity and the volume required by propulsion equipment.
On May 1, 1995, the S5G prototype was shut down after 30 years of successful testing and training activities. Approximately 12,000 Navy and civilian students were trained at S5G during the 30 years of operation.
|Join the GlobalSecurity.org mailing list|