The History of Reactor Safety: The Soviet Naval Experience

Chernobyl was part of a systematic failure of Soviet reactor safety. This failure is very evident if the history of accidents with Soviet naval reactors is compared with the United States Naval experience with Nuclear power. We now have access to accounts of Soviet submarine reactor accidents, and they are very suggestive of the safety weaknesses in the Soviet Naval reactor program.

The first nuclear accident to occur on a Russian submarine was on the Northern Fleet's ballistic missile submarine K-19 (Project 658 - Hotel class). On July 4, 1961, during exercises in the North Atlantic, a leak developed in an inaccessible part of the submarine K-19's primary cooling circuit. The leak was specifically located to a pipe regulating the pressure within the primary cooling circuit. The leak caused a sudden drop in pressure, setting off the reactor emergency systems.

To prevent overheating of the reactor, superfluous heat must be removed, and this is done by continually circulating coolant through the reactor. There was no built-in system for supplying coolant to the primary circuit, and it was feared that an uncontrolled chain reaction might start. An improvised system to supply coolant to the reactor was devised. This required officers and midshipmen to work for extended periods under radioactive conditions in the more remote areas of the reactor compartment as they attended to the leak in the primary circuit. The radiation in this case came from noxious gases and steam. All of the crew were exposed to substantial doses of radiation, and eight men died of acute radiation sickness after having undergone doses of 50 to 60 Sv (5000 - 6000 rem). The crew was evacuated to a diesel submarine, and K-19 was towed home to base on the Kola Peninsula.

Here the problem lay in the lack of redundancy in Soviet naval reactor design. Parts were simply assumed to not fail, while American Naval reactors were designed with multiple back ups in case of failure.

The second nuclear accident to occur was in February 1965 aboard the Project 627 - November class submarine K-11. The submarine lay in dock at the naval yard in Severodvinsk and work was underway to remove the reactor core (Operation No. 1). On February 6, the reactor lid was opened, and the following day, the lid was lifted without having first secured the control rods.[583] Releases of radioactive steam were detected with an abrupt deterioration of conditions. Radiation monitors were going off scale, and all personnel were withdrawn. No work was done on the submarine over the course of the next five days while the specialists tried to discover the reason for the problem. The wrong conclusions were drawn, and the raising of the reactor lid was attempted again on February 12. Once again, the control rods had not been secured, and when the reactor lid was raised, there were releases of steam and a fire broke out. There are no data on radioactive contamination levels or radiation exposure of the personnel. The reactor was finally retired and replaced.

This story might be called "The Three Stooges work on a nuclear submarines." First The connection of the reactor lid to the control rods was a fundamental design defect, but beyond that the shipyard staff did not understand the reactor design, and were so poorly trained that even after an accident had occurred, they drew the wrong conclusions, and as a consequence the accident was repeated. Other reactor accidents occurred in ship yard settings:

In August 1968, the Project 667 A - Yankee class nuclear submarine K-140 was in the naval yard at Severodvinsk for repairs. On August 27, an uncontrolled increase of the reactor's power occurred following work to upgrade the vessel. One of the reactors started up automatically when the control rods were raised to a higher position. Power increased to 18 times its normal amount, while pressure and temperature levels in the reactor increased to four times the normal amount. The automatic start-up of the reactor was caused by the incorrect installation of the control rod electrical cables and by operator error. Radiation levels aboard the vessel deteriorated. . . . .

On September 30, 1980, the submarine K-222 was at the factory in Severodvinsk due for a thorough reactor check. During the course of work, the submarine's crew left for lunch leaving the factory personnel on board the vessel. As a result of a breach in the pertinent procedural instructions, power was sent through the safety rod mechanisms without the controls also being engaged. Following a failure in the automatic equipment, there was an uncontrolled raising of the control rods with a subsequent uncontrolled start up of the reactor. As a result of this, the reactor core was damaged.

Reactor safety flaws were ignored, even after they lead to serious accidents, and accident patterns were repeated.

On August 10, 1985, the Project 671 - Victor-I class submarine K-314 was at the Chazhma Bay naval yard outside Vladivostok. The reactor went critical during refuelling operations because the control rods had been incorrectly removed when the reactor lid was raised. The ensuing explosion led to the release of large amounts of radioactivity, contaminating an area of 6km in length on the Shotovo Peninsula and the sea outside the naval yard. Ten people working on the refuelling of the vessel died in the accident. The damaged reactor compartment still contains its nuclear fuel

The problem was not simply due to the difficulty mastering nuclear safety technology:

The existing framework of project development, building and delivery of military technology (navy) and ammunition is not regulated by law, but by decree of defunct authorities, such as the Central Committee of the Communist Party, various councils of Soviet ministers, the military-industrial complex, as well as joint decisions handed down by the Ministry of Ship Building and the Ministry of the Navy. The administrative body of the military industrial complex, led by the vice-chairman of the council of ministers, itself issued the documents that established the norms, and it was this same body that monitored and enforced the norms that it had itself created. The practice of merging the functions of public agencies contributed to the fact that the Navy itself did not take part in working out quality control and safety requirements for nuclear submarines. Even if the Navy politely refused to receive equipment that they knew in advance to be defective, it could nonetheless be forced to accept it through a common resolution issued by the authorities.

This structure of resolutions and decrees has followed the delivery of all new nuclear submarines to the Soviet Navy. Soviet nuclear submarines were built under enormous time constraints. If the Central Committee of the Communist Party had determined that a particular submarine was to be built by the close of a certain year or a particular season, the submarine yards could not postpone delivery, even if the vessel had not been completed or undergone sea trials. Hence nuclear submarines were often delivered to the Navy without all the necessary safety equipment having been installed. Furthermore, the procedural guidelines and the specifications of the contract were modified and simplified. It was not uncommon for a nuclear submarine to be delivered to the Navy from the building yard with missing or defective parts. In 1989 there were 529 complaints of nuclear submarines being delivered with faulty equipment. In 1990-91 a new nuclear submarine was returned to the building yard due to numerous defects in the mechanical equipment. Another submarine was delivered without light switches having been installed in the cabins or in the missile compartment.[607]

The servicing and repair of nuclear submarines was carried out at naval yards that fell under the jurisdiction of different authorities. This system was established at the dawning of the age of nuclear submarines and it came to the full during the cold war. Almost 25 nuclear submarine projects were initiated and developed during this period. The lack of sufficient standardisation led to problems in the planning stage, in the competence levels of the crew and in an unavailability of spare parts. The quality and safety of the equipment was compromised, and this has been one of the most important contributing factors to the higher incidence of accidents amongst Soviet nuclear submarines as compared to for example American vessels.

it is clear then that the Soviet system of designing, building, servicing and operating naval nuclear power plants was fundamentally flawed:

1. The frequency of accidents was increased as early as the planning stage due to technological deficiencies in a number of areas (information, securing secrets of propulsion and means of carrying out research) and deficiencies in construction. One of the main problems was the poor quality of the metals and materials that were used.

2. At the construction stage, breaches in the technological standards by the builders affected the quality of the finished product such that the finished submarines that were actually delivered to the Navy fell short of the quality of their design.. Furthermore, the schedule for delivery of various systems and parts, as well as the order in which operations were completed and breaches in the proper technical procedures, all contributed to lowering the quality of the submarine. The quality of the work was poor due to a lack of technical understanding amongst the workers. In some instances, there was not even enough technical equipment at the navy shipyards and floating bases.

3. During the testing and approval stage, there were interruptions to the schedule due to delays in deliveries and installation of parts and systems. Under outside pressure from concerned parties, the submarines were approved even before the equipment on board had been tested.

3. The frequency of accidents also increased during the submarines' operational life due to poor maintenance, a disregard of the directions for use and improper procedures for technical equipment and ammunition. The crew were assigned to tasks other than their direct responsibilities, and therefore were lacking in training, especially with regards to measures designed to ensure the survivability of the submarine. At one time, one submarine commander sent in a complaint that eleven of the 28 new members of the crew could not speak Russian and therefore were not qualified to work with the nuclear reactor. Many of the new crew for the nuclear submarines had received only six months training, training that often was insufficient or irrelevant to the situations they might confront on a nuclear submarine.[608] There was a widespread irresponsible attitude from incompetent outside specialists. The crews were too inexperienced to be able to foresee potential dangers that could lead to critical situations while the submarine was in operation. The search and rescue bases, which have shown a lack of co-ordination in conducting searches, weakened overall preparedness in a number of accidents. The absence of formal conclusions at inquests and a lack of ready information in response to the questions about the nature of the accidents and their frequency resulted in a failure to implement measures to improve the conditions on board the nuclear submarines.

Thus the record of Soviet nuclear technology is by no means evidence of the potential for the safe operations of modern nuclear power plants In contrast to the Soviet record, although there has been serious accidents involving American nuclear submarines, none appear to have been due to reactor failure. Thus it is by no means impossible that a Light Water Reactor could undergo a major accident that would involve human deaths and or injuries, as the Soviet experience demonstrates. But given a acceptance of safety constraints by military or civilian regulators, serious accidents are unlikely, but never impossible. With the development of increasingly safe Light Water Reactors serious accidents become much likely. In fact current reactor designs are far safer than NRC guidelines would require. The NRC safety standard calls for one premature death for every 5 million years of reactor operation. A typical pressurized water reactor is expected to produce one premature death every 20 million years of operation. While the same reactor is expected to produce a cancer death every 200 million years. Boiling water reactors are even safer, with fatalities occurring once every 50 billion years, or several times the current age of the universe. it should be noted that recent Light Water Reactor designs are even safer than the designs included in this evaluation. Thus by all rational standards nuclear safety should not be an obstacle to the implementation of current nuclear technology.

Even accidents of the magnitude of the Three Mile Island accident can be considered rare, and grow increasingly improbable with the passage of time. New Light Water Reactor designs are considerably safer than older designs. The Westinghouse AP-1000 reactor is estimated to have a core meltdown once every 2.4 million years. A core meltdown does not signify a Chernobyl type radiation release, as the Three Mile Island accident demonstrated.