The entry says that the thing the stop button did at Chernobyl on 26 April 1986 had been measured in 1983 and described in a letter. This piece sets out what the record shows was known, by whom, and when, as the commission of the Soviet regulator reconstructed it in 1991 from the design documents and the correspondence, and as the IAEA’s advisory group summarised it in 1992. It is a short history of a warning that was written down at every stage and acted on at none.

The rod

The draft design of the RBMK, in 1965, had 212 control and protection rods with absorbers seven metres long, the full height of the core, and 68 of them reserved for emergency protection. By the technical design the absorbers were six metres on 146 rods and shorter on the rest, and the emergency rods were down to twenty. In the working design that was built, the absorbers were five metres long, and the number of rods was later raised to 211 without any change in their design. The commission of 1991 called this “a long evolutionary process” at the end of which the rods “did not cover the entire height of the core”, in a reactor whose critical height could be as little as 0.7 metres in some states.

Below each absorber hung a graphite displacer, joined to it by a telescope, so that when the rod was raised the channel would hold graphite rather than water, which absorbs neutrons. The displacer was 4.5 metres long; with the rod at its upper stop it sat in the middle of the core with 1.25 metres of water above and 1.25 below. When the rod came down from that position, the absorber entered the top of the core and the displacer pushed the lower water column out of the bottom. “Since the absorbing capacity of the graphite displacer,” the commission wrote, “is lower than that of the water being forced out of the lower part of the channel when an RCPS rod moves down from the upper position, local introduction of positive reactivity in the lower part of the core occurs. At a particular core composition and power density field profile, this could cause the formation of a local critical mass.”

Leningrad, 1975

On 30 November 1975 the first unit at Leningrad, the first RBMK-1000, suffered an accident with a release of radioactivity. The officially declared cause was a fuel channel destroyed by a manufacturing defect. The commission of 1991 did not find that convincing, and pointed to what the ministry did next. On 5 and 6 May 1976 the Ministry of Intermediate Sized Machinery set up a commission to develop the basic requirements for the design of RBMK plants, and its resolution spoke of “the unresolved difficulties” of keeping the fuel cool when the water supply was disrupted, of “the variation in neutron power following activation of EPS-5”, and of the need for fast termination of the chain reaction by the emergency protection to compensate for the positive reactivity released when steam filled the core. The Kurchatov Institute’s position in that resolution was that a faster emergency protection system should be developed. “Regrettably,” the commission of 1991 wrote, “the recommendations have still not been implemented, although there is no doubt about the urgency of the proposals made back in 1976.” The reactor at Leningrad had been running, before the accident, with an operating reactivity margin well below fifteen rods. Those who ran the other RBMK plants were not told the real causes. INSAG’s conclusion in 1992 was that the Leningrad accident “is even considered by some to have been a precursor to the Chernobyl accident”, and that “the operating staff at Chernobyl were not aware of the nature and causes of the accident at Leningrad Unit 1.”

In 1977 the Chief Design Engineer’s institute made a proposal, numbered 264 and dated 22 February, to drop the shortened absorber rods into the lower part of the core on an emergency signal, which would have put absorber where the displacers were putting reactivity. At most RBMK units, including Chernobyl Unit 4, it was never fitted. A technical proposal for an experimental rod with a seven-metre absorber and a longer telescope was not carried out either.

In 1980 the same institute made a study, later used in the safety analysis for the third stage of the Chernobyl plant, that listed the factors bearing on nuclear safety. Among them: that a fall in the operating reactivity margin shifted every coefficient of reactivity but one in the positive direction; that an increase in fuel burn-up carried the total coefficient from negative to positive; and that at low power “large irregularities in breeding properties may occur, which can cause large power density distortions with a power peaking factor of more than 10”, so that a rod in the peak region might be worth tens of times more than one elsewhere. The commission of 1991 read this as proof that “the reactor designers were well aware of the possible dangerous consequences of the reactor characteristics and understood how the safety of the RBMK-1000 reactor could be improved.”

Ignalina, 1983

In November and December 1983 the first unit of the Ignalina plant in Lithuania, an RBMK-1500, went through its physical start-up, the sequence of low-power measurements made on a new core, and so did Chernobyl Unit 4. In both, the positive effect of the rods’ insertion from the upper position was observed experimentally. The physical start-up commissions proposed three measures: limits on the withdrawal of the manual control rods to their upper stops; a change in the rod’s design to get rid of the lower water column; and water-film cooling in the rod channels, which would let the displacer go. None was carried out before the accident.

On 23 December 1983 the Kurchatov Institute, as Scientific Manager, wrote to the Chief Design Engineer’s institute, in letter No. 33-08/67, drawing attention to what it called the extremely dangerous nature of the effect. The commission of 1991 quoted the passage: “When the reactor power decreases to 50% (for example, when one of the turbines is switched off), the reactivity margin is reduced as a result of poisoning and distortions of the axial field are observed up to [power peaking factor] Kz ≈ 1.9. Triggering of the EPS in this case may lead to the introduction of positive reactivity. It seems likely that a more thorough analysis will reveal other dangerous situations.” Three proposals followed: to redesign the rods so that there was no water column under the displacer in the upper position; to analyse the reactor’s transient and accident conditions with the rods’ real measured characteristics; and, until that was done, to change the regulations to limit the number of rods that could be completely withdrawn from the core. “The following proposals were then made,” the commission wrote, “which, had they been implemented, might have prevented the Chernobyl accident on 26 April 1986.”

On 2 February 1984, in letter No. 050-01/1-120, the Chief Design Engineer’s institute answered. It had recognised the possibility of a positive reactivity excursion and had its own technical remedies in hand: more shortened absorber rods, a longer telescope, or a return to the draft design’s rods without displacers and with water-film cooling. None of these was carried out. What it proposed instead was an organisational measure, and it informed the other organisations and every RBMK plant of it: “that the number of rods which may be fully withdrawn from the core (up to the upper limit stop switch) should be limited to 150 for the RBMK-1000 reactor and that the other, partially inserted rods must be inserted into the core by at least 0.5 m”. Even that, the commission noted, would have allowed a configuration in which a scram raised the reactivity at the bottom of the core, and would have been consistent with a margin of three to five rods, below the fifteen the operating rules already required. It was in any case never written into the rules. INSAG’s summary in 1992: “Such restrictions were never imposed and apparently the matter was forgotten.”

On 28 December 1984 the Interdepartmental Science and Technology Council on Nuclear Power approved the proposals of two expert commissions it had set up to bring the existing RBMK-1000 units partially into line with the safety regulations. The commissions, the 1991 report says, “ignored certain characteristics of the RBMK-1000 reactor which, as it turned out, played an important role in the initiation and subsequent development of the accident on 26 April 1986.”

What the operators had

The operating procedures for Units 3 and 4, issued in Moscow in 1984, treated the margin as a tool for keeping the power distribution steady. Section 9 said that at nominal power the margin should be not less than 26 to 30 rods, that operation below 26 needed the chief engineer’s authorisation, and that “when the ORM falls to 15 rods, the reactor should be shut down immediately.” Nothing said why. The margin was not among the parameters that triggered an alarm or a trip; there was no instrument for it, only the plant computer, fifty metres away, with its cycle of ten to fifteen minutes. “The Operating Procedures did not draw the attention of the personnel,” the commission wrote, “to the importance of the ORM as an essential parameter for ensuring the effectiveness of the emergency protection system.” Its conclusion was that “right up to the time of the accident, the personnel at units with RBMK reactors had no idea that the ORM value (for the design of the RCPS rods which existed at the time of the accident) not only determined the ability to control the reactor power density field but, primarily, determined the ability of the reactor’s EPS to perform its functions.”

Dyatlov, asked in effect the same question, answered it in 1991 from the other side of the panel. Were the Kurchatov Institute and the design institute aware that at a low margin the emergency protection became “the exact opposite of a protection system, ie a device that drove the reactor out of control”? “Apparently yes – and there are documents to prove it. But I feel they did not have a clear notion. Otherwise their inertness right up to the time of the accident looks strange, to say the least.”

After the accident the rods were rebuilt without the lower water column, and their upper stops were moved so that even a fully raised rod kept 1.2 metres of absorber in the core; a fast scram of twenty-four rods was added; and the margin below which a reactor must be shut down was raised from fifteen rods to thirty. The commission’s list of what was done in the first six weeks ends with a sentence that is also its verdict on the paper trail: “These measures obviously do not tally with the official version, which blamed the accident entirely on personnel errors.”

Sources

• International Nuclear Safety Advisory Group, The Chernobyl Accident: Updating of INSAG-1 (INSAG-7), IAEA, Vienna, 1992: sections 2.2, 4.2, 5.4 and 6, and Annex I (the Shteynberg commission), sections 1-3.8, 1-3.12, 1-3.13, 1-4.1 and 1-4.8, with the references to the letters of 23 December 1983 (Kurchatov Institute, No. 33-08/67) and 2 February 1984 (Scientific Research and Design Institute for Power Technology, No. 050-01/1-120) and to the institute’s report on the physical start-up of Ignalina Unit 1. www-pub.iaea.org
• Anatoly Dyatlov, “How it was: an operator’s perspective”, Nuclear Engineering International, November 1991. neimagazine.com
• US Department of Energy, Report of the U.S. Department of Energy’s Team Analyses of the Chernobyl-4 Atomic Energy Station Accident Sequence, DOE/NE-0076, November 1986, Appendix G, for the rod’s construction. osti.gov