The Chernobyl Atomic Energy Station of Vladimir I. Lenin was widely considered the best power plant in the country (Ivanov) (Filipchuk). Superlatives aside, Chernobyl was inarguably the "flagship of MinEnergo" (Karpan, 314), which refers to the sprawling ministry that operated all RBMK plants except for Leningrad and Ignalina. Of the plant's four reactors, Unit 4 was regarded as the most advanced, boasting safety systems (such as the Emergency Core Cooling System or ECCS) which Units 1 and 2 lacked. According to prevailing opinion among its operators, Unit 4 also exhibited the highest construction quality, having been fabricated largely on the spot by brigades who had three other reactors under their belts. At the same time, KGB attachés quietly filed reports of corroding wall panels in the reactor building and misaligned crane rails in the turbine hall (KGB-I, 65, 165).
 |
Photo Credit: Chidori@emptytriangle.com
|
As for the operators themselves, a post-accident psychological study characterized them "as a fairly typical, mature and stable group of specialists with qualifications regarded in the USSR as satisfactory. They were no better, but no worse, than the personnel at other plants" (INSAG-7, 31). At the management level, things were more unsettled. Chief Engineer Nikolai Fomin had no background in nuclear power, and had been appointed with the expectation that he would lean heavily on the expertise of his deputies, such as Anatoliy Dyatlov and Anatoliy Sitnikov (friends and colleagues who had worked on submarine reactors in the Far East). In the months leading up to the accident, Fomin had often been out of work for extended periods due to a severe spinal injury, shifting yet more of the workload onto his subordinates.
Would any of the above details and factors prove to be relevant in the accident? That is hard to say. But without a doubt it was the nature of the reactor itself that proved most decisive. During its less than three years of operation, Chernobyl Unit 4 followed the same path into danger as every RBMK before it. When first launched, the reactor was loaded with 'fresh' fuel assemblies of 2% enrichment. To counterbalance the enormous energy potential at start-up, many of the reactor's 1661 technological channels were devoted not to fuel, but to static boron rods known as Additional Absorbers. Only in this initial configuration was the void coefficient of reactivity negative, providing a modicum of safety and stability. However, the equilibrium of the core would shift as the fuel was 'burned up' by fission over months and years, leading to a positive void coefficient. Furthermore, as the degree of fuel burnup increased, Additional Absorbers were removed to compensate. Eventually the reactor reached steady-state operation, where a few channels would be reloaded with fresh fuel assemblies on a daily basis. As had been standard practice since the late 1970s, the boron Additional Absorbers were all but completely removed in order to maximize fuel efficiency. In this regime, the void coefficient of reactivity reached a terrifyingly high value (4-5 β)(Kuzmin, 165), turning the reactor into a powder keg. At this magnitude, the void coefficient could override all the other coefficients of reactivity which were meant to counterbalance it, initiating a vicious cycle of steam and reactor power.
Meanwhile, RBMK plants went on recording the same drastically understated estimates of their void coefficients, which they derived from simple experiments carried out on the fly. Although everyone understood that increased steam production led to increased reactivity, operators were falsely led to believe that the reactor's fast power coefficient would always remain negative. In short, that an increase of power could not initiate a feedback loop of ever-increasing reactivity. As the Unit 4 night shift would discover to their cost on April 26th, at low power the RBMK could undergo a sudden transition from liquid coolant to steam. The result was just such a feedback loop of reactivity, meaning that the power coefficient was strongly positive.
This unfounded confidence in the reactor's safety no doubt led Chernobyl's managers to ignore a letter regarding a flaw in the control rods that could briefly increase reactivity during a scram (the so-called tip effect). The letter (grudingly circulated by the RBMK's NIKIET design bureau) provided little detail, mandated no procedure changes, and was written to sound as reassuring as possible. As a result, the operators were not informed and no regulations were rewritten.
Finally, even though the RBMK's designers and scientific overseers were aware of the above safety issues, the reactor's ability to actually explode would prove a shock to all. While the void coefficient of reactivity was known to be positive, a related, much more critical parameter was regarded as negative. The void effect of reactivity refers to the result of the reactor's coolant flashing to steam in its entirety, a total voiding of the core. A tragic error, borne out of the limited capabilities of 1970's supercomputers, led to the ill-fated graph at right. It was no surprise that as coolant turned to steam, reactor power would increase. However, as the density of the coolant fell more drastically, it was calculated that the core would begin to regulate itself, stifling reactivity. This is depicted in curve #1, which becomes negative as it approaches zero. In reality, reactivity in this scenario followed curve #2, rapidly and steadily increasing until the reactor exploded. (Curve #3 depicts the void effect of modern-day reactors.) Worse yet, this nightmare scenario of total core voiding was all too easy to realize. If only three of the RBMK's 1661 technological channels burst simultaneously, the superheated fuel assemblies would become directly exposed to their coolant water and create enormous overpressure in the reactor space. The energy released would be enough to dislodge the 2000-ton Upper Biological Shield, shearing the tops from all remaining channels (Ignalina sourcebook). At that point, reactor runaway and subsequent explosion was inevitable.
How much of this did Chernobyl's managers and operators understand or suspect as they prepared Unit 4 for its first major period of planned maintenance? This question will never be answered. But Nikolai Karpan of the plant's Nuclear Safety office perhaps spoke for all his colleagues:
"We knew it was bad. But not that bad!"
 |
| Photo credit: chidori@emptytriangle.com |
Sources Cited:
- Oleg Bazhan et al, Chernobyl Dossier of the KGB, 2020 (ISBN 978-617-684-255-2).
- Ignalina Sourcebook (see also INSAG-7, page 69).
- INSAG-7
- A. N. Kuzmin, Forty Years with the RBMK, History of Atomic Energy in the USSR and Russia, Part 3, History of the RBMK, 2003.
- Nikolai Karpan, Revenge of the Peaceful Atom.
- Natalia Filipchuk, You Are Building the Station In A Cursed Place.
- Viktor Ivanov, Catastrophe in Polesia
Meanwhile, RBMK plants went on recording the same drastically understated estimates of their void coefficients, which they derived from simple experiments carried out on the fly. Although everyone understood that increased steam production led to increased reactivity, operators were falsely led to believe that the reactor's fast power coefficient would always remain negative. In short, that an increase of power could not initiate a feedback loop of ever-increasing reactivity. As the Unit 4 night shift would discover to their cost on April 26th, at low power the RBMK could undergo a sudden transition from liquid coolant to steam. The result was just such a feedback loop of reactivity, meaning that the power coefficient was strongly positive.
Hi, I wondered what was the cause of the belief that void coefficient would turn back down below some water density (curve 1) ? Was it because they believed that the modicum of moderation provided by the water would decrease enough as it turned to steam to compensate the loss of neutron absorption water usually provides ? Thank you
ReplyDeleteThat is possibly the explanation, but in this case I can only rely on what INSAG-7 has to say. It would be very nice if all of that report's cited source were available on the internet so that more qualified people could look into nuances like this one. But very few of them are.
Delete