Reactor conditions before the test
Published 12 October 2026
The plant's design is necessary to understanding the test, but the design explanation must not be confused with what the operators knew. INSAG-7 describes the RBMK as a graphite-moderated reactor cooled with boiling water. Water passed through individual pressure tubes containing fuel, then the steam-water mixture moved to separators. Steam travelled to the turbines and condensed water returned to the circulation system. The coolant and the graphite had different roles in that arrangement. The later safety report examines how steam content, fuel condition, rod position and power distribution interacted. It does not describe a reactor in which the loss of water had one simple effect under every operating condition. The sign and strength of feedback changed with the condition of the core. C021, pub.iaea.org, The Chernobyl Accident: Updating of INSAG-1
INSAG calls the steam bubbles in the coolant voids. The void fraction is the proportion of coolant volume occupied by steam. Changing that fraction changes reactivity, and the ratio of those changes is the void coefficient. Its sign matters: a positive coefficient means that increasing the steam fraction can add reactivity rather than restrain the chain reaction. The report identifies the void effect as the main component of the RBMK power coefficient, while explaining that other effects also enter the total response to a power change. The language describes a feedback relationship, not a fixed quantity of heat or an inventory of radioactive material. A reactor can become harder to control when the physical changes following increased power tend to raise power further. C021, pub.iaea.org, The Chernobyl Accident: Updating of INSAG-1
The coefficient depended on the configuration of the reactor. INSAG lists the number of rods inserted, additional absorbers, fuel enrichment and fuel burnup among the important variables. It says the void coefficient could range from negative to positive as core composition and operating regime changed. In normal operating conditions, the fast power coefficient remained negative despite that variation. At the time of the accident, however, both the void coefficient and the power coefficient were positive. That last statement concerns the accident condition, not an assertion that every RBMK operated with the same feedback at every moment. It explains why a disturbance during this particular test could become dangerous. The state into which the reactor had been brought mattered alongside the design from which it had been built. C021, pub.iaea.org, The Chernobyl Accident: Updating of INSAG-1
The control and safety rods were intended to change that reactivity. Most entered from above, while shortened rods entered from below to shape the power distribution. A graphite displacer was connected to the absorber through a water-filled separation. With a rod fully withdrawn, the displacer occupied the middle of the fuelled height and water remained above and below it. As the rod began moving down on a shutdown signal, graphite displaced water in the lower part of the channel before the absorbing section completed its travel. INSAG says this initial movement could insert positive reactivity locally in the lower core. Its magnitude depended on the distribution of power and the operating regime. The same system that was supposed to reduce the chain reaction therefore had a dangerous initial effect in some configurations. C021, pub.iaea.org, The Chernobyl Accident: Updating of INSAG-1
This was not simply a matter of a button being ignored. The later report describes a protection system whose physical response could work against its purpose when rods had been fully withdrawn and the core had an unfavourable power distribution. An instruction to shut down assumes that the shutdown mechanism reduces danger. At Chernobyl, that assumption did not hold under all relevant conditions. INSAG's conclusions say the addition of positive reactivity through rod insertion was probably a decisive contributor, while retaining uncertainty about the exact event that began the final power excursion. The technical defect does not eliminate the operating decisions that created the condition. It changes the meaning of those decisions because the consequences were not adequately represented in the system or in the information available to operators. C021, pub.iaea.org, The Chernobyl Accident: Updating of INSAG-1
The rods also moved slowly compared with the speed at which the reactor could become unstable. Starting from their upper limit, full insertion took eighteen seconds. INSAG attributes the slow movement largely to the narrow rod channel and the resistance of the cooling water, which acted as a damper. That time was a feature of the original protection design. It should not be presented as a delay chosen by someone after the emergency button was pressed. The report's account places a physical limitation inside the protection system itself. A signal could be sent immediately while the machinery still required time to place the absorbing material throughout the core. Under the accident conditions, the early positive effect and the subsequent travel time both belonged to the problem. C021, pub.iaea.org, The Chernobyl Accident: Updating of INSAG-1
Power control became more difficult at low output. The reactor had an internal power-density system and a control system with detectors both inside and outside the core. Their coverage depended on the power range. INSAG says that at low power, with the relevant internal systems inactive, operators had to rely on detectors outside the core. Those detectors could not show the neutron-flux distribution inside it or its average distribution over the height, because they were located at the middle plane. The operator could see measurements related to total power without receiving the full picture of where that power was concentrated. The report describes a resulting dependence on experience and intuition. This is an equipment-and-information limit, not proof that the people at the console deliberately chose to operate without any instruments. C021, pub.iaea.org, The Chernobyl Accident: Updating of INSAG-1
A low-power reactor after prolonged operation was also different from a reactor starting with little xenon poisoning. Xenon-135 absorbs neutrons, and its uneven presence could make power distribution difficult to control. INSAG says large differences in power density could arise across the height and radius of the poisoned reactor at low output. Those circumstances existed to a marked degree during the final test, and operators had little or no experience controlling the reactor in that condition. The statement identifies the condition in which their judgment was being exercised. It is not an excuse supplied by this account, nor a claim that the reactor could never be run at low power. The later analysis shows why instructions, instrumentation and training had to address a difficult state, rather than assume that lower total output meant a uniformly easier or safer reactor. C021, pub.iaea.org, The Chernobyl Accident: Updating of INSAG-1
The operating reactivity margin was another quantity whose practical meaning was not fully conveyed. INSAG defines it in terms of the additional reactivity that would appear if all control and safety rods were withdrawn, expressed through the worth of a standard rod. The system calculating it was about fifty metres from the control console and drew on roughly four thousand input points. A measurement-and-calculation cycle took ten to fifteen minutes. The design used the result to guide steady-state control of power distribution. It was therefore not an immediate, continuously displayed warning tailored to the seconds of the final transient. The margin had safety significance through its relationship with rod configuration and feedback, but the later report says operators were not sufficiently aware of that significance during the critical period. C021, pub.iaea.org, The Chernobyl Accident: Updating of INSAG-1
The physical size of the core made the spatial problem important. INSAG gives a height of seven metres and a diameter of 11.8 metres. It says the chain reaction in one part was only loosely coupled to distant parts, so control had to address different regions almost as though they were separate reactors. Close to the accident, the upper and lower halves were behaving nearly independently, with strong xenon poisoning between them. In that condition, an initial positive effect from rod insertion could shift power towards the bottom. A single total power number could not describe all of those internal conditions. The later report uses the dimensions and the distribution to explain why the geometry of the protection rods was relevant to the final excursion, not merely an engineering detail outside the operating story. C021, pub.iaea.org, The Chernobyl Accident: Updating of INSAG-1
Operators could also alter some protection arrangements. INSAG describes accessible terminals through which jumper wires could disable systems, bypass automatic shutdown signals, or reset and suppress alarms. Procedures permitted some of this under defined circumstances. That qualification is necessary: the existence of a bypass does not establish that every use was unauthorised. The later reassessment would specifically withdraw some earlier allegations of rule violations. The design still left important choices available locally, and INSAG regarded the ease with which protection could be changed as part of the safety problem. Equipment, procedure and staff action were joined. The relevant question was not only whether someone touched a switch, but what the procedure permitted, what protection remained, and whether the reactor's condition made that permission unsafe. C021, pub.iaea.org, The Chernobyl Accident: Updating of INSAG-1
Cooling arrangements created another route for the test to affect the reactor. Each of two primary loops cooled half the core. Each loop had four pumps, normally using three with the fourth available as backup. Just before the accident, all eight were running. Four took power from the turbine that remained connected, while four received outside electricity. INSAG says the resulting flow exceeded the usual full-power flow and reduced an already low steam fraction. At the same time, the low reactor output meant the water entering the core was only slightly below its boiling temperature, and might not have been below it at all under the precise flow and pressure settings. The system was therefore approaching the test in a hydraulic condition different from ordinary full-power operation. C021, pub.iaea.org, The Chernobyl Accident: Updating of INSAG-1
The water's route through the plant helps explain this sensitivity. INSAG describes pump discharge entering a common header and then twenty-two distributor headers in each half of the reactor. Individual channels had flow-control valves to adjust the cooling distribution across the core. Water boiled as it passed through the fuelled pressure tubes; the resulting steam-water mixture then travelled to drum separators. Steam passed from the separators to the turbines, and condensed steam returned as feedwater that mixed with the recirculating water at the pump intakes. That returned feedwater was the cooler component of the mixture. Reducing reactor power reduced its contribution and raised the temperature at the pump intake and core inlet. During ordinary startup or shutdown, fewer pumps and throttled flow helped keep the intake cool enough to avoid cavitation and maintain the intended distribution of steam production along the channels. This connects the turbine, pumps and reactor through both water movement and temperature, before the additional effect of slowing turbine-powered pumps during the test. C021, pub.iaea.org, The Chernobyl Accident: Updating of INSAG-1
When the turbine was disconnected, the pumps supplied from it began slowing as turbine speed and generator voltage fell. Reduced coolant flow allowed steam content to rise. With a strongly positive void coefficient, that change added reactivity and formed part of the accident mechanism described by INSAG. The report retains questions about whether the pumps' ability to circulate water deteriorated further or whether cavitation occurred. It also notes the Soviet commission's studies concluding that the pumps did not cavitate. A faithful account cannot turn an unresolved possibility into a settled cause. The report's firmer point is that the reactor, under these conditions, was highly sensitive to pumping disturbances because of the positive feedback between steam and reactivity. C021, pub.iaea.org, The Chernobyl Accident: Updating of INSAG-1
The containment arrangement was limited as well. INSAG describes separate enclosed parts of the reactor and cooling circuit rather than one full containment around the entire system. Pressure-relief pipes led to pools intended to suppress pressure. Around the core, a heavy cover plate was penetrated by fuel channels. The report says the confinement could accommodate steam from two simultaneous channel ruptures. More ruptures could exceed the capacity of the pressure-relief route, lift the cover and sever the remaining channels. These were limits of the design as described after the disaster. They explain why a rapid internal excursion could become a destruction of the reactor building and a release to the environment, rather than remain an isolated broken channel repaired inside an intact pressure boundary. C021, pub.iaea.org, The Chernobyl Accident: Updating of INSAG-1