On April 25th, 1986, No.4 reactor was scheduled to be closed for regular maintenance. It is decided that in this cooperation, in order to test the capability of the reactor turbine generator, sufficient power will be provided to the safety system power supply (especially the water pump) of the reactor in case of power loss. Like Chernobyl, the reactor has a pair of diesel generators, which can be used as backup, but can't be started immediately-the reactor will be used to rotate the turbine, and then the turbine will be separated from the reactor and rotate under its own inertia. The goal of the test is to determine whether the turbine can fully supply pump power during the reduction stage when the generator is started. This test has been successfully carried out in other units before (all safety supplies have been started), and the result is failure (that is, the power generated by the steam turbine is not enough to provide power for the pump during the depressurization stage), but other improvements indicate that other tests are needed. In order to test in a safer and lower power environment, the energy output of Chernobyl No.4 reactor was reduced from 3.2 GW of normal power to 700 MW. However, due to the delay in the start of the experiment, the reactor controller reduced the energy level too quickly, and the actual power output dropped to only 30 million watts. Therefore, the neutron-attracted fission product xenon-135 has increased (this product is usually consumed in higher power reactors). Although the scale of power reduction is close to the maximum allowed by safety regulations, the managers of the staff team chose not to shut down the reactor and continue the experiment. Later, the experiment decided to "cut corners" and only increased the power output to 200 MW. In order to overcome the neutron absorption of residual xenon-135, the control rods far exceeding the safety regulations were pulled out of the reactor. On the afternoon of April 26th 1: 05, as part of the experiment, the water pump driven by the turbine generator was started. Because of this action, the flow of water exceeds the safety regulations. The water flow rate increases at 1 h 19 a.m.-Because water will also absorb neutrons, it is necessary to manually remove the control rods to further increase the water flow rate, which will lead to extremely unstable and dangerous operating conditions. In the morning 1: 23: 04, the experiment began. The unstable state of the reactor shows nothing on the control panel, and it seems that all reactor employees are not fully aware of the danger. The power of the water pump is turned off, driven by the inertia of the turbine generator, and the water flow speed is reduced. When the steam turbine is separated from the reactor, the steam level in the reactor core will rise. Because the coolant is heated, separate steam is formed in the coolant pipe. The special design of Chernobyl RBMK graphite mitigation reactor has a high positive vacancy factor, which means that the power of the reactor increases rapidly due to neutron absorption when there is no water. In this case, the operation of the reactor becomes unstable and more dangerous. At 1: 23: 40 in the morning, the operator pressed the AZ-5 ("fast emergency defense 5") button which ordered "emergency shutdown"-all control rods were inserted, including those that were accidentally taken away before. It is not clear whether this is an emergency measure or just a routine method to shut down the reactor when the experiment is completed (the reactor is planned to be shut down for routine maintenance). This usually means that the emergency shutdown command is a response to an unexpected rapid increase in power. On the other hand, Anatoly Diart, the chief engineer who was at the Chernobyl nuclear power plant at the time of the accident, wrote in his book: "Before the centralized control system of 1: 23: 40 ... no parameter changes that could justify the emergency shutdown were recorded. According to this statement, the Committee ... will collect and analyze a lot of materials. In its report, it was not sure why the emergency shutdown was ordered. There's no need to find an excuse. After the experiment was completed, the reactor was simply shut down. " Due to the control rod insertion mechanism (18 completed slowly to 20 seconds), the hollow part of the rod and the coolant temporarily shifted and escaped, resulting in an increase in the reaction rate. The increased energy product leads to the deformation of the control rod pipeline. After the rod is inserted, it gets stuck and can only enter one third of the pipeline, which can't stop the reaction. At 1: 23: 47, the output of the reactor soared to about 30 GW, which was ten times that of normal operation. The fuel rods began to melt, and the steam pressure increased rapidly, which led to the explosion of large steam, which displaced and damaged the top of the reactor, and the coolant pipe burst, blasting a hole in the roof. In order to reduce the cost and its volume is too large, the reactor is built in a single protective layer. This led to radioactive pollutants entering the atmosphere after the main pressure vessel burst due to steam explosion. After a part of the roof was blown up, oxygen flowed in-combined with extremely hot reactor fuel and graphite moderator-causing graphite to catch fire. The fire spread radioactive materials and polluted a larger area. Because the eyewitness reports are inconsistent with the radio records, there are some claims that the actual incident occurred at 1: 22: 30 local time. According to this theory, the first explosion occurred at about 1: 23: 47, and the operator ordered "emergency stop" after 7 seconds.