Chernobyl Nuclear Power Plant Unit 4 Before the Explosion: The Last Fatal Steps, or Why No One Wanted to Say "Stop"
Let's briefly recap what happened before the tragedy. The personnel fell behind schedule for vibration testing before the scheduled maintenance. It became known (according to Dyatlov's testimony) that the TG-8 vibration test on April 25, 1986, was not conducted, first due to the TG-8 itself being unprepared, and then due to the dispatcher's prohibition on transferring the load to TG-7. Consequently, the vibration test had to be combined with a coastdown, which fatally increased the risk of an accident.
Then, the Kyivenergo dispatcher's ban on April 25, 1986, disrupted the vibration testing schedule and the experiment itself, which also tragically affected the entire chain of events.
There is an extremely important (though not proven) hypothesis already considered about the possibility of xenon poisoning of the reactor, created as a result of both the planned reduction of power by 50% on the eve of the experiment and the subsequent delay in its reduction due to the ban by KyivEnergo.
A. Dyatlov then ordered the reactor power to be reduced to 200 MW, despite this being inconsistent with the test program. This was a fateful step, which largely predetermined the possibility of an accident. This power level was an intermediate level, poorly studied, and at this level, the reactor behaved unpredictably (a fact the Chernobyl Nuclear Power Plant management was aware of). At low reactor power, the water temperature became close to the saturation (boiling) temperature, yet the technical documentation did not specifically prohibit operation at this power, although its plot suggested this.
As our leading nuclear power plant safety specialist, V. Asmolov, noted:
It stabilized at 200 megawatts. That's less than 10 percent... For us in the project, this was an unexplored power level. Because a reactor operates at this power level either when it's starting up or when it's shutting down.
According to Fomin's testimony in court:
I believe the violations of program provisions are the main cause of the accident. First and foremost, the reduction of the reactor's power to 200 MW.
Prosecutor: Who, in your opinion, is the main culprit of the accident?
Fomin: Dyatlov, Akimov, who allowed deviations from the program.
Prosecutor: Who, in your opinion, is the main culprit of the accident?
Fomin: Dyatlov, Akimov, who allowed deviations from the program.
Perhaps 200 MW was needed to legally disable the reactor shutdown protection system (bypassing the requirements of the main Regulations governing the operation of this protection) by shutting down the reactor shutoff valves of both turbines in accordance with the "Regulations for Switching Keys and Cover Plates...," which, according to this document, is disabled at power levels below 100 MW electrical. However, this document, to which Dyatlov refers, is not publicly available.
Perhaps, as V. Dmitriev previously indicated, the personnel believed that 760 MW(t) for conducting vibration tests of the turbine generator was too much, so they reduced it to 200, and Metlenko also said in court that he needed 35–45 MW(el), i.e. 200 MW(t).
INSAG-1 believes that Dyatlov also needed to turn off the protection so that he could repeat the experiment (in case of failure), although competent experts, including those who advised the author, acknowledged that under these conditions, repeating it would have been absolutely impossible.
Finally, there's a hypothesis that the power reduction was necessary to conduct vibration testing. This possibility is also discussed by the author of the textbook, V. L. Gurachevsky (Introduction to Nuclear Energy, Rosatom Library):
Combining the coastdown and vibration tests had two catastrophic consequences: a power reduction, which increased the risk of reactor shutdown; and the shutdown of the reactor protection system, which shut down both turbine generators, which, as we will see below, was also a contributing factor to the accident. This was done so that the turbine could continue to idle during the tests without being shut down by the protection system. Disabling this protection system, an important safety measure, led to a loss of reactor control and, ultimately, an explosion. The two programs are incompatible and contradictory. If not for the vibration tests, the accident could have been avoided and the coastdown could have been carried out using the old scheme, without disabling the protection system for both turbine generators.
Following errors by the SIUR in reactor control, or as a result of xenon poisoning, or a power reduction for vibration testing, the reactor shuts down. The moment the reactor shut down actually became a turning point. Fate offered a real chance to avoid the accident—if the personnel had shut down the reactor, the accident would not have occurred!
However, according to Komarov, an official from a department (or sector) of the Central Committee of the CPSU may have given instructions to A. Dyatlov to increase the power and to conduct the experiment, threatening him with retirement.
Dyatlov then orders a power increase, which was established during the judicial investigation, although he denies it. The reactor begins to be raised from the pit—this was done quite quickly, from 12:38 AM to 12:42 AM—within four minutes, the power was raised to 160 MW, and a little later, by 1:03 AM, to the planned 200 MW. Conducting the experiment at this power level was a gross violation of the test program, which called for a coastdown of 700 MW.
Most importantly, the increase in power to 200 MW due to xenon poisoning of the reactor was achieved by removing the maximum possible number of control rods, while their number in the core became unacceptably low, a gross violation of the Regulations. The reactor was poisoned, and its condition was uneven. And this was a fatal error on the part of the personnel. However, there are a number of caveats to this, discussed below.
We would like to emphasize that the assumptions and conclusions we have made are not adequately reflected in any official document regarding the investigation into the causes of the accident, which must continue – all documents are kept in secret archives in Moscow.
The staff then moves on to preparing for the coastdown tests, taking the final, fateful steps.
Disabling protection for blocking both TGs (for closing the SRV of both turbines) and more
During the tests in 1982–85, steam stopped flowing to the turbine (to ensure pure inertia of the turbine, steam access was closed through valves—stop and control valves (SRVs)), but at the same time, the reactor, according to the experimental design, was automatically shut down (!) by special protection for closing the SRVs of both turbines.
In contrast, the 1986 tests were conducted with this protection disabled, ostensibly because the personnel intended to repeat the experiments in the event of failure. According to the report (No. 1 INSAG-1):
The existing A3 for shutting down the recovery valves of two turbogenerators (Turbine Generator No. 7 was shut down on the afternoon of April 25, 1986) was blocked to allow for a repeat test if the first attempt was unsuccessful. This constituted yet another deviation from the test program…
According to the NIKIET report:
Since it was not planned to shut down the reactor, in order to ensure that automatic protection systems did not interfere with the testing, the formation of the AZ-5 mode for emergency exceeding the specified power in a small range (AZMM) was disabled, the formation of the AZ-5 mode for emergency increase in the rate of power increase in the starting range (AZSP) was disabled, a blocking was introduced for the formation of an emergency signal for disconnecting one of the turbogenerators, the LAR system, designed for automatic regulation of power in the energy range, was disabled - instead, the backup automatic power regulation system of the reactor, including AR1 with automatic overcompensation of its rods by moving the ZAR-AR rods, was turned on, protection for a decrease in the water level in the BS was not introduced.
As Academician A. Alexandrov aptly noted: "The only thing missing [at the unit] was protection against a fool who might decide to disable the protection for his own experiment." The regulations prohibited overriding protections other than those described in the control procedures. The main criticism that can realistically be leveled at the reactor designers is that such a possibility existed.
Why they didn't activate the safety system later remains unclear. Perhaps they didn't consider it important. Dyatlov (according to him) mentioned shutting down the reactor at the beginning of the experiment during the briefing; this cannot be verified, as there were no witnesses. Failure to shut down the reactor was a fatal mistake, and why Dyatlov left the reactor running at full power is now a matter of conjecture.
It is also suggested that the theory about the shutdown of both transmission lines is an excuse used by A. Dyatlov to justify his mistake: in reality, the personnel would not have been able to repeat the experiment. This hypothesis was put forward by specialists who advised the author during the writing of this article.
The tests themselves involved shutting down some equipment (turbine, main circulation pump, and fuel pump), which would otherwise have had to be restarted. This is a lengthy process, especially with modified electrical circuits. The test program made no mention of repeat tests.
Another possibility is that the staff allegedly "forgot" to turn it on. Here's what V. Zhiltsov, a participant in the accident investigation, had to say about this:
To conduct vibration tests on the turbogenerator, they removed one of the safety switches, and after completing the tests, they forgot to re-enable it... The reactor was supposed to shut down automatically upon the "two turbines shut down" signal. But one turbine was already running, and the eighth turbine, where the ill-fated "coastdown" test was being performed, had its safety switch locked because they "forgot" to unlock it after the vibration tests were completed. This was a serious fault on the part of the personnel. Therefore, the reactor continued to operate for almost 30 seconds after the turbine was disconnected, after which an attempt was made to shut it down using the AZ-5 button. According to the entry in the SIUR Operational Log, this was done by L. Toptunov. A few seconds later, he also turned the "clutch power off" key.
These 30 seconds were enough for the reactor to run away. If this is indeed the case, then the official explanation for the personnel's actions regarding the shutdown of the protection systems may not reflect their actual logic.
Perhaps the unit's personnel didn't fully appreciate the risks of shutting down the recovery valve on the remaining turbine, even though their more experienced peers should have understood. Perhaps they assumed that, with the reactor operating at 200 MW, the pressure increase would be slow. Steam could be discharged into the condensers via the BRU-K bleeder (for excess steam when pressure rises).
The main steam drain in an operating reactor is through the turbine. If the turbine is shut down, the drain is shut off, but the reactor continues to generate steam, leading to a pressure increase. This protection was created to prevent a pressure surge in the primary circuit associated with a turbine shutdown, which stops consuming steam.
As a result, the blocking of protection to block both turbogenerators under specific conditions of the accident development (see below – a sharp decrease in feedwater consumption, a slowdown in operation and a breakdown of the main circulating pump) was one of the key factors in the development of the accident.
As stated in the court ruling: “Fomin, Dyatlov, Kovalenko… did not regulate the removal of excess steam from the circuit.”
Dyatlov at the trial:
It was stated that the AZ-5 shutdown should not have been initiated after two turbine generators were shut down. However, this was done in accordance with the operating procedures and did not affect the progression of the accident. At power levels below 100 MW (electrical), this protection should be disabled. Therefore, there was no violation of the operating procedures.
Both the State Panic Administration and INSAG-7 do not consider the shutdown of this protection following A. Dyatlov to be a violation for formal reasons, which seems very strange.
However, as a member of the IXBT forum believes, this "...is true for the TG load shedding protection, and only at the moment it's activated. The AZ-5 protection for TG shutdown (failure of two of the four SC) is activated immediately after connection to the grid and is deactivated only when the reactor is shut down. This was the case before 1986, and it remains the case today."
In fact, according to the Regulations, this protection is activated immediately after the TG is connected to the grid and deactivated only after the TG is disconnected. This means that disabling this protection during the entire period the power unit is operating with an electrical load is prohibited. However, Dyatlov cites a certain Regulation on Switching Keys and Cover Plates, which is absent from the documents, or is attempting to justify his error.
At the same time, Nikolai Fomin, the chief engineer of the Chernobyl Nuclear Power Plant, who was responsible for safety at the Chernobyl Nuclear Power Plant and who became the main "conductor" of the notorious "experiment," openly admits that the disabling of this protection was the most important cause of the accident:
Prosecutor: In your opinion, what could have prevented the accident?
Fomin: If the AZ-5 shutdown from the shut-down valve hadn't occurred, the unit would have remained intact... In 1982, '84, and '85, during the program, the AZ-5 signal to the reactor was sent by shutting down the shut-down valve on the turbine. And in 1986, changes were made to this aspect. Now it's clear to me that the program should have been coordinated with specialists. There was no point in keeping the unit running at power if all the turbine generators were at a standstill.
According to Yu. A. Potemkin, former deputy head of the GVP (Main Military Prosecutor's Office) of the USSR: "...they shut down both turbines while the reactor was still running. This is madness, frankly speaking. This led to the explosion."
From Metlenko's testimony:
"Prosecutor: Was it necessary to shut down AZ-5 when 2 TGs stopped?
Metlenko: "No, absolutely not. We said that with our program, the reactor needs to be shut down."
Fomin: If the AZ-5 shutdown from the shut-down valve hadn't occurred, the unit would have remained intact... In 1982, '84, and '85, during the program, the AZ-5 signal to the reactor was sent by shutting down the shut-down valve on the turbine. And in 1986, changes were made to this aspect. Now it's clear to me that the program should have been coordinated with specialists. There was no point in keeping the unit running at power if all the turbine generators were at a standstill.
According to Yu. A. Potemkin, former deputy head of the GVP (Main Military Prosecutor's Office) of the USSR: "...they shut down both turbines while the reactor was still running. This is madness, frankly speaking. This led to the explosion."
From Metlenko's testimony:
"Prosecutor: Was it necessary to shut down AZ-5 when 2 TGs stopped?
Metlenko: "No, absolutely not. We said that with our program, the reactor needs to be shut down."
SIXTH TRAGIC ERROR (FATAL) – DISABLING THE PROTECTION TO BLOCK BOTH TGs.
Connecting additional main circulation pumps
Technical news feed:
From 00h 43mm 35s to 44m 40s — signal “1PK up”.
From 00:43:36 to 51:23 — signal of malfunction of the measuring part of the AR-2.
00h 43m 37s — protections for increasing the condensate level in the SPP compressor station and for increasing the level in the evaporator are activated.
/The protection data was also removed in violation of the test program in order to prevent the reactor from being shut down/
From 00:43:37 to 51:45 — emergency level deviations in the base station.
From 00:49:19 to 51:23 — signal “1PC down”.
00h 51m 23s — activation of BRU-K1 TG-8.
00h 52m 27s — emergency deviations of the LEVEL in the BS.
01h 00m 02s — signal of malfunction of the measuring part of AR-2.
01h 00m 04s — emergency level deviations in the left half of the BS.
Between 12:43 a.m. and 1:00 a.m., personnel received at least four emergency signals. In forum discussions, several participants wrote that nothing alarming occurred. According to Dmitriev, the reactor was operating unstably, as there were alarms for water level deviations in the reactor compartment and the activation of the BRUK-K (excess steam pressure).
01 h 02 min 20 s — increase in feedwater flow rate from 104 to 424 t/h on the left half to increase the water level in the BS.
To stabilize the water level in the boiler system and reduce the pressure in the circuit by cooling the circuit water, the staff sharply (almost fourfold) increases the feedwater flow rate in the circuit. The feedwater regulator was turned off.
01h 03m — N(T) = 200 MW. Disconnection of TG-8 from the grid, measurement of vibration XX with the generator disconnected.
01h 04m - MCP-12 is turned on.
01h 06m 02s — emergency level deviations in the left half of the BS:
increasing the feed water flow rate from 192 to 1170 t/h (left) to raise the water level in the BS.
01h 07m - MCP-22 is turned on.
01h 09m — water flow rate reduced to 100 t/h on the left and right.
01h 09m 45s — emergency level deviations in the BS.
As part of the test program, additional main circulation pumps (MCPs) 12 and 22 were activated to provide additional reactor cooling in the event of a slowdown in the main circulation pumps. The reactor began to operate erratically, and the water level in the separator drums dropped to the emergency level. To prevent a reactor shutdown, personnel disabled several safety devices.
According to G. Medvedev,
…the total water flow through the reactor increased to 60 cubic meters per hour, compared to the standard of 45… per hour, which is a serious violation of operating regulations. Under these operating conditions, pumps can disrupt flow, and vibrations in the circuit pipelines can occur due to cavitation (boiling of water with strong hydraulic shocks). The sharp increase in water flow through the reactor led to a decrease in steam generation, a drop in steam pressure in the separator drums where the steam-water mixture from the reactor is fed, and undesirable changes in other parameters.
According to INSAG-1 Report for the IAEA:
Operators attempted to manually maintain the reactor's key parameters—steam pressure and water level in the steam separator—but were unable to do so fully. During this period, steam pressure drops of 0,5–0,6 MPa and water levels below the emergency setpoint were observed in the steam separator. To prevent a reactor shutdown under these conditions, personnel blocked the emergency control signals for these parameters.
According to the report of the State Nuclear Academy of Sciences (1991), the increase in consumption was a violation of the Regulations: “Increase in consumption for individual main circulation pumps to 7500 m3/h (violation of clause 5.8 of the TR).”
Let's see how reactor designer Nikolai Dollezhal described this situation:
...Of the four pumps, two were fed by the auxiliary busbars, meaning they had normal voltage and frequency. Two were fed by the busbars of the generator being tested, with a steadily declining speed and, consequently, a declining frequency, which affected their performance characteristics—i.e., speed, flow rate, and developed pressure. A situation arose in which four pumps with different, constantly changing characteristics were operating simultaneously. Operation of such a system can be accompanied by the most unexpected manifestations. It should also be noted that the generator being tested was not in freewheel mode at all, as it was loaded, albeit slightly, by the motors of the four main circulation pumps: this slowed it down and forced it to reduce speed...
To stabilize the water level in the reactor coolant system and the pressure in the circuit by cooling the circuit water, the crew sharply (almost quadrupled) increased the feedwater flow rate in the circuit. After 30 seconds, the operator managed to maintain the level, increasing the feedwater flow rate more than threefold. However, as soon as the cold water from the reactor coolant system reached the core, steam generation significantly decreased, causing a decrease in the volumetric steam quality. This caused the automatic control rods to move upward, and the automatic control system began withdrawing the rods from the core (01:19:39 – signal "1 PK up", reactor power is falling), reducing the ORM.
The combination of two factors: reactor poisoning and increased feedwater consumption, led to the fact that at 1:22:30 a.m., according to the PRIZMA printout, there were only 6–8 rods in the active zone, calculated as fully immersed.
The reduction in steam generation led to a slight pressure drop in the BS. About a minute later, the high-speed pressure-reducing device (BRU-K), through which excess steam was vented into the condenser, closed. This contributed to a slight reduction in the rate of pressure drop. However, the pressure continued to decline slowly until the start of the tests. During this period, it dropped by 5 kgf/cm².
The general characteristics of the neutron field at that moment in time are as follows: in the radial-azimuthal direction it is practically convex, and in height, on average, it is “double-humped” – with a higher energy release in the upper part of the active zone.
This field distribution was a consequence of the reactor's current state: a burned-out core, almost all control rods at the top, a significantly higher steam quality in the upper part of the core than at the bottom, and greater 135Xe poisoning in the central parts of the reactor than in the peripheral parts. Operating the reactor at 200 MW(th) resulted in a large accumulation of 135Xe, which would not have occurred at a reactor power of 700-1000 MW(th).
THE SEVENTH TRAGIC ERROR IS CONNECTING ADDITIONAL RCPs AND BLOCKING THE PROTECTION DEVICE BY SIGNALS OF INSUFFICIENT WATER LEVEL AND STEAM PRESSURE.
A sharp decrease in feedwater consumption
When, in the opinion of the operator, the reactor parameters returned to normal, he sharply reduced the feedwater flow rate, practically to zero, which turned out to be a fatal step, since it led to an increase in the coolant temperature at the reactor inlet almost to the boiling point, i.e., an additional increase in steam production.
A vapor phase begins to form in the process channels, starting from the upper sections of the core and spreading downward. The reactor begins to accelerate. The operating regulator attempts to suppress the power increase, descending and reaching the lower limit switch. An automatic switchover occurs to the backup regulator, which also begins to descend, as recorded by the DREG rapid diagnostic and parameter recording program. However, the efficiency of the four regulator rods is insufficient, and the reactor power continues to slowly increase.
According to the INSAG-7 report (1993):
At 1:09 a.m., the feedwater flow rate was sharply reduced to 90 t/h on the right side and to 180 t/h on the left side, with a total flow rate in the circuit of 56,000–58,000 t/h. As a result, the temperature at the suction [author's note — inlet] of the main circulation pump was 280,8°C (left side) and 283,2°C (right side).
This level—90 tons/hour—is practically zero, within the instrument's margin of error. The water temperature at the reactor inlet has become close to the saturation (boiling) temperature.
However, according to the State Nuclear Academy of Sciences report (1991), “…this was a return of feedwater flow to some average flow rate corresponding to a reactor power of 200 MW and equal to approximately 120 tons/hour on each side of the reactor.” This point of view is shared by a number of specialists.
According to A. G. Tarapon, G. E. Pukhov Institute for Modeling Problems in Power Engineering, National Academy of Sciences of Ukraine (Kiev, Ukraine): "...the capacity was increased only in one (southeastern) quadrant, while in the others it remained at 13,5 MW (residual heat release)..."
At this time, the personnel were placed in extremely difficult conditions. Due to the high steam effect, the power and steam quality, and consequently, the pressure in the steam trap at low power levels, fluctuated up or down due to any disturbances in the multiple circulation circuit. For example, when the BRU-K valves opened, the steam trap pressure dropped, the steam quality briefly increased, and the water level in the steam trap rose. The control unit had to reduce the feedwater flow rate because the water level was +300°C, meaning the turbine was shut down to prevent water from entering the turbine's flow path. Conversely, if the auxiliary main circulating pump was activated, the steam quality dropped, and the water level in the steam trap dropped, requiring an increase in feedwater flow rate, etc. Therefore, the control unit alternately sharply increased and decreased the feedwater flow rate.
THE EIGHTH FATAL ERROR IS A SHARP DECREASE IN FEED WATER CONSUMPTION.
Fall in the OZR level below the permissible level
From 01:12:10 a.m. to 01:18:49 a.m., the DREG program was not running. This was due to the DREG tape being reset by the SDIVT to record the parameters during the "Coast-Down Program..." run.
01:15 am — PN-3,4 are switched on for recirculation (according to the run-down program).
01:18 - TG-8 is synchronized and connected to the network (via the Skala teletype tape),
at 01:18 No.lTG-8 = 32,7 MW (according to the operational log of the NSS at 01:20).
01h 18m 49s - the DREG program is turned on.
01:18:52 AM — 1MPA and 3MPA signals were checked. A 2MPA signal was generated (via the DREG).
01h 18m 54s — AR-2 shutdown due to a malfunction, BRU-K1 TG-8 activation, protection against increasing the level in the BPG to the 2nd limit is disabled; protection against increasing the level in the BTS to the 2nd limit is disabled.
01h 19m 39s — signal “1 PC up”.
01:22:30 — Recording of parameters (before coastdown) on the Skala magnetic tape. According to IAE calculations (performed after the accident), the reactivity margin at this point was ORM = 6–8 st. RR. /According to other data, the ORM margin ranged from 2 to 12 rods with a minimum of 16 RR, which was a gross violation of regulations. The reactor had to be shut down. Additionally, to prevent this from happening, protections were activated to increase the level in the reactor core to the 2nd limit and to increase the level in the reactor coolant system to the 2nd limit./
The operating reactivity margin was significantly lower than normal due to xenon poisoning of the reactor. Out of 211 control rods, according to various estimates, 6–8 were missing; according to Komarov, 1,5 rods; according to V. Fedulenko (I.V. Kurchatov Institute of Atomic Energy), and based on the DREG tapes, only 2 rods (!) were missing, with a minimum of -16.
According to the report (No. 1 INSAG-1), this led to the reactor's emergency protection system (EPS) losing its effectiveness: "Meanwhile, the reactor's reactivity continued to slowly decline. At 1:22:30 a.m., the operator saw on the printout of the rapid reactivity margin assessment program that the operational reactivity margin had reached a value requiring an immediate reactor shutdown. However, this did not stop the personnel, and the tests began... At 1:22:30 a.m., the reactivity margin was only 6-8 rods. This is at least half the maximum permissible margin established by the operating procedure. The reactor was in an unusual, non-scheduled state."
…The work of personnel with an unacceptably small operational reactivity margin led to the fact that almost all the remaining absorber rods were located in the upper part of the active zone.
Under the current conditions, violations committed by personnel led to a significant reduction in the effectiveness of the A3 [emergency protection].”
According to the State Nuclear Regulatory Authority (1991) report, this was a violation of the Regulations, namely: “Operation of a reactor plant with an ORM of 15 RR rods or less during the period… approximately from 01:00 on 26.04.86 until the time of the accident (violation of Chapter 9 of the Regulations)…”
According to the interpretation of INSAG-7, which also testifies to the possibility of xenon poisoning of the reactor: “The consequence of the unsuccessful actions taken to control the reactor was that, in order to compensate for the additional negative reactivity that arose due to xenon poisoning of the core during a power reduction, as well as during the subsequent increase in power to 200 MW, it was necessary to remove some of the operating reserve rods from the reactor, which, according to the Commission, with or without awareness of the severity of the consequences, the personnel transferred the reactor to an unscheduled state, in which the emergency protection ceased to be a guarantee of quenching the nuclear reaction…”.
At 01:22:30 a.m., the SKALA central control system recorded the unit's parameters to magnetic tape. No operational calculations were performed using the PRIZMA program at that time. These were performed after the accident using the magnetic tape removed from the central control system, using the PRIZMA-ANALOG program, outside the Chernobyl Nuclear Power Plant (at the Smolensk Nuclear Power Plant). Control room personnel and SKALA personnel did not have the results of the operational calculations and were unaware of the calculated parameters, including the ORM value, at that time.
Before the accident, the reactor personnel operated the reactor as if blind. They didn't notice the ORM drop below 15 rods due to the length of their calculations. The actual ORM was discovered after the accident, after analyzing Skala's calculations at 1:22 am. SIUR may not have had time to estimate the ORM based on the number of rods inserted (which they can do), as they were very busy operating the reactor at the time. Indeed, it appears they weren't paying attention to the ORM at the time. As the court found, Chernobyl NPP personnel often worked at "limit" levels.
But when they were raising the reactor from the pit, the personnel should have had some understanding of the situation. The rundown was supposed to be carried out by one, the most experienced shift, but by a fateful coincidence, the most inexperienced one was assigned to the job.
According to the State Nuclear Academy of Sciences report (1991), immediately before the tests at 01:22:30: “Under the conditions that arose, a small increase in reactor power (for any reason) due to low subcooling of the coolant to boiling point could lead to an increase in the volumetric steam content in the lower part of the core… Thus, before the start of the tests, the parameters of the core caused an increased susceptibility of the reactor to a self-accelerating process in the lower part of the core… This condition arose not only because there was an increased coolant flow rate than usual…, but primarily due to the low value of the reactor power.”
It should be noted that self-accelerating processes in the reactor are included in the design due to the gigantic positive steam coefficient of 4–5 beta, as well as the absence of a prohibition on operating at low power in the power mode.
While the main circulation pumps (MCPs) were running on coastdown (36,2 s), there were no significant changes. However, after the shutdown of four main circulation pumps (MCPs) powered by the coasting turbogenerator, the positive reactivity created by the reduced water flow in the multiple circulation circuit (MCC) led to an increase in steam quality. The resistance of the fuel channel (FC) increased, and the flow rate through the fuel channel (FC) began to decrease, leading to a further increase in steam quality, and consequently, a rise in reactivity. This triggered an emergency increase in reactor power and the subsequent shutdown of the reactor by the SIUR (Siur-Ion Control System) using the AZ-5 button.
The staff held out hope until the very last moment for the "saving" AZ-5 button; the Chernobyl Nuclear Power Plant often operated "on the brink" in modes with a low ORM margin. Moreover, the reactor's designers insisted that their reactor was safe and that nuclear accidents were completely impossible. I remember those fateful times well. But who could have imagined that it wouldn't work, that the human hand wouldn't be able to keep up with the rapid nuclear processes? As A. Dyatlov, the direct test supervisor, himself admits, the staff was late in pressing the button: "Why Akimov delayed giving the command to shut down the reactor, now it's impossible to find out..."
According to the State Panorama of Russia:
However, it was no longer possible to use the emergency protection system of the reactor of this design under the conditions of the permitted decrease in the ORM, either by emergency signals or manually after the completion of the tests without damage to the core, apparently starting from 00:30 on 26.04.86, which requires verification by additional studies.
How cavitation “finished off” the reactor
Let us also consider the hypothesis of the occurrence of cavitation, which is put forward by specialists from NIKIET.
N. Dollezhal again:
"The uncertainty in the stability of the continuous circulation loop, as well as the fact that one minute before the start of the generator test, the operator sharply reduced the feedwater supply to the separators, apparently caused the occurrence of cavitation."
Cavitation is the formation of cavities (cavitation bubbles, or caverns) in a liquid filled with steam... Moving with the flow... the cavitation bubble collapses, emitting a shock wave and a significant release of energy.
Oh, Yu, Novoselsky:
During reactor startup in the MKU mode, the appearance of nonequilibrium steam at the fuel channel inlets poses no danger: the core is "packed" with absorber rods, and the void effect is negative. In our case, unfortunately, only the reactor's thermal hydraulics resemble MKU mode... After the feedwater supply is almost completely cut off, boiling begins in the upper portion of the fuel channels, and a steam-water mixture with a low steam content (~1%) begins to flow from the core into the separators. After approximately 20 seconds, most of the water volume in the separators is replaced by water at saturation temperature. At this point, testing begins: the safety valve closes, and at 1:23:04, steam supply to the turbine is cut off. The pressure in the circuit begins to increase. Saturated water enters the downcomers from the reactor core, and after 25-30 seconds, cavitation develops in the SAM system due to low superheat. With the appearance of cavitation steam in the lower part of the fuel cell, a powerful positive feedback mechanism is activated: an increase in steam quality causes an increase in energy release, primarily where the steam arrives, i.e., in the lower part of the fuel cell. This, in turn, leads to an increase in steam quality. With an increase in neutron flux density, another positive feedback loop comes into play—accelerated depoisoning of the core. The role of the "end effect" in all of this is poorly understood.
The theory of coolant cavitation in the SAM system is based on experimental data proving that when the coolant temperature at the reactor inlet is close to the boiling point, the coolant boils in the SAM grooves, and the resulting steam enters the reactor fuel channel inlets.
The length of the water communication pipelines from the air defense missile system to the entrances to the technological channels ranges from 5,5 to 24 m; the steam-water mixture travels this distance in one to six seconds and then enters the entrances to the various fuel channels.
Before the explosion: increased vaporization and the void effect
Then, a sharp increase in steam formation in the channels occurs due to the positive steam effect of reactivity, which is associated with the fact that steam, due to its low density, is a poorer neutron absorber than water. This triggers a sharp increase in reactor power due to prompt neutron runaway. This process is accelerated by a powerful positive feedback mechanism: an increase in steam content causes an increase in energy release, which in turn provokes even greater steam production, and so on. Furthermore, the runaway triggers the "burnout" of xenon, which also intensifies the chain reaction. These processes cause the destruction of the fuel channels and a reactor explosion.
A. N. Rumyantsev, one of the experts who warned the designers about the reactor's shortcomings in advance, also identifies this process as the beginning of the accident, since it entailed the introduction of "hot" water into the reactor with a lag of 28 seconds:
The report states that at 1:22 a.m. on April 26, 1986, the operator abruptly reduced the feedwater flow rate, which resulted in an increase in the water temperature at the reactor inlet with a delay equal to the time it took for the water to pass from the separator drums to the reactor… 28-45 seconds. The report states that at 1:23:04 a.m. the shutoff and control valves (SCVs) of turbogenerator #8 were closed and the tests began. By this time, water with an increased temperature had already been supplied to the fuel channels with the longest NVCs for… 19-36 seconds. The reactor continued to operate at a power of 200 MW… in just 36 seconds from the start of the tests, the reactor power, slowly increasing, reached the level that caused the AZ-5 release. …At the time of the AZ-5 release, the TKs in the local supercriticality zone already had an average power level for the zone of 1.2 times the nominal power. Their power continued to increase. Under all the accompanying circumstances, an accident became inevitable.
N. Dollezhal:
At 1:23:04 a.m., the steam supply to the… turbogenerator was cut off [coastdown tests began]. At this time, the emergency protection system should have been activated, but it was disabled; the reactor continued to produce steam for an unknown consumer. Naturally, the pressure in the coolant circuit and the water temperature at the reactor inlet began to rise. In other words, the conditions were precisely those that most likely would lead to the development of powerful hydrodynamic cavitation. At 1:23:40 a.m., i.e., 36 seconds after the second turbogenerator was shut down, the reactor exploded: an accident occurred.
Rumyantsev's calculations reveal the timing of the accident. The second explosion at Unit 4 of the Chernobyl Nuclear Power Plant, which, according to some versions, occurred at 1:23:39 (±1 second), is believed to have occurred. However, the official chronology of events states that the reactor's emergency runaway began at 1:23:43, and the reactor's destruction occurred at 1:23:50–55.
According to INSAG-1 report No. 1, the increase in steam pressure against the background of a sharp drop in the supply of feedwater to the reactor with a decrease in the flow of water through the reactor caused an increase in the reactor power, which, most likely, was the reason for pressing the emergency button.
From the story of witness Lysyuk G.V. (the foreman of the EC), it follows that before Toptunov pressed AZ-5, he managed to shout: “The reactor power is growing at an emergency rate!”
In December 1987, the Swedish Nuclear Power Inspectorate (cited by O. Novoselsky) proposed a similar scenario for the Chernobyl accident: steam bubbles formed during main circulation pump cavitation enter the fuel channels (FCs). Their formation is caused by low water subcooling (i.e., water temperature close to the boiling point) at the pump inlet. Given the high void effect reactivity present in the reactor at the time, this steam triggered a powerful fission flare (reactor runaway). This localized energy surge resulted in the destruction of numerous fuel channels in the lower section. The second explosion was caused by steam displacing water from the upper section of the core. In other words, a large positive void effect again played a role. Over time intervals of several seconds, the sign and magnitude of the equivalent fast reactivity power effect is almost entirely determined by the steam (void) effect. The remaining effects do not have time to make a significant contribution to reactivity.
According to M. Fedulenko (I.V. Kurchatov Institute of Atomic Energy) (in 1986, head of the I.V. Kurchatov Laboratory), steaming the lower and middle sections of the reactor channels did not require a significant increase in power, as the water temperature was practically equal to the saturation (boiling) temperature. This caused the water to be quickly and completely expelled from the reactor channels and replaced by steam, resulting in a rapid, major reactivity spike that triggered a prompt reactor runaway. This runaway "exploded" the fuel assemblies in the lower half of the reactor.
The INSAG-7 final report also effectively acknowledges that core steaming and the void coefficient of reactivity were, at least in part, the cause of the accident:
…Due to the reduced reactor power level at that time, coolant subcooling [the difference between water temperature and its boiling point, in this context water is close to its boiling point] at the core inlet was only insignificant and… could have been zero. These conditions led to the onset of boiling at or near the bottom of the core. Under the operating conditions that existed at that time, the steam coefficient of reactivity was significantly positive, and the core was in a state of increased susceptibility to increased positive steam reactivity feedback in the event of a power increase. Moreover, with the increased coolant flow rate, the margin before cavitation of the circulation pumps decreased.
After the turbine was shut down, the four pumps powered by it began to slow down as the turbine's speed decreased and the voltage of the associated generator dropped. The decreasing flow rate through the core caused an increase in steam quality in the core and triggered the initial positive reactivity feedback, which was, at least in part, responsible for the accident.
After the turbine was shut down, the four pumps powered by it began to slow down as the turbine's speed decreased and the voltage of the associated generator dropped. The decreasing flow rate through the core caused an increase in steam quality in the core and triggered the initial positive reactivity feedback, which was, at least in part, responsible for the accident.
They told the truth, hiding it only partially with words.
IXBT Forum on the causes of the accident:
After all, it was precisely this combination: the pit, then 30 minutes of variable load during unscheduled vibration tests (with active boiling, manually adding large quantities of water, and deliberately cranking up the boil-over protection settings), then a rundown with the reactor running—all of this combined to "overturn" the core's neutron field so dramatically that the reactor's runaway, unchecked by the control and safety systems, began. Had any of these unscheduled experiments been removed, the reactor would have shut down, just as it had during the previous rundown attempts in November 85.
The accident at Unit 4 of the Chernobyl Nuclear Power Plant occurred, on the one hand, due to reactor defects and incomplete technical documentation, and on the other hand, due to some incomprehensible pressure exerted on the staff by a department (sector) of the CPSU Central Committee. A second, more important factor was the staff's lack of knowledge about the specifics of the RBMK reactor, their lack of understanding of the thermal-hydraulic and neutronic processes occurring in the multiple circulation circuit (MFCC) and the reactor, and the fact that the unit was manned by the least experienced shift at the time. The staff had problems with risk assessment—they acted with extreme overconfidence, hoping to trigger the life-saving AZ-5 alarm in time.
THE NINTH FATAL ERROR – THE ORM LEVEL FALLING BELOW THE ACCEPTABLE LEVELS. THE REACTOR HAD TO BE SHUT DOWN.
We only have two questions left to answer: what came first—the reactor runaway and, as a result, the pressing of the AZ-5 shutdown button; or was the reactor shut down after all at the conclusion of the experiment, and assess the role of the end effect under conditions where the majority of the control rods had been removed from the core. And second, the accident itself, the description of which is quite extensive. We will attempt to describe these important points in the next note.
The author expresses gratitude to colleagues for their valuable comments made during the writing of the article.
Links:
Gurachevsky V. L. Introduction to Nuclear Energy.
Chernobyl: KOMAROV'S TESTIMONY
Information on the Chernobyl Accident and Its Consequences Prepared for the IAEA, Report No. 1 (INSAG-1)
O. Yu. Novoselsky, Yu. M. Cherkashov, K.P. Checherov, TECHNICAL ASPECTS OF THE ACCIDENT AT UNIT 4 OF THE CHERNOBYL NUCLEAR POWER PLANT
The truth about Chernobyl lies... in Moscow, Sergei YANKOVSKY, Zerkalo Nedeli No. 16 (441) Saturday, April 26 - May 7, 2003
CHERNOBYL COURT
FOURTH BLOCK, Investigators' Version
Grigory Medvedev. Chernobyl notebook, M Izvestia 1989.
Information on the Chernobyl Accident and Its Consequences, Prepared for the IAEA, Report No. 1 (INSAG-1)
Dollezhal N.A. At the origins of the man-made world. Notes of a designer. — 2010
CHERNOBYL ACCIDENT: SUPPLEMENT TO INSAG-1: INSAG-7
A.G. Tarapon, RECONSTRUCTION OF THE CAUSE OF THE ACCIDENT AT THE CHERNOBYL NUCLEAR POWER PLANT AND THE PROCESSES OF DESTRUCTION OF THE REACTOR AND PREMISES OF UNIT IV
Chernobyl: KOMAROV'S TESTIMONY
V.M. Fedulenko, in 1986, Head of the Laboratory of Thermal Calculations of Channel Reactors, Department 33 of the I.V. Kurchatov Institute of Atomic Energy
On the causes and circumstances of the accident at Unit 4 of the Chernobyl Nuclear Power Plant on April 26, 1986, Report of the USSR State Industrial Supervision Commission
The principle of cavitation heating changes the concept of thermal energy production
O.Yu. Novoselsky, leading researcher at the N.A. Dollezhal Research Institute of Power Engineering until November 2014, The Legend of the Emergency Protection System That Exploded a Nuclear Reactor
Novoselsky O.Yu., Continuation of the "Legend"
A.N.Rumyantsev, Chernobyl in 2009
N.V. Karpan, CHERNOBYL: REVENGE OF THE PEACEFUL ATOM, http://www.physiciansofchernobyl.org.ua/rus/books/Karpan.html, chapter 6
Quoted from O. Novoselsky, Nuclear News, December 1987, p. 67-68
V.M. Fedulenko, On the Causes and Development of the Accident at Unit 4 of the Chernobyl Nuclear Power Plant
CHERNOBYL ACCIDENT: SUPPLEMENT TO INSAG-1: INSAG-7
Information