Anatomy of a nuclear woops.
12-12-2017, 07:50 PM (This post was last modified: 12-20-2019, 04:11 PM by Dartz.)
12-12-2017, 07:50 PM (This post was last modified: 12-20-2019, 04:11 PM by Dartz.)
From: Split-Personality
To: "The Glow in the Dark Brigade"
Subj: An anatomy of an atomic fuckup.
Hot off the presses from the Active Reactor Safety Commission, The Official Report on the Friggan Accident is Out at last and it makes for the most amusing reading one can have in such a nuclear group as ours.
As I'm reading through it, I'm summarising:
The Background
The reactor cores on Frigga are old Lensherr-type Stellarators of the sort that were built everywhere in the belt just before the Boskone War. These 'First-Generation' fusion reactors were popular among the mundane groups as most of the power systems and support hardware could be bought from mundane sources. The turbogenerators, pumps and everything beyond the 'black box' of the core could be standard hardtech of the sort keeping the lights on in mundania for a century.
These reactors, unlike every other fusion core since, run on deuterium-tritium fusion - which produces a spray of fast neutrons the turn everything they touch into white-hot radiation. The net result of this being, that after any sort of lengthy period of operation, the reactor structure becomes hideously radioactive. The core of a recently shut-down Stellarator has been described as the 'single most intense man-made radiation source known' in one publication
Needless to say, most of these have since been decommissioned and quietly entombed in place to decay peacefully away.
Ten units currently remain in use in Fenspace, of which 4 were installed on Frigga. After nearly fifteen years of service, it was decided to run life-extension tests on Frigga's power-plant, to verify its continued safety. This test series was supervised by Lensherr engineers, and was coming to a close when the accident happened.
The Core
Stellarators rely on a precisely shaped magnetic field to maintain the fusion reaction. So long as the field is maintained and the core has fuel - the reaction runs. As soon as the field is removed, the reaction collapses and the cooling system is ramped down. In the Lensherr Stellarators this field was maintained by superconducting magnets, cooled by liquid helium.
In theory, removing the magnetic field was achieved by heating the magnets to the point that they lost their superconductivity and collapsed the field.
In practice, it takes time for the liquid Helium to actually boil off once the heater coils are turned on - up to 60 seconds in normal use. After years of operation, the heaters and their emergency batteries started to degrade to the point where the time taken to quench the magnetic fields in the vent of an emergency SCRAM became unacceptably high, potentially leading to 'operating limits' being exceeded under certain conditions.
The final test to be performed on Frigga Unit 04 was a full power test of the reactor's ability to safely SCRAM. The test called for excitation power to be removed from the main turbogenerator, to simulate a power system fault, and for a reactor SCRAM to be triggered. The emergency generators would come on-line and keep the main cooling pumps in operation. The auxiliary generators supplied enough power to operate the pumps to meet the shut-down decay heat demand of the reactor.
They could not meet the full-power core demands.
This test was otherwise known as 'poking the bear'. Five operating reactors had failed this test in the previous six months and had become subject to shut-down orders.
The Accident
Prior to the test, Unit 04 had been in full power operation for 4 weeks, exceeding the TPS' minimum required spec' by 2 weeks. Core temperatures, neutron flux and radiation levels had all achieved a steady state, suitable for the test. Fuelling levels were rich, but below the upper limit. The reactor normally contained enough fuel for five minutes full-power operation.
This is taken directly from the report:
At 14:00 station time, a grid switching fault was simulated which resulted in a loss of grid connection at the turbogenerator. The turbine began to overrev immediately.
At 14:00:03, the station's automatic control system triggered a SCRAM as per expected Turbine Trip procedures. Steam inlets to the turbine were closed. Divert valves to the condensers were switched to direct. Fuel injectors were closed leaving the core running on internal fuel. The emergency generator start sequence was begun. Cooling pumps began to slow. The field coil heaters were activated automatically.
It should have taken fifteen seconds for the coil heaters to reach operating temperature.
At 14:00:20 The emergency generator failed to correctly sequence and tripped. This was logged as an abnormality, but within the expected bounds of the test.
At 14:00:25 An operator switched reactor cooling pump power supply back to the coasting turbogenerator. Sufficient excitation voltage remained to operate the cooling pumps at a reduced capacity for another thirty seconds if needed. The reactor continued to operate at full power. Core temperature began to increase as coolant flow dropped.
At 14:00:40: The emergency generator was manually switched on-line. No appreciable heating had yet been achieved in the superconductors. Helium pressures showed no sign of boil-off.
At 14:00:45. With indication of magnet heating yet to be achieved, the TPS specified that the test be aborted at this point. Instead the test is continued. With cooling pumps operating at reduced capacity, core temperature increases rapidly.
At 14:00:58. An oxygen depletion alarm sounds in a vent shaft. This is understood by the operators to be an indication that boiloff is happening as expected,
At 14:01:03 Core feed water boils at an increasing rate. Steam pipe pressures increase within the core. A low water alert from the steam seperator system triggers a second automatic SCRAM from the Differential Shutdown System. The reactor control system now has two seperate SCRAM commands in effect.
At 14:01:15. Pressure sensors on the magnets begin to indicate expected values. Increasing temperatures and pressure in the core result in an increased thermal power output as the fusion reaction accelerates.
At 14:01:20. Core temperature exceeds red-line. The test abort sequence is begun.
At 14:01:27. An attempt is made to switch steam flow back to the main turbogenerator and regain power as prescribed in the proceedure. This is rejected by the control system due to the still-active automatic SCRAM from the low-water alert.
At 14:01:36. The safety valve on the top of the steam generators opens to relieve system pressure. Radiation alarms sound in the reactor hall.
At 14:01:44. Core temperature is now 20% above maximum permitted. Core pressure is maintained at maximum by the relief valves.
At 14:01:50. An operator is able to sequence full power to the cooling pumps from an external circuit using an unnaproved proceedure. Cooling pumps achieve full power. SCRAM automatic control commands maximum revolutions from the pumps.
At 14:02.01 Reactor cooling pump 1 starts to over-rev followed by pumps 2, 3 and 4. A commanded trip is overriden by the an operator. Vibration alarms sound. It is thought that this is due to cavitation in the pump, caused by the high temperature of the feedwater. All coolant flow stops. Core temperature spikes. Water remaining in the core boils to steam.
At 14:02:09. Magnet temperatures have still not yet begun to rise. Based on current draw data it is estimated that only 50% of Helium coolant has boiled at this point.
At 14:02:13, An attempt is made to manually reduce pump RPM and achieve flow. This is overridden by the DSS system commanding maximum pump RPM due to a zero-flow condition
At 14:02:15: Core temperature is now 50% above redline. Temperature sensors reach their limit. Temperatures above this point are not recorded. Core power still remains at maximum.
At 14:02:20, A second attempt is made to manually reduce pump RPM and achieve flow. The DSS system is overriden with a patch cable. An operator takes manual control of the cooling pumps and reduces power.
At 14:02:30, Coolant flow is restored, but only at low levels in three of four circuits. The fourth circuit remains locked.
At 14:02:33. With a meltdown rapidly imminent. the emergency coil destruct button is pushed, triggering explosive charges on the field coil supports to destroy the magnetic field coils. It will take ten seconds for the detonators to charge.
At 14:02.35. Control system shows a rapid increase in flow into the core on circuit four. This is taken as an indication that the pumps are working.
At 14:02.37. Power Rate Increase Warning. Sector 270 to 290. Neutron Rate increase Warning. Sector 270 to 290. Local power exceeding maximum: Sector 270 to 290. Reactor power exceeding maximum: 10GW. It is theorised that damaged sensors are giving false readings as cables short-circuit.
At 14:02:38. Chamber pressure exceeds atmospheric.
At 14:02:41. Control system reports loss of signal from 275 of 292 functioning sensors in the reactor chamber.
At 14:02:49. Fire Alarm. Reactor 4 Turbine Hall. Reactor 4 Chamber.
----
Yes, you read that right. The reactor core exploded. The core exploded and caught fire.
I'll leave you all to sink your teeth into that, for the time being.
##Next Time.... How they handled the reactor accident, some intriguing discrepancies, and some damning conclusions.
To: "The Glow in the Dark Brigade"
Subj: An anatomy of an atomic fuckup.
Hot off the presses from the Active Reactor Safety Commission, The Official Report on the Friggan Accident is Out at last and it makes for the most amusing reading one can have in such a nuclear group as ours.
As I'm reading through it, I'm summarising:
The Background
The reactor cores on Frigga are old Lensherr-type Stellarators of the sort that were built everywhere in the belt just before the Boskone War. These 'First-Generation' fusion reactors were popular among the mundane groups as most of the power systems and support hardware could be bought from mundane sources. The turbogenerators, pumps and everything beyond the 'black box' of the core could be standard hardtech of the sort keeping the lights on in mundania for a century.
These reactors, unlike every other fusion core since, run on deuterium-tritium fusion - which produces a spray of fast neutrons the turn everything they touch into white-hot radiation. The net result of this being, that after any sort of lengthy period of operation, the reactor structure becomes hideously radioactive. The core of a recently shut-down Stellarator has been described as the 'single most intense man-made radiation source known' in one publication
Needless to say, most of these have since been decommissioned and quietly entombed in place to decay peacefully away.
Ten units currently remain in use in Fenspace, of which 4 were installed on Frigga. After nearly fifteen years of service, it was decided to run life-extension tests on Frigga's power-plant, to verify its continued safety. This test series was supervised by Lensherr engineers, and was coming to a close when the accident happened.
The Core
Stellarators rely on a precisely shaped magnetic field to maintain the fusion reaction. So long as the field is maintained and the core has fuel - the reaction runs. As soon as the field is removed, the reaction collapses and the cooling system is ramped down. In the Lensherr Stellarators this field was maintained by superconducting magnets, cooled by liquid helium.
In theory, removing the magnetic field was achieved by heating the magnets to the point that they lost their superconductivity and collapsed the field.
In practice, it takes time for the liquid Helium to actually boil off once the heater coils are turned on - up to 60 seconds in normal use. After years of operation, the heaters and their emergency batteries started to degrade to the point where the time taken to quench the magnetic fields in the vent of an emergency SCRAM became unacceptably high, potentially leading to 'operating limits' being exceeded under certain conditions.
The final test to be performed on Frigga Unit 04 was a full power test of the reactor's ability to safely SCRAM. The test called for excitation power to be removed from the main turbogenerator, to simulate a power system fault, and for a reactor SCRAM to be triggered. The emergency generators would come on-line and keep the main cooling pumps in operation. The auxiliary generators supplied enough power to operate the pumps to meet the shut-down decay heat demand of the reactor.
They could not meet the full-power core demands.
This test was otherwise known as 'poking the bear'. Five operating reactors had failed this test in the previous six months and had become subject to shut-down orders.
The Accident
Prior to the test, Unit 04 had been in full power operation for 4 weeks, exceeding the TPS' minimum required spec' by 2 weeks. Core temperatures, neutron flux and radiation levels had all achieved a steady state, suitable for the test. Fuelling levels were rich, but below the upper limit. The reactor normally contained enough fuel for five minutes full-power operation.
This is taken directly from the report:
At 14:00 station time, a grid switching fault was simulated which resulted in a loss of grid connection at the turbogenerator. The turbine began to overrev immediately.
At 14:00:03, the station's automatic control system triggered a SCRAM as per expected Turbine Trip procedures. Steam inlets to the turbine were closed. Divert valves to the condensers were switched to direct. Fuel injectors were closed leaving the core running on internal fuel. The emergency generator start sequence was begun. Cooling pumps began to slow. The field coil heaters were activated automatically.
It should have taken fifteen seconds for the coil heaters to reach operating temperature.
At 14:00:20 The emergency generator failed to correctly sequence and tripped. This was logged as an abnormality, but within the expected bounds of the test.
At 14:00:25 An operator switched reactor cooling pump power supply back to the coasting turbogenerator. Sufficient excitation voltage remained to operate the cooling pumps at a reduced capacity for another thirty seconds if needed. The reactor continued to operate at full power. Core temperature began to increase as coolant flow dropped.
At 14:00:40: The emergency generator was manually switched on-line. No appreciable heating had yet been achieved in the superconductors. Helium pressures showed no sign of boil-off.
At 14:00:45. With indication of magnet heating yet to be achieved, the TPS specified that the test be aborted at this point. Instead the test is continued. With cooling pumps operating at reduced capacity, core temperature increases rapidly.
At 14:00:58. An oxygen depletion alarm sounds in a vent shaft. This is understood by the operators to be an indication that boiloff is happening as expected,
At 14:01:03 Core feed water boils at an increasing rate. Steam pipe pressures increase within the core. A low water alert from the steam seperator system triggers a second automatic SCRAM from the Differential Shutdown System. The reactor control system now has two seperate SCRAM commands in effect.
At 14:01:15. Pressure sensors on the magnets begin to indicate expected values. Increasing temperatures and pressure in the core result in an increased thermal power output as the fusion reaction accelerates.
At 14:01:20. Core temperature exceeds red-line. The test abort sequence is begun.
At 14:01:27. An attempt is made to switch steam flow back to the main turbogenerator and regain power as prescribed in the proceedure. This is rejected by the control system due to the still-active automatic SCRAM from the low-water alert.
At 14:01:36. The safety valve on the top of the steam generators opens to relieve system pressure. Radiation alarms sound in the reactor hall.
At 14:01:44. Core temperature is now 20% above maximum permitted. Core pressure is maintained at maximum by the relief valves.
At 14:01:50. An operator is able to sequence full power to the cooling pumps from an external circuit using an unnaproved proceedure. Cooling pumps achieve full power. SCRAM automatic control commands maximum revolutions from the pumps.
At 14:02.01 Reactor cooling pump 1 starts to over-rev followed by pumps 2, 3 and 4. A commanded trip is overriden by the an operator. Vibration alarms sound. It is thought that this is due to cavitation in the pump, caused by the high temperature of the feedwater. All coolant flow stops. Core temperature spikes. Water remaining in the core boils to steam.
At 14:02:09. Magnet temperatures have still not yet begun to rise. Based on current draw data it is estimated that only 50% of Helium coolant has boiled at this point.
At 14:02:13, An attempt is made to manually reduce pump RPM and achieve flow. This is overridden by the DSS system commanding maximum pump RPM due to a zero-flow condition
At 14:02:15: Core temperature is now 50% above redline. Temperature sensors reach their limit. Temperatures above this point are not recorded. Core power still remains at maximum.
At 14:02:20, A second attempt is made to manually reduce pump RPM and achieve flow. The DSS system is overriden with a patch cable. An operator takes manual control of the cooling pumps and reduces power.
At 14:02:30, Coolant flow is restored, but only at low levels in three of four circuits. The fourth circuit remains locked.
At 14:02:33. With a meltdown rapidly imminent. the emergency coil destruct button is pushed, triggering explosive charges on the field coil supports to destroy the magnetic field coils. It will take ten seconds for the detonators to charge.
At 14:02.35. Control system shows a rapid increase in flow into the core on circuit four. This is taken as an indication that the pumps are working.
At 14:02.37. Power Rate Increase Warning. Sector 270 to 290. Neutron Rate increase Warning. Sector 270 to 290. Local power exceeding maximum: Sector 270 to 290. Reactor power exceeding maximum: 10GW. It is theorised that damaged sensors are giving false readings as cables short-circuit.
At 14:02:38. Chamber pressure exceeds atmospheric.
At 14:02:41. Control system reports loss of signal from 275 of 292 functioning sensors in the reactor chamber.
At 14:02:49. Fire Alarm. Reactor 4 Turbine Hall. Reactor 4 Chamber.
----
Yes, you read that right. The reactor core exploded. The core exploded and caught fire.
I'll leave you all to sink your teeth into that, for the time being.
##Next Time.... How they handled the reactor accident, some intriguing discrepancies, and some damning conclusions.
I love the smell of rotaries in the morning. You know one time, I got to work early, before the rush hour. I walked through the empty carpark, I didn't see one bloody Prius or Golf. And that smell, you know that gasoline smell, the whole carpark, smelled like.... ....speed.
One day they're going to ban them.