ML19207A662
| ML19207A662 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 08/03/1979 |
| From: | Ingram F NRC OFFICE OF PUBLIC AFFAIRS (OPA) |
| To: | |
| References | |
| 79-134, NUDOCS 7908220080 | |
| Download: ML19207A662 (24) | |
Text
fDf
/.... 5, UNITED STATES NUCLEAR REGULATORY COMMISSION
[(.[' W.1
. Q/
Office of Public Affairs Washington, D.C. 20555 No.79-134 FOR IMMFCIATE RELEASE
Contact:
Frank L.
Ingram (Mailed - August 3, 1979)
Tel.
301/492-7715 NOTE TO EDITORS Attached are copies of the Foreword and summary sections of a Nuclear Regulatory Commission report "Investi-gation into the March 28, 1979 Three Mile Island Accident by Office of Inspection and Enforcement."
Attachments b
798 223 O
790 82 20 0TrQ
\\
FOREWORD-On March 28, 1979, the Three Mile Island Unit 2 Nuclear Power Plant experienced the most severe accident in U.S. comercial nuclear power plant operating history.
This report sets forth the facts concerning the events of the accident deterTnined as a result of an investigation by the NRC Office of Inspection and Enforcement.
The IE investigation, which is based on the information available at this time, is limited to two aspects of the accident:
1.
Those related operational actions by the licensee during the perica from before the initiating event until approximately 8:00 p.m.,
March ?8, when primary coolant flow was re-established by starting a reactor coolant pump, and 2.
Thosesteps taken by the licensee to control the release of radio-active material to the off-site environs, and to implement his emergency plan during the period from the initiation of the event to midnight, March 30.
These investigation periods were selected because they include the licensee acHons which most significantly affected the accident sequence and its results.
The results of the IE investigation supports the reported population dose from the accident, developed by an an hoc dose assessment group, which included representatives of various cognizant Federal agencies.
In its report dated May 10, 1979, this group concluded that, " Based on the current assessment.
. the off-site collective dose associated with the radioactive mat:.-ial released during the period of March 28 to April 7, 1979, represents minimal risks (that is, a very small number) of additional health effects to the off-site population." At the same time, the IE investigation identifies several inadequacies in the inplant radiation protection activities of the licensee and criticizes the measurements of off-site radiation levels made by the licensee.
In spite of these ider.ti-fied flaws, no glaring inconsistencies have been found which would significantly alter the conclusions reached by the ad hoc group.
The IE investigation also substantiates earlier conclusions concerning the underlying causes of the accident and those factors that contributed to its severity.
Inadequacies in six major areas have been confirmed:
1.
Equipment performance (failu'res and maloperation).
2.
Transient and accident analyses.
3.
Operator training and performance.
798 224 4.*
s.2
2 4.
Equipment and system design.
5.
Information flow, particularly during the early hours of the accident.
6.
Implementation of emergency planning.
Perhaps the most disturbing result of the IE investigation is confirmation of earlier conclusions that the Three Mile Island Unit 2 accident could have been prevented, in spite of the inadequacies listed above. The design of the plant, the equipment that was installed, the various accident and transient analyses, and the emergency procedures were adequate to have pnnented the serious consequences of the accident, if they had been permitted to Snction or be carried out as planned.
For example, had the operators allowed the emergency core cooling system to perform its intended function, damage to the core would most likely have been prevented. There are other-examples set forth in the report where, had a particular operator action been taken, the consequences of the accident could have been significantly mitigated. On the other hand, had certain equipment been designed differently, it too, could have prevented or reduced the consequences of the accident.
The results of the investigation make it difficult. to fault only the actions of the operating staff. There is considerable evidence of a " mind set "
not only by TMI operators but by operators at other plants as well, that overfilling the reactor coolant system (making the system solid) was to be avoided at almost any cost. Undue attention by the TMI operators to avoiding a solid system led them to ignore other procedural instructions and indications that the core was not being properly cooled.
Without this
" mind set" they might well have acted to preclude or better mitigate the accident. Subsequent actions have been required by f:RC to retrain all licensed operators in an effort to preclude recurrence.
Upgraded procedural instructions have also been required.
It is clear that substantial effort is needed, by both the NRC and the industry, to assure that these lessons learned concerning the TMI accident are implemented at other facilities. Within the NRC, early action has been taken to inform other nuclear power plant licensees of the circumstances surrounding the Three Mile Island accident and to require immediate imple-mentation of compensatory measures to prevent occurrence of similar accidents elsewhere.
In addition, a special Lessons Learned Task Force was established in the NRC Office of Nuclear Reactor Regulation. This group has studied the Three Mile Island accident and has issued a report (NUREG-0578) containing short-term recommendations that will significantly improve continued safe operation of licensed nuclear power plants.
The IE investigation adds further emphasis to the need for such plant and procedural modifications.
798 225
3 Because they have the benefit of hindsight, most retrospective investi-gations like this tend to emphasize areas where people and equipment did not perform as desired.
The IE investigation team made a concerted effort to evaluate the reasoning processes of the people who were operating the plant during the course of the accident.
The report con-tains the team's conclusion as to whether or not the operating staff's actions were appropriate in light rf the trai*ing and factual information dVailable to them at the time they had to make decisions as to what course of action to follow.
Further study is clearly needed with resp?ct to the contributions of various other organizations that influence the operation of nuclear rower plants, including designers, reviewers, builders, vendors and regulatory agencies.
These various studies a e now underway; most notably the Presidentially appointed Kemeny Commission, as well as a wide-ranging internal NRC study under Mr. Mitchell Rogovin.
A full assessment of all the underlying causes of the Three Mile Island accident must await completion of these studies.
The findings of this IE investigation will be the subject of appropriate enforcement action in accordance with the Commission's regulations (Part 2, Title 10, CFR).
A icto
$110, J r.1 Director Office of Inspection and Enforcement 798 226
TMI INVESTIGATION
SUMMARY
OF OPERATIONAL ASPECTS PREACCIDENT CONDITIONS On March 28, 1979, during the first 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> of the 11-7 shift (2300 hours0.0266 days <br />0.639 hours <br />0.0038 weeks <br />8.7515e-4 months <br />, March 27, 1979 to 0700 hours0.0081 days <br />0.194 hours <br />0.00116 weeks <br />2.6635e-4 months <br />, March 28, 1979), the Three Mile Island Unit 2 facility was operating at approximately 97% power with the Integrated Control System in full aut'omatic.
Normal makeup, reactor coolant pump seal injection, and letdown were in operation.
The reactor coolant system (RCS).
boron concentration was 1026 ppm, with the pressurizer spray throttled open and pressurizer heaters energized to equalize RCS and pressurizer boron concentration.
All system and core physics surveillance testing required by Technical Specifications was current, and the facility was in one identified Limiting
~
Condition for Operation ACTION statement of those specifications.
The borated water storage tank to spent fuel pool isolation valve (DH-V157) was open to permit BWST recirculation.
This ACTION statement time limit would have expired at 1500 hours0.0174 days <br />0.417 hours <br />0.00248 weeks <br />5.7075e-4 months <br /> on March 29, 1979.
The RCS leakage as calculated by the licensee was within Technical Specification limits, with the identified leakage being that accumulated in the reactor coolant drain tarA (RCDT).
This leakage was identified as being from the electromatic relief valve (EMOV) and/or one or both pressurizer code safety valves.
A review by the investigators of the RCS leakage procedure showed the procedure to be in error, and the facility was actually operating with an unidentified leakage in excess of Technical Specification limits.
798 227 G
+
The assumption by the licensee staff of EMOV and safety valve leakage appears reasonable based on EMOV and safety valve discharge pipe temperatures.
The leakage was sufficient to cause the temperature of the discharge pipes
(~200 F) to be in excess of specified limits (130 F) in plant procedures, and the facility had been operating contrary to these limits for an extended period of time.
The operating staff on duty in the control room during the 11-7 shift of March 28, 1979, was in accordance ith Technical Specifications.
Each licensed staff member was satisfactorily current with regard to the require-ments of the licensee's requalification program for licensed operators.
Shift conditions as regards personnel behavior and activities were normal, and the only unusual log book entries indicated an increase in the amount of water being added to the makeup tank when compared to that of previous shifts.
No evidence was found showing any maintenance on safety related components was in progress at that time.
The shift foreman and two auxiliary operators were engaged-in trans-ferring resin from condensate polisher tank No. 7 to the resin regeneration tank.
This activity was a carryover frcm the previous shift, and a total of about 11 hours1.273148e-4 days <br />0.00306 hours <br />1.818783e-5 weeks <br />4.1855e-6 months <br /> had already been expended attempting to complete this transfer.
Difficulty was being encountered in this transfer and was attributed by licensee staff to a resin blockage in the transfer line, and snift personnel were involved in an attempt to clear it.
TURBINE TRIP AND PLANT RESPONSE At this time, and probably as a result of actions taken to clear the resin blockage in the transfer line, the plant experienced a total loss of feedwater initiated by a loss of condensate flow with an almost simultan-798 228 2
eous trip of the main turbine at 04:00:37.
All emergency feedwater pumps started as designed, the reactor continued to operate at full power in accordance with its protection system design, and RCS temperature and pressure increased for approximately 8 seconds.
The EMOV opened as designed at its setpoint of 2255 psig.
The reactor automatically tripped when the high RCS pressure trip setpoint was reached.
With the trip of the reactor, the RCS experienced an expected coolant contraction, loss of inventory, cooldown, and the attendant reduction in RCS pressure.
The EMOV failed to close when its closure setpoint was reached about 13 seconds later.
This failure was not recognized by the operating staff for more than 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />.
At approximately one minute after the start of the accident, the pressurizer level stopped decreasing and began to rise.
This rise continued until approximately 6 minutes after the accident, when the level went off scale indicating that the pressurizer was completely filled with water (a " solid" pressurizer).
Operator efforts to control the level of the pressurizer, included throttling high pressure
~
injection woich initiated automatically at 2 minutes, and increasing letdown flow to the maximum extent possible.
These efforts were largely unsuccess-ful.
RCS pressure began to increase moderately as the pressurizer went solid.
At this time, the RCS temperature was also increasing.
This increase would also contribute to the pressure rise since saturation conditions now existed in the loops.
This indication of high pressurizer level was caused by voids, either discrete or distributed, that formed in the reactor coolant system coupled with the open EMOV.
The open EMOV vented the steam space of the pressurizer causing a rapid insurge into the pressurizer-At 8 minutes into the accident, an operator, upon seeing continuing low once-through-steam generator (OTSG) levels and decreasing OTSG pressures, searched his panels for the cause.
798 229 3
c
The block valves on the emergency feedwai.er headers were found closed.
Upon opening the block valves, which admitted emergency feedwater to the OTSG, a rapid cooldown of the RCS and corresponding RCS pressure decrease occurred.
At approximately 14 minutes, the Reactor Coolant Drain Tank (RCOT) rupture disc burst, discharging water and steam into the reactor building and causing a further increase in building pressure (which had begun with lifting of the RCDT relief valve) from the continued discharge of reactor coolant into the open RCDT.
Reactor coolant inventory loss continued with the RCS under saturation conditions.
This continuing loss was caused by the discharge of coolant to the reactor building through the open EMOV, coupled with sustained low rates of coolant injection. Reactor coolant pump (RCP) apparent output flow rate decreased while they continued to be operated outside pressure operating limits.
The staff secured the RCPs in the B loop at 74 minutes, and the A loop RCPs at 101 minutes, with the staff expecting that natural circulation would occur. However, the plant parameters were outside defined pressure /
temperature limits for natural convection.
After the trip of the B loop RCPs, the operating staff believed that the B OTSG had developed a secondary-to-containment leak and this generator was isolated.
By this time, after receiving initial early notification of the trip, plant management had become aware of the worsening situation and called for key individuals to come to the site.
RCS pressure continued to decrease and temperature increased as a result of the failure of natural circulation to develop because both loops were vapor bound; the lack of any other adequate heat sink being available to accommodate the core decay heat; the continuing unrecognized reactor coolant loss through the EMOV; and the 4
798 230
throttled high pressure-injection flow.
This throttled high pressure injection flow was justified by the operators because of the apparently satisfactory but actually misunderstood pressurizer level.
The RCS pressure decreased to a low point of 660 psig at 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> and 19 minutes when the leaking EMOV was diagnosed and closed, and RCS pressure began to increase.
The pressure increase appears to be associated with the heating of the remaining RCS contents by core decay heat, a change in RCS inventory made by modest increac9s in high pressure injection, and a zirconium-water reaction.
The operating staff believea up to this time, that no substantive inventory lost from the RCS had occurred.
This b_elief was based on the misunderstood pressurizer level, without regard for the low RCS pressure.
At approximately 2-1/2 hours into the accident, substantial fractions of the reactor core were uncovered and had experienced sustained high temperatures. This condition would be expected to result in fuel damage, substantial releases of core fission products, and hydrogen generation.
The magnitude of these conditions were not recognized by the plant staff.
With the arrival of senior management, the declaration of a General Emergency, an emergency command team was established with the Station Manager as Emergency Director.
Additional unsuccessful attempts to establish sustained forced cooling with one or more RCPs were made.
The plant staff was faced with the following conditions:
o an inability to achieve forced or natural circulation in the RCS o
high incore and loop temperatures, which were considered to be too high to be realistic o
an apparent inability to collapse the voids in the loops despite
' the increased systert pressure and the high pressure injection flow which by now was increased.
.o 798 231
- e 1
5
Unsuccessful efforts to collapse the voids in the loops were continued for approximately 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> at about 2000 psig.
Because of a growing staff concern over the ability of the EMOV block valve to continue to remain functional under a high use rate, the decision was made to reduce system pressure ?nd float the core flood tanks on the RCS as an assurance of adequate core coverage and as a preliminary step in initiating the use of the decay heat removal system.
This depressurization was accomplished in approximately one hour, using the EMOV flow path to the reactor building, and the RCS was held in this low pressure condition for the next 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />.
The failure of the core flood ' tanks to inject a substantial fraction of their volume was interpreted as an indication that the core was covered.
The piping from these tanks contains large loop seals that prevent them from being used effectively for the purpose of ensuring satisfactory core coverage.
The design function of these tanks is to supply water to the vessel in the event of a large break LOCA, which did not occur during this accident.
The extended period of low pressure appears to have assisted in the release of hydrogen gas from the RCS.
This hydrogen resulted from a significant metal-water reaction with the zirconium fuel cladding.
Some of this gas burned in the reactor building at about 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> after the accident producing a rapid pressure spike (28 psig) in the containment.
This pressure spike received relatively little attention from tiie majority of the plant staff, with many of them being unaware that it had taken place.
However, it is also possible that the release of this noncondensible gas from the RCS contributed to the later apparent success of the staff in collapsing the voids in at least one of the reactor loops (A loop, to which the pressurizer is connected).
This increasing success in establishing what appeared to be some degree of natural circulation, despite continuing high temperatures in 6
portions of the system, led the plant staff to conclude that they had achieved a reasonably stable set of conditions.
The Station Manager left the Emergency Control Center (Unit 2 control room) at approximately 1400 hrs (10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> after the start of the accident).
He was absent for approximately 2-1/2 hours to attend a meeting with the Lt. Governor of Pennsylvania.
During his absence, another staff ramber acted as Emergency Director under additional general guidelines dictated by the Station Manager.
After the return of the Station Manager, the plant staff was directed by corporate management to take the RCS to high pressure to collapse the remaining voids.
During this final repressurization, tne decision was reached to attempt another start of a reactor coolant pump to establish forced circulation.
This was successfully achieved at 1950 bcurs, on March 28, 15 hours1.736111e-4 days <br />0.00417 hours <br />2.480159e-5 weeks <br />5.7075e-6 months <br /> 50 minutes after the s' tart of the accident.
SHIFT CREW ACTIONS When the original turbine trip occurred, the shift crew on duty took the appropriate initial response actions indicated for a comoined turbine trip / reactor trip initiated as a result of a loss of main feedwater.
These actions included control manipulations, verifications of automatic actions, and notifications of appropriate personnel.
The misunderstood pressurizer level, and the conditioning instilled in th9 operators by their training and experience to avoid a solid pressurizer condition at all times caused the shift crew, and those who responded early in the transient to provide assistance, to take e series of actions that were contrary to procedural requirements and/or to prudent operating practices.
9 798 233 2'
7 a
These actions led directly to a sufficient loss of reactor coolant inventory to cause core damage.
For a period of as much as 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />, a reversal of these actions could have prevented the extensive core damage that occurred, although some degree of damage may still have been experienced.
Among the actions taken that contributed to the. accident were:
The throttling of high pressure injection to a minimum, averaging o
only 70 gpm net input to the RCS for the first 31/2 hours of the accident; The continued operation of RCPs at RCS pressures-below the procedural o
requirement which requires they be tripped which maintained a water supply at the pressurizer surge line and resulted in a sustained higher mass flow rate through the EMOV; o
Tho failure to isolate the EMOV after the RCS pressure continued to fall, the RCDT rupture disc had blown, and the reactor building sump pump operation indicated a large discharge of water from the building; o
The failure to establish the conditions for natural circulation when the combined RCS pressure and temperature conditions were outside the procedural requirements.
Other actions were taken by the shift crew members during the early hours of the accident that did not directly contribute to the accident, but would have severely impaired the response of safety-related equipment had other plant conditions developed.
Among these were:
k 7-C)B 8
+
o Disabling the automatic start features of the emergency diesel generators making them unavailable for rapid starting in the event of a power failure during the course of the accident.
(This condition was noted after plant management arrived, but was only partially corrected by restoring control room start capability.)
o Isolating the core flood tanks early in the event so they were not available to discharge their contents into the vessel.
The RCS pressure dropped to within 60 psi of the core flood tank pressure just before the EMOV was isolated, and the core flood tanks apparently had been isolated prior to this time based on the continuing belief of the plant staff that no loss in intrentory had occurred.
EMERGENCY STAFF ACTIONS
~
With the arrival of plant management and the establishment of an Emergency Organization, one of the initial actions taken was to increase high pressure-injection flow rates to allow ECCS to function as it would if operators were not present.
This apparently resulted in eventual reflooding of the core.
The actions taken over the next 13 hours1.50463e-4 days <br />0.00361 hours <br />2.149471e-5 weeks <br />4.9465e-6 months <br /> that eventually led to the successful operation of one of the reactor coolant pumps (RCP-1A) have been summarized earlier.
The Emergency Director (the Station Manager) formed a management team for overall conduct of the emergency by assigning specific individuals responsibility for different functional areas.
A system of periodic meetings with that team for status review and decision-making was established.
Decisions were ultimately made by the Emergency Director following consultation witt. that team, with input from offsite management.
Team members then conveyed the decision to the plant staff for implementation.
798 235 as 9
Within hours of its formation, the management team found themselves between two desired stable states of forced circulation, being unable to use the R. cps at high pressure or the decay heat removal system at low pressure.
Natural circulation was similarly unattainable because of vapor binding in the loops.
Their efforts throughout the course of the accident were to move toward one or the other of these desired conditions before the borated water storage tanks (BWST) inventory was exhausted and they would be forced to use the water on the reactor building floor.
Plant parameter information was utilized by the team in planning courses of action to move toward either of these desired conditions with several notable exceptions:
The persistent disbelief of high temperature data from incore o
thermocouples and system RTDs.
This was based on the rationale that the "ormer were not safety grade equipment, while the latter
~
were outside the calibrated range of the detectors; The failure tc recognize the fact that a full pressurizer did not o
provide assucance of core coverage; The failure to recognize the significance and pursue evidence of o
the pressure spike that. occurred in the reactor building; o
The failure to recognizc-the fact that small decreases in core flood tank level did not provide assurance of core coverage; 4
For each of the above areas, the investigation did not attempt to conclude whether the course of subsequent evants would have or could have been altered.
In the case of the high temperatures, acceptance of the temperature data as valid might have prompted a higher high pressure-injection flow rate and a reluctance to subsequently depressurize +ae plant to use the core flood tanks.
However, had that occurred, it cannot be ascertained whether RCP operation could ever have been established in light of the then unrecognized inventory of noncondensibles (hydrogen) that was in the loops and reactor vessel as a result of the zirconium-water reaction.
Similarly, the general recognition of the pressure spike in the reactor building might have led the Station Manager to conclude that cond,'tions were not sufficiently stable to justify leaving the site.
His remaining on the site might have altered the subsequent actions taken, or the timing of those actions.
0FFSITE TECHNICAL SUPPORT The provision of substantive technical support to the management team directing emergency actions on operational matters suffered primarily as a result of communication difficulties.
This was evidenced in three ways:
o Information (both data and plans) transmitted to offsite support, which had been hurriedly mobilized, suffered from time delays.
Thus, the offsite groups were dealing with historical and limited data.
798 237
+a 4'
11
o The. individuals who had to provide data to offsite groups had concurrent duties pertaining to the management of the emergency.
The emergency duties always took precedence as would be appropri-ate.
o The physical communications facilities were inadequate to handle the volume of information requests and transmittals that this kind of accident required.
The investigation has concluded that these communication problems are related to the misconception that the envelope of the analyzed major accidents for this facility are the limiting events.
The duration of these analyzed events are projected to occur in a relatively short time frame.
The provision of the mechanisms needed to mobilize and communicate with substantial offsite technical support on a real-time basis as an accident progresses had, therefore, not been warranted as a part of emergency planning.
a
}g8 12
SUMMARY
OF RADIOLOGICAL ASPECTS Health physics operations at TMI Unit 2 were routine prior to 0400 hrs March 28, 1979.
The normal complement of radiation protection staff was on site.
The emergency plan and implementing procedures had been rehearsed and evaluated during seven drills conducted in the past year.
Most plant personnel had received training in their emergency plan duties.
- However, some workers who would comprise Emergency Repair Party Teams and Radiolog-ical Monitoring Teams had not received adequate training in use of emergency survey instrumentation and in radiation protection procedures.
Routine retraining of radiation / chemistry technicians was not up to date.
While radiation protection training of the plant staff had been sufficient to maintain personnel radiation exposures witSir. limits during normal operations (when radiation levels were low), it had not prepared workers to cope with the high radiation levels that would soon exist inside the Unit 2 auxiliary and fuel handling buildings.
Less than half of the portable radiation survey instruments were operational.
Several installed area radiation monitors and airborne radio-activity monitors, which were not essential for normal operations, but would have been useful during the emergency, were out of service for repair.
Fifty self-contained breathing devices and 175 half and full-face respirators were on site.
Large quantities of protective clothing were available.
All essential communications systems were operational.
Three emergency environmental monitoring kits containing survey and counting instrumentation and personnel monitoring devices were in place.
One of the three kits was later found to have an inoperable instrument for field measurement of radioactive iodine.
Environmental air samplers were operating at eight offsite locations, and environmental TLDs were in position at 20 locations.
Tanks in the liquid radwaste system were filled to about 60% of capacity.
Valves were aligned to pump the reactor building sump to the auxiliary building sump tank.
Ventilation exhaust from fuel handling and auxiliary buildings was through high-efficiency filters and charcoal adsorbers.
798 239 u
5
. y %'
3
At 0400 hrs, the Unit 2 turbine and reactor tripped.
At 0411 hrs, there was a reactor building sump high water level alarm.
By 0415 hrs, the reactor coolant pressure had dropped from 2435 psig at the time of the reactor trip to approximately.1275 psig.
This pressure was below the setpoint for emergency core cooling system initiation (1600 psig).
At 0415 hrs, there was a pressure rise of 1.4 psig inside the reactor building.
A site emergency should have been declared, based on these indications and criteria in the Site Emergency Plan.
However, because the drop in reactor pressure was believed to be under control, and the reactor building pressure increase was considered to be slight, and because there was no evidence of a release of radi c civity from the station, an emergency was not declared.
Subsequently, there were several radiation monitor alarms indicative of an emergency situation, but
~
no emergency was declared.
At 0622 hrs the first radiation monitor response to cladding flilure occurred.
Radiation levels continued to increase and a site emergency was declared at 0655 brs based on these alarms.
The emergency organization was promptly activated following the declara-tion of a Site Emergency.
The Station Manager arrived in the Unit 2 control room at 0705 hrs and relieved the Shif t Supervisor as Emergency Director.
Initially, the emergency organization approximated the planned organization described in the TMI Emergency Plan.
An exception was that Repair Parties were assembled and controlled by both the Emergency Control Center (ECC) in the Unit 2 control room and the Emergency Control Station (ECS) in the Unit 1 health physics / chemistry 1.ab area.
According to Emergency Plan Implementing Procedures, the Repair Darty was to assemble only at the ECS, under the direction of the Supervisor of Maintenance, and coordinated through the Supervisor of Radiation Protection.
Offsite consequences were assessed by performing dose rate calcula-tions.
Because of errors in these calculations, the dose rates initially predicted (10 and 40 rem /hr at Goldsboro) were higher than actual dose rates.
Radiation measurements by survey teams revealed actual doses were 14 798 240 -
low (less than 0.001 rem /hr at Goldsboro).
Offsite agencies and support groups were notified of the Site Emergency by telephone.
At 0724 hrs, a General Emergency was dec'lared based on radiation levels inside the reactor building.
Again, offsite agencies and groups were phoned.
Following the turbine trip, about 8000 gallons of reactor coolant were pumped from the reactor building sump to the auxiliary building sump tank.
This transfer was terminated at 0438 hrs and was not resumed.
The auxiliary building sump tank overflowed to the auxiliary building sump, causing water containing a relatively low concentration of radioactivity to back up through' floor drains onto the fuel handling building and auxiliary building floors.
Following fuel damage, the concentration of radioactivity in the reactor coolant increased by several orders of magnitude.
A flow of this highly contaminated reactor coolant was maintained through the makeup and purification system for several days following the accident.
This flow was the principal pathway by which radioactivity was transferred from the damaged reactor core to the auxiliary and fuel handling buildings, and ultimately to the environment.
Gases evolving from reactor coolant in the makeup and purification system were collected in the waste gas system.
Small leaks in these systems were of little radiological significance during normal operation.
- However, following fuel damage, radioactive gas leaks caused very high concentra-tions of airborne radioactivity inside the auxiliary and fuel handling buildings and resulted in much higher than normal environmental releases via ventilation exhausts from these buildings.
Radiation levels in the vicinity of some makeup and purification system components exceeded the 1imits of the 1icensee's measurement capability (i.e., greater than 1000 R/hr).
High radiation levels inside the Unit 2 auxiliary building caused full scale readings on several station effluent monitors. A full scale reading for the plant vent gas monitor is equal to 2.8 E-2 s i/cc of xenon-c 133.
The particulate and iodine monitors were off-scale due to interference from the large amounts of radioactive noble gases.
798 241 la
>c 15
As hazaads from direct radiation and airborne radioactive material developed rapidly in the Unit 2 auxiliary and fuel handling buildings, the licensee attempted to control the in plant radiation protection program in accordance with Emergency Plan Implementing Procedures.
The ECS was established in the Unit 1 chemistry and health physics area according to Emergency Plan Implementing Procedures.
The activities of Radiological Moaitoring and Repair Party Teams were to be directed b, the Supervisor, Radiation Protection (ECS Director) from this location.
A Unit 2 reactor coolant sample was col'ected in the nuclear sample room at about 0845 hrs without the knowledge of the ECS Director.
The nuclear sample room and primary chemistry laboratory are Ir<.uted in Unit 1 near the ECS.
Collection of this sample resulted in an immediate increase in radiation and airborne radioactivity levels at the ECS, causing the ECS to be evacuated to the Unit 2 control room.
The high radiation levels disabled the Unit 1 counting room, which contained the only instrument on site capable of performing gamma isotopic analyses.
The individuals who collected and analyzed this sample did not take appropriate precautions.
Sample containers were handled directly without use of remote tools or shielding to reduce hand exposure, extremity dosimetry was not worn on hands, and no air sample was collected.
If the sample lines had been properly recirculated or flushed prior to sampling, the individuals would likely have received significantly greater radiation exposure.
Shortly after the ECS was established in the Unit 2 control room, airborne radioactivity began to increase, as measured by the control room incoming air monitor.
At about 1017 hrs, personnel were requested to put on respiratory protective devices (particulate filter masks), based on an alarm of the control room air monitor and an air sample that indicated high gross beta radioactivity.
Control room personnel remained in respiratory praiective devices for about six hours.
Isotopic analysis of an air sample
' J98 242
~
16
would have likely shown that respirators were unnecessary; but, the isotopic analysis capability had been lost.
The ECS was relocated to the Unit 1 control room at 1012 hrs, maintaining the responsibility for coordination of the onsite and offsite environmental survey teams but relinquishing control of the inplant radiation protection program to the Supervisor, Radiation Protection and Chemistry who remained in the Unit 2 control room.
At 1110 hrs, all nonessential personnel were evacuated from the site.
Evacuees were surveyed for contamination at the assembly areas, exit gates, and at an area established offsite at the 500 kV substation.
Several individuals were found to be contaminated.
During the evacuation, the auxiliary building access control point was relocated from outside the auxiliary building entrance to the Unit 2 control room because of increasing airborne radioactivity in the auxiliary building.
This left no positive control over entries into the auxiliary building.
Although the Supervisor, Radiation Protection and Chemistry briefed some individuals and, at times, directed radiation / chemistry technicians to accompany Repair Party Tea:s into the auxiliary building, several entries were made without his knowledge.
These entries were made into areas of high airborne radioactivity and whole-body exposure rates in excess of 100 R/hr.
In at least one instance, survey intruments were not used.
Two individuals who entered the auxiliary building received a whole-body dose of radiation in excess of a regulatory limit; others became contaminated and received unnecessary doses.
At times, high-range pocket dosimeters could not be located and were not worn.
Items of protective clothing such as hoods, when not readily available, were not worn, resulting in several instances of head contamination.
Extremity monitoring devices were not worn.
Air sampling was not performed in the auxiliary building, in the anc where workers were exposed during the period from about 0900 hrs on March 28 through midnight on March 30.
Appropriate respiratory protective devices were not always worn.
798 243 17 Y
5-
\\
In at least two instances, individuals failed to leave high radiation areas in the auxiliary building when their radiation survey instruments failed or deflected full scale.
In one of these instances, this resulted in a whole-body exposure in excess of regulatory limits.
An example which indicates that the radiation protection and chemistry staff was not adequately trained to cope with the hazards which existed occurred during the sampling of reactor coolant on March 29.
The sampling was at the direction of the Supervisor, Radiation Protection and Chemistry, and was performed by a chemistry foreman with assistance from a radiation protection foreman.
Although the need for a reactor coolant sample was known for several hours, less than one hour was devoted to planning and preparation for taking it.
The two foremen entered the nuclear sample room to survey the area and make the valve lineup to recirculate the sample at about 1600 hrs March 29.
They wore protective clothing, full-face respirators with iodine adorbing cartridges, and high-range pocket dosimeters.
No air samples were taken to evaluate airborne radioactivity, and no one was assigned to time their exposure.
Neither remote valve operating nor sample handling tools were used.
The exposure rate to operate a sample valve was remembered to be 90 R/hr.
About 300 ml of reactor coolant was collected in a hand-held polyethlyene bottle.
A 100 ml aliquot of this sample in a graduated cylinder produced a radiation exposure rate of 400 R/hr at a distance of 1 foot.
A second sample was collected in a beaker.
A portion was removed and placed in a small vial.
The remainder was titrated with hydrochloric acid in preparation for a boron analysis.
Another chemistry foreman, wearing a particulate filter respirator and no extremity monitoring, performed the boron analysis.
After the operation, the three individuals were found to be contam-inated.
Decontamination was incomplete, and residual contamination remained 18
on small areas of one individual's skin for over 30 days.
The licensee reported one chemistry foreman received a whole-body exposure, as measured by his TLD, in excess of NRC limits.
The NRC evaluated the handling of this sample and concluded that, in addition to the repcrted whole-body dose, doses to the hands, forearms, and a small area on the skin of the head of the chemistry foreman and to the hands and forearms of the radiation protection foreman exceeded NRC limits.
Prior to and during the emergency, the licensee performed his own onsite personnel dosimetry program.
No one individual was assigned program-matic responsibility for this program.
During the incident, some radiation /
chemistry technicians processed their own TLD badges.
Beginning March 29, one radiation / chemistry technician, who had not operated the system in over a year, worked without procedures for over 40 continuous hours.
The Emergency Plan Implementing Procedures did not address sustained in plant radiation hazards.
The licensee's radiation protection and chemistry staff was not adequately trained to deal with this degree of hazard, and supplies of equipment and instruments were not sufficient to minimize dose to the workers.
During March 28-30, the licensee's land-based onsite and offsite monitoring teams made abcut 500 direct radiation measurements.
These measurements were made primarily to confirm the predicted location of.the noble gas effluent plume and to determine the dose rate produced by the plume.
The rate of release of radioactivity (source term) from the station was periodically calculated based on dose rate measurements in the plume and meteorological conditions cxisting at the time of measurement.
The calculated source terms were used to predict dose rates in other areas when meteorological conditions changed.
Monitoring team survey results were also used to assess the need fcr protective actions and to supplement thermoluminescent dosimeter (TLD) results in assessment of accumuiated dose.
These dosimeters were in place at 15 locations within 3 miles and at 5 locations ranging from 9 to 15 miles from the site prior to the accident.
798 245 19
?;
4 b >,
These TLDs were used to perform an after the fact assessment of direct radiation doses to the public.
In general, the licensee's onsite and offsite survey teams performed surveys in appropriate areas at appropriate times.
However, during a five and one-half hour period from 1700 hrs to 2238 hrs on March 28 and a two-hour period from 0340 to 0540 on March 29, no offsite surveys were performed in the plume.
Both of these periods of time were within the interval when the majority of the noble gases were released and when a plume was well defined because of suff.icient wind speed and almost constant direction.
Radiation levels on March 28, with the exception of 50 mR/hr measured at 1548 hrs on Pennsylvania Rt 441, about 1500 feet south of the North Gate, were not above background until 2238 hrs when a radiation level of 13 mR/hr was measured near Kunkel School (5.6 mi NNW).
Several other radiation levels above background were noted in this general area prior to midnight.
- However, the 13 mR/hr value was the highest one measured, until 30 mR/hr was measured in Goldsboro at 0600 hrs on March 29.
Radiation levels during the remainder of March 29 were generally less than 1 mR/hr, with the maximum noted as 3 mR/hr in Royalton at 2355 hrs.
Offsite radiation levels measured on March 30 were also generally below 1 mR/hr, with the maximum noted as 15 mR/hr at 1ccation S-ll (one mile south of the plant) at 0906 hrs.
The highest radiation level measured onsite (outside of the plant) during March 28-30 was 365 mR/hr (s, y) at 2325 hrs on March 28 at a location 1,000 feet northwest of the Unit 2 station vent.
Although not a pre planned consideration in the licensee's Emergency Plan, helicopter-based survey teums were used to track the noble gas plume.
Up to three helicopters chartered by the licensee were used during March 28-30, with the majority of surveys taking place on March 30.
Over 300 radiation measurements were made by the helicopter teams.
The high st measurements reported were 3000 mR/hr (p,y) at 15 feet above the plant vent at 1410 hrs on March 29 and 1200 mR/hr (p,y) at 130 feet above the Unit 2 reactor building at 08Cl hrs on March 30.
w 4
798 246 20
A total of 57 samples were collected on March 28-30 for the purpose of assessing radioiodine concentrations in the environment.
Air samples collected on March 28, which were counted in the field with a single chanael analyzer having a sodium iodine detector, indicated that "radiciodine" was present at offsite locations with concentrations ranging up to 2.3 E-7 pCi/cc.
The "radiciodine" was subsequently shown to be xenon-133 and xenon-135 at 1400 hrs on March 28 (the time at which the first gamma spectro-metry of one of these samples was completed by the Pennsylvania Bureau of Radiological Health).
Forty of the fifty-seven samples collected were analyzed by gamma spectrometry, and no radioiodine,was detected.
Results of samples from certain portions of the licensee's routine radiological environmental monitoring program collected on March 29 (TLDs, radioiodine in water, and radiciodine in air) were available around mid-day on March 30.
The sample results confirmed that the offsite radiological impact was no worse than earlier estimates made using data gathered by the monitoring teams.
These data supported the conclusion that radioactive noble gases released to the atmosphere were the principal cause of exposure for individuals in the plant environs.
798 247 y
21
UNITED STATES NUCLEAR REGULATORY COMMISSION f
]
WASHING TON. D. C. 2C555 posyaos ANo rags paso U.s. NUCLea R RgGULAyORY OPPICI AL BUSINESS coMuestION PEN ALTY POR PRIVATE USE, $300 U
L J
O e
e N
en 9
9 e
s s
.