05000000/LER-1977-040-01, /01T:on 771213,two Hydrogen Detonations Occurred in Offgas Sys & Stack Base,Resulting in Momentary Radioactive Release & Injury to One Employee.Initial Detonation Caused by Arc Strike on Instrument Tubing
| Event date: | |
|---|---|
| Report date: | |
| 0001977040R01 - NRC Website | |
text
i
^
~
.L-UCENSEE EVENT REPORT
,.
- A N
)
1 coNTRoi. stocol l
l l
l l
tm.sasa mNr au.asouisso iNPo w ten) 1 6
ucemet NUMSEA i
@ l ClT lM lN l 5 l1 l l0l0l~l0l0l0l0l0l-l0l0l l4l1 l1 l1 l1 l l0l1l
'SS 14 15 25 26 30 31 32 N[T,,j
' cArsoom so occNGT NUMSER EVENT oATE REpoRf oAIS
' QCON'T l l
lLl l 01 Sl 01-10 l2 l4 l5 l l 112111317-l 71 l11212l717l7l l
' s 37 Sdl SS 50 81 68 69 74 73 80 EVENT oESCMIT1oN igl On December 13, 1977, two hydrogen detonations occurred. in the Milistone Unit 1 -
l-ae so
- @[
Off Gas. system and stack base. The detonations resulted in a momentary radioactive l as ao
,@l release and-injury to one employee. As a result of the detonations, a site l
ee ao
@l emergenc9 was declared.
I d f (RO-50-245/77-40/lT) 89 seut 80 COO 4 CoMpoNSNT Coog WE.ATON E IM l B I [AJ l Il2 lZ lZ l Zl Zl (Z_]
lZl91919I LN_l 89 to 11 12_
17 43 44 47 48
CAUSE
oESCM8710N
@l T'he initial hydrogen detonation was caused by a welder causino an are strike on l
ee ao g[
instrument tubing associated with the Off Gas system.
l ae so gl l
E "oecoveE s powen amer status oncovtav osscarrCN
'3 [JJ l0 l8 l 9 l l N/A l
lAl l
N/A l
s,s,,
to 12 13 44-45 4e ao op astaIsal See attached response l l
Base of stack.
- l a "stsasso aMouNr op acnvrry
.ccAnom os nasAsa 3
[MJ d
8 9
1o 11 44 48 SQ PERSONNEL EXPOSURES NuMaan tves ossenetCN
@ l010 l1l y l Any mentionable exposure was received by injured employee. (See I
aa 11 12 13 attached)eo PERSONNEL INJURES NuMSEM oESCNFTON 3 l010l1l l No significant injuries.
Slicht concussion and multiole skin abrasions.
l 89 11 12 60 oFFSITE CONSEQUENCES
,3[
See attached response.
l l
89 80 LOSS oR oAMAGE To FACIUTY Tv96 oESCRWfCN 3yl See attached response.
l 89 10 80 PUBUCITY gl Numerous media accounts and press releases.
I as 00 AcolTioNAL FACTORS gl See attached response to letter from B.H. Grier to 0.C. Switzer, December,1977.
I es 80 8301100153 821207 s
l ae P. J. Przekop 203/447-1791 p
~
.a S
RESPONSES TO LETTER - B.H. GRIER TO 0.C. SWITZER, DECEMBER - 1977 Item a.
Review the sequence of events, determine safety significance 1
and define specific measures to prevent recurrence.
~
i l
a.1) Sequence of Events l
On December 13,1977, at 0930 hours0.0108 days <br />0.258 hours <br />0.00154 weeks <br />3.53865e-4 months <br />, a hydrogen detonation in the Unit 1 Off Gas System occurred. This hydrogen detonation resulted I
irr a; local emergency being declared to evaluate the-incident and take necessary corrective. actions. Following the hydrogen detonation, airborne radiation levels in the reactor building increased. indicating that the i
off gas pipe to reactor building loop seal was lost as. a result of the j
explosion.
Personnel were dispatched at 0935 hours0.0108 days <br />0.26 hours <br />0.00155 weeks <br />3.557675e-4 months <br /> to refill the off gas systen loop seals in the reactor building and in the stack basement.
i The reactor building loop seal was promptly re-established without further complications. Although action was taken to re-establish the stack loop seal and this action was initially anticipated to have been successful, subsequent investigations into the incident indicate that this attempt did not successfully re-establish this necessary barrier.
Although plant operation remained stable throughout this event, erratic readout on stack gas radiation monitoring equipment prompted plant management to initiate a power reduction from the initial load of 89 percent rated output to 50 percent rated output.
Initiation of this reduction occurred at 1041 hours0.012 days <br />0.289 hours <br />0.00172 weeks <br />3.961005e-4 months <br />.
Following the power reduction to 50 percent rated output cnd with no corresponding reduction in stack gas monitor readings, it became apparent that background activity at the stack had caused this monitoring systen to essentially become inoperable.
l At 1145 hours0.0133 days <br />0.318 hours <br />0.00189 weeks <br />4.356725e-4 months <br />, with this information in hand, a management decision was made to initiate a controlled shutdown.
At 1300 hours0.015 days <br />0.361 hours <br />0.00215 weeks <br />4.9465e-4 months <br />, during the controlled shutdown, a second hydrogen i
detonation occurred at the base of the stack. This condition prompted i
the Supervising Control Operator to immediately nanually scram the 3
reactor.
The main steam isolation valves were closed and a controlled cooldown was established using the Isolation Condenser system. At 1308 hours0.0151 days <br />0.363 hours <br />0.00216 weeks <br />4.97694e-4 months <br />, a site emergency was declared and personnel proceeded to assemble l
in the designated areas. The Primary Emergency Control Center was also established to determine the magnitude of the incident and site access and egress became controlled. -
t Immediately after the incident, it was. discovered that the second hydrogen detonation had resulted. in a momentary release of radioactive materials. Additionally, an employee passing the stack at the time, was injured and contaminatad as a result of the explosion and required ambulance transportation to the local hospital.
Personnel in the Emergency Control Center coordinated activities to determine the magnitude of the i
l release and minimize the possible spread of contamination. Site monitoring i
teams were dispatched to determine the extent of the release.
l l
i
__.____x,___.___
i v
3 8
Resp 0NSES TO l.ETTER - B.H. GRIER TO 0.C. SWITZER, DECEMBER - 1977 At 1513 hours0.0175 days <br />0.42 hours <br />0.0025 weeks <br />5.756965e-4 months <br />, after an evaluation of information provided by the site monitoring teams, a decision was made to gomence releasing personnel from the site.
Control points were established to suryey each individual for contamination prior to leaving the site. At 1900 hours0.022 days <br />0.528 hours <br />0.00314 weeks <br />7.2295e-4 months <br />,.all personnel had been released, or cleared for release, and the Emergency Control Center was secured.
After a thorough investigation of the incident, it is suspected that the initial hydrogen' detonation, occurring at 0930, December 13, 1977, was caused-by a welder welding in the area outside of the steam l
jet air ejector room. An are strike ort instrument tubing, that is used by steam jet air ejector flow instrumentation, was identified. At the time of the detonation,. welding was being performed adjacent to this 4
tubing and it has been established that the are strike that occurred would have been sufficient to cause the hydrogen detonation.
a.2) Safety Significance The health and safety of the public were never jeopardized by the hydrogen detonations of December 13, 1977.
i The initial detonation was confined within the volume of the off gas piping. The design basis for the off gas piping system a'dequately recognizes the hydrogen detonation potential and the resulting combination of the hydrogen and oxygen components within the piping provided no direct threat to the health and safety of either the public or station personnel.
The second detonation, taki.ng place in the base of the stack, did provide a physical threat.to personnel in the immediate vicinity. While no significant radiological threat existed, even for nearby personnel,
.the sizable energy release of the detonation produced missiles in the form of a metal door and its associated hardware which had the potential to cause severe injuries to personnel who may have been in the path of this debris.
The location of this door in the stack precluded a trajectory which would allow impact upon nuclear safety associated plant structures. A possiblity exists that some of these components may have impacted upon adjacent 345 KV transmission lines with the potential for removing one offsite power supply from Millstone Unit 2.
This event has no safety concerns as additional offsite power supplies are available in addition to onsite diesel generators.
A buildup of hydrogen-oxygen to an explosive concentration, even on a local basis, is considered very unlikely within the reactor building.
i Reactor building construction is such that a single interconnected air i
space exists throughout and is constantly exhausted by a high volume ventilation system or by the Standby Gas Treatment system. The prompt re-establishment of the reactor building loop seal precluded any possibility of a hydrogen detonation in this volume.
l e
m RESP 0NSES TO LETTER - B.H. GRIER TO D.C. SWITZER, DECEMBER - 1977 Analyses of the sequence of events following the first detonation indicates that there was no additional offsite consequence.
There was no evidence to indicate that radioactivity greater than that which would
~
nomally be discharged was released. The design of the discharge piping and the status of the stack filters preclude any consideration of the potential for a large discharge of radioactivity as a result of this event. The recombinatica of the hydrogen and oxygen gases and the creation of a vacuum in the delay pipe resulted in further decreases in the amount.of radioactivity released subsequent to this event.
i The second detonation did result in potential offsita radiological consequences greater than nomal since the release was at ground level.
Using the-meteorology at the time of the incident, a. calculated potential dose at the downwind site boundary of 0.94 mrem to the whole body and l
6.0 mrem to the thyroid was calculated.
Since a number of conservatisms were used in the calculation, it is estimated to. be conservative by a i
factor of 5 to 10. Thus a realistic assessment shows the consequences i
are well below one millirem. Thus, the radiological consequences of the incident are well within federal-limits and guidelines and a small fraction of the average annual background dose in the area of 125 mrsm.
It should also be noted that the downwind direction was south, or out to the Long Island Sound. Therefore, it is unlikely that any member of the public was in the path of the radioactive release.
Dose calculations were performed to determine the dose received by the individual injured in the blast. The calculations determined that the whole body dose and skin dose were less than 60 mrem and 323 mrem, respectively.
In addition, a whole body count was performed and it detemined that the internal uptake was negligible and the expected
}
maximum dose to internal organs would be less than 1 mrem.
Evaluations were also made of other personnel that were involved in reducing localized increases-in radiation levels or that became contaminated as a result of the hydrogen detonations. The evaluations e
concluded that no reportable exposures occurred and internal uptake was negligible.
Environmental radiological samples of snow and air particulates and themoluminescent dosimeters were collected from various locations after the two events.
No radioactivity above nomal background was observed. This further confirms that the offsite consequences were insignificant and that the above analyses are conservative.
l As a consequence of the detonations and related events, the radwaste building ventilation was shut down. Subsequently, moisture drainage resulted in small quantities of radioactivity from the ventilation l
ducts betag deposited on building structures.
Rainfall during the i
period 12/13/77 through 12/21/77 washed some of this activity into yard l
drains and eventually into the onsite quarry.
Conservative calculations based on chemistry analyses of the liquid in yard drains and on building l
i t
i l l
S RESPONSES TO LETTER - 8.H. GRIER TO D.C. SWITZER, DECEMBER - 1977 structures and the amount of rainfall indicate that a maximum of 2.3 j
millicuries total radioactivity was discharged during this period. The predominant isotopes are cesium-134 and cesium'-137. At no time was the discharge concentration above allowable discharge limi~ts (10,CFR20, Appendix B, Table 2). The offsite consequences of this radioactivity have been evaluated and found to be insignificant.
a.3) Measures to Prevent Recurrence In order to prevent. recurrence of this incident, the following corrective actions will have beert completed prior to returning Millstone Nuclear Power Station, Unit 1 to service.
1.
The ceiling slab at elevation 26 feet 0 inches, which was severely damaged by the hydrogen detonation, will not be replaced. Originally, this ceiling slab served as a barrier to allow personnel to enter the stack base while discharges were being made. The ceiling slab was determined to serve no structural purpose and by administrative 1y eliminating personnel access during future operation, it is not needed.
Removal of this slab will eliminate any volume in which combustible gas mixtures could form.
In order to preclude any possibility of out leakage through the stack door, as a result of eliminating the ceiling slab, full seals will be installed on the new door.
In addition, a second leakage barrier will be installed on the outside of the door frame.
2.
The ventilation system carrying the exhaust from all plant buildings, and entering the base of the stack, has been modified to allow air flow through the base of the stack.
In addition, procedure changes have been made to assure that at least one main exhaust fan is running at all times the plant is operating at power. These changes will eliminate any possibility of an explosive mixture of hydrogen and oxygen from fonning in the stack base.
'3.
The stack off gas pipe loop seals have been eliminated. The majority of the off gas pipe is sloped back to the reactor building and any moisture buildup in this section will be removed by the off gas pipe's reactor building loop seal.
The short section of off gas piping which exists within the stack will now be exposed'to the warm atmosphere provided by main buildi.ng exhaust air flow, thereby eliminating the need for a loop seal at this location.
This modification will eliminate any possibility of hydrogen entering the stack base through an unfilled loop seal.
i
. ~.
3 1-RESPONSES TO L'ETTER - B.H. GRIER TO 0.C[ SWITZER, DECEMBER - 1977 4
. In order to preclude the possibility of the reactor building loop seal from becoming lost, a valve has been installed downstream of the loop seal.
This valve will normally be closed and only opened periodically in accordance.with plant procedures to drain any excess water.from the off gas pipe.
A sight glass will also be provided to monitor the flow from the off gas pipe during periodic draining operations 5.
All piping that required replacement as a ' result of the-detonation and all piping that was possibly affected by the detonation has been appropriately leak tested and inspected.
.The results have been found to be satisfactory; all piping is intact and no degradation has been found.
i 6.
All piping.. that is accessible during operation that could cause propagation of an ignition source to the Off Gas system has been identified. This piping has been conspicuously marked and labeled with appropriate signs.
i Item b Consider the need for additional instrumentation to detect combustible or exolosive caseous mixtures in enclosed spaces, and for plant eouipment to remove or dilute such mixtures; In response to the concern relating to the need to evaluate enclosed spaces for the potential of retaining gaseous mixtures,' which could concentrate to a combustible or explosive point, the following plant areas have been analyzed.
1.
Steam Jet Air Ejector Room - This mom contains piping and air removal components, which contain a mixture of hydrogen and oxygen.
In addition to having its own ventilation system this room is connected via a passageway to the main turbine building and its exhaust ventilation systen.
Instrumentafon lines leaving this room are likewise in a well ventilated area.
Piping and components in this area are designed to withstand the forces of a system detonation.
l 2.
Pipe Chase Below SJAE Room - The pipe chase below the SJAE room is open and freely ventilated by the Turbine Building
- Exhaust system. There is approximately 3 feet of off gas pipe in the chase and any possible leakage would be rapidly dispersed l
by existing ventilation.
3.
Reactor Buildini-Holdup Pipe loop Seal Area - Loss of water i
sealing by either " blow-out" or " suck-back" following a detonation could lead to emission of an explosive gas mixture.
Such mixture could not concentrate in this area as it is completely ventilated by the Reactor Building Exhaust Ventilation system and/or the Standby Gas Treatment system.
Further these loop seals have been modified to preclude the loss of sealing wa ter.'
1
- - - - - - - - - - - - ' - - ~
~ - - - ~ ~ ~ ~ ~ ~ " ~ ~ ~ ~ ~ ~ ~
^
'~
,ee,
..;..s 3
RESPONSES TO LETTER - B.H. GRIER TO D.C.' SWITZER, DECEMBER - 1977 4.
Xenon-Krypton Buildino - Current off gas system operations-i preclude the utilization of the piping and components located in this-building.
Tie-in points between this building and the existing off gas system do exist, and isolat' ion valves are red i
tagged closed. Potential valve leakage is of no consequence-since the tie-in pipi.ng and components are maintained under instrument air pressure at a level higher than the pressure in the existing off gas system.
Furthermore,.the building is ventilated' as described in the July 20, 1975, D.C. Switzer letter-to. E Lear,. titled. " Interim Off Gas Systenr."
5 Stack Gas Sampling Room - This enclosed space contains stack i
gas. sampling equipment and off gas filter differential pressure instrumentation, which is connected to the base of the stack i
by instrument tubing, via sealed penetrations.
The potential for penetration leakage and subsequent buildup of an explosive mixture has been eliminated by installation of ventilation equipment designed to maintain the stack sample room at a higher pressure-than the instrument sensi.ng points in the base of the stack.
6.
Main Stack Base Area - Elimination of loop seals in this area results in consideration of only a short run of 4" off gas pipe and the filter casings as potential sources of input for explosive gas mixtures. Both are designed to withstand the forces of a system detonation.. Should a failure occur in t
l these-components, an exglosive concentration is precluded by the design changes made to improved ventilation air flow as described a.3 above. The addition of dilution air to the base
'of the stack is more than adequate to preclude concentrations of gases to the detonation point.
t The need for additional instrumentation to detect explosive gaseous mixtures in enclosed spaces has been considered.
Addition of hydrogen detection equipment has been deemed unnecessary in that previously described methods for elimination of explosive mixtures, are more effective and reliable.
Furthennore, such equipment measures hydrogen concentration at specific sample points and not all areas of an enclosed space.
In addition, area radiation monitors in these areas are sensitive to Off Gas system leakage.
~
Item c.
Review the adequacy of procedural and management controls for preventing and responding to events of this nature, including but not limited to the actions taken in this case I
to review the initial detonation and to assure proper performance l
of corrective actions such as refilling loop seals.
l A review of procedural and management controls for preventing and responding to events of this nature has been accomplished. A comparison I
f i
1 y
s RESPONSES T01.ETTER - B.H. GRIER TO 0.C. SWITZER, DECEMBER - 1977 of existing procedures against license requirements and Regulatory Guide 1.33 revealed no discrepancies or inconsistancies.
A review of personnel actions during and subsequent to these events indicated that confonnance to the requirements of these procedures was maintained ~throughout.
However, as a result of these reviews and of knowledge gained subsequent to thesa events, additional precautions and prerequisites have been identified and will be incorporated into station procedures.'
These procedural changes. will provide additional assurances against the recurrence of a similar event.
Item d Identify and correct instrumentation, equioment, piping and structural damage resulting from the detonations.
' After the incident, thorough inspections were performed to assess the extent of the damage.. These inspections were not limited solely to the Off Gas system and stack, but also included associated equipment and systems, such as instrumentation, electrical equipment, ventilation duct work and miscellaneous piping.
Following is a list of items, requiring corrective action, that were discovered during these inspections.
9 1.
Stack base door - severely damaged as a result of the hydrogen detonation.
2.
Stack interior ceiling slab - excessive damage consisting of bent support beams and exposed re-bar.
3.
Stack surface - vertical cracks on interior and exterior surfaces that.may or may not have been caused by the hydrogen detonation.
4.
Off Gas system isolation valve (FCV 8-5) - located in stack.
Damage to valve yoke and valve operator stem.
5.
Off gas discharge line (4"-0G-14) - located in stack. Damage consisted of bent pipe, distorted elbow and broken anchor bolts.
t 6.
Gland seal exhauster discharge line (10"-0G-17) - located in stack.
Damage consisted of bent and distorted pipe and broken anchor bolts.
7.
Standby Gas Treatment Exhaust system discharge line (12"-SGT-1) - located in stack.
Damage consisted of bent and distorted pipe and broken anchor bolts.
)
8.
Off gas filter inlet valves (1-0G-51a and Sib) - located in stack.
Damage to reach rod operators.
9.
Off gas filter outlet valves (1-0G-53a and 53b) - located in stack.
Damage to reach rod operators.
e i
~
a
' RESPONSES'TO LETTER - B.H. GRIER TO D.C. SWITZER, DECEMBER - 1977
- 10. Augmented Off Gas systent return 1ine (2"-0G-49) - located in stack. Not in service at this time.
Damage consisted of distorted pipe and broken pipe hanger.
l 1T.
~
Plant ventilation system exhaust turning vane - located in
~
stack turning vane found to be bent and distorted.
l
- 12. Main exhaust fan fHVE'l-C) - located in heating and ventilation room.
Internal r b broken loose.
l
- 17. Liquid radwaste exhaust systent (HVE-7A and 78) - Minor seam split on exhaust duct.
- 14. Steam jet air ejector flow transmitter - discovered inoperable after first hydrogen detonation.
- 15. Off gas filter differential pressure indication - appeared to be damaged by first detonation.
16.
Iso-kinetic probe for the stack gas sampling system - located in stack. Supporting beasrdislocated and probe mounting beam bent.
.17.
Steanr det Air Ejector Rupture Disc - ruptured by first I
detonation.
i
~
The corrective actions for several of the above itens have been I
specified in other areas of this report.
In all other cases the corrective-action required to return the above items to an operable status will I
have been taken prior to returning the Unit to operation. An engineering i
evaluation has been completed by Northeast Nuclear Energy Company and
. EBASCO Services Incorporated and has concluded that no stack structural damage has occurred that would reduce its capacity to withstand design i
loads.
In addition, an inspection of the stack foundation mat did not reveal any conditions that would be indicative of a damaged base slab or support piles.-
i i
i I
I 1.