ML20045A365

Forwards Attachments A,B & C.Attachment a Represents Important Insights from ABWR Severe Accident Analysis.Info Will Be Used to Develop Tier 2 Documentation

ML20045A365
Person / Time
Site:	05200001
Issue date:	06/04/1993
From:	Fox J GENERAL ELECTRIC CO.
To:	Poslusny C Office of Nuclear Reactor Regulation
References
NUDOCS 9306100160
Download: ML20045A365 (17)

Text

_ _

[

+

GE Nuclear Energy Geretal(lectuc Comrsty I75 Cunnet Avenue. San Jose. CA 95125 June 4,1993 Docket No. STN 52-001 Chet Poslusny, Senior Project Manager Standardization Project Directorate Associate Directorate for Advanced Reactors and License Renewal Office of the Nuclear Reactor Regulation

Subject:

Insights from the ABWR Severe Accident Analysis

Dear Chet:

Please find the enclosed Attachments A,B & C. Attachment A represents the important insights from the ABWR Severe Accident Analysis. This information will be used to develop Tier 2 documentation. It will also be used as the basis for Tier 1 and ITAAC. However, only a small portion of this information should be elevated to Tier 1.

As an example, the Tier 1 and Tier 2 documentation for the COPS system has been included as Attachments B and C.

Sincerely, 4 yp Jack Fox Advanced Reactor Programs cc: C. E. Buchholz (GE)

Norman Fletcher (DOE)

JIWim i

9306100160 930604 PDR ADOCK 05200001

/

l l

+

A

~

PDR

/

1 i

Kkud A ABWR Features to Mitigate Severe Accidents The ABWR has been designed to prevent the occurrence of a core damage accident.

In fact, the probability of a core damage accident is less than I chance in 1 million.

l This represents an improvement in severe accident prevention when compared to current plants. In the extremely unlikely event of a core damage accident, the ABWR containment has been designed with specific mitigating capabilities. These capabilities not only mitigate the consequences of a severe accident but also address uncertainties in severe accident phenomena. The capabilities are listed below along with a discussion of the specific severe accident phenomena that the mitigation devise is addressing. The severe accident issues addressed are consistent with the issues discussed in SECY 90-016.

Firewater Addition System This system not only can play an important role in preventing core damage, it is the primary source of water for flooding the lower drywell should the core become damaged and relocate into the containment. The primary point ofinjection for the firewater addition system is the LPFL header inside the vessel. Flow can also be delivered through the drywell spray header to the upper drywell. The drywell spray mode of this system not only provides for debris cooling, but it is capable of directly cooling the upper drywell atmosphere and scrubbing airborne fission products. This system has sufficient capacity to cover the core debris ex-vessel and provide debris cooling and scrub fission products released as a result of continued core-concrete interactions.

The firewater addition system operating in the drywell spray mode will also reduce the consequences of a suppression pool bypass or containment isolation failure. This is due to the fission product removal function performed by this mode of operation.

Fission products will be scrubbed by the sprays prior to leaving the containment.

The firewater addition system has been sized to optimize the containment pressure response and slow the rate of containment pressurization. The system is capable of delivering water to the containment up to the setpoint pressure of the COPS system.

3 3

The flow rate, nominally 0.055 m /sec at runout and 0.044m /sec at the COPS setpoint, is sufficient to allow cooling of the core debris, while maximizing the time until the water level reaches the bottom of the vessel, at which point it is turned off.

Lower Drywell Flooder The lower drywell flooder system has been included in the ABWR design to provide alternate cavity flooding in the event of core debris discharge from the reactor vessel and failure of the firewater addition system. This system is actuated from the melting of a fusible plug. The temperature set point for the plug is 533 K. The system consists of ten 4 inch diameter lines located about 4 m below the normal suppression pool water level discharging into the lower drywell about I m above the floor. The expected flooder flow is 10.8 kg/s per valve. Only two of the valves are required to open to remove decay heat energy and the energy from zirconium-water reaction and allow for quenching of the debris. The passive flooder will not open until after vessel failure.

CEB9S17-2

c By flooding after the introduction of core material, the potential for energetic core-water interactions during debris discharge is minimized. The flooder will cover the core debris with water providing for debris cooling and scrubbing any fission products released from the debris due to core-concrete interactions.

(Further design and testing detail is not included as it is not an important insight from the PRA. However, this information will be included in the Tier 2 design description.]

Containment Overpressure Protection The COPS is part of the atmospheric control system and consists of an 8-inch diameter overpressure relief rupture disks mounted on a 14-inch line which connects the wetwell airspace to the stack. This system will provide for a scrubbed release path in the event that pressure in the containment cannot be maintained below the structural limit. The system includes two reclosable valves which may be used to re-establish containment isolation as a part of post accident recovery. These valves should be normally open and be designed to fail open.

This controlled release will occur at a containment pressure of 0.72 MPa (90 psig). The setpoint of the COPS system is based on the competing goals of minimizing the probability of containment structural failure and maximizing the time of any fission i

product release. The setpoint was assumed to be reliable to within +/- 5% of the actuation pressure at nominal temperature. The effect of temperature on the rupture disk should be small, the analysis assessed the variability of about 2% per 56 K (100 F).

The area of the rupture disk is designed to permit the COPS system to be effective in mitigating the pressure increase during an ATWS event in which the operator controls the injection flow. This provides ample margin to steam generation rates related to decay heat generation. Analysis of the blowdown of the containment following rupture disk operation indicates that the pool swell and the blowdown loads will not threaten the piping, and that significant entrainment will not occur.

This system is beneficial for several of the severe accident issues. In cases with continued corc<oncrete attack, or those with no containment heat removal operational, the containment will pressurize. The COPS pro ides a controlled release path preventing containment structural failure and mitigating fission product release.

The COPS system reduces the effect of uncertainties in severe accident behavior, e.g.

debris coolability, in the ABWR design.

Vessel Depressurization The ABWR reactor vessel is designed with a highly reliabic depressurization system.

The nitrogen supply and battery capacity are sufficiently to allow depressurization after RCIC failure during a long-term station blackout. This system plays a major role in preventing core damage, however, even in the event of a severe accident, the RPV depressurization system can prevent the affects of high pressure melt ejection. If the reactor vessel would fail at an elevated pressure, fragmented core debris could be transported into the upper drywell. The resulting heatup of the upper drywell could pressurize and fail the drywell. Parametric analyses performed in Section 19AE of the CEB9317-3

>wi

+

e op se,

.ABWR SSAR indicate that even in the event of direct containment heating, the probability of early dr>well failure is low. The RPV depressurization system further decreases the probability of this failure mechanism.

Lower Drywell Design

)

The details of the lower drywell design are important in the response of the ABWR containment to a severe accident. Seven key features are described below.

Sacrificial Concrete The floor of the ABWR lower drywell include a 1.5 meter layer of concrete above the containment liner. This is to insure that debris will not come in direct contact with the containment boundary upon discharge from the reactor vessel. This added layer of concrete will protect the containment from possible early failure.

- Basaltic Concrete The sacrificial concrete in the lower drywell of the ABWR has been constructed of low gas content concrete. The selection of concrete type is yet another example of how the ABWR design has striven not only to provide severe accident mitigation, but to also address potential uncertainties in severe accident phenomenon. Here, the uncertainty is whether or not the core can be cooled by flooding the lower drywell.

For scenarios in which the lower drywell flooder is unable to cool the core debris, the concrete type selected will result in a very low gas generation rate. This translates into a long time to pressurize the containment. This is important because time is one of the key factors in acrosol removal.

Pedestal The ABWR pedestal is formed of two concentric steal shells with webbing between them. The space between the shells is filled with concrete. The thickness of the pedestal structure is 1.7 m. A parametric study of core concrete interaction was a

performed which indicated a very small potential for pedestal failure even in the event of continued interaction. Furthermore, any potential failure will not occur Ihr approximately one day.

Sump Protection The lower drywell sumps are protected by corium shields such that core debris will not enter them. This maximizes the upper surface area between the debris and the water and maximizes the potential to quench the core debris. The shields are made of alumina which is impervious to chemical attack from core-concrete interaction. The walls of the floor drain sump shield have channels which permit water flow, but which will not permit debris flow. The equipment drain sump shield has no such channels. The height and depth of the shields has been specified to ensure that debris will not enter the sumps in the long term.

CEB931N

_ = _ _ _ -.

)

j e

Increased Floor Area J

The floor area of the lower dqwell has been maximized to improve the potential for debris cooling. The lower drywell floor area of 88 m2 exceeds the ALWR Utility 2

Requirements Document criteria of 0.02 m /MWth.

Wetwell-Drywell Connecting Vents The flow area between the lower and upper drywell has been designed in a way to-allow adequate venting of gases generated in the lower drywell. The connecting vents 2

flow area is 11.25 m. This is important when considering the steam generation rates associated with fuel-coolant-interactions in the lower drywell. The interconnections between the lower dr>well and the wetwell is at elevation -4.55,11.7 m above the floor of the suppression pool. Thus, approximately 2.E6 kg of water must be added from outside the containment for the pool to overflow into the lower dqwell.

The path from the lower to the upper dnwell includes several 90 degree turns. This tortuous path enables core debris to be stripped prior to transport into the upper drywell minimizing the consequences from high pressure melt ejection. Also important when considering high pressure core melt scenarios, the configuration of the connecting vents will result in the transport of some core debris directly into the suppression pool. This is preferable to transport into the upper drywell and would result in the debris being quenched with only a slight increase in the suppression pool temperature.

Solid Vessel Skirt The vessel skirt in the AllWR does not have any penetrations which would allow the flow of water from the upper drywell directly to the lower drywell. This ensures a very low probability that water is in the lower drywell before the time of vessel failure.

Thus, large scale fuel-coolant interactions are precluded.

Inerted Contalmnent One of the important severe accident consequences is the generation of combustible gasses. Combustion of these gasses could increase the containment temperature and pressure. The AllWR containment will be operated inerted to minimize the impact from the generation of these gasses.

Containment Isolation The AllWR containment design has striven to minimize the number of penetrations.

This impacts the severe accident response due to a smaller probability of containment 1

isolation failure. All lines which originate in the reactor vessel or the containment have dual barrier protection which is generally obtained by redundant isolation valves. Lines which are considered non-essential in mitigating an accident isolate automatically in response to diverse isolation signals. Lines which may be useful in mitigating an accident have means to detect leakage or breaks and may be isolated should this occur.

I i

1 CE1193-17-5 i

1

t Upgraded Low Pressure Piping I

The low pressure piping in the ABWR has been upgraded to withstand higher pressure. This reduces the probability of an interfacing system L.OCA and the severe accident consequences associated with such an event.

Drywell-Wetwell Vacuum Breakers The ABWR contains eight 20-inch diameter vacuum breakers which provide positive position indication in the control room. Failure of the vacuum breakers to close as designed can potentially lead to increased source terms and early containment failure. The vacuum breakers have been located high in the wetwell to reduce potential loads occurring during pool swell. The analysis in the PRA assumes that the position switch which provides annunciation in the control room can sense a gap between the disk and the seating surface greater than 0.9 cm. Additionally, the vacuum breakers will be tested during periodic outages to ensure operability. The result of the vacuum breaker design in the ABWR is to reduce the potential for suppression pool bypass.

Residual IIcat Removal System The RHR system is the primary mechanism for the removal of decay heat from the containment. This system is capable of pumping saturated water up to the pressure of the COPS setpoint. Recovery of a single loop of RHR is adequate to remove decay heat in the long term. The RHR system also has a drywell spray functions which may be important in preventing high temperature failure of the containment in an accident in which debris is entrained to the upper drywell. The wetwell spray may be used to mitigate the effects of suppression pool bypass.

Overall Containment Performance The design of the ABWR containment provides for holdup and delay for fission product release should the containment integrity be challenged. The design basis containment leak rate is 0.5% per day at containment design pressure. Leakage is expected to be of this magnitude in a severe accident. Ixmg term containment pressurization is governed by the generation of decay heat and non-condensable g<ses. The primary source of non-condensable gas generation is metal-water reaction or the zirconium in the core. This is accommodated by a relatively large containment volume and a high containment pressure capability. The mitigating systems discussed above ensure that the decay energy results in steam production. The suppression pool absorbs this energy, resulting in very slow containment response which ensure ample time for fission product removal.

The containment strength was evaluated. The limiting structure is the drywell head.

Service Level C was found to be greater than 97 psig. This is adequate to withstand the generation of 100% metal water reaction. The median ultimate strength of the containment was found to be 134 psig. Ultimate strength capability is important for 4

very rapid containment challenges such as direct containment heating and rapid steam generation. Evaluation of both these phenomena indicate early containment failure from these mechanisms is unlikely.

CEB93-17-6 i

. ~ - -

4 Key Severe Accident Modeling Parameters Table 1 provides a list of key severe accident modeling parameters. This list has been derived from the discussions presented above and from 'a variety of ABWR severe accident evaluations.

t I

5 5

h I

l I

i l

CEB9317-7

.l

.\\

Table 1 Key Severe Accident Parameters Parameter Description Value Relates to What Feature?

Core Power

~

3926 MW Containment Performance El. of Top of Fuel 9.05 m Containment Performance Normal Water Level 13.26 m Containment Performance ADS Area 0.07m2 Vessel Depressurization Containment Leak Rate 0.5% per day Containment Performance Containment Senice Level C 97 psig Containment Performance Containment Ult. Strength 134 psig Containment Performance Total Zr in Core 72,550 kg Containment Performance Sacrificial Concrete Concrete Type llasaltic Basaltic Concrete Height of Layer 1.5 m Sacrificial Concrete Pedestal Thickness 1.7 m Pedestal Compartment Volume Lower Drywell 1860 m3 Containment Performance Upper Drywell 5490 m3 Containment Performance Wetwell 9585 m3 Containment Performance Floor Area Lower Drywell 88 m2 Lower Drywell Tolerance of Vacuum 0.9 cm Vacuum Breaker Ilreaker Position Switch Overflow Elevation LDW to Wetwell

-4.55 m I.ower Drywell UDW to Wetwell 7.35 m Lower Drywell LDW to UDW vent area 11.25 m2 Connecting Vents Lower Dgwell Flooder Elevation

-10.5 m Lower Drywell Flooder Area per valve

.0081 m2 Lower Drywell Flooder Plug Temperature 533 K Lower Drywell Flooder 6

Suppression Pool Mass 3.6 x 10 kg Containment Performance COPS Diameter of Disk 0.2 m (8")

COPS Diameter of Piping 0.36 m (14")

COIS Setpoint 0.72 MPa (90 psig)

COPS Tolerance at nom temp 5%

COPS EITect of temp on setpoint 2% per 100 F COPS Firewater Addition System injection Locations Vessel and Drywell Firewater Addition System Runout flow 0.055 m3/kg Firewater Addition System 3

Flow rate at 90 psig 0.044 m /kg Firewater Addition System Corium shield Height 0.4 m Corium shield Depth 0.4 m Corium shield Material Alumina Corium shield Channel I.cngth for 1m Corium shield Equipment Sump Shield CEIL 93-17-8

/hTMNkLb 23 A6100AS Standard Plant

/

REV.A

~

/

(4enetrations_ichgJrywell rather th L

19E.2.i\\

Sesere Acci ent Design

~

Feature Consideratiois failure.)JTo corppare the consequences of severe g

accidents resul/ng in fission product releases via Although thafrequency of core damage is very low drywell head failure to those with releases through the in the ABWS design, features were added to the design COPS, MAAP%as used to simulate a series of severe to ensurrfs robust response of the containment to a accident sequences for both release mechanisms. These seve[ accident. This section discusses the important severe accident sequences are described m Section

?.9E.2.2. Failure pressure of the drywell head was cysiderations for the severe accident design features.

assumed to be equal to its median ultimate strength, s

1.025 MPa (134 psig). The results of these runs show

-19 Edit;t-Containment Oserpressure releases of volatile fission products, after 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />, for Protecuon 4 stem the COPS cases to be several orders of magnitude less than for the corresponding drywell head failure cases. J ABWR has a sery low core damage frequency-iFC~si release fractions _ are compared in Tabl)

S Furthermore, in the unlikely event of an accident 9E.2-25.jMost accident sequences show this large resulting in core damage, the fission products are difference in releases between drywell head failure and typically trapped m, the containment and there is no COPS cases.

release to the environment. Nonetheless, in order to mitigate the consequences of a severe accident which M.K1T Pressure Setpoint results in the release of fission products and to hmit the tietermination effects of uncertainties in severe accident phenomena, Educ acudd ABWR is equipped with a Containment Overpressure Protection System (COPS). This system is intended to

. Sem 1 factor 7%m considered m. determmmg the provide protection against the rare sequences in which pumum pmssu

, for the rupture disk. The structural integrity of the containment is challengedp results of the prw,e set eanalysis show that it is desirable to avoid drywell head failure. This can be assured by oyrrassurizationfit-hs been determined that tiicse

" rare sequ' enc'es comprise only lto be provided later)))

providing a rupture disk pressure setpoint below the of the hypothesized severe accident sequences.

pressure that would begin to challenge the structural v

integrity of the containment. However, as the pressure f The COPS 6 art of the mmospheric contro

%m uce c sne to comainment Mm and fission product release is also reduced. Thus, the ste m _ a_

consists of 8-inch diameter overpressure relief rupture disk /inounteddn (cries 6 setpoint of the rupture disk must optimize these competing factors: minimizing the probability of 14-inch line w hich connects the wetwell airspace to the drywell head failure while maximizing time before stack. The COPS provides a fission product release fission product release to the environment.

point at a time prior to containment structural failure.

Thus, the containment structure will not fail. By engineering the release point in the wetwell airspace,

' e service level C capability of the containment the escaping fission products are forced through the serves as one indication of the structural integrity of the suppression pool, in a core damage event initiated by a c ntainment. As shown in Appendix 19F, the senice

)

transient in which the vessel does not fail, fissiori level C f r the ABWR is 97 psig, hmited by the products are directed to the suppression pool via the drywell head. Thus, it is desirable to set the rupture SRVs, scrubbing any potential release. In a severe M setpoint Mow Ws value.

j accident with core damage and vessel failure or in a LOCA w hich leads to core damage, the fission products The distribution of drywell head failure pressure wdl be directed from the vessel and drywell through the nd the distribution of rupture disk burst pressure were drywell connecting vents and into the suppression pool also considered m determmmg the burst pressure. As again insunng any release is scrubbed. Eventually,if stated in Attachment A to Appendix 19F, the drywell the containment pressure cannot be controlled, the head failure pressure is assumed to have a lognormal rupture disk opens. Any fission product release to the distnbution with a median failure pressure equal to its environtnent is greatly reducc4 by the scrubbing ultimate strength of 1.025 MPa (134 psig). The provided by the suppression pool.

variability of rupture disk opening pressures is best modeled with a normal or Gaussian distribution.

In the absence of the COPS, unmitigated Typical high quality rupture disks exhibit a tolerance of overpressurization of the containment will result in 59 of the mean openmg pressure. Tests have shown that this 5% tolerance spans 2 to 12.5 standard failure of the drywell head for mostjescre accident R scenariog[5cnie fiiih-pressure core melt sequences deviations of the rupture disk population. Tiu,s analysis 1

(resuit in fission product leakage through the moveable f the Containment Overpressure Protection System

^"

• U 19E.2-42.15

MMIn%d(3

'2-ABWR m ouis Stan lard Plant REY.A conservatively assumes that only 2 standard deviations are included within the 15% tolerance.

The clapsed time to rupture disk opening was A critical parameter in determining the risk of within 0.8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> of the base case value of 20.2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> drywell head failure before rupture disk opening is the for both cases tested. Higher rupture. disk temperatures pressure difference between the drywell and wetwell.

(i.e. lower pressure setpoints) reduce the time to rupture L. ate in an accident the drywell is at higher pressure disk opening and lower rupture disk temperatures (i.e.

than the wetwell. For a given rupture disk setpoint, the higher pressure setpoints) increase the time to rupture probability of drywell head failure increases as the disk opening. There were no significant changes in pressure difference increases. The maximum drywell to fission product release. For both cases the Csl release wetwell pressure difference is 0.1 MPa (14 psi). This fraction at 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> remainsi less than lE-7.

pressure difference occurs for cases in which firewater spray was activated after vessel failure but terminated Another parameter affected by the variation in the before containment failure. Cases without firewater rupture disk temperature is the probability of drywell spray have pressure differences of no more than 0.05 head failure prior to rupture disk opening in a severe MPa (7 psi).

accident. Using the rupture disk and drywell head failure distributions, it was determined that the probability of A rupture disk setpoint of 0.72 MPa (90 psig) at drywell head failure prior to rupture disk opening 366 K (200 F) was chosen. The residual risk of drywell increased from about 2% for the base case to about 3%

head failure may be calculated by combining the two for the case with the rupture disk temperature of 311 K distributions with an offset corresponding to the (100 F). With a rupture disk temperature of 422 K pressure difference between the wetwell and the drywell.

(300 F), the probability decreased to about 1.5%. The A 90 psig setpoint results in a 5% probability of rupture disk temperature variation has a similar effect drywell head failure prior to rupture disk opening for a on the severe accident sequences in which 'he firewater 0.1 MPa (14 psi) drywell to wetwell pressure spray system is activated. The probability of drywell difference. For a drywell to wetwell pressure difference head failure prior to rupture disk opening increases from of 0 05 MPa (7 psi), the drywell head failure about 5% for the base case to about 6.5% for the case probability prior to supture disk opening is 2%. This with the rupture disk temperature of 311 K (100 F) and is judged to be an acceptable level of risk.

decreases to about 4% for the case with the rupture disk temperature of 422 K (300 F).

@Fd.fhM Variability in Rupture Disk Setpoint The results of this sensitivity study show that variations in rupture disk temperature, which cause Nickel was chosen as the material for the rupture small variations in rupture disk opening pressure, have disk for evaluation purposes due to its relative a minor effect on the performance of the ABWR insensitivity to changes in temperature. At Containment Overpressure Protection System.

temperatures above room temperature the opening pressure of a typical nickel rupture disk will decrease by 49E2,%3 Sizing of Rupture Disk about 2% for a 56 K (100 F) increase in temperature.

Thus, in order to estimate the uncertainty due to The size of the rupture disk has also been vanations in the temperature of the ABWR rupture optimized. If the rupture disk is too small, it could be disk, a sensitivity study was performed in which the incapable of venting enough steam to prevent further pressure serpoint of the rupture disk was varie containment pressurization. On the other hand, if the O

rupture disk is too large, level swell in the suppression The nominal pressure setpoi ' f the rupture disk pool could introduce water into the COPS piping. If is 0.72 MPa (90 psi qLX/ - (200 F). Two cases this were to occur, the piping could be damaged or there were examined um MMLrn this sensit,ivAiy.

For both cases thcT/2hP-PF-R sequenceWas used a]scould be carryover of waterborne fission products f 8

the containment.

the base case. First, the rupture disk pressure setpoint was reduced to 0.708 MPa (88 psig) which corresponds An eight. inch rupture disk was selected. This is to a rupture disk temperature of 422 K (300 F); and, sufficient to allow 35 kg/sec of steam flow at the second, the pressure setpoint was increased to 0.735 opening pressure of 90 psig (0.72 MPa-a) and MPa (92 psig) which corresponds to a temperature of corresponds to a energy flow of about 2.4% rated 311 K (100 F). This temperature range, from 311 to power. For virtually all severe accident sequences, the 422 K (100 to 300 F), bounds all anticipated rupture rupture disk would not be called upon until about 20 disk temperatycs_

hours after scram. The decay heat level at this time is tot O LN assare

}MS UC coh/ OM f p we. b

)

o MM j

Amendment 77 19E.2-42.16 i

A4 tac %p y 3

j ABWR 2 M 00AS Standard Plant REV.A less than 0.5E Thus, there is ample margin in the sizing of the rupture disk for severe accidents.

The potential for increased risk due to the rupture disk opening early has been considered. It is assumed An additional accident was considered in the that recovery of RHR capability is sufficient to selection of the rupture disk size. In the event of an terminate containment pressurization and prevent ATWS with the additional failure of the standby liquid drywell head failure. In the 3.9 hours1.041667e-4 days <br />0.0025 hours <br />1.488095e-5 weeks <br />3.4245e-6 months <br /> between rupture control system, the operator is directed to lower water disk opening and hypothetical drywell head failure for level to control power. Analysis has shown that the the LCLP-FS sequence, the probability of recovering RHR system is capable of removing the energy RHR capability is only 4% (see Subsection 19.3.2.7).

generated by the ATWS from the containment This represents the probability that the COPS was (Subsection 19.3.1.3.1). If the additional failure of opened unnecessarily since RHR would have been containment heat removal is assumed, a simple recovered in this time period.

calculation indicates that an the rupture disk area is just safficient to limit the containment pressure below For cases with passive flooder operation, the service level C.

fission product release occurs about 6 to 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> sooner than it would have if the drywell head was allowed to Calculations were also performed to investigate the pressurize to 1.025 MPa (134 psig). For the range of potential cffccts of pool swell and fission product severe accident sequences described in Section 19E2.2, carryover at the time of COPS operation. These the probability of RHR recovery in a similarly defined analyses (Subsection I?E.2.3.5) indicate that pool time window is about I11 swell does not threaten the integrity of the COPS piping and that no significant entrainment of fission For both cases, there is a small probability that products will occur due to carryover.

RHR will be recovered before the time at which containment would fail if the rupture disk setpoint has

<-19th2 &lt Comparison of ABWH been surpassed. In light of this fact and given the Performance With and Without COPS difference in magnitude of the fission product release, it waa accM is clearly preferable to direct the fission products The results of the MAAf* calculations for the through the rupture disk.

various accident scenarios were investigated in Section1 19E.2.2 and he reicases arc summarired-in 49&2&id Suppression Pool Bypass

.L ble l E omparisons of Csl release fraction at 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> show large differences between the COPS and A comparison of performance for cases i drywell head failure cases. Csl release fraction at 72 suppression pool bypass flow through an o.

vacuum g--

hours for drywell head failures is on the order of 0.1%

breaker valve was also considered. Case *ere run with o, sM to 15E For all cases with release via the COPS, bypass effective area varying from to 2030 c d 4 b.A h MAAP_ predicts release fractions of less than lEh 60054-10 2.19 nbA4ully oppn vacuum breaker had a F19E.2-26 summarizes several critical parameters ) a+ffectie-arn.of-203OrmDThe dominant the Loss the dominant low pressure core melt scenario. j of All Core Coolant with Vessel Failure at Low Pressure sequence was considered with Passive Flooder i

There is, of course, some reduction in the elapsed Operation since previous analysis has shown that the time to fission product release for the COPS cases firewater system is capable of mitigating bypass.

when compared to the drywell head failure cases. For i

the dominant accident sequences in which the operator No credit was taken for aerosol plugging of the initiates the firewater spray system prior to bypass leakage in this analysis; and, therefore, the overpressurization, the time difference between rupture results are conservative. Also, it was assumed that the i

disk opening and drywell head failure is only 3 to 4 bypass leakage was present from the beginning of the hours. A typical example is the Loss of All Core accident sequence. As the bypass area increases, the l

Coolant with Vessel Failure at Low Pressure with fraction of fission product aerosols which pass through Firewater Spray addition sequence (LCLP-FS), as the suppression pool decreases. Thus, the benefit of a described in Subsection 19E.2.2.1. For this sequence wetwell release of fission products is significantly the wetw cll pressure will reach 0.72 MPa (90 psig) and reduced as the bypass area increases.

the rupture disk will open at 31.1 hours1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />. Without the rupture disk, the drywell will reach 1.025 MPa (134 For bypass effective areas less than 50 cm (.054 2

psig) at 35.0 hour0 days <br />0 hours <br />0 weeks <br />0 months <br />s-2 ft ), Csl releases at 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> from the COPS cases were smaller than for the corresponding drywell head failure cases. However, the differences in Csl releases at Amendment ??

19E.2-42.17

AkkAQ V

4 ABWR u o mnAs Standard Plant REV.A i\\

72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> were only factors of 2 to 4 rather than several init'iated by the triggering a fusible plug at the line exit orders of magnitude. The time difference between (LD\\ side). Since four inch diameter fusible disks may be drywell head failure and rupture disk opening was 4 to 8 cominercially available, the flooder line diameter was hours for these small bypass areas. For bypass effective chose'n as four inches.

2 2 (.054 ft ), Csl release

\\

areas greater than 50 cm fractions at 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> are on the order of 10% for both Th(minimum acceptable flow rate for the floode the drywell head failure cases and the COPS cases. O -

system corresponds to the flow rate which can just the other hand, the time difference between rupture disk absorb the, heat generated in the debris bed. Minimum opening and drywell head failure is only 2 to 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> acceptable \\ flow is calculated in Section 19E.2.8J.2.

for these larger bypass areas. These relatively small The expected How rate in the flooder system catt be time differences will not significantly affect the obtained by hpplying Bernoullfs equation to the lldcxler X&lO geometry. This calculation is presented in etion o.M @pgnitude of the offsite dose. StibnGunubntitteditmet1992fhas a complete discussion of 19E.2.8.2.3. \\

- m Ag%

pr suppression pool bypass flow through vacuum breaker

\\

v tves.

19 E.2.8.2.2 Minimum Acceptable flow

/ Rate s

1

.2.8.1.6 Summary

/

Heat is Fenerated in the debris d by fission s

A product decay and zirconium oxidatio/L Any flooder for the(wetwell pressure setpoint of 0.72 MPa (90 psig qverpressure relief rupture disk meets the stated flow in excess of\\the amount reqdred to remove design goal. The 5.1% maximum probability /of generated heat will participate in qudiching the debris containment failure in a severe accident combined pith and establishing a water pool abov 4.he debris bed. As the alread[ low core damage frequency produces an shown in Attachment (9EC, the ti required to quench extremely low probability of significant fission product the debris is not a entical para eter in determining release. In add,ition, the clapsed time to rupture disk containment performa'nce. The efore, the minimum opening is greater than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> for mgit severe acceptable flow rate for tbe lowe drywell flooder sy~ tem accident sequenedt

/

is the rate which will cbmpi cly absorb all the heat 6

generated in the debris bed 19 E.2.8.2 Ilower Drywell Flooder

\\

/

The decay heat gener -ion rate at the time when 19 E.2.8.2.1 Introduction /

debris is expected to first plit'er the lower drywell during

\\

/

credible accident scenari is approximately one percent This section providds the basef for sizing the lower of rated power (39 MW. Thirty-nine megawatts can be drywell ikxxter system. The system is described in detail used as a first app ximatikm of the decay heat i

in Section 9.5.12 of the ABWR.$SAR.

generation rate of th debris be in the lower drywell.

/

This assumption is aighly con. rvative because tt.e 1

The lower drywell flocker provides an alternate entire core mass w)fl never completely relocate into the source of water to the lowti drywell once it contains lower drywell. Furthermore.

• oble' passes and volatiles will escape fro / he molten debris, decay heat as )ciated with these\\ carrying away the core debris. The primary watet source is the firewater t

addition system. Waterdresenkin the lower drywell stwo constituents i

only cools the core debris and establishes a water pool (approximatel : 20 percent of total).

above the debris. Wate'r absorbs heht by first heating up I

to saturation conditirins and then boiling away. Debris Heat can also be generated in the tA by exothermic cooling requires thaf the water absortithe heat generated reactions the debris constituents. Theinost energetic in the debris bed and the latent and senAible heat released reactions ' volve oxidation of zirconium by water vapor by the debris as'its temperature decreases. Quenching and car n-dioxide. The only source ol; significant prevents or mifigates core concrete interaction (CCl),

amount of oxidizing agents is the concreteyeneath the 1

An overlyingwater pool scrubs fission pr'oducts which debris

. The water above the bed will notyontribute may be released from the d bris bed.

\\

signi cantly to oxidation because the surface af the bed

\\

will/orm a crust which will quickly be depleted of The/flooder system is comprised of teq piping zir90nium. NUREG-5565 indicates that a t 'pical lines. Each line ori;inates in one of the ten ' vertical a ation rate for concrete is two inches per hou The pipey'which are part of the drywell to whtwell neration rate, assuming that the H O and 02 contiecting vent system. The vents are arra' ged F, leased during ablation completely react 2

n with sydimetrically around the perimeter of the lower irconium, is 3.6 MW. Combining these two sourc s j

Jrywell. The flow through each flooder line will he

/

Amemhnent ??

192.1 j

l

hch&- f M I5*#

- hdrop bo% cM h Hydrogen burning and detonation are not a concern for the ABWR containment because the containment is inerted with hydrogen. There could be a potential for burning in the COPS system and the stack after the rupture disk opens. However, due to the design and operation of the COPS system, this issue does not have an impact on risk.

Hydrogen burning and detonation will be precluded in the piping associated '

with the COPS system. The piping will be inerted during operation with rupture disk located at the inlet of the stack. This, combined with initial purging of the piping, will ensure that the inertion of the containment will extend out to the stack, and prevent burning of hydrogen in the portion of the COPS system which is within the reactor building. Therefore, there will be cio concern of the leading edge of the containment atmosphere mixing with the gas in the piping and causing a burn. After passing of the leading edge of the gas flow, the mixture in the piping will be identical to that in the i

containment. The gas flow through the system will prevent the backflow of air into the COPS piping.

Hydrogen burning could occur in the plant stack as the gas flow enters the stack. The stack is a non-seismic structure located on top of the reactor i

building. Because of this configuration, the reactor building has been designed to withstand the loads associated with the collapse of the plant stack.

Furthermore, no credit is taken in the analysis for the plant stack to reduce the offsite dose by providing for an elevated release..All releases were presumed to occur at the elevation of the top of the reactor building. Therefore, hydrogen burning or detonation in the stack will have no impact on the consequences of a severe accident as modeled in this analysis.

No burning will occur within the COPS piping. Furthermore, no credit was i

taken for the plant stack to reduce the source term to the environment and the reactor building can withstand the collapse of the plant stack. Therefore, hydrogen burn or detonation in the COPS system will have no impact on risk and no further consideration of this phenomenon is required.

GB93-1410 l

A "M

ABWR

)

m iooAs Standard Plant ggy 4 N

i fractio at'72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> are on the order of 104 for both[ reduced as a result of the COP.S 'mplementation into the design. ih asaHs md d yta p ed.l.

t..e drywell head failure cases and the COPS cases:Ord cf dp mdm h reducu! h as mg%g the other hand, the time difference betseen rupture disk ct openmg and drywell head failuie:is only 2 to.1 hours1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />

-19E/ 8J 1.ower Drywett:Flooder for these larger bypass areasTThese"relauvely small tt y eref.Rur, time differences will-not significantly affect the 19 E.2.8.2.1 Introduction magnitude of be bifsite dose. Attachment 19EE has a K

I complete diicussion of suppression pool bypass flow This section provides the bases for sizing the lower through vacuum breaker valves-drywell flooder system. De system is desenbed in detail in Section 9 5112 of the ABWR SSAR.

19 E.2.8.1.6 Summary

\\

J The lower drywell flooder provides an alternate!

A wetwell pressure setpoint of 0.72 MPa (90 psig) source of water t6 the lower drywell once it contains' for the overpressure rehef rupture disk meets the design core debris. The pnmary water source is the firewater goal. The probability of containment structural failure addition system. Wher present in the lower drywell is minimized while maximizing the ume to fission cools the core debris and establishes a water pool ab,dve product release in a severe accident. The 5.lc the debris. Water abso'rbs heat by first heating up to maximum probability of containment structural failure saturation conditions arid then boiling away. Debris if the pressure reaches the rupture disk setpoint in a cooling requires that the wher absorb the heat geryerated severe accident, combined with the already low core in the debris bed and the late'n,t and sensible heatfeleased damage frequency and reliable containment heat by the debris as its temperature decreases. Qdenchir g removal, produces an extremely low probability of prevents or mitigates core concrete interaction (CCl),

significant fission product release. In addition, the An overlying water pool also ' scrubs iission products elapsed time to rupture disk opentag is greater than 24 which may be released from the dhbns bed. /

hours for most severe accident sequences.

\\

/

t j he w o M ad, ne flooder system is comprised $f ten piping The net risk reduction associated with the lines. Each line originates in one o the ten vertical f pipes which are part of the dry (s are I

implementatiorf of the COPS system in the design of well to wetwell connecting vent system. The vent the ABWR$s-summanzed4rt-Table-19E.2Pand Figure 49E.2-22f All sequences which would result in symmetrically around the perimeter \\of the lower COPS operation were assumed to lead to failurs of the drywell. The flow through each floode'r line will be s

drywell head. This may slightly overpredict the initiated by triggering a fusible / plug at the line exit probability of drywell head failure since there will be (lower drywell side). Since four' inch dianieter fusible somewhat more time available for the recovery of disks may be commercially available, the ' ooder line Aq as contammeg heat removal if the COPS system were diameter was chosen as fourinches.

d AM

/not present. Table 49E.2 26 indicatesdiIow probability

/

i of RHR recovery in the interval between the time of De teflon disk esides'between the staint ss steel (g'y~ J COPS initiation and the time of drywell head failure if disk and the fusible plug in the flooder valke. Its

.l M h

/

COPS were not present. For the case with firewater purpose is to insulate /the fusible plug fro'tn the

}

addition to the containment, the probability of RHR relatively cold suppression pool water. If insulauon recovery dunng the period of interest is 4E Therefore, was not provided, melting of the plug might n6t be i

uniform and operatio' of the flooder valve mighl be no significant error is introduced into the calculatio n.

n

% 34ctdq impaired. The disk will not melt or stick in the va vc

. Table:19Ei,27 indicates that the probability (J because teflon has a softening temperature f drywell head failure increases by a factor 50 for approximately 400*C and a maximum continuous sequences with core damage (Classes I and III) if the operating temperature of 288*C both of which l

l COPS system is not present. For Class 11 sequences, above the plug melting temperature of 260 C the loss of containment heat removal may lead to core Furthermore/ teflon he high chemical resistance and damage for those sequences which have drywell head will not adhere to the stainless steel plug nor the failure. Since the probability of drywell head failure fusible pi, g.

u increases by a factor of 100 without the COPS system, N core damage probability associated with Class !!

ge/m nimum acceptable flow rate for the flooder events also increases by a factor of 100. Figure system corresponds to the flow rate which can just 19E.2 22 shows the probability of exceedence versus absorb the heat generated in the debns bed. Minimum whole body dose at 1/2 mile for the ABWR and for the ace'eptable flow is calculated in Section 19E.2.8.2.2.

ABWR without the COPS system. The offsite dose is ne expected flow rate in the flooder system can be l

19E.2 4210 g

y GC WRLd b

.(

/

ABWR oesign Documsat 2.14.6 Atmospheric Control System Design Description The Atmospheric Control (AC) System consists of a nitrogen supply, injection lines, exhaust lines, bleed line, valves, controls, and instrumentation. The AC System also has the containment overpressure protection system. Figure 2.14.6

_y

(

sho s the basic system con 6guration and scope.

The AC System provides an inert atmosphere within the primary containment during plant operation.

Except for the primary containment penetrations isolation valves, and suppression pool level sensors, the AC System is classined as non-safety-related.

The AC primary containment penetrations, isolation valves, and suppression pool level sensors are classined as Seismic Category I. Figure 2.14.6 shows the ASME Code class for the AC System piping and components.

AC System components are located in the Reactor Building, except for the nitrogen supply.

Figure 2.14.6 shows the Class 1E divisional power assignments for the AC System

(

components. In the AC System, the independence is provided between the Class lE divisions, and also between the Class IE divisions and non-Class 1E equipment.

The main control room has control and open/close status indication for the containment isolation valves.

AC System components with display interfaces with the Remote Shutdown System (RSS) are shown on the Figure 2.14.6.

The safety-related electrical equipment located in the Reactor Building is quali6ed for a harsh environment.

The two valves in the containment overpressure protection system fail open on loss of pneumatic pressure or loss of electrical power to the valve actuating 06 %

solenoid. The other pneumatic valves shown on Figure 2.14.6 fail close on loss of pneumatic pressure or loss of electrical power to the valve actuating solenoids.

Inspections, Tests, Analyses and Acceptance Criteria Table 2.14.6 provides a dennition of the inspections, tests and/or analyses,

(

together with associated criteria, which will be undertaken for the AC System.

L]J 5/21/93 2.14.6

O O

O

.~

50 w<$

TO PLANT STACK PLANT HPIN STACK JLNNS AC NNS g TO REACTOR BLDG FROM HVAC EXHAUST I

NIT ROGEN TO SGTS SUPPLY t

4k AN x-,

~

HVAQ

~

2 P

l 2lNNS g

2q l

i_ _ _ _ _ _T g

q

- - - - - - - - - H J ( l-> TO SGTS l

[

l 21SGTS Ily i

l 6

l g-

</ / /// / / //// / / / / / / / /,

1 I

I l6 6

5 l

r- - T ~' I~ T ~ ~ ~5 9

l L

_j DAYWELL 6--

,d)(b T H >#F-b 3

FROM

/

l I

N

/

NITROGEN RUPTURE NNSl l

SUPPLY

/

DISK

^2 l

'""'j

"'"j l

q

/

BLEED LINE VALVE l

6 j

j i

6 l

6 h

l l

/

/

/

l I

p-2 --1 ; r t

--I >f M- $" # "&^c ' "

l l_ _ _ _; j r ;__

I l

I j

HVAC DIVISION I - R 2

NNS

/

/

/

/

l l DIV. til '

M/

/

r DIVISION 11 - R G/////u/n//////////

3 DIE IV CONTAIN N

OVERPRESSURE SUPPRESSION PROTECTION POOL SYSTEM NOTES:

1. INBOARD CONTAINMENT ISOLATION VALVES ARE POWERED FROM CLASS 1E DIVISION 11

/

OUTDOARD CONTAINMENT ! SOLATION VALVES ARE POWERED FROM CLASS 1E DIVISION 1 EXCEPT AS NOTED WITH "" WHICH IS POWERED FROM CLASS 1E DIVISION lil

{

Figure 2.14.6 Atmospheric Control System m-

f i

O l

\\

l Table 2.14.6 Atmospheric Control System

3 Inspections, Tests, Analyses and Acceptance Criteria Design Commitment inspections, Tests, Analyses Acceptance Criteria l

1.

The basic configuration of the AC System 1.

Inspections of the as-built AC System will 1.

The as-built AC System conforms with the 7-l is as shown on Figure 2.14.6.

be conducted.

basic configuration shown on l

Figure 2.14.6.

i 2.

The ASME Code components of the AC 2.

A pressure test will be conducted on those 2.

The results of the pressure test of the System retain their pressure boundary code components of the AC System ASME Code components of the AC System integrity under internal pressures that will required to be pressure tested by the conform with the requirements in ASME be experienced during service.

ASME Code.

Code Section Ill.

3.

In the AC System, independence is 3a. Tests will be performed in the AC System 3a. The test signal exists only in the Class 1E provided between Class 1E divisions, and by providing a test signal in only Class 1E division under test in the AC System.

h between Class 1E divisions and non-Class division at a time.

-t-1E equipment.

3b. Inspection of the as-installed Class 1E 3b. In the AC System physical separation f

divisions in the AC System will be exists between Class 1E divisions. Physical

$~

performed.

separation exists between these Class 1E divisions and non-Class 1E equipment.

g 4.

Main control room displays and controls 4.

Inspections will be performed on the main 4.

Displays and controls exist or can be provided for the AC System are as defined control room displays and controls for the retrieved in the main control room as p

in Section 2.14.6.

AC System.

defined in Section 2.14.6.

s 5.

RSS displays provided for the AC System 5.

Inspections will be performed on the RSS 5.

Displays exist on the RSS as defined in d

are as defined in Section 2.14.6.

displays for the AC System.

Section 2.14.6.

Q 6.

The two valves in the containment 6.

Tests will be conducted on the as-built AC 6.

The two valves in the containment overpressure protection system fail open System pneumatic valves.

overpressure protection system fail open 4

on loss of pneumatic pressure or loss of on loss of pneumatic pressure or loss of electrical power to the valve actuating electrical power to the valve actuating solenoid. The other pneumatic valves solenoid. The other pneumatic valves shown on Figure 2.14.6 fail close on loss of shown on Figure 2.14.6 fail close on loss of pneumatic pressure or loss of electrical pneumatic pressure or loss of electrical power to the valve actuating solenoids.

power to the valve actuating solenoids.

P 2

in

=me-

"'i r-

-W