ML20237A720
| ML20237A720 | |
| Person / Time | |
|---|---|
| Site: | Catawba  |
| Issue date: | 12/07/1987 |
| From: | Tucker H DUKE POWER CO. |
| To: | NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM) |
| References | |
| NUDOCS 8712150194 | |
| Download: ML20237A720 (16) | |
Text
-
e*
i Duxn POWEIt GOMPANY P.o. DOx 331f10 CIIAllLOTrE, N.O. 2fl242 1
IIAL IL TUCKER resanown vsos rue.einsent (704) 373-4M1 wims. man paoamenom j
December 7, 1987 U. S. Nuclear Regulatory Commission Attention: Document Control Desk l
Washington, D. C. 20555 l
Subject:
Catawba Nuclear Station Docket Nos. 50-413 and 50-414 l
Steam Generator Tube Rupture Analysis
Dear Sir:
License Conditions 16 and 10 of Facility Operating Licenses NpF-35 and 57, respectively, require that Duke power Company submit steam generator tube rupture analyses and propose any necessary changes to the Technical Specifications.
This it, sue was pursued generically by the Westinghouse Owners Group Steam Generatcr Tube Rupture (SGTR) Subgroup. On March 30, 1987 the NRC Staff issued a Safety Evaluation Report accepting the Subgroup's analysis methodology documented in WCAp-10698, "SGTR Analysis Methodology to Determine the Margin to Steam Generator Overfill", December 1984.
I Section D of Enclosure 1 of the Staff's SER required additional plant specific input for each utility referencing WCAp-10698.
The five items are addressed in l
Attachments 1 through 5.
I On December 4, 1987 Duke submitted a proposed Technical Specification change that would add limiting conditions for operation and surveillance requirements for the steam generator power operated relief valves.
No other changes to the l
Technical Specifications were identified as a result of the SGTR analysis.
In accordance with 10 CFR 170.21, a check for $150.00 is enclosed.
Very truly yours,
' g/c Hal B. Tucker
\\
ROS/1060/sbn 8712150194 071207 PDR ADOCK 05000413 i
1 Attachments P
PDR i
fC,CY LdhKIL NS'
r U. S. Nuclear Regulatory Commission December 7, 1987 Page Two xc:
Dr. J. Nelson Grace, Regional Administration U. S. Nuclear Regulatory Commissior.
Region II 101 Marietta Street, NW, Suite 2900 Atlanta, Georgia 30323 Mr. P. K. Van Doorn NRC Resident Inspector Catawba Nuclear Station Dr. K. Jabbour Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington, D. C.
20555
Catawba Nuclear Station Training programs for SGTR Item (1)
Each utility in the SGTR subgroup must confirm that they have in place simulators and training programs which provide the required assurance that the necessary actions and times can be taken consistent with those assumed for the WCAp-10698 design basis analysis. Demonstration runs should be performed to show that the accident can be mitigated within a period of time compatible with overfill prevention, using design basis assumptions regarding available equipment, and to demonstrate that the operator action times assumed in the analysis are realistic.
Response
Steam generator tube rupture is one of the many abnormal and accident scenarios that are covered during the simulator portion of the operator training program.
Currently this training is conducted on the McGuire/ Catawba simulator located at Duke's Technical Training Center near the McGuire Nuclear Station. Studies of operator action times on this simulator in late 1983 indicated that the times were consistent with analysis assumptions.
Duke recently accepted delivery of a' plant-specific simulator at the new Catawba Training Center. The simulator should be ready for use in operator training in early 1988.
The simulator and training programs will be reviewed against the plant specific analyses which are discussed in response to Items (2) and (5) below. Demonstration runs will be performed #n show that the SGTR accident can be mitigated within a period of time compatible with overfill prevention, using design basis assumptions regarding available equipment, and to demonstrate that the operator action times assumed in the analysis are realistic.
Results of this effort will be provided to the NRC by June 1, 1988.
i i
i
\\
l
7
)
Site Specific Offsite Dose Calculation The thermal hydraulic analysis of the steam generator tube rupture event was performed to determine the inputs necessary for the site specific offsite dose calculation. The following summary describes the computer code used, the code model, the analysis assumptions, and the results of the dose calculation.
COMPUTER CODE j
i The analysis was performed with the RETRAN-02 computer code (Reference 1).
The j
code version used differs from that reviewed by the NRC in two respects, cor-rected errors and the addition of the multiple state control rod model.
This model was not used in this analysis.
Therefore, the version used actually differs from the reviewed version only due to the correction of known errors in the reviewed version. The NRC SER on RETRAN-02 (Reference 1) listed general limitations of the code as well as further limitations for PRR analyses. These J
limitations are individually addressed as they apply to this analysis.
a.
Multidimensional neutronic space time effects are not important for steam generator tube rupture.
]
b.
This analysis is not started from suberitical or with zero fission power.
Boron transport is not important for steam generator tube rupture, c.
and therefore no attempt is made to model it.
s d.
The point kinetics model is used with a conservative treatment of control rod reactivity insertion as discussed in Reference 2.
e.
Metal-water reaction is not important for steam generator tube rup tur e.
f.
Equilibrium thermodynamic modeling of regions outside the pres-surizer has been shown to be sufficiently accurate by comparison to plant data in Reference 2.
Equilibrium treatment of the steam generator secondary side with the injection of subcooled auxiliary feedwater flow is conservative since it results in a lower secondary pressure and therefore increased primary-to-secondary leakage.
g.
One dimensional thermal hydraulics are sufficient to simulate a steam generator tube rupture.
hl.
A BWR limitation not applicable to this analysis, h2.
The drift flux correlation is not used in this analysis,
- i. Refer to n3.
Page 2
- j. The optional gap linear thermal expansion model is not used in this analysis.
k.
Air is used only in the separated cold leg accumulator volumes, which do not inject and therefore do not af fect this analysis.
1.
This analysis is completely within the suberitical region for
- water, Single phase heat transfer or two phase pre-CHF heat transfer m.
dominate in this analysis, nl.
The Beattie two-phase multiplier option is not used in this analysis. Use of the dynamic slip option was not restricted to the Bennett data base conditions. However, the results using dynamic slip with similar nodalizations in the past have been more physically reasonable than those obtained with an HEM model, n2.
The algebraic slip option is not used in this analysis.
n3.
Refer to nl.
o.
The pressurizer empties but does not fill in this analysis.
Com-parisons to plant data for an empty pressurizer is presented in Reference 2.
J p.
The nonmechanistic separator model is not used in the analysis, q.
As discussed in Reference 2, the homologous pump curves used in this analysis are a composite from three different sources.
The pump model performance using these curves has been compared with plant data in Reference 2.
r.
A BWR limitation not applicable to this analysis.
s.
The turbine model is not used in this analysis.
t.
The subcooled void model is not used in this analysis.
u.
The bubble rise model is used in this analysis as described in Reference 2.
Comparisons of the model performance va. plant data are also presented in Reference 2.
v.
This analysis uses the temperature transport delay model as described in Reference 2.
The dominant flow direction in the loops is forward due to forced circulation prior to reactor trip and natural circulation after the loss of offsite power.
Page 3 w.
The standalone auxiliary DNBR model is not used in this analysis.
x.
In general, enthalpy transport is used only in homogeneous volumes.
In the lone exception, enthalpy transport is turned off before significant separation occurs.
Enthalpy transport was used in multidirectional, multifunction volumes as discussed in Reference 2.
y.
The local conditions heat transfer model is not used in this analysis.
z.
The heat transfer surface area adjustment and the feedwater inlet enthalpy bias were reviewed for this analysis.
These adjustments were less than 0.1% and 0.1 Btu /lbm respectively, i) Two phase flow was not encountered in the primary loop.
Results of the secondary side transient analysis were reviewed to confirm that they were physically reasonable,
- 11) Refer to o.
PLANT MODEL AND INITIAL CONDITIONS The analysis used the Catawba Unit 2 RETRAN Model as described in Reference 2.
Unit 2 was selected based on its similarity to the post reactor trip level behavior of the reference plant. This behavior is important since it results in later identification of the ruptured steam generator and therefore a longer duration for pirmary-to-secondary leakage. The base plant model was reinitialized to appropriate conditions for this analysis:
1) core power: nominal + 2%
- 2) average RCS temperature: nominal + 4 F
- 3) random SG tube plugging:
10% (random)
- 4) RCS flow:
nominal thermal design flow
- 5) pressurizer level: nominal + 9%
- 6) pressurizer pressure: nominal + 30 psi
- 7) narrow range steam generator level: nominal - 9%
The first, second, third, and fourth changes were made to maximize heat input and time required for cooldown. The fifth and sixth changes were made to maximize time to reactor trip and therefore maximize the amount of primary-to-secondary flow occurring at a high pressure difference.
The seventh change was made to minimize initial steam generator secondary (low activity) inventory.
l
Page 4 SGTR OFFSITE DOSE CALCULATION The transient analyzed was a double ended rupture of a single steam generator tube at the tubesheet on the cold leg side. Assumptions were made to maximize the time until reactor trip and the amount of primary-to-secondary leakage.
For dose calculation purposes, it is conservative to delay reactor trip since this will maximize the total duration of primary-to-secondary leakage and the steam generator activity. Credit was taken for operator action at 30 minutes to manu-ally initiate safety injection, which also trips the reactor. Other operator action times were taken from Table 2.3-2 of WCAP-10698.
Preliminary analysis results show that releases are within the criteria of 10 CFR 100 for NUREG-0800, Section 15.6.3, Cases I and II.
REFERENCES 1.
EPRI NP-1850s-CCM, "RETRAN A Program for Transient Thermal-Hydraulic Analysis of Complex Fluid Flow Systems", Electric Power Research Institute, Revision 2, November 1984.
2.
DPC-NE-3000, " Thermal-11ydrau11c Transient Analysis Methodology", Duke Power Company, July 1987, i
l i
Catawba Nuclear Station Structural Analysis of Main Steam Lines Item (3)
An evaluation of the structural adequacy of the main steam lines and associated supports under water-filled conditions as a result of SGTR overfill.
Response
The Catawba main steam lines and associated supports were evaluated (Calculation CNC-1206.02-71-2046) to determine their structural adequacy under water-filled conditions with all supports active. All pipe stresses were found to be well within code allowables.
l J
s i I
________.____m
At tachment 4 Steam Generator Tube Rupture Equipment List INTRODUCTION On March 30, 1987 the NRC issued a Safety Evaluation Report (SER) on WCAP-10698, "SGTR Analysis Methodology to Determine the Margin to Steam Generator Overfill".
Section D of Enclosure 1 of the SER listed the plant specific submittal requirements.
Item 4 requires "A list of systems, components, and instrumentation which are credited for accident citigation in the plant specific SGTR E0P(s).
Specify whether each system and component specificed is safety grade.
For primary and secondary PORVs and (block) valves specify the valve motive power and state whether the motive power and valve controls are safety grade.
For non-safety grade systems and components state whether safety grade backups are available which can be expected to function or provide the desired information within a time period compatible with prevention of SUTR overfill or justify that non-safety grado components can be utilized for the design basis event. Provide a list of all radiction monitors that could be utilized for identifi-cation of the accident and the ruptured steam generator, and specify the quality and reliability of this instrumentation if possible.
If the E0Ps specify steam generator sampling as a means of ruptured SC identification, provide the expected time period for obtaining the sample results and discuss the effect on the duration of the accident."
The Catawba emergency procedures do not specify steam generator sampling as a means of ruptured SG identification.
The remainder of the response to Item 4 is given in the to11owing sections.
EQUIPMENT LIST The following list gives the equipment credited in the Catawba emergency proce-dure, EP/1 or 2/A/5000/1E, for accident mitigation of the design basis steam generator tube rupture. The list is in the format of Table 4.4-2 from WCAP-10698 and lists both primary and backup equipment.
l J
4 E -
Page 2 Design Basis Equipment List for Cetawba EP Principal Bac kup Reference Stet; Equipment Equipment 1.
Identify Ruptured AFW flow control valves AFW isolation valves SG(s)
SG 1evel instrumentation Redundant channels 2.
Perform I:esets N/A to Allow MSIV Bypass Valves to be Opened 3.
Isolate Steam MSIV for ruptured SG MSIVs for intact SGs Flow from Turbine stop valves Ruptured SG(s)
Steam dump valves Reheater steam supply valves Auxiliary steam supply valve Steam line drains Steam traps Condenser air ej ector valves Isolation valve for AFW pump turbine trip steam supply line to and control valves AFW pump turbine SG safety valves 4.
Check Ruptured AFW valve control switches AFW isolation valver SG(s) Level AFW control valves MFW flow control valves MFW isolation valves 5.
Verify Secondary N/A Integrity (No Faulted SG(s))
6.
Check Intact SG AFW flow control valves AFW isolation valves Levels 7.
Check Pressurizer N/A PORVs and Block Valves 8.
Establish Operator N/A Control of the Following by Reset
Page 3 EP Principal Backup Reference Steg Equipment Equipment 9.
Check NC Pump N/A Status 10.
Initiate RCS PORV on intact SG(s)
Multiple intact SGs Cooldown 11.
Verify Ruptured N/A SG(s) Pressure Stable or Increasing and At Least One SG Without a Tube Leak, Isolated from all Ruptured SGs, and Available for RCS Cooldown 12.
Establish Instrument N/A Air to Containment
~
13.
Depressurize NC Pressurizer PORV and block
' Redundant pressurizer System to Mini-valve PORV and block valve mize Leakage 14.
Monitor RCS N/A Depressurization 15.
Stop RCS Pressurizer PORV Pressurizer PORV block j
Depressurization valve 16.
Check if SI Should N/A be Terminated 17.
To Terminate SI SI pump switches SI pump breakers HHSI isolation valve HHSI pumps controls SAFETY GRADE STATUS OF EQUIPMENT Of the equipment in the preceeding list, all principal equipment is safety grade except for the MFW and AFW feedwater flow control valves and the isolation valves for the steam supply to the AFW pump turbine.
The backup equipment in each case is safety grade.
In addition, the isolation function of the MFW control valves is safety grade.
4 Page 4 f
PORV AND BLOCK VALVE POUER The normal motive power for the primary and secondary PORVs is the Instrument Air System which is not safety grade.
Safety grade backup air motive power is avail-able for two primaty PORVs and all secondary PORVs. The primary and secondary PORV block valves are electric actor operated and are supplied with emergency power from the diesel generators. The primary and secondary PORVs and block valves all have safety grade control power.
RADIATION MONITORS The radiation monitors which could be used to diagnose a steam generator tube rupture accident are:
Monitor 1987 Availability through September (% of time) 1 EMF 26 Unit 1 Steam Line A 96 1 EMF 27 Unit 1 Steam Line B 74 1 EMF 28 Unit 1 Steam Line C 99 1 EMF 29 Unit 1 Steam Line D 78 2 EMF 10 Unit 2 Steam Line A 98 2 EMF 11 Unit 2 Steam Line B 98 2 EMF 12 Unit 2 Steam Line C 97 2 EMF 13 Unit 2 Steam Line D 98 1 EMF 33 Unit 1 Condenser Air 95 Ej ector Exhaust 2 EMF 33 Unit 2 Condenser Air 98 Ejector Exhaust 1 EMF 34 Unit 1 Steam Generator 10 Wa ter Sample 2 EMF 34 Unit 2 Steam Generator 71 Water Sample Of these monitors, the first eight could be used to identify the ruptured steam generator on the applicable unit.
The radiation monitors are high quality instrumentation but are not safety grade.
1 1
i 4
Overfill Evaluation for Steam Generator Tube Rupture INTRODUCTION On March 30, 1987 the NRC issued a Safety Evaluation Report (SER) on WCAP-10698.
"SGTR Analysis Methodology to Determine the Margin to Steam Generator Overfill" (Reference 1).
Section D of Enclosure 1 of the SER listed the plant specific submittal requirements.
Item 5 requires "A survey of plant primary and ' balance-of-plant' systems design to determine the compatibility with the bounding plant analysis in WCAP-10698. Major design differences should be noted. The worst single failure should be identified if dif ferent from the WCAP-10698 analysis and the ef fect of the dif ference on the margin to overfill should be provided."
RELATIONSHIP TO REFERENCE PLANT Section 4.1 of Reference 1 discusses the selection of a reference plant.
The plants are categorized into groups according to fuel rod length, number of loops, safety injection characteristics, and steam generator model.
These categories are given in Table 4.1-1 on page 4-4 of Reference 1.
The Catawba units are in the 412 HP SI group, Unit I having D3 steam generators and Unit 2 having D5 steam generators. The Catawba steam generators have smaller tube diameters and larger secondary steam volumes than the reference plant and therefore, other things being equal, a larger margin to overfill.
SINGLE FAILUR.E EVALUATION Section 4.4 of Reference 1 identifies 1) the minimum set of equipment for which credit must be assumed in the design basis analysis in order to prevent steam generator overfill and 2) the most detrimental single failure, relative to over-fill, from this list of design basis equipment.
The discussion in Sections 4.4.1 and 4.4.2 is generally applicable to Catawba with exceptions as noted below.
Auxiliary Feedwater (AFU) Flow Control Valve The Catawba AFW supply design is different from the reference plant.
There are four steam generators, two motor-driven AFW pumps, and one turbine-driven AFW pump. The motor-driven pumps are each aligned to two steam generators, A to A and B, B to C and D.
The turbine-driven pump can be aligned to any or all steam generators but is normally aligned to B and C.
Each steam generator thus has two separate AFW supply lines, one from a motor-driven pump and the other from a tur-bine driven pump. Each supply line has a separate flow control valve and an isolation valve. Therefore, failure of an AFW valve on the ruptured steam generator to close could be mitigated by closing its companion isolation valve.
The operator would not, in contrast to the reference plant, be required to stop the motor-driven pump supplying that generator.
Since flow to the intact gene-rators would not be reduced, the LOFTTR1 analyses performed to investigate this failure on the reference plant bound Catawba.
Page 2 Steam Generator Power Operated Relief Valve (PORV)
Although the discussion of steam generator PORV failure in' Reference 1 is generally applicable to Catawba, the Catawba plant emergency procedures (EPs) isolate a failed open PORV within the tube rupture procedure and therefore do not transfer out and back in to the first step of the procedure. Also the Catawba I
EPs include faulted steam generator isolation guidance only within the procedure, not on the fold-out page.
The stems generator PORVs are safety related and are seismically and environmen-tally qualified to achieve RCS cooldown to ND initiation temperature.
In addi-j tion, these valves close in 20 seconds or less after receiving a closure signal.
The PORVs can be actuated by either a pneumatic piston operator or local hand-wheel.
The pneumatic operator has two modes of operation. One mode provides non-adjustable automatic pressure control.
This mode of operation is non-safety.
The safety grade mode of operation is provided by the use of a nitrogen control system. All components of this system are environmentally and seismically quali-i fied.
Nitrogen is supplied by seismically mounted cylinders located in the Dog-
)
house.
The two PORVs in each Doghouse have independent nitrogen supplies with j
solenoids and controllers powered from independent essential electrical trains.
Safety grade control for the nitrogen control system is provided in the control room. Control for these valves is also provided from the auxiliary feedwater pump turbine control panel.
Each PORV is provided with an upstream electric motor operated block valve whose primary purpose is to allow PORV isolation for repair or maintenance.
The block j
valve can be used to isolate a partially or fully stuck open PORV. The motor j
operators fail "as is" and can be controlled from the control room.
The steam generator PORV block valves power supplies and controls are IE.
The PORVs on steam generators A and B receive Train A emergency power while those on steam generators C and D receive Train B power.
Because Catawba is a four loop plant, a single failure of a PORV on an intact steam generator will still leave two PORVs and their associated intact steam generators available for cooldown.
Thus the three loop reference plant generic analysis bounds Catawba.
Because Catawba has safety related PORVs and remotely operable emergency powered block valves, the generic analysis also bounds Catawba with respect to isolation times for failure of the PORV on the ruptured steam generator. No common mode PORV failures have been identified.
MSIV for Ruptured SG The layout of the main steam lines at Catawba is dif ferent from the reference plant.
There is no manifold so each steam line is separate from its MSIV to its turbine stop valve.
There is an equalization header, connected to each steam line, from which the various auxiliary steam extractions are taken.
The differences in layout do not affect the operation of the Main Steam System with respect to isolation of flowpaths following a tube rupture event, and the generic analysis is applicable to Catawba ij
l Page 3 Steam Generator Safety Valves There are five safety valves per steam line at Catawba.
The generic analysis is applicable with respect to these valves.
Isolation Valve for Steam Supply Line to Turbine-Driven AFW Pump Two of the four Catawba steam generators, B and C, have steam supply lines to the turbine-driven auxiliary feedwater pump. There is only one normally closed trip valve in each supply line from the steam generator.
However, as stated in the generic analysis, the turbine stop and control valves can be closed as a redun-dant means of isolating steam flow from the ruptured steam generator.
Catawba does have check valves in the steam supply lines, thus allowing this mode of isolation to remain ef fective as the intact steam generators are depressurized during recovery.
Feedwater Flow Control Valve Failure lhe two Catawba main feedwater pumps have a design flow of 18040 gpm and a design head of 2040 ft.
As stated in Section 4.4.3 of Reference 1, results of the gene-3 ric analysis (1400 ft pressurizer) are expected to be representative of other Westinghouse plants (Catawba has an 1800 ft3 pressurizer) to within 1 minute for margin to overfill.
SI Reset Device At Catawba SI reset is not necessary to stop AFW pumps. Although SI reset is necessary to reset the diesel generator load sequencers, instructions are included in the Catawba EPs to remove control power if the sequencer cannot be reset. This would allow rearrangement of loads on the emergency power busses without SI reset. As explained below, removal of bus power is not necessary at Catawba as a contingency for failure of an SI pump switch.
Therefore, the three reasons for including the SI reset device in the design basis equipment list for the reference plant are not applicable to Catawba.
Pressurizer PORVs and Block Valves A contingency procedure no longer exists at Catawba for terminating primary to secondary leakage without pressurizer pressure control.
This procedure was deleted per 10 CFR 50.59 af ter a plant specific reliability assessment of the pressurizer pressurc control systems.
This deletion was reviewed by the NRC as noted in Reference 2.
Page 4 SI Pump Switches The time required to locally operate the breaker for an SI pump whose reset switch fails to work is estimated to be less than the 2 minutes allowed in the generic analysis.
Local operation of the relevant breaker during such an event is addressed in operator training.
Loron Injection Tank (BIT)
Catawba no longer has a BIT.
As a part of the modification that removed the tank, the inlet isolation valves were removed also. Also, reactor coolant pump seal injection at Catawba is normally provided by the centrifugal charging pumps (CCPs), the same pumps that inject through the high head safety injection (HHSI) l lines. Stopping both CCPs stops HHSI flow but also simultaneously stops seal injection flow. Therefore this method of compensating for a failed open outlet isolation valve is only used for the design basis case, where the loss of offsite power renders the reactor coolant pumps inoperable anyway. For the more likely case where the reactor coolant pumps are available to provide pressurizer spray, it is assumed that the failed outlet isolation valve can be locally closed.
Since the valve is normally closed, failure of the valve to close requires that it has already successfully opened.
This indicates that the valve should be mechanically operable and that the failure to close is a problem with remote operation of the valve. In that event the operator is instructed in the Catawba EPs to locally close the valve while using normal spray to prevent repressuriza-tion. A calculation has been performed to show that the time required to locally close the valve is less than the time available before pressurizer overfill In summary, for the design case the operator can stop the remaining CCP
]
occurs.
within 2 minutes to compensate for a failed open outlet isolation valve.
For the case with pressurizer spray, the operator can locally close the failed valve l
before pressurizer overfill. The pressurizer spray have sufficient capacity to offset the flow from one CCP through the HHSI lines and prevent repressurization.
CONCLUSION The generic analysis in Reference 1 is generally applicable to Catawba. The exceptions have been noted above and do not affect the result, namely that there is margin to overfill for the design basis steam generator tube rupture event.
REFERENCES 1.
WCAP-10698-P-A, "SGTR Analysis Methodology to Determine the Margin to Steam Generator Overfill", R. N. Lewis, et.al., August 1987.
2.
August 6,1987 letter from the NRC (C. A. Julian) to Duke Power Company (H. B. Tucker) transmitting Inspection Report Numbers 50-413/87-13 and 50-414/87-13.
)