ML20203F802
ML20203F802 | |
Person / Time | |
---|---|
Issue date: | 04/17/1986 |
From: | Office of Nuclear Reactor Regulation |
To: | |
Shared Package | |
ML20203F799 | List: |
References | |
NUDOCS 8604280068 | |
Download: ML20203F802 (9) | |
Text
h .
SAFETY EVALUATION REPORT 0
Topical Report
Title:
Desiga Bases for the Thermal Overpower AT and Thermal Overtem e erature AT Trip Functions Topical Report Numbers: WCAP-8745 Topical Report Date: March 1977 !
- 1. INTRODUCTION This report describes the bases for the overpower and overtemperature AT trip functions in Westinghouse reactors, and the analytical methods used to derive the limiting safety system settings for the trips. These trip functions are designed to provide primary protection against departure from nucleate boiling (DNB) (overte,mperature AT), and fuel centerline melt (overpower AT) through excessive linear heat generation rates (LHGR) during postulated transients.
Since AT, the coolant temperature difference between vessel outlet and inlet, is (to a good approximation) proportional to the core power, and since the core power level is an important determinant of both DNBR and LHGR, the indicated AT serves as a useful primary parameter for these trip functions. Other parameters such as the average coolant temperature, the pressurizer pressure and the axial power offset modify the AT trip setpoint and thoreby account for the effects of pressure on DNB, and of the power shape on both DNB and LHGR. In addition, delays in signal propagation are accounted for with rate-lag or lead-lag compensation.
The overpower and overtemperature AT trip functions involve the Westinghouse (i) design bases and methods for evaluating fuel centerline temperature and DNBR, (ii) calculational methods for core power distribution using coupled core and systems transient codes, and (iii) and the application of the computer codes: THINC, THINC-IV, LOFTRAN, TWINKLE and PANDA. However, the review of these design and calculational methods, and computer codes is considered outside the scope of the present evaluation. This review focused, instead, on the applicability of the methods and the results to the derivation of the overtemperature and overpower setpoint limits.
8o04290068 860417 PDR TOPRP EPfVWEST C pg
I
- 2. Summary of Topical Report Section 1 of the report presents a short backgrodod, specifies the primary purpose of the overtemperature and overpower AT trips, and provide's a
~
summary of the report. The protection system and methods for set point determination described in the report are stated to apply to Westinghouse plants that reference RESAR-35 and operate under the guidelines of constant axial offset control.
Section 2 of the report specifies the design bases for core protection during normal operation, operational transients, and postulated transients occurring with moderate frequency, and describes the functional form of the thermal overpower and overtemperature trips. The general design criteria are specified as (i) UO 2 melting temperature will not be exceeded for 95% of the fuel rods at the 95% confidence level, and (ii) at least a 95% probability that DNB will not occur at the limiting fuel rod at a 95% confidence level. These criteria are to be met by ' restricting the calculated fuel centerline temperature to less than 4700 F, and by limiting the minimum DNBR to 1.3. A third design limit, namely that the hot leg temperature be maintained below saturation temperature enables the vessel average inlet / outlet coolant temperature difference (AT) to be used as a measure of the core power.
The overpower and overtemperature trips are activated on a two-out of three logic for three-loop plants, and on a two out-of-four logic for two and four-loop plants. The indicated vessel AT is continuously compared with the setpoint for each channel, which is calculated by analog circuitry programmed to evaluate the four-term setpoint representation. The leading term in the expression for the overpower AT setpoint is an adjustable, preset value of coolant temperature rise that is independent of the process variables. The second term is dependent on the average coolant temperature, and applies rate-lag compensation for pipe and thermal time delays. The third term accounts for the effects of coolant density and heat capacity on the relationship between aT and core power. The last term reduces the AT setpoint to account for adverse power distribution effects, and is a function of the axial flux difference. For the
~
2
overtemperature AT setpoint, the leading terni is also a preset adjustable value of AT independent of the process variables. The second term accounts for the effect of temperature on the desig, limits, and is lead / lag compensated for instrumentation and piping delays. The third term accounts for the_ effects of pressure on the design limits. As with the overpower AT trip, the last term accounts for tha effects of adverse power distributions, and is dependent on the axial flux difference.
Section 3 of the report presents the procedures for calculating the safety setpoints for the overpower AT trip. The calculation proceeds in four steps:
(1) A trip setpoint independent of the power distribution, typically at 118%
nominal power level, is selected, (2) power level and distribution during control bank and boration/ dilution system malfunctions are evaluated using a static nuclear core model without feedback, (3) the limiting LHGR occurring during these transients is compared to the threshold for fuel centerline melt, and (4) if the threshold is exceeded, either the trip setpoint is appropriately lowered or, more frequently, a trip reset function f(AI) is determined such that highly skewed power distributions are eliminated, and the threshold for fuel centerline melt is not exceeded. The evolution of Westinghouse methodology for calculating core power distribution effects is described in Section 3.2.
The basic method consists of calculating the envelope of maximum Fq , as a function of axial offset, for expected and unexpected plant maneuvers.
Originally all maneuvers that satisfied the control rod insertion limits were admitted, and an f(AI) function was generated based on the peaking factor analysis. In response to concerns regarding fuel densification, the f(AI) trip reset function was made appropriately more restrictive, the trip setpoint of 118% was sometimes reduced, and operating restrictions on part length control rods were introduced. Later, the constant axial off-set control (CAOC) method of plant operation was introduced in response to the requirements of the loss-of-coolant accident (LOCA) emergency core cooling. CAOC operation maintains the axial power distribution within a specified band, diminishes the adverse effects of xenon transients and serves to lower core peaking factors.
Therefore, f(AI) trip setpoints established prior to the introduction of the added CAOC constraints are considered conservative under CAOC operation. This situation exists for some 14x14 and 15x15 fuel assembly plants. For 16x16 and 17x17 fuel assembly plants operating under CAOC, Westinghouse analyses have 5
indicated that no f(AI) function is required to preclude fuel centerline
, melting during overpower transients because the thermal overpower limit of 118 percent of rated reactor power alone provides adequate protection against fuel melting. '
Section 4 of the report presents the procedures for calculating the safety setpoints of the overtemperature AT trip. The effects of core-wide parameters such as thermal power level and vessel average temperature are separated from power distribution effects by determining the former with a reference chopped ccsine shape and accounting for the latter through the f(AI) portiun of the trip. Assuming the reference power distribution, limits of safe operation are defined in the space of thermal power level, coolant inlet temperature, and primary system pressure. These limits of safe operation are determinea by the conditions that the vessel exit temperature be less than the saturation temperature, and that the minimum DNBR be above 1.3. To account for the effects of adverse axial power shapes, a set of " standard power distributions" (a set of limiting shapes having various values of axial offset), are generated using three-dimensional static nuclear calculations. For each power shape, the power level that gives a minimum DNBR of 1.3 is determined by iterative use of the THINC code. The procedure generates an envelope of allowable power vs.
axial offset for a given pressure and inlet temperature. Two envelopes, one at the inlet temperature corresponding to 118% power and the second corresponding to 80% power, are generated. The envelopes consist of positive and negative deadband regions of zero slope (in allowable power vs. axial offset), and regions of positive and negative slope. The widths of the deadband and the slopes are utilized to generate AT trip reset as a function of the axial offset, and hence determine f(AI).
Section 5 of the report describes analyses of anticipated transients with a coupled-core-system transient model used by Westinghouse to verify that the i methodology of standard power distributions described in Section 4 is I applicable under current plant operating procedures. The coupled-core system model is a combination of the lumped parameter single-loop system code, LOFTRAN, and the three-dimensional spatial neutron kinetics code, TWINKLE. DNB evaluations were performed with the THINC and equivalent codes. DNBR was 4
calculated using the axial power distribution predicted with LOFTRAN/ TWINKLE, a 4
4
.- . _ _ _ ,_...,y._,.._ _.-._..-_.-,,,,,._ - -- -. . .- - %-,. - - - , ,c
control-bank position-dependent value of F"AH, and the coolant conditions
, present at the moment. Five DNB-related transients were analyzed with the LOFTRAN/ TWINKLE model: (1) uncontrolled bank withdrawal at power, (2) step increase in steam flow caused by equipment malfunction, (3) inadvertent opening
~
of turbine throttle valve, (4) uncontrolled boron dilution at power with manual rod control, and (5) uncontrolled boration/ dilutions with automatic rod control. Worst pre-accident core conditions (i.e. , power level, control bank position and xenon distribution) to be used in the LOFTRAN/ TWINKLE analyses were determined by analyzing a complete set of initial conditions with the static nuclear model. Sensitivity to such variables as bank worth, bank withdrawal speed, automatic versus manual rod control, moderator feedback and Doppler feedback were analyzrd with the LOFTRAN/ TWINKLE model to identify -
limiting transients and conservative assumptions.
The adequacy of the standard power shape methodology can be established if the f(AI) functions generated using this methodology can be shown to be conservative khen compared with the results of the LOFTRAN/ TWINKLE analyses.
Section 6 of the report presents these comparisons which demonstrate that the f(AI) trip reset function generated using the standard shape methodology is conservative with respect to the results of the LOFTRAN/ TWINKLE analyses.
T
- 3. Summary of Technical Evaluation The evaluation of WCAP-8745 was based on an assessment of the general methodology presented, the scope and applicability of the methods discussed, uncertainties in the trip function design bases, and verification of the standard power shape methodology with the LOFTRAN/ TWINKLE model. The following sections address each of these concerns.
3.1 General Methodolo g The design bases and criteria for the overpower and overtemperature AT trip have been clearly defined and are consistent with Westinghouse general t,afety limits pertaining to ma<imum fuel temperature and minimum DNBR. The threshold for fuel centerline melt has been correlated with a limiting value of 5
kw/ft. The correlation includes the effects of burnup, flow rate, power distribution asymmetry and initial fill gas pressure level, and is based on an approved PAD analysis and is therefore acceptable. The minimum DNBR of 1.3 assumed in the analyses is an acceptable thermal safety limit. The functional forms of the trip setpoints appropriately account for effects such as coolant density and pressure variation, adverse core power distribution and instrumentation and piping delays (in addition to the variations in core power level), and for monitoring LHGR and DNBR.
3.2 Scope and Applicability Although Section 1 of the topical report specifies its applicability to Westinghouse plants that reference RESAR-35 and operate under CAOC, Westinghouse has indicated that they consider WCAP-8745 applicable to all Westinghouse plants that employ overpower and overtemperature AT trip for core protection. Westinghouse has stated that new methods and technology developed after the submittal of WCAP-8745 are described in separate topical reports, and do not invalidate the conclusions of WCAP-8745. As examples of such new methods, Westinghouse has cited changes in DNB analysis methodology (Improved Thermal Oesign Procedure and WRB-1 and WRB-2 correlations), fuel design (Optimized Fuel Assembly), and plant operating procedure (Relaxed Axial Offset Control), ar.d referenced topical reports describing these changes. While we agree that the basic design philosophy described in WCAP-8745 is not invalidateJ by changes in DNB analysis methodology, fuel design, and plant operating procedure, the application of this methodology must account for changes in system design and operation. The adequacy of the standard power shapes in establishing the core DNB protection system must be evaluated whenever changes are introduced that could potentially effect the core power distribution.
3.3 Uncertainties in Trip Function Design Bases In response to a request for information regarding uncertainties in the trip function design bases, Westinghouse has provided the error allowances 6
included for bistable error, signal linearity and reproducibility, calorimetric
. error, error in the T measurement and error in the pressure measurement.
Uncertainty in flow is accounted for by the use of a minimum technical specification flow in the analysis. Theerrorallowancesarearithyetically summed to obtain a total error allowance. Currently Westinghouse has ;
introduced a method of statistically combining error allowances, and has verified the conservatism of the old error allowance methodology by several plant specific statistical setpoint calculations. The statistical method has l been reviewed and approved by the NRC staff. Since the error allowance l methodology of WCAP-8745 has been demonstrated to be conservative with respect to the statistical method, we find it acceptable.
3.4 Verification of the Standard Shape P'ethodology with LOFTRAN/TWINKEL In support of the setpoint methodology, Westinghouse has provided the core axial offset, peak-to-average power, and shape cf the standard power shapes used in the standard shape methodology. The adequacy of the standard power shape methodology was demonstrated by establishing that the f(AI) functions used in this methodology are conservative (for the prediction of DNBR) when compared with the results of the LOFTRAN/TWINKl.E analyses. In the l comparison, five DNB-related transients were chosen aftei sensitivity to bank worth, bank withdrawe' speed, control rod operation mode, and l moderator and Doppler feedback were analyzed to identify the limiting I
transients and conservative initial conditions. The F 3g versus rod position function used in the DNB analysis had been demonstrated to be conservative for 50 different rod insertions in nearly 30 different plants. The power shapes l used in the DNB verification covered the entire cycle life. We therefore l conclude that the comparison between the standard power shapes methodology and the LOFTRAN/ TWINKLE analyses is sufficiently comprehensive in the choice of transients studied and in the applicability of the results to different f Westinghouse core designs studied at sufficient points in the cycle life.
7
i
~
- 4. Recommendation i t
We have reviewed the Westinghouse design bases for the thermal overpower and overteenperature AT Trip functions described in WCAP-8745, e.nd 'f,ind them acceptable for referencing by Westinghouse in licensing documents for plants that operate under constant axial offset control.
s 4
s 1
i 8
REFERENCES i
- 1. F.E. Motley, et al., "New Westinghouse Correlation WRB-1 for ,
Predicting Critical Heat Flux in Rod Bundles with Mixing Vane Grids,"
WCAP-8762 (Proprietary), July 1976.
- 2. H. Chelemer, et al., " Improved Thermal Design Procedure," WCAP-8567 (Proprietary), July 1975.
- 3. J. Skaritka, et al., " Fuel Rod Bow Evaluation," WCAP-8691, Rev. 1 (Proprietary), July 1979.
- 4. S.L. Davidson, ed., " Reference Core Report 17x17 Optimized Fuel Assembly," WCAP-9500-A, May 1982.
- 5. R.W. Miller, et al., " Relaxation of Constant Axial Offset Control,"
WCAP-10216-P-A (Proprietary), June 1983.
- 6. S.L. Davidson, ed., " Reference Core Report Vantage 5 Fuel Assembly,"
WCAP-10444-P-A (Proprietary), September 1985.
9