ML20209D965

Safety Evaluation of Topical Rept WCAP-14750, RCS Flow Verification Using Elbow Taps at Wesstinghouse 3-Loop Pressurized Water Reactors. Changes to TS Bases Acceptable

ML20209D965
Person / Time
Issue date:	07/07/1999
From:	NRC (Affiliation Not Assigned)
To:
Shared Package
ML20209D945	List: ML20209D942 ML20209D965
References
NUDOCS 9907140098
Download: ML20209D965 (12)

Text

,

earg i

g *t UNITED STATES g }

4 NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 20556 0001 1

\*****/

SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION RELATED TO WCAP-14750. "RCS FLOW VERIFICATION USING ELBOW TAPS AT WESTINGHOUSE 3-LOOP PRESSURIZED WATER REACTORS"

1.0 INTRODUCTION

By letter dated November 26,1996 (Ref.1), Southern Nuclear Operating C,ompany (SNC), the  !

licensee for the operation of Joseph M. Farley Nuclear Plant Units 1 and 2, submitted for staff I review a Westinghouse Owners Group (WOG) Topical Report, WCAP-14750, "RCS Flow l Verification Using Elbow Taps at Westinghouse 3-Loop PWRs," which describes a 1 methodology of using the cold-leg elbow tap pressure differential (AP) measurements for the reactor coolant system (RCS) flow surveillance verification.

Technical Specifications (TS) Limiting Conditions for Operation (LCO) of the Westinghouse- '

oesigned pressurized-water reactors (PWRs) require the RCS flow to be maintained greater than u equal to a specified minimum measured flow (MMF) rate during MODE 1 operation.  ;

This LCO MMF is an input value in the design basis transients safety analyses that use a i statistical core design method (such as the improved Thermal Design Procedure (ITDP) or the Revised Thermal Design Procedura (RTDP) (Ref. 2)) to demonstrate that the departure from nucleate boiling ratio (DNBR) limit is not violated during normal operation and anticipated operational occurrences (AOO). Surveillance requiremente (SR) require that the RCS total flow be verified within its limit by control board RCS flow indicator reading at least once per 12 l

hours, and be verified by a precision heat balance at least once per 18 months.

In the precision heat balance measurement, calorimetric menurements are made on the steam generator (SG) secondary side with the feedwater flow rates measured by venturi meters. The RCS flow rate is calculated from the calorimetric measurements in conjunction with the enthalpy rise across the reactor vessel as indhated by the hot- and cold-leg resistance temperature detectors (RTDs). Each hot leg has three thermowell RTDs installed around a cross-section to j determine the bulk hot-leg temperature. However, due to the use of low leakage core loading l patterns that f asult in changes in the core radial power distribution, the phenomenon of  !

increased hoteg temperature streaming has been observed in many plants, including Farley 4 Units 1 and 2. As a result of the increased temperature streaming, the bulk hot-leg temperature '

as measured by the three RTDs in each hot-leg is erroneously high, resulting in a calculated j RCS flow lower it.an the actual value. Licensees expressed a need to use the cold-leg elbow j tap flow rneasurement as an alternate method for the RCS flow verification because of the '

inherent limitation of the calorimetric based method. The use of the cold-leg elbow tap measurements foc the 18-month RCS flow surveillance has been approved by the NRC for 1 McGuire Nuclear Station (Ref. 3), Catawba Nuclear Station (Ref. 4), and South Texas Project )

Electric Generating Station (Ref. 5). j 9907140098 990707 ENCLOSURE PDR TOPRP EMVWEST C PDR

F .

L 2.

WCAP-14750 describes the methodology of using the cold-leg elbow tap Ap measurement for

' the RCS flow measurement, and the application of this elbow tap Ap methodology as an attemate method for satisfying the TS 18-month RCS total flow surveillance for the Westinghouse-designed 3-loop PWRs for which Farley Nuclear Plant is the lead plant.

2.0 EVALUATION-

'.The staff evaluation of WCAP-14750, as discussed in the ensuing sections, includas the following:

.A generic evaluation of the appropriateness of the cold-leg elbow tap flow measurement.

• The' procedure for converting elbow tap AP measurements to RCS flow.

The best estimate hydraulics calculation for confirming RCS flow measurement.

The flow measurement uncertainty evaluation.

. . The recommended Standard Technical Specifications (STS) changes to implement slbow tap measurement methodology.

A review of the process implementation for the Farley units 2.1 Elbow Tap Flow Measurement Methodology 2.1.1 Elbow Tap Flow Measurement Cold ' leg elbo' w tap flow meters are used by Westinghouse plants, including Farley Units 1 and  ;

2, for surveillance verification of the RCS flow through the centrol board indication every 12 l hours The purpose'of the 12-hour elbow tap surveillance reading is to verify that the full power l steady state flow has not decreased below its limit during the fuel cycle. The principle of l operation of an elbow meter is based on the centrifugal force of a fluid flowing through an elbow creating'a 6P betweer. the outer and inner radii of the elbow. The relationship between the volumetric flow rate through an elbow, Q, and AP between the pressure taps at the outer and  ;

inner radii of the elbow can be expressed as Q = C APu2. The elbow mater coefficient C is a i function of elbow bend ard cross-section radii, and is affected by the location of pressure taps, upstream and downstream piping, and other factors. The cold-leg elbow tap - flow element is not calibrated in advance in a laboratory, but the measurement is typically normalized against j the established RCS flow rate from the precision heat balance calorimetric flow measurement at  !

the start of each fuel cycle. The cold-leg elbow taps are typically used as an indM.ation of

' relative changes in the RCS flow rather than a measurement of absolute value of the RCS flow. The cold-leg elbow tap AP also provides a measure of the reduced RCS flow rate for the

. low-flow reactor trip.

Figure 4-1 in WCAP-14750 shows' the configuration of the Prairie Island Unit 2 cold-leg elbow taps, which is a standard configuration used in other Westinghouse PWRs including the Farley  ;

units. The elbow taps are located in a plane 22.5' around the first 90* elbow tum in each of the 4 cold legs. Each elbow has three low pressure taps spaced 15' apart on the inside pipe radius and one high-pressure tap on the outside pipe radius used as the common tap. The pressure

. taps are connected to three differential pressure transmitters to obtain AP data. As the elbow i

- taps in the cold legs are fixed, the elbow meter coefficients in each elbow tap configuration l should remain unchanged. The topical report also cited an American Society of Mechanical Engineers (ASME) publication (Ref. 6) stating that tests have demonstrated that elbow tap flow  !

1 L 1- J

F .

1 l

measurements have a high degree of repeatability, and are not affected by changes in the elbow surface roughness. l To confirm elbow tap flow measurement repeatability, Section 4.1 of WCAP-14750 provides comparisons of the data between the RCS flow measurements using the elbow taps and ultrasonic leading edge flow meters (LEFM) from the Hydraulic Test Program at Prairie Island Unit 2 (PI-2). The PI-2 Hydraulic Test Program was in place since 1973 and the test data covered 11 years of plant operation, during which a significant change in system hydraulics was made. The data showed that the elbow tap measurements agree to within 0.3% of the LEFM flow measurements. Various processes or phenomena for possibie effects on the elbow tap flow measurements were eva'uated, including the effects of fouling, erosion, upstream velocity i distribution, and SG tube plugging and replacement. The Test Program concluded the I following:

1. Fouling conditions are not present in the cold-leg elbow since there is no change in crcss section to produce a velocity increase and ionization.

2. Stainless steel elbow surface erosion is unlikely and the flow velocities are not large I relative to the conditions that cause erosion.

3. The upstream velocity distribution, including the distribution in the elbow tap flow meter, i remains constant so the elbow tap flow meter AP versus flow relationship does not change.

4. The plenum velocity head approaching the outlet nozzle is small compared to the piping l velocity head; therefore, SG tube plugging does not affect elbow tap flow measurement repeatability.

5. Replaced SG configurabon is the same, and the same difference in plenum and nozzle velocity heads will exist, therefore SG replacement will have no impact on the elbow tap flow coefficient.

The elbow taps at Farley were not calibrated and the elbow meter coefficients were not determined. However, RCS flow mecsurements using elbow taps were normalized against the precision heat balance flow measurements at the start of each fuel cycle. The staff concludes that as the elbow meter coefficients remain constant, the relative changes of flow rate through the cold-leg elbows can be correlated with the relative changes in the elbow tap AP.

2.1.2 Elbow Tap Flow Measurement Procedure Section 4.2 of WCAP-14750 describes the procedure for determining the RCS flow from elbow tap AP measurements. This procedure relies on the total basciine calorimetric flow (BCF),

which is based on the calorimetric flow measurements from early fuel cycles. The future cycle flow (FCF) will be determined from the BCF multiplied by the elbow tap flow ratio (R). Section 4.2 of WCAP-14750 defines the elbow tap flow ratio (R), as R = (K/B)", where B is the "beseline elbow tap total flow coefficient" defined as B = APs x v , K is the " future cycle elbow tap total flow coefficient" defined as K = AP, x vp, and v is the specific volume of the coolant, s

Su tion 4.2 indicated in the paragraphs under " Baseline Calorimetric Flow" that the BCF would be based on either (1) one baseline cycle such as the first fuel cycle, or (2) the average of multiple cycles with consistent measurements (i.e., the calorimetric measurements of the averaged cycles are adjusted for th3 effects of known changes in system hydraulics to the same hydraulic configuration as the beseline cycle). The SNC (WOG), in response to staff

l

\

4 questions (Question 4, Ref. 7), provided the criteria to be used in choosing the calorimetric flow measurements frorn early cycles for averaging, and indicated the use of only one calorimetric measurement to be the preferred procedure. In response to a further staff request for guidelines in defining the baseline flow from one baseline cycle or the average of multiple cycles (Question 1, Ref. 8), SNC (WOG) revised the paragraphs of " Baseline Calorimetric Flow." These new paragraphs provide a revised procedure for the determination of the BCF, including the acceptance criteria in the choice of early cycle calorimetric flow measurements, and tha determination of the BCF from the chosen cycle data. The staff ' . reviewed the revised procedure for defining the BCF, and found it acceptable. SNC M 3) indicated that they will use the revised procedure in place of the existing procedure in WCAP-14750 Section 4.2.

The baseline and future cycle " flow coefficients" B and K, are calculated based on the average AP from all cold-leg elbow taps. For each individual elbow tap, the elbow meter coefficient C in the elbow meter equation would be constant. Therefore, the ratio of the volumetric flow rates through the elbow tap between two fuel cycles can be expressed in terms of the square root of the AP ratio. The AP ratio would be the same for the three elbow taps in the same cold leg, barring measurement uncertainties. The staff questioned (Question 2b, Ref. 7) whether it would be appropriate to define the elbow tap flow ratio (R) based on the average of the square root of the AP ratios from all elbow taps, rather than the AP average. In response, SNC (WOG) calculated R based on the average of the square root of the AP ratios and compared that with R calculated using the AP average method described in WCAP-14750 Section 4.2. SNC (WOG) used Farley Unit 1 and 2 indicated transmitter AP values for each fuel cycle to do this.

The results show insignificant difference between the two calculations. The staff, therefore, concludes that using average AP is acceptable.

SNC (WOG) also asserted, in response to a staff question (Questions 2c,2d, Ref. 7), that there is no need to include an additional allowance to the future cycle flow ratio R to account for the AP ratio distribution among the elbow taps using an one-sided tolerance limit to provide a 95 percent probability at 95 percent confidence level. The overall RCS flow determination procedure based on the elbow tap AP measurements includes the following:

1. A calculation of the future cycle flow ratio R based on determining the ratio (between the future and baseline cycles) of the average indicated AP values.

2. A separate comparison with the predicted system flow to account for the system hydraulic effects such as SG tube plugging.

3. A separate uncertainty calculation to account for the flow measurement uncertainties.

The entire process assures a conservative RCS flow surveillance verification because of the j conservative uncertainties to the baseline calorimetric flow measurement and plant process ,

computer indication, and a one-sided acceptance criterion for flow measurements confirmation with best estimate calculations. The staff review of the best estimate flow confirmation and the flow measurement uncertainty calculation will be discussed below in Section 2.1.3 and 2.2.1 of this report, respectively. Based on the above, the staff concludes that the elbow tap RCS flow measurement procedure described in Section 4.2 with the revised BCF procedure is acceptable.

m ,

)

u -5 L 2.1.3 Best Estimate Flow Confirmation s  ! The elbow tap flow measurement procedure includes a requirement that utilities are to perform a best estimate (BE) hydraulics analysis to confirm the future total RCS flow determined from

^

a i

. the elbow tap flow measurement.- The BE RCS flow calculation, described in Section 5 of -

WCAP-14750, is based on the flow resistances of various components in the reactor coolant i loops and the reactor coolant pump performance characteristics. Therefore, changes in the RCS flow rate can be evaluated based on system hydraulic changes in the plant (e.g., plugging )

and sleeving of SG U-tubes, reactor coolant pump wear, and changes in the fuel design).

i

~

The BE hydraulic analysis confirmation procedure specifies that utilities are to compare the

. elbow tap flow ratio (R) to an estimated future cycle flow ratio (R'). R is based on the elbow tap i AP measurements as previously discussed. R'is the ratio of the estimated future cycle RCS flow to the estimated initial baseline cycle flow based on the flow analysis of known RCS )

hydraulics changes, such as SG tube plugging or fuel design changes. If the measured R is greater than (1.004 x R'), R will be limited to (1.004 x R'). The multiplier 1.004 applied to R' is a measure to prcvide an allowance of 0.4% for elbow tap flow measurement repeatability.

The 0.4 percent repeatability value was determined by combining appropriate instrument uncertainties for two different cycle measurements of RCS flow at 100% rated thermal power using all of the cold-leg elbow tap channels. A derivation of the repeatability value of 0.4 percent flow for Farley 1 and 2 was provided in response to staff requests for additional information (Question #3, Ref. 7, Question #3, Ref. 9). The repeatability allowance is implicitly L included in the elbow tap flow measurement uncertainty calculations because all of the instrument uncertainties included in the repeatability derivation are common with those in the

. elbow tap flow measurement uncertainty calculations. The SNC (WOG) States that since the

- elbow tap flow measurement uncerteinty includes this repeatability allowance, the measured flow ratio R can be 0.4 percent higher than the estimated flow ratio R' and still define a

~

= conservative flow.

The BE RCS flow analysis employs an RCS flow calculational procedure developed by Westinghouse in 1974 using BE values of the RCS component flow resistances and pump performance with no margins applied, so the resulting flow calculations define a true best estimate'of the actual flow In the analysis, the flow resistances of the RCS loops, which are comprised of the reactor vessel,' reactor coolant piping, and SGs, are used in conjunction with the reactor coolant pump head-flow performance to define individual loop and total RCS flows.

The component hydraulic design data and hydraulic coefficients are determined from analyses

- of test data.l The flow resistance of the reactor vessel, consisting of the reactor core, vessel internals and vessel nozzle, is daterm'ined from the AP measurements of a full size fuel assembly hydraulic test, and hydraulic model test data for each type of reactor vessel. The reactor coolant piping flow resistance combines the resistances of the hot-leg, crossover-leg and cold-leg piping. The flow resistance is based on analyzing the effects of upstream and downstream components on elbow bydraulic loss coefficients, using the results of industry

- hydraulic tests. The flow resistance is defined in five parts; inlet nozzla, tube inlet, tubes, tube outlet, and outlet nozzle. Section 5.1 of WCAP-14750 indicates that numerous component flow resistance tests and analyses (including the overall flow resistance confirmed by the Prairie island Unit 2 Hydraulics Test Program) have confirmed that this hydraulic analysis procedure has an uncertainty of 2 percent flow. This indicates that actual flow is expected to be within 2

. percent ofina calculated BE flow, Utilities have used this hydraulics analysis procedure to

P .

I 6 l estimate RCS flows at all Westinghouse plants, including Farley Units 1 and 2.

SNC (WOG) stated, in response to a staff question (Question #8, Ref. 9), that comparison of the elbow tap flow measurement results to a BE flow model predicted value is for the purpose of a cross check and does not provide a direct input to verification of thm ,afety analyses RCS flow assumption. The best-estimate flow analysis defines the expectec change in fivw for a new cycle. If the elbow tap measured flow is greater than the BE flow by more than the repeatability uncertainty for the elbow taps, then the more conservative (smaller) of the two values is used to define the RCS flow for the cycle. The staff finds that the BE hydraulic analysis will be used merely as a confirmation of the elbow tap flow measurement and will not change the TS surveillance requirement for a flow measurement, and is, therefore, acceptable.

2.2 Elbow Tap Flow Measurement Licensing Considerations Plant TSs require that the LCO RCS MMF be verified for compliance through calorimetric measurements performed every 18 months at the beginning of each fuel cycle. and qualitative verification thereafter using the installed flow instrumentation every 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> to assess potential degradation. The LCO MMF limit and the low-RCS flow reactor trip setpoint are inputs to the safety analyses to demonstrate that the DNBR limit is not exceeded during normal operation and anticipated transients. The RCS flow measurement uncertainty associated with the low flow reactor trip is accounted for in the reactor trip setpoint allowance. The flow measurement uncertainty associated with the MMF surveillance is accounted for in the safety analyses either deterministically or statistically. In the deterministic safety analyses, the initial RCS flow is assumed to be the thermal design flow which is the MMF minus the measurement uncertainty, in the statistical method, e.g., the ITDP or RTDP, the initial RCS flow is assumed to be the MMF with the flow measurement uncertainty accounted for statistically in the DNBR safety limit.

Section 7.0 of WCAP-14750 discusses licensing considerations associated with applying the elbow tap Ap measurement as an alternate method for performing the 18-month RCS flow surveillance currently performed with the precision calorimetric measurement. As the RCS flow measurement uncertainty will likely increase with the elbow tap Ap methodology, an evaluation must be made on the flcr.v measurement uncertainty and its impact on the existing safety analyses. An increase in the flow uncertainty beyond those currently established in the low-RCS flow trip setpoint allowance or the LCO MMF uncertainty assumed in the safety analyses will require a revision to the TS to reflect the uncertainty changes. The increase the flow uncertainty will also require a new safety analyses to demonstrate that the flow uncertainty and the MMF assumed in the safety analyses will not result in the DNBR safety limit being exceeded during normal operation and anticipated transients.

2.2.1 Flow Measurement Uncertainties Appendix B of WCAP-14750 provides a sample elbow tap flow measurement safety evaluation (SE) for the Westinghouse-designed 3-loop PWRs. The sample SE includes a ciiscussion of the uncertainty calculation, and a proposed STS markup associated with using the elbow tap AP method. With the elbow tap methodology, the RCS albow tap AP measurements are correlated to the precision calorimetric measurements performed during an earlier fuel cycle when the hot-leg streaming effects were minimal. The RCS flow measurement uncertainties include uncertainties associated with the following:

y.

u .

l-l 7 a

calorimetric measurement of the RCS total flow for the baseline cycle AP transmitters plant process computer indication for the current cycle RCS flow measurements using the cold-leg elbow taps The sample evaluation stated that the uncertainty calculation performed for this method of flow measurement is consistent with the methodology described in NUREG/CR-3659 (Ref.10), with the only significant difference being assuming correlation to previously performed RCS flow calorimetrics. This is accounted for by adding certain instrument uncertainties previously considered to be zeroed out by the assumption of normalization to a calorimetric performed each cycle.

The calculations account for the plant instrumentation, test equipment, and procedures which were in place at the time the calorimetric was performed. The calculations also include

. additional instrumentation drift uncertainties to reflect the correlation between the elbow tap AP measurements and the calorimetric flow measurements. Uncertainty calculations are performed for the indicated RCS flow (computer) and the RCS low-flow reactor trip.

The uncertainty methodology of NUREG/CR-3659 uses a statistical uncertainty combination technique, i.e., those groups of components which are statistically independent are statistically combined, and those errors which are not independent are combined arithmetically to form independent groups, which can then be statistically combined. As the elbow tap AP measurements were correlated to the calorimetric measurements of the baseline cycles, the overall RCS flow measurement uncertainty is a statistical combination of the baseline cycle calorimetric measurement and elbow tap measurement uncertainties. This uncertainty calculation method has been accepted in connection with the ITDP and RTDP evaluations.

Appendix A of WCAP-14750 contains sample uncertainty calculations performed using Farley-specific inputs. The calculations include the following uncertainties:

BCF measurement instrumentation uncertainties flow calorimetric sensitivities overall calorimetric flow measurement uncertainties cold-leg elbow tap flow measurement uncertainties for the process computer

- low-flow reactor trip uncertainties The ;aff evaluation of the Farley-specific uncertainty calculation will be addressed in Section 2.3.2 of this SE. Each licensee referencing WCAP-14750 should perform a similar plant-specific calculation of the measurement uncertainties associated with the elbow tap Ap method.

The staff will review this calculation on a plant-specific basis.

2.2.2 Modifications To Standard Technical Specifications Section 7.3 of WCAP-14750 recommends that, for the Improved STS, the LCO and the surveillance requirements include two different DNB flow requirements - one to be used if a precision calorimetric measurement is performed and one to be used with an elbow tap AP measurement. The applicable Bases section will also require revisions to include a description of the elbow tap AP method of flow measurement. Attachment 1 to Appendix B provides a sample impruved STS markup. TS Table 3.3.1-1, " Reactor Trip System Instrumentation,"

Item 10, " Reactor Coolam Flow - Low" trip function will be modified with the trip setpoint and L_

I

, i

\

l i

l

i.  !

i  !

8  !

allowable value consistent with the safety analysis limit assumption and the elbow tap AP flow  !

measurement uncertainty for the low flow reactor trip. Item c of LCO 3.4.1, and both SR 3.4.1.3 and 3.4.1.4 of the 12-hour and 18-month RCS flow verification will be modified to allow for the l

use of either the precision heat balance method or the elbow tap AP measurement method with a respective MMF limit specified for each method. Each of the limit values contains a flow q

mea surement uncertainty calculated for the precision calorimetric measurement, or the elbow tap AP measurement. Basis B 3.4.1 will also be modified to reflect the alternate methods of precision calorimetric and elbow tap AP measurements for the RCS flow verification, and their respective measurement uncertainties. The staff finds these recommended TS modifications to be acceptable. However, each licensee who references WCAP-14750 should ensure that ,

(1) the RCS flow measurement uncertainties associated with the calorimetric flow measurement '

and the elbow tap flow measurement, respectively, are equal to or greater than those obtained l frem the respective uncertainty calculations, and (2) the MMF and uncertainties specified in the TS are consistent with those assumed in the safety analyses.

2.3 Imp!ementation of Cold Leg Elbow Tap Flow Measurement Procedure l

2.3.1 RCS Flow Performance Evaluation  !

Section 6.0 of WCAP-14750 describes the evaluation of Farley Units 1 and 2 RCS flow ,

performance. For each unit, the evaluation includes the following:

determining the baseline cycle calorimetric flow (BCF) l

+

determining the elbow tap total flow coefficient (B)

+

evaluating the elbow tap flow ratio (R) based on the current cycle elbow tap flow coefficient calculated from the elbow tap AP measurements

, A best-estimate RCS flow prediction is also made based on the known hydraulic changes. SNC (WOG) provided the analytical model, including the RCS hydraulic network diagram, and component flow resistance values, are provided in response to a staff request for additional information (Question 3c, Ref. 7). The analyses determined the baseline cycle 1 initial startup flows of both units based on the baseline hydraulic designs. Hydraulic changes during subsequent cycles, including pump impeller smoothing, steam generator plugging, and fuel, design changea are modeled to determine best-estimate flow rates of various cycles. The BE flow prediction is used to confinn the elbow tao measured RCS flow. The evaluation process follows the procedure described in Section 4.2 of WCAP-14750, and is therefore acceptable with the exceptions discuscad in the following paragraphs.

Section 6.1 discussed the evaluation of the BCFs for Farley Units 1 and 2, and Tables 6.1 and 6.2, respectively, provide the Farley Units 1 and 2 early cycle calorimetric flow measurement data and the calculated BCF values for Farley Units 1 and 2, respectively. This BCF evaluation was based on the original 3CF deterrnination procedure described in Section 4.2 of WCAP-14750. As discussed in Section 2.1.2 of this SE, this BCF determination procedure has been replaced with a revised procedure described in the SNC (WOG)'s response to Question 1, Reference 8. In the same document, the SNC (WOG) also provided a revised calculation of the BCFs using the revised BCF procedure. SNC (WOG) states in the footnote that applicable information in Section 6 and Appeadix D of WCAP-14750, which summarizes the elbow tap measurement procedure, will be revised to reflect these changes. The staff has reviewed and found the revised BCFs for Farley Units 1 and 2 to be acceptable.

l 1

y

.t e

9

'2.3.2 Uncertainty Evaluation Appendix A of WCAP-14750 contains uncedainty calculations which were performed using Farley specific inputs.' Tables A-1,' A-2, and A-3, respectively, provide the values of the baseline calorimetric flow measurement instrumentation uncertainties, flow calorimetric sensitivities, and calorimetric flow measurement uncertainties. Tables A-4 and /-5, respectively, provide the cold-leg elbow tap flow measurement uncertainties for the process computer, and low-flow reactor trip uncertainties. The uncertainties for a calorimetric measurement or the elbow tap measurement consist of uncertainties from all components in the measurement channel. These include non-instrument-related measurement errors (such as temperature stratification of a fluid in'a pipe) and instrument-related errors (such as errors due to metering devices,2 calibration accuracies of sensors, process rack, and readout devices,

~ drift, temperature and pressure effects, etc.). These uncertainty components are combined to derive a channel. statistical allowance using the statistical combination technique consistent with the methodology described in NUREG/CR-3659 (Ref.10).

Table A-4 in WCAP-14750 shows an overall RCS flow uncertainty of 2.3 percent for the process computer. Table A-5 shows the calculated channel statistical allowance for the reactor trip function'is lower than the total allowance of 4 percent flow span assumed for the low-flow

' reactor trip function. In response to a staff question (Question 3b, Ref. 7), SNC (WOG)

~

asserted that the uncertainty input values relative to 1) the reference accuracy,2) pressure and

' temperature effects, 3) calibration accuracy for sensors and process racks, and 4) sensor and rack drift magnitudes, are 20 (standard deviation) or better. Therefore, the overall uncertainty

~ for RCS flow utilizing the cold-leg elbow tap methodology and used for the RTDP analyses is a 95/95 probability / confidence value.

SNC (WOG) considered drift of the instruments and process racks. SNC (WOG) did this since

~ they do not normalize cold-leg elbow tap measurement against a precision heat balance flow measurement at the beginning of each fuel cycle. In response to a staff question (Question 5,

- Ref. 7), SNC (WOG) explained that they included sufficient drift allowances in the instrument uncedainties shown in Tables A-4 and A-5 for the weillance interva!. The staff concludes that this explanation is acceptable.

2.3.3 Farley Technical Specification Modifications The current Farley TS SR 4.2.5.2 requires SNC to determina that the RCS total flow is witMn its limit by measurement at least once per 18 months. Th'e SR does not specify the method to measure RCS flow. TS Bases 3/4.2.5, "DNB Parameters," states that the 18-month surveillance of the total RCS flow rate is a precision measurement that verifies the RCS flow

' requirement at the beginning of each fuel cycle. Therefore, there is no need to change the Farley TS to implement RCS flow measurement using the cold-leg elbow tap AP measuren. ants, except for modifying the Bases.

Attachment 1 to Appendix C of WCAP-14750 provides markups of proposed modifications to Farley TS Bases 3/4.2.5. The changes reflect using cold-leg elbow tap AP measurement as an alternate method for the 18-month RCS flow surveillance. The change references WCAP-14750 regarding the method to correlate the flow indication channels with selected precision calorimetrics. The Bases is also revised, in Response to Question 4 in Reference 8 to state that the indicated total RCS flow rate is based on a measurement utilizing two indication 1

r. 2 10 channels per loop and an uncertainty of 2.4 percent flow. The flow uncertainty value, which includes 0.1 percent flow uncertainty for feedwater venturi fouling, is consistent with the calculated elbow tap flow measurement uncertainty of 2.3 percent shown in Table A-4 of WCAP-14750 for the process compu'er. The staff finds tne changes to the TS Bases acceptable.

3.0 CONCLUSION

The staff has reviewed the cold-leg elbow tap RCS flow measurement methodology, including the following items:

a correlation of the indicated APs to the baseline calorimetric RCS flow rate the flow measurement uncertainty calculation a

proposed Improved STS changes Farley-specific flow measurement uncertainty calculation TS changes assouated with implementing the elbow tap flow measurement Based on its review of the technical bases regarding the cold-leg elbow tap RCS flow measurement procedure and the measurement uncertainty calculation provided in SNC (WOG)'s submittal, the staff finds WCAP-14750 acceptable for referencing to support licensing actions. The staff notes that this acceptance is based, in part, on the revisions described in the response to Question 1, Reference 8, regarding the procedures for determining the baseline calorimetric flow, as well as the BCF calculations for Farley Units 1 and 2.

t

f.3 3

f 11 REFERENCES '

1. Letter, Dave Morey (Southern Nuclear Operating Company) to US Nuclear Regulatory Commission, " Joseph M. Farley Nuclear Plant, Request for NRC Review of WCAP-14750,

'RCS Flow Verification Using Elbow Taps at Westinghouse 3-Loop PWRs'," November l 26,1996.

2. WCAP-11397-P-A, " Revised Thermal Design Procedure," Westinghouse Electric Corporation, April 1989.

3. Letter from Victor Nerses (USNRC) to T. C. McMeekin (Duke Power Company), " Issuance I of Amendments - McGuire Nuclear Station, Units 1 and 2, Reactor Coolant System (RCS) l Flow Rate measurement (TAC Nos. M88659 and M88660)," January 12,1995.

l l

4. Letter from R. E. Martin (USNRC) to D. L. Rehn (Duke Power Company), " Issuance of l Amendments - Catawba Nuclear Station, Units 1 and 2, Reactor Coolant System (RCS)

Flowrate measurement (TAC Nos. MS8480 and M88658)," February 17,1995.

5. Letter, Thomas W. Alexion (USNRC) to William T. Cottle (STP Nuclear Operating Company), " South Texas Project, Units 1 and 2 - Issuance of Amendments, Re: Reactor l Coolant System Flow Monitoring (TAC Nos. M99245 and M99246)," April 19,1999. '

l

6. " Fluid Meters, Their Theory and Application," 6th Edition, Howard S. Bean, ASME, New York,1971.

7.- Letter, Dave Morey (Southern Nuclear Operating Company) to U. S. Nuclear Regulatory Commission, " Joseph M. Farley Nuclear Plant, Response to Request for Additional laformation Related to WCAP-14750, 'RCS Flow Verification Using Elbow Taps At Westinghouse 3-Loop PWRs'," February 2,1999.

8. Leder, Dave Morey (Southern Nuclear Operating Company) to U. S. Nuclear Regulatory  ;

Commission, " Joseph M. Farley Nuclear Plant, Response to Request for Additional J Information Related to WCAP-14750, 'RCS Flow Verification Using Elbow Taps At i Westinghouse 3-Loop PWRs'," June 7,1999. I

9. Letter, Dave Morey (Southern Nuclear Operating Company) to U. S. Nuciear Regulatory -

4 Commission, " Joseph M. Farley Nuclear Plant, Response to Request for Additional .

Information Related to WCAP-14750, 'RCS Flow Verification Using Elbow Taps At Westinghouse 3-Loop PWRs'," October 1,1997.

10. NUREG/CR-3659, PNL-4973, "A Mathematical Model for Assessing the Uncertainties of l Instrumentation Measurements for Power and Flow of PWR Reactors," February 1985.

l

, l l

Westinghouse Owners Group . Project No. 694 I cc:

l Mr. Nicholas Liparulo, Manager Equipment Design and Regulatory Engineering Westinghouse Electric Corporation )

Mail Stop ECE 4-15 P.O. Box 355 Pittsburgh, PA 15230-0355 1 Mr. Andrew Drake, Project Manager l Westinghouse Owners Group Westinghouse Electric Corporation Mail Stop ECE 5-16 I P.O. Box 355 l Pittsburgh, PA 15230-0355 l l Mr. Jack Bastin, Director Regulatory Affairs -

Westinghouse Electric Corporation 11921 Rockville Pike Suite 107 1 Rockville, MD 20852 l Mr. John Galembush, Acting Manager Regulatory and Licensing Engineering Westinghouse Electric Corporation PO Box 355 Pittsburgh, PA 15230-0355 l

L: