ML20073C779

Forwards Proprietary Response to Request for Addl Info Re Reactor Sys Flow Measurement Methodology.Encl Withheld

ML20073C779
Person / Time
Site:	Catawba, McGuire, Mcguire
Issue date:	09/15/1994
From:	Tuckman M DUKE POWER CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
Shared Package
ML19311B363	List: ML20073C779
References
NUDOCS 9409260161
Download: ML20073C779 (67)

Text

l 11 l DukeIbwerCornpany M S Tin 113 P.O. Box 1006 Senior Vicehesident Charlotte,NC282011006 Nuclear Generation (704)3102200 Oflice (701)382-4360 Fax DUKEPOWER l

l I

September 15, 1994 j U. S. Nuclear Regulatory Commission Washington, D. C. 20555 Attention: Document Control Desk

Subject:

McGuire Nuclear Station Docket Numbers 50-369 and -370 Catawba Nuclear Station

~

Docket Numbers 50-413 and -414 Technical Specification Revision to Change Method of Measuring Reactor Coolant System Flow Rate; Supplemental Information By letter dated January 10, 1994, Duke Power Company submitted a license amendment application to change the method by which reactor coolant flow is measured for Technical Specification surveillances ,'

at McGuire and Catawba Nuclear Stations. The Staff responded with a request for additional information (Reference letter, May 3, 1994, R. E. Martin to M. S. Tuckman). Enclosed are responses to the questions presented in the RAI.

Please note that Attachment 1 to the Enclosure contains information which is proprietary to Westinghouse Electric Corporation. This information, together with affidavits supporting the proprietary designation, was originally submitted in support of license amendments to remove the RTD Bypass System at McGuire (submittal dated October 25, 1985) and Catawba (submittal dated July 22, 1987). This present calculation, to support the change of RCS flow rate measurement technique, uses many of the same data, which are considered to remain proprietary. These data are indicated by brackets. Those numbers which were developed by Duke to support

& this submittal need not be considered proprietary. Attachment la j@ of the Enclosure contains a non-proprietary version of Attachment pgo- 1.

oo

$$ Please note also that margin to the Technical Specification minimum flow limit at McGuire and Catawba remains small; Catawba Unit I has Id already failed to meet minimum flow as measured by the present 5@ method, and has received approval to use the proposed method for 1 gC the current cycle; this approval allowed the unit to reach 100%

& power after being restricted to 98% for several weeks at the SE beginning of the cycle. Both of these units will face RCS flow

& A a- '

measurements in the upcoming months. It is requested, therefore,

,_ _ canas ur. M i (Aking /iA'rW / //J/' }

U. S. Nuclear Regulatory Commission September 15, 1994 Page 2 that review of this amendment request' proceed as expeditiously as Staff resources will permit, to minimize potential derating caused by inadequate measured flow.

If any additional information is required, please call Scott Gewehr at (704) 382- 581.

f f . \u2 %x M. S. Tuckman cc: Mr. V. Nerses, Project Manager Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Mail Stop 14H25, OWFN Washington, D. C. 20555 ,

Mr. R. E. Martin, Project Manager Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Mail Stop 14H25, OWFN Washington, D. C. 20555 Mr. S. D. Ebneter, Regional Administrator <

U.S. Nuclear Regulatory Commission - Region II 101 Marietta Street, NW - Suite 2900 Atlanta, Georgia 30323 G. F. Maxwell Senior Resident Inspector McGuire Nuclear Station R. J. Freudenberger Senior Resident Inspector Catawba Nuclear Station

,4

- . , _ s. n aa m. s o .s a m a v.u. n .e a . ._x ,.

i l

1 l

a t

REQUEST FOR ADDITIONAL INFORMATION REACTOR COOLANT SYSTEM (RCS) FLOW MEASUREMENT METIIODOLOGY j (1) Ilow are the elbow tap transmitters calibrated? What is the elbow tap transmitter calibration experience at Catawba and McGuire? The question pertains to changes  ;

found during calibrations, the ermdition of the transmitters and ccmnecting tubing, ,

and procedures pertinent to the calibration, i i

Response to Question 1: )

The reactor coolant flow transmitters at Catawba and McGuire are calibrated each 18  :

months per Technical Specification 3/4.3, typically each refueling outage. The method )

used for calibration is a standard technique which is used for all I&C equipment. Two {

techniques are available for use in calibration of the transmitters, the wet rig (wet j calibration) or the dry calibration (evacuation of all liquid in transmitter cell and apply i direct pressure). Standard calibration procedures and practices are followed to ensure i transmitter calibration accuracy. Transmitter calibration and data collection requires the  !

disconnecting of instrument tubing from the transmitter for the connection of calibration  !

equipment. This may allow small amounts of air to be intmduced to the impulse lines and f transmitter when using either of the above techniques. This small amount of air will not }

i affect the accuracy of the calibration since the air will rapidly be absorbed by the highly pressurized (>2000 psia) reactor coolant system, f The experience at McGuire and Catawba with the elbow tap transmitter drift shows that the transmitter drift fnnn cycle to cycle is generally very small. Since "as found" and "as j left" data is recorded during each calibration of these transmitters, excessive drift in individual transmitters can be detceted and evaluated to detennine if a transmitter should be repaired or replaced. To date no excessive transmitter drift has been noted. Some #

minor adjustments are made during the calibration to " tighten up" the transmitter output even though it may not be out of tolerance; and relatively few failures have occurred which have required transmitter replacement.

1 (2) What is the uncertainty associated with the indicated differential pressure, including j

tubing run influence, the transmitter, and any other data processing that is used in the management of data between the transmitter and the point-of-use.

I Response to Question 2: .!

l i

Since no coolant flow is occurring in the instmment tubing between the laps and the  ;

r transmitter, tubing run influences have little if any effect and are not a concem. As '

discussed in the response to question I above, small amounts of air which may be l introduced into the instrument tubing will not affect the calibration accuracy. The . }

uncertainties associated with the flow measun: ment channels are pmvided as a response to i questions 14 & 15 below. j i

f i

t f

i

l i

(3) In regard to hot leg temperature nonuniformities:

• Please provide w hatever information is available to the licensee regarding the expected temperature distribution in the hot leg as a function of power and as a function of core loading (including burnup).

= What is the experience of" thermal / flow" switching in the upper plenum as it affects hot leg temperatures? What is understood regarding this phenomenon?

. Ilow are differences hetween true average temperature and indicated temperature considered in using temperature data as indicated by the RTDs during operation? For example, what is done to compensate for changes between indicated temperature and average temperature with respect to reactor trip and the other uses of but leg temperature for plant operation?

P Response to Question 3:

The only infonnation available regarding the temperature distribution in the hot leg is from the hot leg RTDs themselves. This infonnation does not provide enough data points to give an accurate temperature profile. The RTD temperature data indicates that the temperature in the upper section of the hot leg pipe is hotter than the lower section. This is interpreted to indicate that the hotter water exiting from the central region of the core is  ;

tuming without significant mixing and entering the hot leg upper half. Similarly, the '

cohler water in the periphery regions of the core is tuming without significant mixing and entering the lower half of the hot leg. The calorimetrics which have been perfonned at the end of a cycle indicate, as expected, that the flattening of the temperatum pmlile due to burnup does cause a change in the hot leg streaming such that the calculated flows increase from the beginning of cycle calculated flows. This confirms the expectation that changes in the hot leg temperatum profile are related to changes in the core exit temperature profile.

The phenomenon of thermal / flow switching in the upper plenum has occurred at McGuim Unit I and Catawba Unit 2. What is understood about this phenomenon is that for whatever reason hot and cold water is " switched" between adjacent loops causing a spike in temperature indications from one or more RTDs in the adjacent hot legs. This phenomenon has been observed primarily in Westinghouse plants with inverted top hat upper intemals assemblics and is an aperiodic short lived phenomenon, usually lasting only a few seconds to a minute in duration. The upper plenum themial/ flow switching is characterited by a short lived increase in the indicated Thot average in one hiop of an adjacent loop pair and a simultaneous decrease in the indicated That average of the other loop in the pair. In addition, similar but smaller temperature changes are observed by the cold leg RTDs aller the appropriate loop transit time. This indicates that the temperature changes caused by this mmmaly are real and caused by the actual change in hot leg water temperatures resulting from the thermal / flow switching.

No changes or corections to the indicated RTD temperatures is made to compensate for the potential diffemnce between indicated Tavg and actual Tavg during phmt operation.

The other primary uses for temperatum data as indicated by the RTDs during plant operation is the mactor control system and the overtemperature AT (OTAT) and

overpower AT (OPAT) trip setpoints. The reactor control system uses an indicated auctioncemd high Tavg to contml the movement of the contml nxis. Thus, rod control will be conservatively performed utilizing the k>op with the highest indicated Tavg. An increase in hot leg streaming will likely cause an incmase in indicated Tavg for a panicular loop resulting in the plant being controlled to an indicated Tavy which is higher than the actual plant Tavg. This is conservative with respect to DNB and fuel centerline temperature protection events. Events where low Tavy may be a concern have been i evaluated and found to be sufficiently conservative for an indicated Tavg 51.0 F higher than the actual Tavg. A higher indicated Tavg and lower actual Tavg is conservative with respect to the OTAT and OPAT trip functions since during a transient, when the indicated Tavg reaches the trip setpoint, the actual Tavg will be less than the setpoint, pmviding additional margin to DNB.

(4) What is the licensee's experience regarding cold leg temperature nonuniformities? If cold leg temperature differences hase been observed, n hat is the source of those dilTerences?

Response to Question 4: p McGuim and Catawba have one cold leg RTD located in the top and one spare RTD located at various angles, depending on the particular unit and coolant loop, from the top of each cold leg pipe downstream of the reactor coolant pump. These RTDs generally indicate cold leg temperatures within i 1.0 F of the cold leg average temperature as calculated from the average of the two RTDs. This small difference in temperature is  ;

probably attributable to some small temperatum gradient, originating from the different steam generator tube lengths in the bundle, passing through the reactor coolant pump, or possibly some gradient attributable to the pump itself such as pump heating or centrifugal force on water with varying densities. Generally, for an individual loop, these cold leg temperatures change in unison with one another, usually related to actual temperature changes. Any stmaming effect in the cold leg is small and the temperature differences between RTDs do not appear to change significantly from cycle to cycle. .

t (5) Are there any performance data regarding before/after cleaning of the feedwater venturis? What is the basis for assuming the venturi characteristics do not change with time, assuming the venturis are clean? What is the quantitative effect of crud buildup?

I Response to Question 5:

Duke has never donc any precision testing just before and then just after cleaning. The feedwater venturis are inspected during each outage before and after they are cleaned. The inspections have never detected any crosion around the upstream taps or on the venturi itself. The ventmis have thus far been found to be smooth and unmarred by wear or  ;

corrosion, which would likely show up as pits and surface irregularitics. As long as the venturis stay smooth and free of emsion, the flow characteristics will remain unchanged.  ;

Crud buildup or venturi fouling results in an indication of higher than actual feedwater flow, which results in an indication of higher than actual secondary power, which results in

i an indication of higher than actual RCS flow Since this is non-conservative with respect to RCS Ilow, corrections or allowances for venturi fouling are used to ensure the RCS i flow is conservative. ASME flow noules are used as a standani to nonnalize the process feedwater venturi flow measurement and correct forIbuling. These noules are periodically cleaned (typically once per fuel cycle depending on run time) and maintained  ;

in a clean condition by valving these nonles out of service when not in use ihr testing.

Confidence in our ASME flow noule characteristics / flow measurement is based on MNS

& CNS noule calibration histories (which provide evidence that the nonic discharge ,

coef ficient will retum to its original value following cleaning), periodic cleaning / inspection of the nonics and the fact that the nonles are only placed in-service for testing purposes.

Fouling levels on the feedwater venturis have resulted in small increases (typically less  ;

than 0.5%) in calculated flow over the last couple of cycles. This low level of fouling accumulation is attributed to improvements in feedwater chemistry including the use of a reverse-osmosis filtering system.

(6) What data are used Ihr determination of power level with respect to operation and  ;

plant control?

Response to Question 6:

Power level is detennined by continuously perfonning a secondary side calorimetric on the plant compulcr. This is calculated using inlet feedwater flow, inlet feedwater temperature, and outlet steam pressure. Corrections are made for S/G blowdown and cycle heat losses and heat gains such as insulation losses and reactor coolant pump heat addition.

i Feedwater venturi fouling is detennined by trending various parameters such as turbine first stage pressure, main steam flowrate, feedwater pump suction flows and primary .

calculated power vs. secondary calculated power. When trends show an excessive amount '

of fouling, a fouling test is perfonned. This consists of valving in a precision set of ASME noules and comparing their reading with the feedwater venturi readings corrected for any inlet and outlet flows that occur in between. The amount of fouling measured by this test is input to the plant computer to correct the power calculation.

(7) What data are available to support a conclusion that the components exposed to water in the elbow tap differential pressure instrumentation do not change over the life of the plant? (The question does not apply to transmitters.)

1 Response to Question 7:

As discussed in the Technical Specification change submittal, specific phenomena which might affect the elbow meter repeatability were examined. These phenomena wen: found to have little if any shon or long term effects on the repeatability of the elbow tap flow indications. Fouling as expetienced with venturi meters is not a concern since the process which causes this ibuting is not present in the cold leg cibow. Deposits in the RCS from impurities in the reactor coolant are expected to be small or nonexistent. Most deposits of impurities in the reactor coohmt are expected to occur in the hottest portions of the RCS i and in regions experiencing the lowest flow. Any deposits in the RCS piping willaffect the interior of all the RCS piping and not just the region of the elbow taps. This will cause a i

i

real flow change arxl will be reflected in the elbow tap APs. If preferential deposits wem to occur in the region of the laps, the reduction in pipe diameter would be extremely small in comparison to the diameter of the cold leg cibow (31"). Fouling or deposits within the instrument tubing between the elbow tap and the differential pressure instrument is not a concern since no flow is transmitted within the tubing. In addition, crosion (flow accelerated corrosion) is not a concem since the velocity of the RCS fluid is small relative to velocitics known to cause cmsion in stainless steel. Emsion of the RCS piping will be small or nonexistent during plant life. Any changes in clbow diameter as a result of small amounts of crosion will not be significant with regard to the 31" diameter of the cold leg pipe. 'Ihc clbow taps have been positioned on the elbow in such a manner that velocity pressure components and turbulence effects are minimized while not impacting the differential pressure indications. Since the piping between the elbow tap and the differential pressure instrument is used to transmit the pressure signal only, any velocity component of the turbulent RCS flow which is impaned to the tap location will result in random noisc in the pressure signal.

Every attempt will be made to check for effects which may affect the calibration of the elbow meter. Ilowever, unless these effects are large, such as a plugged cibow tap, detec' 'n of small changes which will affect the calibration of the elbow taps will continue to be milicult since the effect on flow will be small and nearly undetectable. Comparisons to the analytical flow model prediction of flow will be used to detennine the extent to which the elbow tap calculated flow reflects actual flow changes.

(H) Please provide a comparison of analyses and observed RCS loop / total flow rate behavior as determined from elbow tap differential pressure indications and from calorimetric testing for both Catawba and both NicGuire units. Sufficient data should be used to establish that:

. Changes during refueling outages, such as steam generator (SG) tube plugging, SG tube sleeving, and core changes are fully compared for the operating history of the plants. ,

e llehavior during operation is fully described.

Use of differential pressure information of the type previously provided for parts of two cycles for Cataw ba Unit I to calculate flow rates for purposes of the comparison is acceptable. Ilowever, the staff requests a more extensive comparistm for Catawba Unit I and also requests that the comparison be done for the other afTected Duke plants. The flow rate comparison should be done on a "best estimate" basis.

In addition, please provide:

• a description of the analysis used for the comparisons, e discussion of difTerences between analysis and behasior based upon plant data, and

. discussion of differences in w hat appear to be identical hmps,if such differences are observed.

Resportse to Question 9:

Analytic HCS Flow Model Description An accurate RCS flow prediction requires a nulel which can determine flow without relying on hot anal coki leg temperature inputs. The reactor coolant flow may be calculated by first detennining the system head loss curve for a reactor coolant kop with a given configuration. Once the system head loss curve has been established it is compamd to the reactor coolant pump perfonnance curve to determine the intersection of the two curves. The intersection of the two head curves will define the system operating point, the point where the kop system head loss matches the head produced by the reactor coolant pump. 'The flow is then obtained fmm the pump flow / head curve. By establishing the system head losses for each loop and accounting for changes in these system losses due to plant changes over time, e.g., steam generator tube plugging, modifications and different fuel designs, a reasonably accurate RCS llow, and therefore the change in flow due to system changes, may be calculated for each plant configuration.

The first step in calculating the RCS flow is to detennine the system head loss for a given assumed flow. Table 1 lists the pressure drops for the RCS cxcept for the steam generators.

Table i HCS I oop Prusure Drops Loon Section Afdni Reference Flow ilot Leg 2.09 10,277 lbm/sec/ loop Pump Suction 40 Elbow I.18 10,277 lbm/sec/ loop Pump Suction Straight Pipe 0.31 10,277 lbm/sec/ loop Elbow Tap 90 Elbow 1.28 10,277 lbm/sec/ loop Pump Suction 90 Elbow 1.28 10,277 lbm/sec/ loop RCP Weir Plate 2.0 100,900 ppm / loop Cold Leg 2.54 10,277 lbm/sec/ loop Inlet Nonle 10.01 10,277 lbm/sec/ loop Downcomer 0.36 39,587 lbm/sec Downcomer exit 2.68 39,587 lbm/sec ,

Core Support 2.20 39,587 lbm/sec Lower Core Plate 7.24 39,587 lbm/sec Bottom Noule 3.2 101,700 gpm/ loop Core 17,3 101,700 gpm/ loop Top Nonle 1.1 101,700 gpm/ loop Upper Core Plate 3.99 39,587 lbm/sec Outlet Nonic 2.33 10,277 lbm/sec/ loop

'lhennal Driving llead -1,30 10,277 lbm/sec/ loop These pressure drops are adjusted to account for the flow differences between an assumed flow and the reference flow by multiplying a ratio of the flows squared to the above pressure drops as shown below.

Since N' = Flow 2

'2 f

New Flow y,"" g,"

( Reference Flows

'Ihis adjustment of the pressure dmps is made for all the above pressure drops except the thennal driving head. which is assumed to be relatively constant for full power operation.

In addition to the above adjustments to the primary side pressure drops, adjustments are also made to account for changes in plant configurations, i.e., barrel baffle region downHow to upilow modification, convening from Westinghouse OFA to Mark BW fuel, etc.

McGuire Units 1 & 2 were originally designed with downward RCS coolant flow in the barrel baffle region. This downward flow parallels the downcomer and produced undesirable balHe jetting due to the differential pressures between the barrel bafHe region and the core region. The core barrels were modified such that the flow in the barrel haftle region is in the upwant direction parallel to the core Dow. The changes in the APs as a result of this modification were largely due to the changes in flow distribution in the downcomer and core regions. The plugging of holes in the upper core barrel might have changed the pressure drop in the downcomer region slightly, but since the pressure drop in the downcomer is small (0.18 psi), the change in pressure drop due to plugging these holes is therefom not significant and the change in pressure drop is neglected. Holes plugged in the lower core plate are in the barrel baffle region amund the periphery of the lower core plate and do not change the geometry of the plate in the core flow path. Therefore, changes in the APs for the lower core plate are made by adjustments to the core flow fraction. The flow fractions used to adjust the APs in the downcomer and the core regions are given in Table 2 below for McGuire Units 1 & 2.

Modification of the McGuire Units 1 & 2 reactor vessel core barrels fmm an upflow configuration to a downDow configuration were accounted for after the Unit i end of cycle 7 refueling outage and Unit 2 c:id of cycle 6 refueling outage.

Table 2 McGuire RCS Flow Fractions l

Downflow Configuration Upflow Configuration Flow Fraction Flow Fraction Unit Downcmner Core Heeion Ibwncomer Core Recion McGuire Unit 1 0.963 0.925 0.964 0.931 McGuire Unit 2 0.963 0.925 0.963 0.930 l l

Catawba Units 1 & 2 were designed with an upilow barrel hafne region and therefom no I adjustment to the downcomer and core region APs is necessary. The flow fractions for Catawba Units 1 & '2 are given in Table 3 below.

, , .~. -. . . .- . - -. - _- -

i

)

)

I Table 3 Catanba RCS Flow Fractions .,

1 Downcomer Core Renion l t

Catawba Units 1 & 2 0.966 0.943 Adjustments to the APs in the core region are also made to account for different fuel -  !

designs. A new fuel vendor (B&W) is providing fuel for McGuire and Catawba Nuclear '

Stations which has replaced the original Westinghouse fuel. Since approximately 1/3rd of the core is replaced during a mfueling outage, the cores consist of 1/3rd,2/3rds, and full cores of B&W fuel as the new fuel is introduced. The B&W Mark-BW fuel with debris-trapping bottom nonics (DTBN) has an approximately 2.4% lower pressure drop than the  !

Westinghouse Standard and OFA fuel. The pressure drops for each fuel type at 101,700 l gpm/ loop are given in Table 4 below.

Table 4 Core Region Pressure Drops (Full Cores) i i

Westinghouse Std & OFA B&W Mark-HW w/DTBNs AP, psi (from Table 1) AP, psi Bottom Nonle 3.2 3.12 Core 17.3 16.88 Top Nonle 1.I 1.07 Total - 21.6 21.07 ,

1 For cores with 1/3nt and 2/3nis Mark-BW fuel a weighted avemge of the pressure drops for cach fuel type is used. The pressun: drops used for each core configuration are given in Table 5 below.

i Table 5 Core Region Pressure Drops (Partial Cores) 1/3rd Mark BW fuel 2/3rds Mark-BW fuel AP, psi AP, psi Bottom Nonle 3.17 3.15 Core 17.16 17.02 ,

Top Nonle 1.09 1.08 l Total 21.42 21.25 j i

1 l

l l

The steam generator pressure drops are then calculated using the fbilowing input information arxl equations.

Table 6 Steam Generator Input For Pressure Drop Calculation Flow rate, w Assumed, ppm ID of steam generator tube, Dnim 0.664 inches Coolant average temperature 590.70 F Average fluid density, pave 43.94 lbm/ft3 @2250 psia Coolant inlet temperature 620.00 F Steam generator inlet fluid density, pin 41.29 lbm/ft3 @2250 psia Coolant outlet temperature 561.30 F Steam genemfor outlet fluid density, pout 46.12 lbm/fl3@2250 psia Inlet nonic ID, Djnn, 31.00 inches Outlet noule ID, Donn, 31.00 inches Percent steam generator tube plugging Initially 0.0 Number of steam generator tubes, nnihe, 4,674 MNS-1, MNS-2, CNS-1 4,570 CNS-2 Friction factor, f 0.014 Average tube length,Initx, 55.9 ft MNS-1. MNS-2, CNS-1 686.64 in. CNS-2 The steam generator pressure drops are calculated using (bnns of the continuity equation and Darcy's fonnula given below.

Tube water velocity,

" '"^

ft / see r,,a., = 0. 408< U ,s,a,' >

s Inlet nonic water vehicity, g,

r,,,,, = 0. 408 , ft / see

( O,,,,,, s Outlet nonic water velocity, f T n,

r,,,,,,. = 0. 408 , ft / see

\

IA,wt v )

Pressure dn>p in tubes, AP,,,,,, = 0.000216 x f x /,,,,,, x p, x ( w/n,",,;)~

psi tui.e The friction factor, f, is obtained frrun a friction factor chart utilizing the information below:

c(drawn tubing) = 0.000005 c 0.000005

-- - 9. 0 x 10-3 D 0.664 in./12 in / ft f = 0.012 for completely turbulent flow Tube entrance & exit pressure drop, AP,,,, = (K,,, x 0.0001078 x p, x v,,,,,,' ) + ( K,,,,, x0.0001078 x p,,,,, x v,,,,,' ) psi where K,,, = 0.50 and K,,,, = 1.0 SG noules pressure drop, APy ,,,,, = (K,,, x 0.0001078 x p,,, x v,,,,,,') + ( K ,,, x 0.0001078 x p,,,,, x r,,,,,,,') psi where K,,, = 0.427 and K',,,,, = 0.266 l

The steam generator tube pressure drops are affected by the number of tubes plugged  ;

and/or sleeved during each outage. Plugging and sleeving of tubes in the steam generators ,

causes a reduction in the flow area which results in an increased pressure drop across the j steam generator tubes. The number of tubes plugged and/or sleeved expressed as a percentage of steam generator tubes (4674 for McGuire Units 1 & 2 and Catawba Unit I and 4570 for Catawba Unit 2) is used to calculate the How area through the steam generators in each loop. The calculated steam generator tube plugging percentage assumes i that 18 sleeves are equivalent to i plugged tube. The steam generator tube plugging l performed through the Technical Specification n' vision submittal date for McGuire and Catawba are given in Tables 7 - 10 below.

l l

l

I Table 7 McGuire Unit 1 SG Tube Pluneine Percentanes -

Equiv Equiv Equiv Equiv Equiv Cum Cum Cum Cum Cum Plug Plug Plug Plug Plug OUTAGE EOC A% 11 % C% D%  % Tubes Aug-81 0.428 0.449 0.428 0.428 0.433 Mar-82 0.471 0.449 0.428 0.428 0.444 Jul-82 0.492 0.449 0.428 0.428 0.449 -

- Nov-82 0.599 0.449 0.449 0.428 0.481 >

Feb-83 1 0.642 0.449 0.449 0.428 0.492 May-85 2 0.642 0.492 0.471 0.449 0.513-May-86 ' 3 2.696 2.546 2.567 3.231 2.760 i Sep-87 4 3.252 3.659 2.931 4.044 3.471  !

Oct-88 5 3.851 4.236 3.594 4.878 4.140 Mar-89 4.172 4.750 3.937 5.285 4.536 Jan-90 6 4.516 6.005 5.991 6.062 5.643 ,

Sep-91 7 5.281 6.585 6.384 6.610 6.215 Jan-92 6.133 7.651 7.560 7.270 7.153 May-92 7.841 8.079 8.330 8.253 8.126 j Apr-93 8 8.587 9.35. ' 9.269 9.833 9.268- l Aug-93 9.528 9.874 9.739 10.67 9.953 i 4

i l

Table 8 l McGuire Unit 2 SG Tube Pluccine Percentanes  !

I' Equiv Equiv Equiv Equiv Equiv Cum Cum Cum Cum Cum l Plug Plug Plug Plug Plug  !

OUTAGE EOC A% 11 % C% D%  % Tubes  :

May-83 0.021 0.(XX) 0.000 0.(XX) 0.005  !

Jan-85 1 0.021 0.(XXI 0.000 0.(KX) 0.005 j '

Dec-85 0.021 0.000 0.(XX) 1.070 0.273 Mar-86 2 2.460 2.439 2.482 2.503 2.471 l May-87 3 3.209 . 3338 3.124 3.231 . 3.225  !

Jun-88 4 4.086 4.086 4.236- 3.487 3.974 (

Jul-89 5 5.327 4.792 4.942 4.536 4.899 .;

Aug-90 6 5.512 5.079 5.229 4.710 5.132 Jan-92 7 5.874 6.023 5.616 5.450 6.741 j May-92 6.128 6.279 5.872 5.705 5.996 l Jul-93 8 7.305 7.306 6.621 6.754 6.996  ;

Sep-93 7.669 8.141 6.941 7.075 7.456 i

Table 9 Catawba Unit i SG Tube Pluonino Percentanes Equiv Equiv Equiv . Equiv Equiv Cum Cum Cum Cum Cum Plug Plug Plug Plug Plug OUTAGE EOC A% 11% C% D%  % Tubes Jun-85 0 0.064 0.150 0.043 0.064 0.080 Jan-86 1 0.064 0.150 0.043 0.064 0.080 Nov-87 2 0.064 0.150 0.(43 0.064 0.080 Aug-88 0.064 0.150 0.043 0.107 0.091 Dec-88 3 0.492 0.385 0.300 0.513 0.423 Feb-90 4 2.2(4 0.471 1.284 1.626 1.396 Apr-91 5 2.598 0.941 4.151 2.884 2.643 Aug-92 6 4.438 1.562 6.025 4.l(4 4.032 Nov-93 7 6.953 3.466 11.168 10.056 7.911 Table 10 Catawba Unit 2 SG Tube Pluccinn Percentanes Equiv Equiv Equiv Equiv Equiv Cum Cum Cum Cum Cum Plug Plug Plug Plug Plug OUTAGE EOC A% 11 % C% D%  % Tubes Aug-86 0.175 0.175 0.175 0.175 0.175 Sep-86 0.197 0.328 0.197 0.2.84 0.252 Feb-88 1 0.284 0.350 0.197 0.328 0.290 Feb-89 2 0.328 0.460 0.219 0.328 0.334 Jun-90 3 0.525 0.481 0.263 0.481 0.438 Oct-91 4 0.678 0.481 0.350 0.503 0.503 Feb-93 5 0.985 0.613 0.635 0.722 0.739 The pressure drops calculated for each loop and steam generator are converted to a head loss using the equation:

f 21 APpsi x 1441 Head, ft = ' ,y, P

ft'

'Ihc k)op and steam generator head losses are then summed together to obtain a total system head loss amund each loop. This total head loss is then used to calculate the flow at the point where the system head curve crosses the reactor coohmt pump curve. The

point at which the two curves cross establishes the system operating flow for the given configuration.

Ec reactor coolant pump head /How curves developed fmm int perfonnance data are given in Table 11 below. An equation for the pump curve is detennined by fitting a curve to the data in the region of 50JXX) to 130,(XX) gpm.

r Table 11 Reactor Coolant Pump Ilead Curves McGuire Units 1 & 2 and Catan ha Unit 2 Catawba Unit 1 Flow. corn llead. feet IIead. feet 0 516 519 10JXX) 495 509 20,(XX) 470 484 30JXX) 437 447 40,0(X) 409 410 50JXX) 391 395 60,(X X) 380 382 70JXX) 384 387 80/XX) 366 371 90JXX) 329 339 100,(XX) 284 292 10ljX)0 279 286 110JXX) 234 240 120jXX) 184 181 130JXX) 127 122 A fi>urth-order curve fit was perli>mied for each of the sets of data atuve in the region of 50JXX) to 130JXX) ppm and the resulting equations are:

McGuire Units 1 & 2 and Catawba Unit 1 2 d y = 2231.8399x- 16.60783x + 0.0486987x' - 0.0000519092x + 28109.989 Catawba Unit 2 y = 1811.4992x - 14.082053x2 + 0.0425535x' -0.0000465846x' + 51565.277 The equations for the reactor coolant pump curves are used to calculate the flow that the pump can pmvide at the head loss detenuined for the input assumed flow. Since the system head must match the head produced by the reactor coolant pump, the input )

assumed flow is compared with the calculated pump flow to detemiine whether they match. if the flows do not match, a new assumed flow, halfway between the previous assumed flow and the calculated pump Dow, is selected and the process is repeated until

the k>op flows are equal. This defines the point where the system head curve intersects the tactor coolant pump head curve and defines the system Dow for a given plant configuration.

Plant Data and Analytical Model Comparistms McGuire Unit 1:

McGuire Unit 1 is the one unit that thus far does not appear to be greatly affected by the hot leg streaming phenomenon. The RCS flow as mea.ured by the calorimetric has trended very closely to the changes in full power AT, as can be seen in Figure 1. In addition, the changes in RCS llow as detemiined by the calorimetrics has generally trended with the elbow tap APs with some small hot leg streaming effects. The trends in flow as detennined fn>m the indicated citx>w tap APs (Figure 2) and the analytical How model (Figures 4 & 5) are in g(xxl agreement with one another (Figures 4 & 5 differ only in the Y-axis scale with the scale on Figure 5 matching the scale of Figure 2 for casy comparison). The individual h>op flows given in Figure 2 indicate an unequal hu>p flow distribution at McGuire Unit I since plant startup. Generally, this is the case for the other three Westinghouse units as well, and is likely the result of differences among the as-installed huip components, e.g.. piping, steam generators, and reactor coolant pumps.

110 wever, the average of the individual h>op flows in Figure 2 trend very close to the  !

individ'ial hw>p Hows as given in Figure 5. In addition, the individual kx>p Dows in Figure 2 show good agreement with the changes in flow expected as the result of plant changes. ,

The hx>p flows in Figure 5 are almost identical during early plant life since the analytical  !

malel h>op flows were detennined using the same reactor ax)lant pump curve and loop j geometry for allloops.

1 All cibow tap transmitters weir replaced during the September 1987 EOC 4 infuchng outage. This resulted in a step increase in the elbow tap AP indications from those of the previous transmitters (Note: the AP flows in Figure 2 have a correction included for calorimetric APs before September 1987). Once corrected, the elbow tap AP ilow (Figure 2) for early plant operation agrecs well with the analytical model (Figure 5) and calorimetric (Figure 1) flows.

The upflow malification was made during the September 1991 EOC 7 refueling outag:

along with the introduction of the first batch of B&W fuel. The upflow modification was expected to change the RCS How distribution through the core region slightly and not effect the total flow result significantly. The B&W fuel was expected to increase flow slightly due the decreased flow resistance in the core. The increase in flow due to the new fuel was expected to be offset by steam generator tube plugging. The flow in the plant as indicated by the elbow tap APs actually increased slightly after this outage. This is likely due to the new fuel having a greater effect on the Dow than anticipated and panially due to l the analytical flow model assumption that 1/3rd of the core would be replaced when slightly more was in fact replaced (= 407e). The April 1993 EOC 8 refueling outage resulted in a flow decrease as indicated by the calorimetric, cibow tap AP flow and analytic 1 flow model, due to any flow increase from the second batch of B&W fuel being more than l offset by substantial steam generator tube plugging.

Figure 6 shows the daily average elbow tap AP data for McGuire Unit I back to June 20.

1992. This data covers cycles 8 & 9 and has been m<xlined to climinate bad data and data at off piwer days (<99% power). The data shows that the elbow tap APs are very steady with no major drilling trends when the transmitters are operating properly. Channel 1, Loops 1 and 3, transmitters failed during cycle 9 and were replaced during the last outage.

The elbow tap data in Figure 6 shows changes due to SG tube plugging following the three outages during this time frame. Unit I went through a refueling outage fmm March,1993 through June 1993 arul steam generator outages from August 1993 through November 1993 and February 1994. The changes in loop flow resistances are reflected in the elbow tap AP changes correspmding to these time periods, with the largest change occurring after the February 1994 outage where significant SG tube plugging occurred.

McGuire Unit 2:

McGuire Unit 2 has experienced a signiDeant flow impact as a result of hot leg streaming.

The RCS flow as measured by the calorimetric has trended very closely to the changes in full power AT as can be seen in Figure 7. The calorimetric perfonned February 1985,just af ter the EOC 1 refueling outage, indicated a large increase in RCS flow which was not accompanied by significant plant changes. The large increase in flow was the result of a decrease in indicated AT of approximately 1 F as can be seen on Figure 7. This decrease in AT was investigated by Duke and Westinghouse and found to be caused primarily by hot leg streaming changes and a reduction in themial power caused by feedwater venturi fouling. For the same calorimetric the individual hiop flows as detennined by the elbow tap aPs (Figum 8) and the armlytical flow model (Figures 10 & 11) do not show a corresponding increase in flow. Note. Figures 10 & 11 differ only in the Y-axis scale with the scale on Figure 1 I matching the scale of Figure 8 for easy comparison. In addition, no changes in plant geometry (SG tube plugging etc.) were made prior to this calorimetric which would indicate a significant flow increase or decrease (Figures 9 & 10).

The calorimetric flows following the February 1985 calorimetric have basically tmoded downwant consistent with the increasing trend in AT, but in excess of that expected by plant geometry changes. Since plant startup the calorimetrics have indicated that the total RCS How has dropped appmximately 11,000 gpm (Figure 7) whereas the elbow tap and analytical model flow has indicated a drop in total flow of approximately 4.500 gpm (Figures 8 & 10). This indicates a significant impact fmm hot leg stmaming as it shows that 6.500 ppm of the calorimetric How decrease can be attributed to this phenomenon.

The upuow modification was made during the August 1990 EOC 6 refueling outage. As mentioned in the discussion of the uptiow modification at McGuire Unit 1, this nuxlification was expected to change the RCS flow distribution through the core mgion and not effect the total flow result significantly.110 wever, the calorimetric flow following this outage decreased approximately 7,500 gpm to just above the then Technical Specification minimum measured flow limit of 385,(X)O gpm (Figure 7). Since the steam generator tube plugging percentage during the EOC 6 refueling outage was small this magnitude of flow decrease was not expected. The elbow tap and analytical flow nxxici for the same time frame indicate a flow decrease of = 500 ppm which is consistent with the percentage of steam generator tube plugging which occurred during the outage.

An end of cycle calorimetric was performed just prior to the Jarmary 1992 EOC 7 refueling outage to detennine if core bumup resulting in the flattening of the core exit temperature profile would increase the calorimetric measured flow. As expected the flow detemiined from this calorimetric did increase (= 3JXX) gpm). His indicates that the unsubstantiated flow decreases measured by 11 calorimetrics are the result of hot leg streaming changes. as suspected.

The first batch of B&W fuel was introduced during the January 1992 EOC 7 refueling outage. The 11&W fuel was expected to increase flow slightly due the decreased flow resistance in the core. This increase in flow due to the new fuel was expected to be offset by steam generator tube plugging as shown in Figures 10 & 11. The flow in the plant as indicated by the elbow tap APs actually increased slightly after this outage. His is likely due to the new fuel having a greater effect on the flow than anticipated and panially due to the analytical flow nuxlel assumption that 1/3rd of the core would be replaced when slightly more was in fact replaced (= 37%). The July 1993 EOC 8 iefueling outage, which involved a second batch of B&W fuel and additional steam generator tube plugging resulted in a flow decrease as measured by the calorimetric. The elbow tap APs and analytic model flows indicate no decrease and a slight decrease respectively.

Figure 12 shows the daily average cibow tap AP data for McGuire Unit 2 back to September 22,1993. This data covers cycle 8 and has been modified to climinate bad data and data at off power days (<99% power). The data shows that the elbow tap APs are very steady with no major dritting trends when the transmitters are operating pmperly.

The elbow tap data in Figure 12 shows changes in hiop flow resistances due to SG tube plugging perfonned during the SG tube leak outage which ended in October 1993.

Catawba Unit 1:

Catawba Unit I has experienced substantial flow impact as a result of hot leg streaming in recent calorimetrics. De RCS flow as measured by the calorimetric has trended very closely to the changes in full power AT as can 1c seen in Figure 13. The calorimetrics perfonned between plant startup and February 1986, following the EOC 1 refueling outage, indicate an unsubstantiated increase in RCS flow which was not accompanied by significant plant changes. The increase in flow was largely the result of AP transmitters which drifted excessively. This increase in flow was indicated by the calorimetric (Figure

13) and the elbow tap AP flow (Figure 14) since the drift in cibow tap AP transmitters affects both melluxis of flow measurement. The analytical flow malci does not predict any flow changes during this period as no changes to plant geometry were made (Figures

15. 16 & 17). Note, l'igures 16 & 17 differ only in the _Y-axis scale with the scale on Figure 17 matching the scale of Figure 14 for casy comparison ne calorimetric flows following the November 1987 calorimetric have basically trended downward consistent with the increasing trend in AT, but in excess of that expected by plant geometry changes. Since plant startup the calorimetrics have indicated that the total RCS tlow has dropped approximately 21JXX) gpm (Figure 13) whereas the elbow tap and analytical model flow have indicated a drop in total flow of appmximately 9,000 gpm and 5JXX) gpm respectively (Figures 14 & 16). Ec difference between these flow indications is the excess elbow tap AP transmitter drift during the early calorimetrics which caused the

unsubstantiated indicated flow increase This indicates a significant impact from hot leg simaming as it shows that 12,(XX) to 16,000 gpm of the calorimetric flow decrease can be attributed to this phenomenon.

The first batch of B&W fuel was intnxtuced during the April 1991 EOC 5 mfueling outage. The B&W fuel was expected to increase flow slightly due the decreased flow resistance in the core. This increase in flow due to the new fuel was expected to le offset by steam generator tube plugging as shown in Figures 16 & 17. In fact, the flow in the plant as indicated by the cibow tap APs and the calorimetric actually incmased slightly after this outage. This is likely due to the new fuel having a greater effect on the flow than anticipated and panially due to the analytical flow nulel assumption that 1/3rd of the com would be replaced when slightly more was in fact replaced (= 37%). The August 1992 EOC 6 and November 1993 EOC 7 refueling outages, which include the second and third  !

batch of B&W fuel and additional steam generator tube plugging. resulted in substantial flow decreases as measured by the calorimetric. The elbow tap AP llows indicate a slight i flow increase resulting fmm the EOC 6 and a decrease in flow fann the EOC7 mfueling outages. The analytic model flows, however, indicate a slight decrease in flow after both outages.

Figures 18 and 19 show the daily average elbow tap AP data for Catawba Unit i back to January 1,1992. This data covers cycles 6 & 7 and has been modified to climinate bad data and data at off power days (<99% power). The data shows that the elbow tap APs are very steady with no major drifting trends when the transmitters are operating properly.

Channels Al, C2, and C3 did have drift problems during cycle 7 and were replaced during the last outage. In addition, the OAC scanner was recalibrated on August 5,1993 which resulted in decrease in AP of approximately 0.3% in cach channel.1he numemus mndom  !

small dips in the AP indications are the result of a bad data point in the daily avemge. l Occasionally a point will read invalid by the computer and this will be recorded as a zero reading for the 5 minute period in question. This causes the daily average for that day to ,

mad slightly low resulting in the small dips in the elbow tap APs. l Catawba Unit 2:

Calawba Unit 2 has experienced a substantial flow impact as a result of hot leg streaming i in early calorimetries. 'lhe RCS flow as measured by the calorimetric has trended very 1 closely to the changes in full power AT as can be seen in Figure 20. The calorimetrics j performed between plant startup and the June 1990, EOC 3 refueling outage, indicate unsubstantiated decreases in RCS flow which were not accompanied by significant plant changes. Between October 1990 and Jarmary 1992 five calorimetrics were performed.

The October 1990 calorimetric was perfonned with a feedwater pressure transmitter of the wrong range, therefore, the calculated elbow tap coefficients for this calorimetric were not i used. Another calorimetric was perfonned in November 1990, but due to questions  !

c(mccming the amount of feedwater venturi fouling present, the calculated cibow tap ]

coefficients for this calorimetric were also not used. Following the claaning of the )

feedwater venturis the third cah,rimetric was perfonned in January 1991. The calculated )

cibow tap coefficients for this calorimetric were used for the remainder of Cycle 4. The I calorimetric pcrfonned following the EOC 4 refueling outage in December 1991 was not used to establish the new cibow tap coefficients since it was discovemd that a dra'n valve l

- - - - .- _ = . _ - _ . -

on a transmitter installed for the calorimetric was leding. The calorimetric performed January 1992 was used to calculate the elbow tap coefficients for Cycle 5. When using ,

these two " goal"' calorimetrics ( January 1991 and January 1992) the calorimetric flow, as indicated by the dashed line on Figure 20, after the June 1989 calorimetric trend slightly '

downward consistent with the increasing trend in AT, but in excess of that expected by the very small amount of steam generator tube plugging (Figure 22). In the period between plant stanup and the October 1990 calorimetric, the calorimetric Hows have indicated that the total RCS Dow has dropped approximately 13,000 gpm (Figure 20) whereas the elbow tap and analytical mmici flow have indicated a drop in total flow of approximately 150  !

ppm (Figures 21,23 & 24). Note, Figures 23 & 24 differ only in the Y-axis scale with the scale on Figure 24 matching the scale of Figure 21 for casy comparison This indicates a significant impact fann hot leg streaming as it shows that 12,850 gpm of the calorimetric flow decrease in this time frame can be attributed to this phenomenon.

The first batch of B&W fuel was introduced during the February 1993 EOC 5 refueling i

outage. The B&W fuel was expected to increase flow slightly due the decreased flow resistance in the core. The flow in the plant as indicated by the elbow tap APs and the analytical How model did not change appreciably, i. c., a very small increase in flow, while the calorimetric flow also showed a slight increase af ter this outage.  !

Figures 25 and 26 show the daily average elbow tap AP data for Catawba Unit 2 back to '

May 1,1992. This data covers cycles 5 & 6 and has been modified to climinate bad data and data at off power days (<99(7e power). The data shows that the elbow tap APs ate very steady with no major drifting in:nds when the transmitters are operating properly.

Channel B2 did have drift pmblems during cycle 6 and was replaced during the last l outage. The numerous random small dips in the AP indications are the result of a bad data i f

point in die daily average. Occasionally a point will read invalid by die computer and this will be n' corded as a zero reading for the 5 minute period in question. This causes the .j daily average for that day to read slightly low resulting in the small dips in the cibow tap APs.

Summary:

As discussed above, the RCS flow measured at McGuire Unit 2 and Catawba Units 1 & 2 using the calorimetric method have been greatly affected by the hot leg streaming ,

phenomenon. Calorimetrics affected by hot leg streaming have exhibited large decreases in the measured RCS How which are not consistent with changes in plant geometry (steam generator tube plugging, fuel changes, etc.). The How changes detemiined by the Dow calorimetric have been shown to trend the increases in AT as indicated by the hot and cold leg RTDs. With unpredictable hot leg streaming patterns, the flows detenuined by the calorimetric become highly unpredictable. At the same time the flows as indicated by the cibow tap AP transmitters trend extmmely well with an analytical flow model and pnxtuce  ;

predictable and irliable flow results. j

-.. 1 (9) Please discuss the use of"on-line" differential pressure indications with respect to the observed behavior as opposed to the averaged data used for comparisons and evaluations.

J i

Respinse to Question 9:

The use of "on line" AP data would provide calculated RCS loop flows much as those l already plotted in Figures 2. 8,14 and 21 with the additional points between calorimetrics and the noise asmciated cith measuring the process APs. This noise in this'  ;

" instantaneous" data is due to the AP fluctuations caused by the turbulent RCS loop flow in the cold leg clbow. The data used for comparisons and evaluations is only averaged to  :

the extent necessary to pmvide useful data for use in calculations. Much of the data used l for these comparisons and evaluations (APs, temperatures, etc.) comes from past  ;

calorimetrics used to perfonn the Technical Specification flow surveillances. Averaged ,

data is primarily used to get a more representative parameter value (a value not at a  ;

momentary extreme) and reduce the process noise and error associated with the measurement of the parameter.

l (10) Please discuss proposed future conduct of calorimetric tests and use of the results.

Resp >nse to Question 10: i Future calorimetric tests for the purpose of flow determination are not planned. Changes in RCS flow will be predicted for changes in plant geometry primarily by the analytical model. These predictions will be compared to the indicated cibow tap AP flow to detemiine if the expected flow changes correspond to the plant changes. If predictions of RCS Ilow do not correspond to the itxlicated elbow tap flow, further investigation into the source of the dif ference will be performed.

l (11) Discuss known information on the degree of fouling on or in the RTD scoops. Discuss any plans for inspection of the scoops and RTDs in forthcoming outages.

l Response to Question 11: l No inspections for fouling of the RTD scoops have been perfonned since the installation of the RTD thennowells during the RTD bypass removal outages at cach plant. Fouling of l the RTD scoops is expected to be small and have little impact on the temperatuit sensing  !

function of the RTD. Major fouling of the RTD scoops would only affect the thermal j response time of the RTDs and no significant change in the response time of the RTDs has been observed. Currently, inspections of the scoops and RTDs for fouling are not planned j for future outages.

(12) Discuss any considerations underway for a redesign of the RTD and/or its scoop, both ,

in terms of geometry, placement, and number to be used. .

Response to Question 12: )

There are no plans for a redesign of the RTD/ scoop configuration. No practical, cost effective redesign of this configuration could be made which would provide adequate temperature indication and not negatively impact the RCS flow resistarce and ALARA.

Since ALARA concems were a major factor in the climination of the RTD bypass system, a redesign of the RTD/ scoop configuration would have to mimimite the ALARA impact.

I i

i

. . . . . - .- -~ - .

A new design of this system could have considerable present and future impact on ALARA. Any plant modification to change this configuration could involve a considerable one time ALARA impact. In addition, designs which incorporate more RTDs and/or scoops may msult more potential crud traps which will have significant future ALARA impact. Therefore, since no practical redesign of the RTD/ scoop configuration can be made, which adequately addresses the above concems and provides an adequate temperature indication, efforts focused on the incasumment of RCS 110w by the more direct method of cibow tap AP indications.

(13) The data presented on the handouts in the February 10,1994, meeting shows that the flow rate indicated by the calorimetric heat balance method approximates that indicated by the elbow taps except for the AleGuire Unit 2 in 1985, Catawba Unit 2 up until 1990 and Catawba Unit 1 in October 1992 and .lanuary 1994. Provide an assessment for each of these divergencies.

Response to Question 13:

The large increase in flow for McGuire Unit 2 in 1985 is the result of a decrease in indicated AT of approximately 1 F. This decrease in AT was investigated by Duke and Westinghouse and found to be caused primarily by hot leg streaming changes and a mduction in thennat power caused by feedwater venturi fouling. As can be seen in Figure 5 the RCS total How was largely the result of this change in AT. For the same calorimetric the individual loop flows as detennined by the elbow tap APs (Figum 6) do not show a corresponding increase in flow. In addition, no changes in plant geometry (SG tube plugging, etc.) were made prior to this calorimetric which would indicate a How increase or decrease (Figures 7 & 8).

The decrease in RCS flow for Catawba Unit I during the October 1992 and Jaimary 1994 calorimetrics was the result of hot leg streaming changes which resulted in an increased l indicated AT. The Catawba Unit 1 plot ofindicated full power ATin Figurc 9 shows that the indicated AT for the October 1992 and the January 1994 calorimetric increased approximately 1 F cach. These two increases in AT represent a decrease in flow of approximately 12,750 gpm while the changes in cibow tap AP indicate a flow decrease of approximately 2,100 gpm. Steam generator tube plugging was perfonned during the j outages prior to these calorimetrics (Figures 11 & 12), however, the magnitude of the I cibow tap AP flow changes is consistent with the analytical model pmdicted flows which  ;

indicates that = 10,650 gpm of the calorimetric decrease is attributable to hot leg stmaming.

The decreases in RCS flow for Catawba Un'.t 2 prior to 1990 were the result of increases j in indicated AT caused by changes in hot leg streaming pattems. The Catawba Unit 2 plot of indicated full power AT in Figure 13 shows that the indicated AT up to 1990 increased j approximately 2 *F. This increase in AT mpresents a decrease in flow of approximately j 13,000 ppm while the changes in cibow tap AP indicate a flow decrease of appmximately j 150 gpm. Very little steam generator tube plugging was perfomied during this timc frame i as indicated in Figure 15, therefore, very little decrease in the RCS flowrate should have i been observed.

l 1

I

a i

l (14) The February 10,1994, meeting handouts provided an RCS Flow Uncertainty l Analysis. That information is requested to be submitted by a DPC letter with the  !

I appropriate non-proprietary and proprietary versions and accompanied by an affidavit, consistent with the requirements of 10 CFR 2.790.  ;

Response to Question 14:

Included with response to question 15 below. See Attaciunent 1.

(15) Duke Power's approach to the issue appears to assume that, notwithstanding the greater variability in hot leg temperature, the uncertainty in the measurement of RCS flow rate for surveillance purposes decreases from the present value of 2.1 % to 1.9%.

Please submit the complete derivations of these values camsistent with the format of previous submittals, identified by DPC,in September 1987 and in Catauba FSAR question 492.7 for the surveillance flow rate. This should also be done relative to the McGuire submittal of October 8,1981, and the Catan ha submittal of.luly 30.1984, for the reactor protection system trip flow measurement.

J Response to Question 15:

The present uncertainty value of 2.2% tlow (2.1% llow measurement uncertainty plus 0.1% feedwater venturi fouling penalty) stated in Figure 3.2.1 of the Catawba Technical Specifications is an uncenainty allowance and not a calculated uncertainly value. The present calculated flow measurement uncenainty value is 1.83% llow, therefore, the j calculated flow measurement uncenainty is increasing by 0.07% to 1.9% llow. The calorimetric and flow uncertaintics for the time period after the RTD bypass removal at McGuire and Catawba are given in Attachment 1. In addition, the flow surveillance and loss of flow setpoint uncertainties for the elbow tap flow methml are also included in Attachment 1.

(16) The DPC 'opical report DPC-NF 2004P-A,on core thermal hydraulic methodology, states in Section 6.4, that the flow uncertainty standard deviation is 1.337% and that the uncertainty is 2.2%. Please discuss the comparability of this methodology of arriving at a flow uncertainty relative to the methodology identified as the McGuire and Catanha licensing basis methodology in the handouts (numbers 35,36,37, etc.)in the meeting on February 10,1994.

Response to Question 16:

The flow uncertainty of 2.2% stated in topical repon DPC-NE-20NP-A is identical to the licensing basis methodology identified in the February 10.1994 handouts. %c standard deviation is determined from the 2.2% uncenainty allowance by assuming a nomial distribution and a 95% one-sided probability level. The probability factor of 1.645 is obtained from nonnal distribution tables. Dividing the uncertainly of 2.2% by the pmbability factor of 1.645 results in a standard deviation of 1.337%.

l (17) To support the proposed Technical Specification amendment, the loss of flow trip measurement uncertainty has been revised to include the following sensor uncertainty allowances: I

- Sensor calibration accuracy

- Sensor temperature effect

- Sensor pressure effect flowever, should a measurement, test, and equipment accuracy term for the sensor also be included in the revised loss of flow Channel Statistical Allowance calculation?

Response to Question 17:

The current Duke setpoint methodology for McGuire an k'atawba includes measurement and test equipment accuracy tenus (M&TE) for the sensor when the calibration accuracy to measurement and test equipment accuracy exceed t4 to I ratio. A review of the calibration procedure and the measurement and test egipment used to calibrate the elbow tap AP transmitters indicate that the ratio of sensor cahoration accuracy to sensor M&TE do not meet the 4 to 1 criteria. Therefore, a conservative sensor M&TE tenn is included in the McGuire and Catawba uncertainty calculations in Attachment 1.

The result of the additional tenn for McGuire is that the calculated value for the RCS flow uncertainty is increased from the present calculated value of 1.7% flow to 1.76% llow instead of the 1.74% llow indicated in the November 16.1993, Technical Specification submittal. This does not change the proposed Tecimical Specification change which increases the present value on 1.7% flow plus 0.1% feedwater venturi Ibuling penalty to 1.8% flow plus 0.1% feedwater vetun o "ag penalty.

Vic result of the additional tenn for Catawba is that the calculated value for the RCS flow uncertainty is increased from the present calculated value of 1.83% flow to 1.88% flow instead of the 1.87% flow indicated in the November 16,1993. Technical Specification submittal. This does not change the proposed Tecimical Specification change which decreases the present value of 2.1% flow plus 0.1% feedwater venturi fouling penalty to 1.9% flow plus 0.1% feedwater venturi fouling penalty.

Le additional sensor M&TE term for the McGuire low reactor coolant flow setpoint uncertainty does not change the setpoint of 91% of minimum measured flow per bop and the allowable value of 90% of minimum measured flow perloop as presented in the November 16,1993, Technical Specification submittal. Ilowever, the channel statistical allowance (CSA) will increase from 3.729 flow given in the November 16,1993, submittal to 3.79% flow.

The additional sensor M&TE term for the Catawba low reactor coolant flow setpoint uncertainty does not change the setpoint of 91% of minimum measured flow per loop and the allowable value of 89.7% of minimum measured flow per loop as presented in the November 16,1993, Technical Specification submittal. Ilowever, the channel statistical allowance (CS A) will increase from 3.37% flow given in the November 16,1993, submittal to 3.42'7e flow.

MNS-1 Comparison Between Full Power Delta T and RCS Flow As Determined By Flow Calorimetric

- 405000 54.50 I -

l 55.00 - --- J02500 55.50 P + 400000 E l

3 l o

56.50

[ 397500C y  ! / 0 o l _.

~~

i 395000 57.50 ,

1 392500 58.00 f

< I 58.50

' 390000 Jul-83 Nov-84 Mar-86 Aug-87 Dec-88 May-90 Sep-91 Jan-93 Jun-94 Feb-82 Date

-C Full Power Delta T Total RCS Flow Figure 1 i

_ __ _ _ _ --____- _ - - -------- -- - - - - - - - - . ~ . -, ,

MNS-1 Individual Loop Flows As Determined By Elbow Tap Delta Ps 104.000.00 -

e .

102,000.00 - e .

imom.m = g- -

< -m .

E  %

98.000.00 f m

96.m0 m 7

" N ./ x -- --. . ._

. x, N

=

94.000.00 +

l 92.000.00 Nov-84 Mar-86 Aug-87 Dec-88 May-90 Sep-91 Jan-93 Jun-94 Feb-82 Jul-83 Date Loop D

^

-*- Loop A 0 Loop B Loop C Figure 2

MNS-1 Individual Loop Steam Generator Tube Plugging Percentages 07 i C S--b--C-G  ?

l 2t x 4 "N 4i '

i it 6

l N'x- s o i k S  ! -*- Loop A \

'}g 8 l

0 Loop B N"g x e

5 10

Loop C 0

E  !

^

Loop D 8 12 i m i I4 T s 4

a b

l 16 4

Feb-82 Jul-83 Nov-84 Mar-86 Aug-87 Dec-88 May-90 Sep-91 Jan-93 Jun-94 Oct-80 l

Date i

i Figure 3 l

MNS-1 Individual Loop Flows As Determined By Analytic Flow Model

,m.sw.m7 g, 3mem -

"N, \n 99,e --

g 5

99.000.00 1

-=- Loop A \.~.\

\  : Loop B S

Loop C \

• 98,500.00 "

Loop D I

98.000.00 i

e 97.500.00 l

L Oct-80 Feb-82 Jul-83 Nov-84 Mar-86 Aug-87 Dec-88 May-90 SeP-9I Jan-93 Jun-94 Date Figure 4

MNS-1 Individual Loop Flows As Determined By Analytic Flow Model 104.000.00 r 102.000.00 7 i,

c n-o-o-c 100,000.00 I a- ~. __

~

T~

E o.

Q N

? 98.000.00 i ' E

$ -*- Loop A E

l 0 Loop B 96,000.00 1 l

0 Loop C

^

Loop D 94,000.00 -

l '.

92.000.00 Oct-80 Feb-82 Jul-83 Nov-84 Mar-86 Aug-87 Dec-88 May-90 Sep-91 Jan-93 Jun-94 Date Figure 5

e g -

ne Ea - ~ mf Om .

fh y9" {%uQ~ r 8D g -

m ,& Sir ." u9* .9m f: f - Qn .

" =pu .,g g. a 9 o

,9u

+f 9*

uh

'y k t $

f wlii '

i '

f',

g

~

1 l

.: . l

I I

.l I

q

- -

• h l

' q '

~:

I W. *

. I ' ) ;, . ' I >

l I

(

I i.l 1

h} ,

ug5 ,'

Y . ;

Ifh ) . !  ! ' I '  !.

8 o.  ; j p ,n N lf ) ,lj I

, l I 1 I y l '

lf

.: .! l ,'

- )s u5g t'lI 1, ( 4 ,

! [ . ,

.f . . I a

] * '(/

[

) d ,] 1

> l . <

\ i l

V.I .,. gs ,;' ][

ft ) ,; l'  ;; ((l;I' . ' 1

ll

.: ! .)l4. " /, .*l . u

\.

,j

' I.  !

@8, IL  ? , '

< ' ('jt

\ -

1 I I.,

. ( ,f4 j ,

d

58. - . . . . . . _ _ . . . . . . _

2 2 3 3 3 a 3 3 3 3 3 4 4 4 4 4 2 9 9 9 9 9 9 9 9/ 9 9 9 ,

2 9

/

9 9

/

0 m

0 m

9 9

8

/

9

/

7 9

/

9 i

7

/ /

7

/

6

/

5

/

4

/

3 2

/

4

/

3

/

2 -

2

/ . *' ' 1

/

2

/

2

/

0

/

1

/

2

/

mi i

/

2

/

0

/

1

/

2

/

0

/

1

/

2

/

0 5

/

1

/

6 6

• 9 0 1 1 2 3 6 7 9 0 1 1 2 3 0 .* 0 1 1 0 0 0 m 0 0 0 1 1 0 0 0 0 0 a

)

3 9 ,* \5 1 r : * :  : l  ;! / ,illl l! < ,l - ;  ;  : >i  !? . i  ! i

MNS-2 Comparison Between Full Power Delta T and 'RCS Flow As Determined By Flow Calorimetric 55.w - 7 4090m 55.50 - i T 404000 56.m = l P I f E Y h399000 l 5

$ 57.00 7 M

$ i 2

t 394000 -

! i S

= 57.50 ' t

• S 2

j j

58.00 h * ' 389000 58.50 7 "

i ,

384000 59.00 Nov-84 Mar-86 Aug-87 Dec-88 May-90 Sep-91 Jan-93 Jun-94 t

Jul-83 Date l

--C Full Power Delta T Total RCS Flow i

Figure 7

MNS-2 Individual Loop Flows As Determined By Elbow Tap Delta Ps 101.000.00 --

i imuumf  ;

~ . _

99,000 m i '

~. ./-

E 98.000.00 - , . ,

@ 97,000.00 " y 96,000.00 95,000.00 -- ,

94,0M.00 Jul-83 Nov-84 Mar-86 Aug-87 Dec-88 May-90 Sep-91 Jan-93 Jun-94 Date i

^

- - Loop A Loop B Loop C Loop D f

l l

l Figure 8

MNS-2 Individual Loop Steam Generator Tube Plugging Percentages 0~ C t

1~

at 2*

3 3 .. -.- Loop A E i

LoopB 5 i .

\ _.- toop C

.N G

6h 'OU O \

$ i 7f'  ?

8i

\

9l ,

Feb-82 " Nov-84 Mar-86 Aug-87 Dec-88 Moy-90 Se # jan-93 3""

Date Figure 9

MNS-2 Individual Loop Flows As Determined By Analytic Flow Model 100,500.00 -

e e-- 2 100.250.00 i

i i

100.000.00 i '

E i

@ 99,750.00 -l

-*- Loop A 's O Loop B E t

! -*- Loop C ^

w 99.500.00 N l

^

Loop D "x,

I l

'I 99.250.00 i 99,000.00 Jul-83 Nov-84 Mar-86 Aug-87 Dec-88 May-90 Sep-91 Jan-93 Jun-94 Feb-82 Date l

Figure 10 l

L_----__-----__--_--_____--____---____ _ _ _ - _ _ _ _

4 9-n u

J i

c t i y _

3 l 9-a n n J a -

A y g 1 B _

9-p d -

S e

e -

i n .

0 -

9-m y r  : a t

e M -

e Dl e 8 8-sd c e

Ao D .

sM w

t D

e a

owo 1

1 7

l 8- er g u Fl A u ig -

pF F o

o -

6 8-L b r

a l

M a e u A B C D d

i p p p p 4 v o o o o 8-o o o o  : v i L L L L o d N I

n -

" 0 2- - - 3 8

S c lu .

J N

M 2 8-

!1 fi  !

7 h!l!" b e

0 0 0 0 0 0 0 0 F 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0, 0 0 0, 1 0 9 8 7 6 5 4 .

0 0 9 9 9 9 9 9 .

1 1 E@ c

e Ehk' e cS $ 0V &

goy Oa- - )o9" t bnt3 u g - goA 9 ~

a2 i -

rM g u9"

$ 3p) i i

l A"8u~o: %wo8nn9"

' i'

.n$

n9" il '

, O:r u lQ" p j g

^' ,

amf M , / 1 T

< u' - [ j d l Q' ~ 4I h '

g:-(, .

i (' # 1)

' l

\7

' i

)

,'; : ii' i' ' :{ !,:!, ,!: j;),i:

l 1

,;i' t j. l ji: - I' hI l:

t {

ell U

)

. . }(i: [,{

aW ji:lI i i lli;ii l .';: j [ l I8:),!

i,it{. : 3s l (

b

{l':,

N+

j i Is 4

' i' h

l

, I' ,'

( i. '

y>7 h '

4 i

j

> g c

t g[ ; i

, I ' '

(f '

]1

' lL ua :

a9 ;5u@ z ggg55: :E = E :i 5 5:ll!5:j[E 5=E:

E = f :s ! 5:E l! 5 :i 5 E 5=E

E 5li

: E 5l[ 5. sl -

il EE == iE !!  !! hjijsljjl[ j 3

9 3 9 / 3 9 // 3 9 3 9/ 3 9/ 39/ 39/ 3 9/ 39/ 3 9/ 3 9/ 39/ 4 9//4 9 4 9 4 9 4 9 4 9 4 9 4 9 4 9 4 9 4 9 4 9/ 4 9/ 4 9/ 49//4 9 /4 9 /4 9 4 49 4 4 9/ 4 9/ 4 9/ 4 9 4 9 4 9 649 4

/

2 /

3 8 2 8 5 0 5 0 8 3 8 3 8 3 //

8 23/2 8// /2/ 7/ 2/// 7 22/27/ 4 9 4 9 /9 8 ////

9 9 1

2 2 9//13 4 3 8 3 9 4 9 5 1 2 2 3 / 2 /1 / / 1 1 1 2 2 / / 1 2 2 3 2 2 2 // 1

/ / / / 21 1/ / / 1 ////

1 1 2 2 // / 2/ 2 32 3 1 1

/ / //4 3 43 // / / 5 5 4/ 4/4 /4 /

///1 1 2 2 2 2 3 3 5 5 5 5 9 0 01 1 1 1 1 1

1 1

1 1 1 1

1 1 1 1 1 1 1 J

%' C ohP 'Y

i' i - - 3l i l i  ;'  ; [' l r !

CNS-1 Comparison Between Full Power Delta T and RCS Flow As Determined By Flow Calorimetric 55.00 T - 405000 l

55.50 it ,

- 400000 56.00 7 i /  ;

i

^

56.50 P ' - 395000 E 57.00 ._

$ 57.50 -f390000f g

. 2 58.00 3 2 i

- 385000 .9 4

58.50 7, 59.00 h, T 380000

' 59.50 f I ' '

l 375000 l

60.00 ' .

Nov-84 Mar-86 Aug-87 Dec-88 May-90 Sep-91 Jan-93 Jun-94 '

Date i

C Full Power Delta T Total RCS Flow Figure 13

CNS-1 Individual Loop Flows As Determined By Elbow Tap Delta Ps 101,000.00 --

100,000.00 -

l "'. ~'

99,000.00 - -

"N ,

'=

E N

a . '

,- 98.000.00 t N e o l 97,000.00 t 96,000.00 -

95,000.00 ,

Mar-86 Aug-87 Dec-88 May-90 Sep-91 Jan-93 Jun-94 Nov-84 Date

^

-*- Loop A

^

Loop B

• Loop C Loop D Figure 14

CNS-1 Individual Loop Steam Generator Tube Plugging Percentages 0- t C m  :

i

,+

2 -

i .

1. I o>

3 h4

-o

, 1,

, 5T S _.- Loop A to 6t  : Loop B

)7k C

1

.: Loop C

=

[9 8t E

^

toop D Pa 9k m

10 -

i 11

\

12 Nov M ' Aug-87 Dec-88 May-90 30P-91 Jan-93 3""'94 Figure IS

CNS-1 Individual Loop Flows As Determined By Analytic Flow Model imanm =

f i _

0 0 4 . v- m.

N. .

=

99,500.00 -

E

-"- Loop A

$ l 0 Loop B

[o 99.000.W ^

Loop C i '

Loop D l

I 98,500.00 i

i 9t.000.00 Aug-87 Dec-88 May-90 Sep-91 Jan-93 Jun-94 Nov-84 Mor-86 Date Figure 16

CNS-1 Individual Loop Flows As Determined By Analytic Flow Model 101.000.00 -

- m- .

100 000.00

e

~ ~ ~

u

~

7 1

i N 99,000.00 E

-*- Loop A

$- 98.000.00 T  ;

0 LoopB o j 0 Loop C 97,000.00 - ^

Loop D I

96,000.00 t i

i 9C 000.00 Mar-86 Aug-87 Dec-88 May-90 Sep-91 Jan-93 Jun-94 Nov-84 Date Figure 17

I_

l i

._ w m w a oo o > >

> 8 i

88 8 88 i l L <

?

$ i 55 29/62/6  !

55 29/12/6 j

J a

l d J i i

' J  ?

i 5[29/61/6

(

!!29/11/6

) ' f

,' ', L 4 51 29/4/6

)  ! ii29/03/5

' l L

!!29/52/5

(

f '

h 0; I

55 29/02/5 j

1 9 ,J si 29/41/5 O i t

==

O,, s, 1

• i } ' gg 29/9/5 N

9 < 1

J( i

!!29/4/5 I '

/ 1 S EE 29/92/4

$ h f [ Y ll ,) J f ll 29/42/4 km j 1

!!29/91/4 o j ) 'l gg 29/31/4 gt

>f n< ( ,

V L f

g y' i i i 5-5 29/8/4 *

$ S

% ) h ) ][29/3/4 (

y '

i b i q 'j ][ 29/92/3 l

[ f a

!!29/42/3

')$

EE 29/81/3

{ ,

l p'

t h h 29/2/3 E [f d  !

k>0 l ] >

I b  !!29/62/2 i

)#

L }

0 9 l i

!!29/12/2 E ,

o d d <

k ' 55 29/61/2 I  ! ,

A

/ f i

'  !!29/11/2 i i

) t EE 29/5/2

) 1 l ==

)

( (

'j )<  !!29/13/1 f

i {i

> /(

') ,d(i EE

==

29/62/1 i L h h, 1 EE 29/12/1 l

29/61/1 J

I l l {

i

!!29/11/1 5 f ( EE 29/6/1

{ l {l < j [ f fl [ ll ( EE I II 8 8 8 8 8 8 8 d

y o p' < $ 3E 5 8 3i u o w w w k

OOOd3 3 8 EE8 5 8

\ L  !

C '

J 11 39/61/01

) t ,

i m f

3

[ 4 d / ,

! !39/5/01

I[ 39/22/9 l '

J I f [ , )l} 1

!I 39/21/9

) ) ' ' '

p '

l q j i 39/2/9 f

d i I (l

) l l

/

(i q'

' i si 39/32/8 39/31/8 i I l i

i, 1 as s

j? ,

) J,,P ,

,h( ,

/ i l fi lI 39/2/8 j[ 39/22/7 2

1 i i we fj h

f :hb ,

l i 39/2/7 3 39/3/6 Q

t  ;

a h.o, 3 a.g i 39/32/5

/

U~ 4 o / t t I sa'al 39/31/5 #

i < $ i g

V ,

h / -

(, E i

i f $$ 39/3/5 P

j 4 '

4 q

',' , ][39/02/4 k g v, '

i

,i i

gg 39/01/4 o

!!E 1

(

h '

' si M 39/13/3

'3 f

4

') i

'I  ; nE i t i 2? 39/91/3 g }  ;

9 f 3@3 g J i d

'  !< ) '

/, is 39/42/2 g l

' l i ss '

3 5 39/41/2 i h i 39/4/2 i ,-

t - i l l

L

][39/42/1

[ ][ 39/31/1

• i

==

$ ni ar 39/2/1

% '} > > g; 29/32/21

'?  !

5 I i 5E l

! / ~

\ b i

3I 29/11/21 l

k -

f j (

I f

i

{

}{29/03/11 l

1 4 ;2 '

4 (r / ][ 29/91/11 i

f -_ -

1 I

l ji y

[ '

1 p

) l  ?

j iE 29/8/11 y 29/92/01 8 8 8 8 a 9 8 h b h 5 8 $

I CNS-2 Comparison Between Full Power Delta T and RCS Flow

As Determined By Flow Calorimetric

- 402000 f

55.00 T; e t 400000 l 55.50 1 398000 P 396000 E .

y 56.50 7 i

i

% 394000 g i E

~.

8f 57.00 f ., ,

i i 392000$ ~

o

-I 57 50 ' '

- 390000 S 2 . \,

~

O 388000 c-58.50 ~ i 386000 59.00

' 384000 Mar-86 Aug-87 Dec-88 May-90 Sep-91 Jan-93 Jun-94 Date O  : Total RCS Row Full Power Delta T Figure 20

CNS-2 Individual Loop Flows As Determined By E! bow Tap Delta Ps 101.000.00 7

i ey % -a 99.000.00 i

0

• 98.000.00 3  !

e 97.000.00 -.

I 96.000.00 1 95.000.00 1 94.000.00 Mar-86 Aug-87 Dec-88 May-90 Sep-91 Jan-93 Jun-94 Date

-*- Loop A Loop B 0 Loop C a loop D Figure 21

_ . _ . _ _ _ . _ _ . - - - - - - - - - - - - - - - - - - - - - - - - - - . - - - - . - - --. ~--.,n --- - - - - < - > - - - - - - - - - - - - - - - - - - - - - - _ _ . _ - - >

CNS-2 Individual Loop Steam Generator Tube Plugging Percentages 07 0.2 a e- e  :

38 IN 0.4 -

1

a. .

~

j 0.6 1 -*- Loop A a -

i

$  ! O Loop B g 0.8 -

I

• Loop C 5

O .!

=

E 0

ji t

^

Loop D 1.2 i .;

i , e .  ;

Aug-87 Dec-88 May-90 Sep-91 Jan-93 Jun-94 Mar-86 Date Figure 22

CNS-2 Individual Loop Flows As Determined By Analytic Flow Model imam 7 i

100A50.00 m x' ~.

c 100A00.00 t N"

-"- Loop A o 0 Loop B "N, 100,350.00 1

• Loop C

^

Loop D 100,300.00 i 100.250.00 Mar-86 Aug-87 Dec-88 May-90 Seo-91 Jan-93 Jun-94 Date Figure 23

_ _ _----- _-_________-_--_____ _ _-- __- .= ._ . . _ . _. . _.

CNS-2 Individual Loop Flows As Determined By Analytic Flow Model 101.000.00 -

c E _ _ _ _ _ - _ .

100,000.00 f 99,000.00 h i

E 98,000.00 g I -=- Loop A h I >- Loop B c 97 000.00 t '

O Loop C a Loop D 96.000.00 1 95,000.00 --

l 94,000.00 l l

Aug-87 Dec-88 May-90 Sep-91 Jan-93 Jun-94 Mar-86 Date

Figure 24 i

m 8 8 m

2 o

?

u U

o U

e U

o U

e N

o 5

m 8 U m 8 5 m

5/1/92 0:00 _._ r ,-r ,

i g

5/6/920:00 5!

I' (

I f 5/15/92 0:00 [h  ; )

5/20/92 0:00 E5 i , l l j

' ]

= - <

5/27/92 0:00 [j j j j )

6/1/920:00 EE::

, i <  : <

l

4 4

l 6/6/92 0:00 [j l } l j [} f

/ l 6/13/92 0:00 i_E ,! ,

6/25/92000 55 i  !

I,

: i

( i 7/1/920:00 gg  !

7/6/920:00 E5 i  : <' L h

4 1 i

7/11/920:00 Eg l l 1 7/18/920:00 E5 i

7/23/920:00 Ej '

i l l n

7/31/92 0:00 5j  !

j ,

d 9.

I 8/7/920:00 E5 i I

' t '

8/12/920:00 h!  ;!

l [ $

8/17/92 0:00 [] i  !

,p 8/25/92 0:00 :]l  : j' }[

8/30/92 0:00 55 l l ,f ( ,

I

$2 M] 9/4/920:00 ll (  ; i

' I 6O 5 9/9/920:00 EE==

) (I

-) J

(

t C

w46 y 9/14/92 0:00 [g }

gm

j ,

p 9/20/92 0:00 55 4 l  !

{; 8E4 D 9/25/920:00 E5 i i

-4 9/30/92 0:00 [g <

- e' j l l$5~:

10/5/920:00 Eg i, j ( ( o

( ,

( s 10/11/92 0:00 [g

} '

j, I

y 10/17/92 0:00 EE i , i f 10/22/92 0:00 Ei ( '

('

10/29/920:00 E5 b -

==

11/3/920:00 EE i

{' 4 11/8/920:00 EE::

! i q  :  : 1 11/13/92 0:00 [g y -

c  !

12/22/920:00 EE::

i- ',

{

l ,)

I, 12/27/920:00 EE  : ,

i 1/1/93 0:00==  !! (

! . (h , '

: 1 , t 1 1/7/93 0:00 EE i; '

, fi f' i

== '

l 1

1/12/93 0:00 =E i

L  :  :

1/17/93 0:00 [g t )  ;

} l i

1/22/93 0:00 =5 1  ! i h "> m"-cp N 8

- - O m m>Mbakd -

eega s L 0 3 \\ E3 39D f 4

[ hf f- h h 00:049/81/4

' [ [ '

![ 00:0 49/9/4 J

f ,' J, j , ,

jy 00:049/13/3  !

i -< ' iS '

I h' I  !! 00:0 49/12/3 i Y, , 1, d!

,, i.i e

00:0 49/11/3 i

(( (f .}ly 00:0 49/2/3 i L l,

y 00:0 49/71/2 h ,

jg 00:049/7/2 t

=i  :

!! 00:0 49/92/1 t [ 00:0 49/02/1

[ ( =i  !

i 4

, , , o ,

, if 00:049/7/1

i

' ' !I 00:0 39/52/21 p

j j , ,

}

![ 00:0 39/61/21 E J 1 r h

ll 00:039/7/21 j }

j. ,

j j[ 00:0 39/72/1i 4 i '

<- l as p ,

4  !! 00:0 39/81/11 p oA ,t ' ' u j 3 g ,

&c ' L l HF 00:039/9/11 -

mR

,P ei i ID [

I j

d

$$ 00:039/13/01 - $!

@5 ,

h,  !! 00:0 39/02/01 3-i 0$

z*

( <

i I ]l 00:0 39/01/01-

/ ) i 07 R I  ?  !! 00:0 39/03/9 gR j[. 00:0 39/51/9 -

c L i

'h h 00:039/6/9 4 ,- =- ~ --

I  ! li 00:0 39/72/8

[g -%

fp ' d ~--

>; :7' ij 00:039/81/8

~

( l[ 00:0 39/9/8 I 'l N ff h 00:039/13/7

'  ! ! 00:0 39/22/7

[ j l )

j i

j?! 00:0 39/31/7- l

/ J '

l l; 00:039/3/7  :

ss

-( y '

I I

i i 00:0 39/32/6 fi{ 00:0 39/41/6 i

][ 00:0 39/3/6 d-

[ '

]! 00:0 39/52/5 I U ~ 00:0 39/61/5

= '-- := 2 =2 = fj '-~ : '- ' :-H

'f' {' - --h[h 00:039/1/5

-~

~,

a, ji 00:0 39/12/4 N s q \' g ;3 s i a g 00:0 39/8/4 m 8 o e o m o e o m S-8

" " @ 8

"-

• W O O M M B B 8 8

Attachment la Non-Proprietary Version r

t l

\

1. Feedwater Venturi Flow Measurement Uncertainty for McGuire (Post RTD Ilypass Removal)

Parameter Sensitivity 1.nop Flow Biratncitt Upcertaintv farint Uncertainty Feedwater Flow Venturi (K) Note ! -

Note i Expansion Temperature

• _3.2 F 0.002%/ F 0.50cle_

Material Density Temperature

• 3.2 F 0.042%/*F 0.13 %

Pmssure 20.7 psi 0.0005%/ psi 0.01 %

Diffemntial Pressure Instmment 0.03 % 0.5%/% 0.02%

Process Fluctuation 1.8% 0.5%/% 0.9 %

• - indicate dependent parameters.

Note 1 Venturis are normalized to ASME flow elements. The ncenainty in feedwater flow using the ASME flow elements is 0.59% flow.

The combination of the random components using SRSS is pmvided below:

= 0.913% flow per venturi tap set L. --

The above feedwater flow uncertainty is conservatively rounded to 0.92% flow. 'Ihis uncertainty is then converted to total feedwater flow uncertainty for 4 venturis with 2 tap sets per venturi using the following relationship:

= 0.33% total feedwater flow ]

Combining the this feedwater flow uncenainty with the uncenainty for the ASME flow elements above gives the following venturi flow constant uncertainty: ]

= 0.68% total feedwater flow l

l l

1

2. Calorimetric Contribution to McGuire Flow Measurement Uncertainty (Post ItTD llypass llernoval)

Parameter Sensitivity Loop Flow I Pararnetet Uncertainty fattat Uncert ainty l l

Feedwater Flow l Constant K 0.68 % 2.0 1.36 %

AP & Fluctuation 0.33 % 2.0 0.66 %

Tempering Flow 15 % 0.0335 %/% 0.50%

Feedwater Enthalpy _ _

Temperature Pressure

~ ~

Steam Enthalpy Pressure 40 psi 0.0046%/ psi 0.18 % l Carryover [ ] l SG Blowdown 20 % 0.0001 %/% 0.002 %

llot Leg Enthalpy Temperature (RTD) [ ]

Temerature (DVM) 0.2 F 1.905%/ F 0.381 %

Pressure

• Streaming

( 1.? *F 1.905%/ F 2.286 %

]  ;

Pump Power [ ]

IIcat Losses 20% 0.0004 %/% 0.008 %

Charging Flow 20 % 0.0015 %/% 0.030 %

Letdown Flow 20 % 0.0015 %/% 0.030 %

Cold Leg Enthalpy Temperature (RTD)** 0.5 *F 1.560%/ F 0.780 %

Temerature (DVM)* *

• 0.2 *F 1.560%/ F 0.312 %

Pressure

• Cold Leg Specific Volume

{ '

a]

Temperature (RTD)*

• 0.5 "F 0.145%/ F 0.073 %

Temerature (DVM)**

• _ 0.2 F 0.145%/ F 0.029 % ,

Pressure

• l - -

L _

l

• ** *** -indicate dependent parameters.

, , l l

'lhe combination of the random components using SRSS is provided below: l l

l l

]= 3.1 M flow per t<wy The above loop flow uncen,ai aty is c;onverted to total RCS flow uncertainly for a four loop plant while also incorporating a psi or P7c flow Barton bias (pressurizer pressure) uncertainty.

-. - - - 1 l

=

i= 1.62% Total primary system flow

_ _ _. J l

.lmi'8 A

3. Current McGuire Total RCS Flow Technical Snecification Surveillance Uncertainty (Post RTD Ilypass Removal) j Parameter Allowance  % Flow

- 1 Process Measurement Accuracy (PMA)

Primary Element Accuracy l i

Sensor Calibration Accuracy (SCA)

Sensor Drift (SD)

Sensor Temperature Effects (STE)  ;

l Sensor Pressure Effects (SPE) j i

Rack Calibration Accuracy (RCA) .

4 Rack Drift (RD)

Rack Temperature Effects (RTE)

Computer Isolator Drift (ID)

Allowance for noisy signal (RDOT)

Analog to digital conversion accuracy (A/D)

The above uncertainties are combined using the equation below:

f ) u CS A =

C -

CS A = 1.47 % flow (Single elbow tap uncenainty)

~

We uncertainty associated with the precision calorimetric (  % flow + ~

% bias)is combined with the RCS flow pmcess instrumentation uncenainty alxic. Assuming two out of the three elbow taps in a given loop are available to measure flow, the nonnalized cibow meter flow uncenainty is detemiined as follows:

d= 1.70% RCS flow W

4. Current McGuire Low itCS Loon Flow Reactor Trin Function Measurement Uncertainty (Post itTD llypass Removal)

Parameter. Allowance  % Flow snan I Process Measurement Accuracy (PMA)

Density effects on AP cell 0.40% flow 0.33 Precision now calorimetric _ l.60% flow l.33 _

Noise Sensor Calibration Accuracy (SCA)

Sensor Drift (SD)

Sensor Temperature Effects (STE)

Sensor Pressure Effects (SPE)

Rack Calibration Accuracy (RCA)

Rack Comparator Setting Accuracy (RCSA)

Rack Drift (RD)

Rack Temperature Effects (RTE) <

Bias, Barton transmitter, (Pressurizer pressure) .

The above unce ainties are combined using the equation below: -.

CSAr. -

I.

t- -

CSA = 1.94 % flow span (2.33 % flow)

'Ihc Allowable Value is calculated below:

Total Allowance (TA) = 2.92 % flow span or 3.50 % tiow.

Tg= _

= 1.07 % 110w span T, e ' .

= 1.32% flow span T[ < T therefore 2 the limiting Trigger Value is 1.07 % now span or 1.28 % flow Since the Trigger Value in Tech. Specs. was already at a conservative value of 1.2 % flow at this time no change in the current Allowable Value was made. 'lherefore, with a McGuire loss of 00w setpoint at 90 % Gow, the Allowable Value was left at 88.8 % tlow.

~

l l

l S. Feedwater Venturi Flow Measurement Uncertaint3.for Catawba  ;

(Post itTD llypass itemoval) ]

Parameter Sensitivity Loop Flow Paramefrt Uncertaintv Jntrint Uncertaint y Feedwater Flow Venturi (K) Note 1 -

Note !

Expansion Temperature

• 1.0 F 0.00~2%/ F 0.002%

Material [ ]

Density Temperature

• 1.0 *F 0.043%/ F 0.043 %

Pressure 30 psi 0.00052%/ psi 0.0156 %

Differential Pressure Instniment 0.72% 0.72%/% 0.52%

Pmcess Fluctuation 0.22 % 0.5%/% 0.11 %

• - indicate dependent panuneters. .

Note i Ventuds am normalized to ASME flow elements. The uncertainty in feedwater flow using the ASME flow elements is 0.59% flow.

De combination of the random components using SRSS is pmvided below:

1 l= 0.54% flow per venturi tap set L J .

This uncenainty is then converted to total feedwater flow uncertainty, for 4 venturis using 1 tap set per ventud, with the following relationship:

= 0.27% total feedwater flow Combining the this feedwater flow uncertainty with the uncertainty for the ASME flow elements above gives the following venturi flow constant uncertainty:

= 0.65% total feedwater flow t

E D

am.

E

6. Calorimetric Contribution to Catawba Flow Measurement Uncertainly (Post ItTD livnass Removall Parameter Sensitivity Loop Flow Parameter Uncert ain(y Inittor IIncert aint y Feedwater Flow Constant K 0.65 % 2.0 1.30 %

AP & Fluctuation 0.27 % 2.0 0.547c Tempering Flow 15 % 0.0335 % 0.50 %

Feedwater Enthalpy Temperature 1.0 F 0.143%/ F 0.143 %

Pressure 30 psi 0.0001%/ psi 0.003 %

Steam Enthalpy Pressure 0.0049%/ psi 0.196 % _

Carryover [_ 40 psi llot i.eg Enthalpy Temperature (RTD) ). j Temerature (DVM) 0.0 F 0.193 %/ 12 0.0%

Pressure

• _ 30 psi 0.00923%/ psi 0.28% ,

Streaming (random)

Streaming (systematic) _

Pump Power /IIcat Losses - -

0.1%

Cold Leg Enthalpy Temperature (RTD)* * [ ']

Temerature (DVM)* *

• 0.2 F 1.563%/'F 0.313 %

Pressure

• 30 psi 0.00258%/ psi 0.077 %

Cold Leg Specifie Volume

[ 0.2 F Temperature (RTD)**

] l Temerature (DVM)**

• 0.146%/ F 0.029 %  ;

Pressure

• 30 psi 0.00138%/ psi 0.041 %

• * * * * * - indicate dependent parameters.

l No SG Blowdown uncertainty is listed above since the uncertainty is small and only affects the combination of the random components in the sixth decimal place.

'The combination of the random components using SRSS is pmvided below:

[

~

= 2.45% flow per loop incor]uration of the )/o Badon bias and h systematic streaming allowance results in the following total primary system flow uncenatiity:

q _ -

I= = 1.75% Total primary system flow d _ _

7. CurrrnLCatawba Total ItCS Flow Technical Snecification Surveillance Uncertain.ty (Post ItTD Ilypass Removal) ,

Parameter Allowance  % Flow Process Measurement Accuracy (PMA)

Primary Element Accuracy Sensor Cahbration Accuracy (SCA)

Sensor Drift (SD)

Sensor Temperature Effects (STE)

Sensor Pressure Effects (SPE)

Rack Calibration Accuracy (RCA) .

Rack Drift (RD)

Rack Temperature Effects (RTE)

Computer Isolator Drift (ID)

Allowance for noisy signal (RDOT)

Analog to digital conversion accuracy ( A/D)

'The above uncertainties are combined using the equation below:

CS A =

CS A = 1.5 % flow (Single elbow tap uncertainty)

~

The uncertainty associated with the precision calorimetric (I.69% flow + 0.06% bias) is combined

~ -

with the RCS flow process instrumentation uncertainty above. Assuming two out of the three elbow taps in a given loop are available to measure flow, the nomialized cibow meter flow uncertainty is detennined as follows:

1.83% RCS flow even

I i

l

8. Current Catawba I,ow RCS Imon Flow Reactor Trio Function Measurement Uncertainty (Post RTD llypass itemoval) i Paratncier Allowance  % Flow snan l

Process Measurement Accuracy (PMA)

Density effects on AP cell OAl% flow 0.33 Precision flow calorimetric 1.60% flo'w 1.33 Noise -

Sensor Calibration Accuracy (SCA)

Sensor Drift (SD)

Sensor Temperature Effects (STE)

Sensor Pressure Effects (SPE)

Rack Calibration Accuracy (RCA) .

Rack Comparator Setting Accuracy (RCSA)

Rack Drift (RD)

Rack Temperature Effects (RTE)

Bias, Barton transmitter, (Pressurizer pressure)

'lhe above uncertainties are combined using the equation below:

\

CS A =l CSA = 1.96 % flow span (2.35 % flow)

'lhe Allowable Value is calculated below: 1 l

Total Allowance (TA) = 2.92 % flow span or 3.50 % flow. -

T=_

i ]= 1.11 % flow span T=

2 = 1.32% flow span T[ < T therefore 2 the limiting Trigger Value is 1.11 % flow span or 1.33 % flow

'lherefore, with a Catawba loss of flow setpoint at 90 % flow, a conservative Trigger Value of 1.2 % tlow is assumed and results in an Allowable Value of 88.8 % flow.  !

l l

C I l

i

9. Pronosed McGuire Total RCS Flow Technical SneciGcation Surveillance Uncertaintv (For Proposed Elbow Tap Flow Surveillance) farameter Allowance  % Flow

• Process Measurement Accuracy (PMA) - -

Density effects on AP cell -

Noise Sensor Calibration Accuracy (SCA)

Sensor Measurement & Test Equipment (SM11!)

Sensor Drift (SD)

Sensor Temperature Effects (STE)

Sensor Pressure Effects (SPE) .

Rack Calibration Accuracy (RCA) .

Rack Drift (RD)

Rack Temperature Effects (RTE)

ComputerIsolator Drift (ID)

Allowance for noisy signal (RDOT)

Analog to digital conversion accuracy ( A/D) ( v.

' Allowance ist AP span' '120% Flow span 120% ' Flow

• Allowance in % Flow =

( 2 ,( 100% AP span ,(100% Flow span, The aluve micertainties are combined using the equation below:

cs A -

7 -

csA k ,

CSA = 1.92 % flow (Single elbow tap uncertainty) I 1

7 ,_ J The uneenainty associated with the precision calorimetric g. To 00w + fo bias) is 1

combined with the RCS flow pmcess instmmentation uncenainty above. The total flow is i

i I

.]

assumed to be calculated using only two of die thme elbow taps per loop. Therefore, the single loop flow uncedainty expressed as a percent of loop flow may be expressed as a percent of total RCS flow by dividing by the , ,

'Ihc resulting RCS tlow uncenainty is:

g= 1.76% flow (Total 4 Imop RCS flow uncenainty)

Ihe flow uncenainties for t!1e McGuire loss of flow setpoint are given in the table below.

.19. Pronosed McGuire law RCS IAon Flow Reactor Trio Function Measurement Uncertainty Parameter Allowance  % Flow *

~ ~

Process Measurement Accuracy (PMA)

Density effects on AP cell , _.

Pmcision flow calorimetric . 3,13 Noise

{ ]

Sensor Calibration Accuracy (SCA) 0.50% AP span 0.36 Sensor Measurement & Test Equipment (SMTE) 0.31% AP span 0.22 Sensor Drift (SD)

Sensor Temperature Effects (STE) 0.65% AP span 0.47 Sensor Pressum Effects (SPE) 0.50% AP span 0.36 Rack Calibration Accuracy (RCA)

Rack Comparator Setting Accuracy (RCSA) 0.35% AP span 0.25 Rack Drift (RD)

Rack Temperature Effects (RTE)

~~

l Bias, llarton transmitter, (Pressurizer pressure)

AH wance n sp n w sp n 0% Row '

• Allowance in % Flow =

t 2 s ( 100% AP span , s 100% Flow span,

'lhe aluve uncenaintics are combined using the equation below:

a I

m W

1

<:sA e CSA = 3.79 % now

'Ilic Allowable Value is calculated below.

Saicty Analysis Limit = 86.5% flow Total Allowance (TA) = 4.50 % flow An increase in the total allowance from 3.50% llow to 4.50% flow was made to maintain the allowable value at or near its current .alue. 'Ihis will prevent the need for excessive transmitter calibrations resulting from an allowable value which is too close to the setpoint.

Tg = (

} = 1.33 % flow .

T, =

T 2= 1.0 % llow T i> T therefore 2 T is2used to calculate the Allowable Value of 90 % flow The table below summarizes the changes necessary to account for the additional uncertainty associated with using the unnormalized elbow taps for flow surveillance.

CSA TS Setpoint/ Allowable Margin Total Allowance Value Existing 2.33% llow 90%/88.8% flow 1.17% flow 3.50% flow Proposed 3.79% flow 91%/90% flow 0.71% flow 4.50% flow j

I l

11. Proposed Catawba Total RCS Flow Technical Specification Surveillance Uncertaintv (For Proposed Elbow Tap Flow Surveillance)

Paramelen Allomt!!Cr  % Flow

• Process Measurement Accuracy (PM A)

Density cliccis on AP cell 0.50% AP span 0.36 Noise

{

Sensor Calibration Accuracy (SCA) 0.50% AP span 0.36 Sensor Measurement & Test Equipment (SMTE) 0.20% AP span 0.14 Sensor Drift (SD)

Sensor Temperature Effects (S111) 0.70% AP span 0.50 Sensor Pressure Effects (SPE) 0.50% AP span 0.36

{

Rack Calibration Accuracy (RCA) '

Rack Drift (RD)

Rack Temperature Effects (RTE)

Computer Isolator Drift (ID)

Allowance for noisy signal (RDOT)

Analog to digital conversion accuracy (A/D)

, Allowance in % Flow = f Allowance in AP span' '120% Flow span Vl

. 120% Flow '

( 2 s s 100% AP span A 100% Flow span j 1he above uncenainties are combined using the equation below:

csA - .

_ J CSAa CS A = 1.89 % flow (Single elbow tap uncenainty)

~

7e n o w + fo bias) is The combined uncedainty wi:h the RCS Dow associated with uncena process instrumentation the precision calorimetric (iniy abo assumed to be calculated using only two of the three elbow taps per loop. De resulting RCS tlow uncenainty is:

l 1

= 1.88% flow (Total 4 Loop RCS flow uncenainty)

I i ne now uncenainues nor tl e Catawba loss of flow setpoint are given in the table below.

12. Proposed Catawba Low RCS Imon Flow Reactor Trio Function Measurement Uncertainty Parametet Allowance  % Flow

• Process Measurement Accuracy (PMA)

Density effects on AP cell Precision flow calorimetric

( )

2.71 Noise

]

Sensor Calibration Accuracy (SCA) 0 0% AP span 0.36 Sensor Measurement & Test Equipment (SMTE) 0.20% AP span 0.14 Sensor Drift (SD) r

_ j Sensor Temperatum Effects (STE) 0.70% AP span 0.50 Sensor Pressure Effects (SPE) 0.50% AP span 0.36 Rack Calibration Accuracy (RCA)

Rack Comparator Setting Accuracy (RCSA)

Rack Drift (RD)

Rack Temperature Effects (RTE)

Bias,13 anon transmitter, (Pressurizer pressure) i

' Allowance in AP span '120% Flow span T' 120% Flow

, Allowance .in % Flow =

t 2 < 100% AP span j (100% Flow span, 7he above uncenainties are combined using the equation below:

CSAsi g l

J p

t i

CSA =

CS A = 3.42 M. Ilow lhe Allowable Value is calculaled below.

Safety Analysis Limit = 86.59 flow Total Allowance (TA) = 4.50 % Ilow.  !

An increase in the total allowance fmm 3.50% flow to 4.50% flow was made to maintain the allowahic value at or near its current value. 'Ihis will prevent the need for excessive transmitter calibrations resulting fann an allowable value which is too close to the r setpoint.

T=1

]= 1.3 % 110w T, = .. t

~

-j T 2= 1.34 % flow T 1< T therefore 2 T was 1 used to calculate the Allowable Value of 89.7 % flow

'Ite table below summarizes the changes necessary'to account for the additional uncertainty associated with using the unnonnalized elbow taps for flow surveillance.

t CSA TS Setpoint/ Allowable Margin Total Allowance l Value Existing 2.35% flow 90%/88.8% flow 1.15% flow 3.50% flow Proposed 3.42% flow 91%/89.7% flow 1.08% flow 4.50% flow t

i n

l a

eo .

1 1

]

i i

i l

l

l This attachinent summarizes the following twelve uncertainty calculations:

1. Feedwater Venturi Flow Measurement Uncertainly fi>r McGuire

2. Calorimetric Contribution to McGuire Flow Measurement Uncertainty

3. Current McGuire Total RCS Flow Technical Specification Surveillance Uncenainty f

4. Current McGuire Low RCS l>>op Flow Reactor Trip Function Measurement Uncertainty

5. Feedwater Venturi Flow Measurement Uncertainly for Catawba

6. Calorimetric Contribution to Catawba Flow Measurement Uncenainty

7. Current Catawba Total RCS Flow Technical Specification Surveillance Uncertainty

8. Current Catawba Low RCS Loop Flow Reactor Trip Function Measurement Uncertainty

9. Prolvised McGuire Total RCS Flow Technical Specification Surveillance Uncertainty

10. Pmposed McGuire 1;>w RCS Loop Flow Reactor Trip Function Measurement Uncertainty

11. Pn> posed Catawba Total RCS Flow Teclutical Specification Surveillance Uncertainty

12. Pmposed Catawba Low RCS Loop Flow Reactor Trip Function Measurement Uncenainty l