ML20217G711
| ML20217G711 | |
| Person / Time | |
|---|---|
| Site: | South Texas  |
| Issue date: | 07/31/1997 |
| From: | HOUSTON LIGHTING & POWER CO. |
| To: | |
| Shared Package | |
| ML19317C566 | List: |
| References | |
| NUDOCS 9708080039 | |
| Download: ML20217G711 (51) | |
Text
ATTACHMENT 6 RCS Flow Measurement Using Elbow Tap Methodology Licensing Submittal July 1997 NON PROPRIETARY f
4 9708000039 970806 PDR ADOCK 05000498 P
RCS Flow Measurement Using Elbow Tap Methodology Licensing Submittal July 1997 NON-PROPRIETARY
i i
Tcble cf Ccatents 1.0 B A C K G R O UN D..........................................................
1 2.0 PRO PO S E D Ci l A NG E....................................................
1 3.0 S A F ETY EVA LU ATI ON...................................................
1 3.1.
INTRODUCTION.........
1 4
h 3.2 S U M M A RY......................
2 3.3 RCS IlOT LEO TEMPERATURE STREAMING 2
3.3.1 Phenomenon 2
3.3.2 l l i st o ry.....................................................
3 3.3.3 llot Leg Streaming impact on RCS Flow Measurements...................
4 3.3.4 Correlation of Changes in Power Distribution and RCS Flow................
4 3.4 ELBOW TAP FLOW MEASUREMENT APPLICATION..,..................... 10 3.4.1 Elbow Tap Flow Measurements............................,...,... 10 3.4.2 Elbow Tap Flow Measurement Procedure 12 3.5 BEST ESTIMATE RCS FLOW ANALYSIS................................ 17 3.5.1
Background
17 3.5.2 Prairie Island Ilydraulics Test Program............................... 17 3.5.3 Additional Prairie Island Tests..................................... 19 3.5.4 System Flow Resistance Analyses................................... 19 3.5.5 Best Estimate RCS Flow Calculations................................ 20 3.6 EVALUATION OF SOUTil TEXAS PROJECT RCS FLOW PERFORMANCE,
20 3.6.1 Best Estimate Flow Predictions...........
21 3.6.2 Evaluation of Elbow Tap Flows.................................... 22 3.6.3 Evaluation of Calorimetric Flows,.................................. 23 3.6.4 Flow Com parisons......................................
24 3.6.5 Power / Flow Correlation for South Texas Project 24 3.7 ELBOW TAP FLOW MEASUREMENT LICENSING CONSIDERATIONS 31 3.7.1 B ac k grou n d................................................. 31 3.7.2 Supporting Calculations.......................................... 31 3.7.3 Potential Document impacts..........................
32 APPENDIX A -
INDICATED RCS FLOW and REACTOR COOLANT FLOW -
LOW REACTOR TRIP INSTRUMENT UNCERTAINTIES.,......
33 APPENDIX B -
NO SIGNIFICANT HAZARDS CONSIDERATION......................
39 APPENDIX C -
MARKUPS TO SOUTH TEXAS PROJECT TECHNICAL SPECIFICATIONS and BASES.........
43 i
List cf Fig:res 3.3 1 UPPER PLENUM AND RCS HOT LEG FLOW PA1 TERNS........................
6 3.3 2 TYPICAL CORE EXIT TEMPERATURE GRADIENT AND RCS IlOT LEG CIRCUMFERENTIAL TEMPERATURE GRADIENT,,..................
7 3.3-3 TYPICAL CORE EXIT TEMPERATURE CilANGE.......................,........
8 3.3 4 CALORIMETRIC FLOW MEASUREMENT BIAS VERSUS DIFFERENCE IlETWEEN AVERAGE SECOND ROW AND OUTER ROW ASSEMBLY POWERS.........
9 3.4 1 LEADING EDGE AND ELBOW TAP FLOW METER LOCATIONS AT PR AI RI E ISLAN D UNIT 2............................................... 16 3.61 SOUTil TEXAS PROJECT UNIT 1 FLOW COMPARISONS.......................... 28 3.6-2 SOUTil TEXAS PROJECT UNIT 2 FLOW COMPARISONS..............
29 3.6 3 FLOW 131AS VS POWER DIFFERENCE.......
30 List of Tables 3.4 1 COMPARISONS of LEFM and ELBOW TAP FLOW MEASUREMENTS AT PRAIRIE ISLAND UNIT 2 15 3.6 1 SOUTil TEXAS PROJECT BEST ESTIMATE FLOW
SUMMARY
,.................,... 25 3.6-2 SOUTil TEXAS PROJECT ELBOW TAP DIFFERENTIAL PRESSURE
SUMMARY
26 3.6-3 SOUTH TEXAS PROJECT CALOR! METRIC FLOW
SUMMARY
27 A-1 BASELINE FLOW CALORIMETRIC INSTRUMENTATION UNCERTAINTIES.....................
34 A-2 FLOW CALORIMETRIC SENSITIVITIES 35 A-3 CALORIMETRIC RCS FLOW MEASUREMENT UNCERTAINTIES..
36 A-4 COLD LEG ELBOW TAP FLOW UNCERTAINTY (QDPS/ PROCESS COMPUTER)........
37 A-5 LOW FLOW REACTOR TRIP..................
38 ii
South Texas Proiect Units 1 and 2 RCS Flow Measurement Usine Elbow Tap Methodoloey 1 icensine Submittal
1.0 BACKGROUND
Current Technical Specifications for South Texas Project Units 1 and 2 have a surveillance requirement to determine the Reactor Coc! ant System (RCS) total flow rate by a precision heat balance measurement at least once per 18 months. The RCS total now limit is the value assumed in the transient and accident analysis (plus measurement uncertainties) required to maintain minimum Departure from Nucleate Boiling Ratio (DNBR). The current surveillance method calculates RCS total flow based on steam generator thermal output from a precision calorimetric measurement, divided by the enthalpy difference across the reactor vessa:1 as indicated by the hot and cold leg ResistanceTemperature Detectors (RTDs), in recent cycles, measurements for both Unit I and Unit 2 have indicated apparent decreases in RCS total flow rates. Ilowever, these decreases are not substantiated by the changes that have occurred in the system hydraulics, and are not confirmed by other indicatiens of loop Dow. Changes in core reload designs have resulted in core exit temperature distributions that, when comtined with incomplete Dow mixing and asymmetric How patterns in the reactor vessel upper plenum, produce varying hot leg temperature indications. The net efTect of these phenomena has resulted in what has been referred to as hot leg streaming. Ilot leg streaming effects directly impact the hot leg temperatures used in the calorimetric based RCS How measurement, resulting in calculated RCS total flow rates that are lower than actual values. The apparent RCS total flow reductions caused by hot leg streaming have resulted in the measured RCS Cow limit closely approaching the Technical Specification minimum, with a minimum RCS total Dow margin as low as 0.37% having occurred in Unit 2.
l 2,0 PROPOSED CHANGE The current Technical Specification Table 2.21 (page 2-4), " Reactor Trip System Instrumentation Trip Setpoints,"
provides the Trip Setpoint and Allowable Value for the RCS Flow Low trip. The Allowable Value is to be changed to reflect the increased uncertainty associated with the correlation of the elbow taps to a previous baseline caloi! metric. In addition, Technical Specification 3.2.5 (page 3/4.2-11), " Power Distribution Limits, DNB Parameters," is to bc changed to allow the RCS total flow to be measured by the elbow tap Ap method. These changes will include modification of surveillance requirement 4.2.5.3, which currently requires performance of a precision heat balance every 18 months, to not specify the method for RCS Dow measurement to be used at the beginning of each fuel cycle. Appropriate Technical Specification Bases sections will also be revised to reflect use of the elbow tap Ap method for flow measurement and to provide clarification. The revised Technical Specifications are in Appendix C.
3.0 SAFETY EVALUATION
3.1 INTRODUCTION
Reactor Coolant System (RCS) secondary calorimetric based How measurements at many pressurized water reactor plants, including South Texas Project Units 1 & 2, have been affected by increases in hot leg temperature streaming. The increases are related to changes in the reactor core radial power distribution resulting from implementation oflow leakage core loading patterns, in some cases, measured flow appears to have decreased to, or below, the minimum measured flow required by the Technical Specifications. Such occurrences require licensee actions to either account for the apparent now reduction in the plant safety analyses or to confirm by other means that RCS flow has not decreased below the specified limit. In many cases, plants have relied on the repeatability of RCS elbow tap flow meters to demonstrate that RCS flow has not decreased. This alternate approach confirms RCS How by a normalization process using both calorimetric and elbow tap flow measurements.
PageI
Sputh Texas Proleet Units 1 and 2 RCS Flow Measurement Usine Elbow Tan Methodoloev I.leensine Submittal Currently, the Technical Specifications require that RCS flow be measured once per fuel cycle to demonstrate that the actual flow is greater than the minimum flow assumed for the safety analysis. This Safety Evaluation justifies use of an altemate method to measure total RCS flow at South Texas Project Units 1 & 2.
The current RCS calorimetric How measurement method based on RCS temperature and secondary calorimetric power measurements has inherent limitations imposed by changes in the core radial power distribution. He proposed attemate method using elbow tap Cow measurements normalized to a measured baseline calorimetric flow minimizes these limitations.
3.2
SUMMARY
The procedure described in this safety evaluation for verifying RCS total Dow with elbow tap flow l
measurements normalized to calorimetric flow measurements has been approved by the Nuclear Regulatory I
Commission for application at other nuclear power plants. Applicability of the procedure has been confirmed by comparing measured RCS elbow tap Dow trends with best estimate flow trends based on analysis and application of RCS hydraulic test data (Section 3.6)
Evaluation of plant operating data from South Texas Project Units 1 & 2 has defined sufficiently accurate baseline parameters for both the elbow tap and calorimetric flow measurements. Flow changes measured by elbow taps obtained over several fuel cycles are consistent with the predicted flow changes due to changes in RCS hydraulics, as shown on Figures 3.6-1 and 3.6-2. Application of the procedure using normalized elbow tap measurements will result in the recovery of the apparent decrease in How attributed to changes in hot leg temperature streaming.
While modifications to the South Texas Project Technical Specifications will be needed to allow use of the alternate RCS flow measurement procedure, no unreviewed safety questions have been identified.
3.3 RCS IlOT LEO TEMPERATURE STREAMING 3.3.1 Lhenomenon The RCS hot leg temperature measurements are used in control and protection systems to ensure temperature is within design limits, and in a surveillance procedure with secondary plant calorimetric power measurements to determine the RCS flow. Uncertainty in the hot leg temperature measurement can have a significant impact on PWR performance.
A precise measurement of hot leg temperature is difficult due to the phenomenon known as hot leg temperature streaming, i.e., large temperature gradients within the. hot leg pipe resulting from incomplete mixing of the coolant leaving fuel assemblies at different temperatures. The magnitude of these hot leg temperature gradients where the temperatures are measured is a function of the core radial power distribution, mixing in the reactor vessel upper plenum, and mixing in the hot leg pipe.
Page 2
South Texas Prolect Units 1 and 2 - RCS Flow Measurement Usinn Elbow Tan Methodolocv Licensine Submittal Prior to application oflow leakage core loading patterns, the largest difTerence in fuel assembly exit temperatures at full power was typically no more than 30 F.
The lowest temperatures wee measured at the exit of fuel assemblies on the outer row of the core. Flow from a fuel assembly in the center of the core mixes with coolant from nearby fuel assemblies as it flows around control rod guide tubes and support columns toward the hot leg nonles. Flow from a fuel assembly on the outer row of the core, separated from the center region Hows by the outer row of guide tubes, has little opportunity to mix with hotter flows before reaching the nonles, so a significant temperature gradient can exist at the noule.
Since hot leg flow is highly turbulent, additional mixing occurs in the hot leg pipe, and the maximum gradient where temperature is measured,7 to 17 feet downstream from the reactor vessel nonle, is less than at the nonie. In 1968, gradients measured on the circumference of the pipe were as high as 7 to 10 F, so turbulent mixing in the pipe did not climinate the gradient introduced at the core exit.
The 1968 tests and subsequent tests showed that the highest temperatures are in the top half of the pipe, while the lowest temperatures are in the bottom half, as expected, since the colder water from the outer row of fuel assemblies is closest to the bottom half of the hot leg nonle.
Figure 3.3-1 illustrates a postulated flow pattern in the reactor vessel upper plenum between the core exit and the hot leg noule. Figure 3.3 2 illustrates typical temperature gradients at the core exit and on the hot leg circumference at the point where the temperatures are measured. Typically, the core exit and hot leg gradients remain relatively stable, changing only slightly as the radial power distribution changes during a fuel cycle.
3.3.2 History Prior to 1968, there were no multiple temperature measurements on hot leg pipes, so temperature streaming gradients were undetected and resistance temperature detector (RTD) locations were based on other criteria.
During startup of a Westinghouse-designed 3-loop plant in 1968, RTDs on opposite sides of the hot leg pipes measured difTerent temperatures. Recalibrations and special tests confirmed that the measurements were valid, so Westinghouse concluded that the hot leg temperature differences resulted from incomplete mixing of Dows leaving fuel assemblies at different temperatures. To confirm this conclusion, thermocouples were strapped to the outside of two hot leg pipes, and gradients were detected that increased as core power increased. The maximum full power gradient was 10 F in one loop and 7'F in the other loop. Since only one RTD was used to define hot leg temperature for control and protection systems, the hot leg temperature measurement was not as accurate as intended.
With additional analyses and development, Westinghouse designed and installed new instrumentation systems at other plants aller 1968 to compensate for hot leg temperature streaming gradients. The new system, called the RTD Bypass System, employed scoops in the hot leg piping at three uniformly spaced locations on the circumference of the pipe. Holes on the upstream side of the scoop collected small sample flows. The three sample flows, which were at different temperatures, were combined and directed through an RTD manifold where the average hot leg temperature was measured.
Page 3
South Texas Project Units 1 and 2 - RCS Flow Measurement Usine Elbow Tan Methodolocy Licensine Submittal To eliminate personnel radiation exposure to RTD Bypass System piping during plant shutdowns, Westinghouse replaced many systems afler 1988 with a system having three thermowell RTDs in each hot leg. The RTDs were installed at uniformly spaced locations, like the RTD bypass scoops, to retain the three measurements on the hot leg. In many cases the thermowell RTDs were installed
.nside the bypass scoops, so the average thermowell RTD measurement was the same as the temperature by the RTD Bypass System.
Subsequent to 1968, additional hot leg streaming measurements were performed at 2-loop,3 loop and 4 loop plants. The results of these measurements were used in several analyses to define hot leg temperature streaming uncertainties used in safety analyses and protection system setpoint calculations. Gradients measured in these tests varied from 7 to 9 F. Afler 1988, the thermowell RTD systems provided hot leg streaming data from the three RTDs in each hot leg. The gradients measured prior to 1991 varied from 2 to 9 F with most of the gradients measured at 5 to 7 F.
3.3.3 Hot i en Streamine Impact on RCS Flow Measurements Before 1988, reports of hot leg temperatut 9 measurement problems were unusual, and no significant changes in streaming gradients were identified. In 1988, the first significant indication of a streaming change occurred at a 4-loop plant, followed by similar occurrences in 1989 and 1990 at three more 4-loop plants, in all four cases, the measured coolant temperature rise across the core (AT = Tu - Tg) had increased from that measured in previous fuel cycles by as much as 3%.
l Since coolant AT is a major input in determining the measured RCS calorimetric flow, a AT l
increase of 3% implied that RCS flow had apparently decreased by 3%. Many other plants, including South Texas Project Units 1 & 2 and several 3 loop and 4-loop plants, have also reported apparent flow reductions. In some cases,the apparent flow wasjust at or above the minimum flow requirement specified in the Technical Specifications, raising a concern that measured flows could be lower in future cycles. In all cases, however, RCS elbow tap flows indicated that the actual How had not significantly changed.
Both units at one plant site in 1990 reported that calorimetric flows appeared to be below Technical Specification requirement. After additional data had been evaluated, data from elbow taps confirmed that RCS flow was adequate. The Nuclear Regulatory Commission was advised of the apparent low calorimetric flow indication and the elbow tap flow data. The Nuclear Regulatory Commission concurred with the licensee's conclusion that RCS flow was adequate for safe operation at full power for the remainder of the cycle.
3.3.4 Correlation of Chances in Power Distribution and RCS Flow At the plants where apparent flow reductions were measured, Westinghouse noted that in all cases the core exit thermocouples measured much larger temperature gradients, approaching 60 F, as shown on Figure 3.3-3, due to much lower exit temperatures at the edge of the core. A review of core radial power distributions indicated that the power generated in outer row fuel assemblies decreased significantly from power levels measured in earlier cycles, confirming the large core exit -
temperature gradients.
Page 4
South Texas Project Units 1 and 2 - RCS Flow Measurement Usine Elbow Tan Methodolouv Licensine Submitta Westinghouse comp. '. radial power distributions and calorimetric flow measurements obtained from seveal cycles L.
weral 3-loop and 4 loop plants, and concluded that the apparent changes in flow correlate with the radial power distribution gradient at the edge of the core
. Figure 33 4 plots apparent low leakage loading pattern induced flow decreases measured at a group of 3 plants versus the difference between the average power generated in second row and outer row assemblies. The apparent flow decreases appear to occur when power difTerences exceed 50%, a condition consistent with low leakage loading patterns. The correlation of power difference versus flow can be represented by a straight line, as shown on Figure 3.3 4. According to this data, the measured RCS flow appears to decrease by 3% as the difTerence between power in second row and outer row assemblies increases from 49% to 78%.
Page 5
. South Texqholect Units I and 2 - RCS Flow Measurement Usine Elbow Tap Methodolony 1.icensine Subnthaj FIGURE 3.31 UPPER PLENUM and RCS liOT LEO FLOW PAT 111RNS
)
RD Upper I leu I
RCS Hot Leg Reactor Core j
4 Outer row of assemblies Center Page 6
South Texas Project Units I and 2 - RCS Flow Measurement Usinn Elbow Tan Methodolony Lleensinn Submittal FIGURE 3.3 2
. TYPICAL CORE EXIT TEMPERA'IURE GRADIENT and RCS IlOT LEO CIRCUMFERENTIAL TEMPERATURE GRADIENT 10 C
L
-> Outer row 2
l
$0 L,,
i -.-
0 m.10 O
e 0
, 20 T
a:
30 0
20 40 60 80 100 Core Center Core Area (%)
C
- r 59 0
~
3-20 2-2 U
a
~
j Top
",3 i
i 180*.
240' 300' 0*
60' 120*
180' Dottom Hot Leg Pipe Circumference cotto.o Piige 7
. South Texas Prolect Units 1 and 2 - RCS Flow Measurement Usine Elbow Tap Methodolony 1.lcensina Submittal FIGUlG 3,3 3 1YPICAl, CORE EXIT TEMPERA 1URE CilANGE o
BOC 4
+
BOC5 20
+
M
\\
{10 o
o g +
g t,
o
'-3
\\
o o
E Rs 8 0
0
+\\g a
s N h.
w s
\\
9
+'s p-10 g
\\
Eo
\\
\\
" 20
\\
\\
g
+
\\
5
\\
o 5
-\\
$-30
\\
s.
t t
t t
t t
t t
1 0
10 20 30--40
- 50. 60.70
- 80. 90 -100 Core Area (% from Center)
+
Page 8
South Texas Prolect Units I rad 2 RCS Flow Measurement Usinn Elbow Tap Methodolony 1.icensing Submittal FIGURE 3.3-4 CALORIMETRIC FLOW MEASUREMENT BIAS
.VERSUS DIFFERENCE BETWEEN AVERAGE SECOND ROW AND OUTER ROW ASSEMBLY POWERS
+1%
O O
i F
0%
O
'o L
O O
W.
O 000 O
O B
-1%
O O
I O
O A
O O
O S
00 00
-2%
O O
O o
O O
-3%
0 2nd ROW - OUTER ROW POWER DIFFERENCE 100 PERCENT POWER I
i Page 9
South Texas Proiect Units 1 and 2 RCS Flow Measurement Usine Elbow Tan Methodoloev l.icensine Submittal 3.4 ELHOW TAP FLOW MEASUREMENT APPLICATION 3 A.1 Elbow Tan Flow Measurements Elbow tap differential pressure (Ap) measurements are being used more frequently in the industry to determine if, or by how much, RCS flow has changed from one fuel cycle to the next. Elbow tap flow meters are installed in all Westinghouse PWRs on the RCS pump suction piping on each loop, as shown for Prairie Island on Figure 3.41. The Ap taps are located on a plane 22.5' around the first 90' elbow. Each elbow has one high pressure and three low pressure taps connected to three redundant Ap transmitters. Elbow taps in this anangement are used to define relative rather than absolute flows, due to the lack of straight piping lengths upstream from the elbow. The Ap measurements are repeatable and thus provide accurate indications of flow changes during a cycle or from cycle to cycle.
The RCS elbow tap flow meters' are a form of centrifugal meter, measuring momentum forces developed by the change in direction around the 90' cibow. De principal parameters defining the Ap for a specified flow are the radius of curvature of the elbow and the diameter of the flow channel through the elbow. Tests' have demonstrated that elbow tap flow measurements have a high degree of repeatability and that the flow measurements are not affected by changes in roughness of the elbow surface.
Specific phenomena that have affected other types of flow meters or that might affect the elbow tap flow meters in the RCS piping applistion have been evaluated to determine if these phenomena would affect repeatability of the flow measurement. In addition, measurements at Prairie Island Unit 2, where the highly accurate ultrasonic Leading Edge Flow Meter (LEFM) is installed, were compared with elbow tap measurements to confirm elbow tap How measurement repeatability. The results of these evaluations and comparisons are summarized in the following paragraphs.
Venturi Fouline Venturi flow meters in feedwater systems are affected by crud deposits (i.e., fouling) that affect surface roughness, local pressures, and flow area through the venturi throat. Fouling is apparently caused by an electro-chemical ionization plating of copper and magnetite particles in the feedwater on the venturi surfaces. The fouling process is directly related to the velocity increase as flow approaches the smaller venturi flow area. This condition is not present in an elbow since there is no change in cross section to produce a velocity increase and ionization, in addition, surface roughness changes as experienced in venturi flow meters do not affect the elbow tap flow measurement.
3
" Fluid Meters, Their Theory and Application",6th Edition, Howard S. Bean, ASME, New York,1971.
Page 10
South Texas Protect Units I and 2 - RCS Flow Measurement Usine Elbow Tap Methodolony 1,1censine Submittal Meter Dimensional Chances The elbow tap flow meter is part of the RCS pressure boundary, so there are only minimal dimensional changes associated with pipe stresses, and pressure and temperature are the same (near full power conditions) whenever flow measurements are made. Erosion of the stainless steel elbow surface is unlikely, and velocities are not large (42 fps) relative to erosion. The effects of a dimensional change or erosion could only affect How by changing elbow radius or pipe diameter, and these dimensions are very large relative to a possible dimensional change. Therefore, elbow tap flow meters are considered to be a highly stable flow measurement element.
Unstream Velocity Distribution Effects The velocity distribution entering the steam generator outlet nonle may be skewed by its off-center location relative to the tube sheet. The velocity distribution entering the 90' elbow where the flow meter taps are located may also be skewed by the out-of-plane upstream 40*
elbow on the steam generator outlet nonle. Ilowever, these velocity distributions, including l
the distribution in the elbow tap Dow meter, remain constant so the elbow tap Dow meter l
Ap/ flow relationship does not change.
Another upstream effect that was considered was steam generator tube plugging. Tube plugging is typically distributed randomly across the tube sheet, so the velocity distribution l
approaching the outlet nonle does not change as additional tubes are plugged. The velocity distribution could change if a large number of tubes were plugged in one area of the tube sheet.1-lowever, the plenum velocity head approaching the outlet nonle is small compared to the pipe velocity head (0.6 fl versus 27 0), and the large change in now area greatly reduces or Gattens an upstream velocity gradient. Therefore, any tube plugging, even if asymmetrically distributed, does not impact elbow tap Dow measurement repeatability.
Also considered was the effect of steam generator replacement on the elbow tap flow measurements. The replacement steam generators will have the same outlet nonle off center location and the same nozzle diameter and taper. Since the configuration is the same and the same difference in plenum and nonle velocity heads will exist, steam generator replacement will have no impact on the elbow tap flow coefHcient. The RCS How will increase since there will be no plugged tubes and the steam generator flow resistance will be reduced; the elbow taps will correctly measure the increase in flow.
Page 11 l
J
South Texas Prolect Units I and 2 - RCS Flow Measurement Usine Elbow Tan Methodolocy I icensine Submittal Flow Measurement Comnarisons The LEFMs installed at Prairie Island Unit 2 provided data ta confirm repeatability of elbow tap flow meters. The comparisons, listed in Table 3.41, covered i1 years of plant operation, during which a significant change in system hydraulics was made. A reactor coolant pump impeller was replaced, and the replacement impeller produced additional flow. The LEFM data after impeller replacement was in agreement with the predicted How change, and the elbow tap flow meters indicated similar changes. The ll year flow comparison shows that the average difference between cibow taps and LEFMs was less than 0.3% Oow. Another comparison of data obtained before and after impeller replacement showed that measurements agreed to within 0.2% Cow on the ratio of ficws with one and two pamps in operation, thus further confirming the relative flow measurements from elbow tap flow r"ters.
3.4.2 Elbow Tao Flow Measurement Procedure The elbow tap flow measurement procedure relies on repeatability of elbow tap Aps to accurately verify RCS Dow. Comparison of elbow tap measurements at or near full power from one cycle to the next provides an accurate indication of any change in Dow. When normalized to calorimetric flows, the elbow tap Aps can accurately verify flow for any future fuel cycle. The elbow tap procedure for verifying RCS flows is described in detail below.
Baseline Ca!orimetric Flow The Baseline Calorimetric Flow is defined as the calorimetric flow which best reprecents the actual plant flow at the beginning of plant life. Calorimetric flow measurements obtained during early fuel cycles before low leakage losding pattern application are expected to be consistent with the best estimate flow predictions, both in total now and in changes in flow resulting from known hydraulics changes, based on the best estimate flow analyses described in Section 3.5.
Any early cycle calorimetric measurement which determines flow for the cycle to be within the specified measurement uncertainty could be used to define the baseline calorimetric flow.
To improve accuracy, calorimetric Dows from all fuel cycles are evaluated for use in defining baseline calorimetric How. If a known hydraulics change (e.g., tube plugging) was made before a cycle, calorimetric flow for the cycle should be adjusted so all Dows have a common hydraulic baseline. The hydraulic configuration that existed at initial plant startup is usually defined to be the common hydraulic baseline. After adjustment, all cycle calorimetric flows should be similar, differing only by a calorimetric measurement repeatability allowance.
Calorimetric flows that fall well outside the allowance (either high or low) should not be used in defining baseline flow. Calorimetric flows appearing to be significantly impacted by low leakage loading patterns and hot leg streaming are not typically considered since the objective of the procedure is to correct for the impact of low leakage loading patterns. Additionally, calorimetric flows that are significantly higher than the best estimate flow should not be included in the baseline flow calculation because they introduce a non conservative bias.
Page 12
South Texas Fiolect Units 1 and 2 - RCS Flow Measurement Usine Fibow Tan Methodolocy Licensine Submittal The accuracy of the baseline calorimetric flow measurement is based on plant specific instrumentation uncertainties that existed when the flow measurements used to define baseline -
Dow were performed, instrument uncertainty calculations, described in Section 3.7, define the total Dow measurement uncenainty. -Included in the baseline calorimetric flow measurement uncertainty is an allowance for non-conservative hot leg temperature streaming based on streaming gradients that existed when baseline flow measurements were perfonned. Although low leakage loading patterns cause larger streaming gradients, the streaming uncertainty becomes more conservative, so a larger, low leakage loading pattem induced streaming uncertainty is not needed.
Baseline Elbow Tan AP Elbow tap Aps obtained in the first cycle define a baseline elbow tap Dow coefficient, which is used in connection with the baseline calorimetric How to define a future cycle flow. The baseline elbow tap flow coefficient (B) is defined by the following equation:
B = Ap,
- y, (Eq.1) where:
B
= baseline elbow tap total flow coeflicient, (inches 110
- f9/ lb),
2 Apu
= baseline average elbow tap Ap (inches 110),
2 3
vn
= average cold leg specific volume (ft / lb).
I l
The baseline elbow tap Dow coefficient, based on the average Ap from all elbow taps, defines
}
the total now to be consistent with the total baseline calorimetric How. Repeatability and l
accuracy are improved when all elbow tap Ap measurements are used.
Flow Verification for Future Cycles Elbow tap Aps will be obtained at the beginning of a future cycle to define the change from the baseline flow. The average of all elbow tap Aps measured at or near full power defines the future cycle elbow tap flow coefficient (K), applying the equation:
K=Ar*Vr (Eq.2)_
P where:
K
= future cycle elbow tap total flow coeDicient, (inches H O
- f9/ lb).
2 App
= average future cycle elbow tap Ap (inches 110),
2 3
r
= average future cycle cold leg specific volume (ft / lb).
v The change in flow from the baseline cycle to the future cycle is defined by the elbow tap flow ratio (R), based c:1 he equation:
R = (K / B)"'
(Eq. 3) where:
R
= ratio of future cycle flow to baseline flow.
1 Page 13 I
- South Texas Proleet Units I and 2 - RCS Flow Measurement Usine Elbow Tan Mcthodolocv 1.lcensine Submittal The future cycle flow is determined by multiplying the baseline calorimetric flow by the elbow tap How ratio (R), applying the following equation:
FCF = R
- BCF (Eq.4) where:
FCF
= total future cycle Dow, gpm, BCF
= total baseline calorimetric flow, gpm.
Best Estimate Flow Confirmation A future total flow detennined from an elbow tap flow measurement is confirmed by comparing the measured elbow tap flow ratio (R) with an estimated flow ratio (R') based on the best estimate flow analysis (described in Section 3.5) of known RCS hydraulics changes such as steam generator tube plugging or fuel design changes. The estimated flow ratio is defined by the following equation:
R' = FEF / BEF (Eq.5) where:
= future cycle estimated flow, the estimated RCS flow, based on actual RCS hydraulics changes, BEF
= best estimate flow, the estimated initial (baseline) cycle RCS flow, based on hydraulics analyses, t
l An acceptance criterion is applied to the comparison of R and R':
If R s (1.004
- R'), the elbow tap flow ratio R is used to calculate the future cycle RCS total flow using Equation 4.
If R > (1.004
- R'), the quantity (1.004
- R') is used to define the future cycle RCS total flow, modifying Equation 4 as indicated below.
FCF = 1.004
- R'
- BCF (Eq. 6)
The multiplier (1.004) applied 'to R' is an allowance for the elbow ts. ;iow measurement repeatability. Since the elbow tap Dow measurement uncertainty includes this repeatability allowance, the measured How ratio [R) can be 0.4% higher than the estimated flow ratio [R']-
and still define a conservative flow.
Application of this acceptance criterion results in definition of a conservative future cycle flow,
{
confirmed by both the elbow tap measurements and the best estimate hydraulics analysis.
Page 14
South Texas Project Units 1 and 2. RCS 1710w Measurement Usinn 1%ow Tan Methodolony 1 icensinn Submittal l
TAllLE 3,41 COMPARISONS of LEFM and ELiloW TAP FLOW MEASURhMENTS AT PRAIRIE ISLAND UNIT 2 RCS FLOW MEASUREMENT COMPARISONS AT FULL POWER gpm/ loop loop / Meter A/LEFM A/ Elbow il/LEFM
!!/ Elbow Feb 1980 97519 97950 Jul 1981 98673 98309 97763 97267 Aug 1991 98724 98$$7 97$43 97607
- Normallred to LEFM Flow RATIO OF FLOW WITil 1 PUMP OPERATING TO FLOW WITil 2 PUMPS OPERATINO loop / Meter
-- A/LEFM A/ Elbow 11/LEFM ll/ Elbow Dec 1974 1,0819 1,0777 1,0852 1.087$
Jul 1981
-1.0794 1,0816 1.0820 1.0820
__=
Page 15 L
South Texas Project linits 1 and 2 I(CS Ilow Measurernent tfs'ne !?lbow Tan Methodoloey I Icentine Submittal 110U111!3.4.I Ll!ADING EDGli Fl.OW Ml?TliR AND El.110W TAP l't.0W Ml!TER !.OCATIONS AT PRAIRIE ISI.AND UNIT 2 l
STEAM GENERATOR 40deg ELDOW L EADING E DGE rLOW uETER REACTOR y/ (d"^$n j S uce COOLANT p
/
i,
)
PUMP APPROX.
5'
(
ELOOW TAPS "d%lEMI?
y RTD BYPASS r RETURN n
l ELBOW TAP 31"1D PIPE e
Page 16
South lea Proleet Units 1 and 2 RCS Mow Measurement Usine Elbow Ton Methodolouv 1.icensine Subruhtpl 3.5 liliST liSTIMATl! RCS FLOW ANALYSIS 3.5.1 Ilackcround Westinghouse developed the best estimate RCS flow calculational procedure in 1974 and has i
applied the procedure to estimate RCS Hows at all Westinghouse designed plants. The procedure uses component now resistances and pump performance with no margins applied, so the resulting flow calculations dcEne a true best estimate of the actual flow.
Uncertainties in the best estimate hydraulics analysis, based on both plant and smponent test data, define a flow uncertainty of t2% flow, indicating that actual How is expected to be within 2% of the calculated best estimate flow.
The best estimate hydraulics analysis was developed and confirmed by numerous component flow resistance tests and analyses. The most signincant input was the test data collected at Prairie Island Unit 2, where ultrasonic Leading !!dge Flow Meters (Ll!FMs) were installed. This program and other tests are described in the following sections.
3.5.2 Prairie Island livdraulles Test Procram lhe LIIFM was installed in 1973 at Prairie Island Unit 2, on both loops as shown on Figure 3.41.
Measurements were obtained during the hot functional and plant startup tests in 1974. In addition to the LIIFM flows, concurrent measurements of reactor vessel and steam generator flow resistances were also obtained, as well as reactor coolant pump dynamic head, input power and speed.
The program collected data during plant heatup from 200'F to nonnal operating temperatures with one and two pumps operating. Full power flow measurements were obtained early in 1975.
Subsequent flow and pump input power merurements were obtained in 1979,1980,1981 and 1991.
The Ll!FM accuracy for the Prairic Island plant uasurements was established by a calibration test at Alden Laboratories and by analyses of dimensional tolerances to be 10.67% of measured flowi The Alden test modelled the piping configuration both upstream and downstream from the metered pipe section. Tests perfonned at several circumferential locations of the ultrasonic transducers defined the optimum location for the transducers in the pipe section relative to the upstream and downstream elbows.
The component Ap accuracy fbr the Prairie Island measurements was established by calibrations to be within 11% of the measured Ap. The sum of the Aps measured across the reactor and steam generator were within 1% of the pump Ap, confinning measurement accuracy.
Page 17
South Texas Prolect Units I and 2 RCS Flow Measurement Usine Fibow Tan Methodolouv I Icensine Submittal lhe flows measured in 1974 75 were 5% higher than predicted, due to the following effects, evaluated in additional analyses.
Reactor Coolant Pumn Perfonnancq Reactor coolant pump performance was higher than predicted from hydraulle model tests, producing an additional 2% flow, partly due to pump impeller thennal expansion and partly due to conservatism in the hydraulics scaleup from the model tests. With How, head, input power and speed data, hydraulle and electrical etliciency were verined. Since the LEFM also measures reverse flows, the resistance of the pump impeller to reverse flow was confirmed to be as originally specified.
Reactor Vessel Flow Resistance The reactor vessel flow resistance was lower than predicted from reactor vessel model tests and fuel assembly flow resistance measurements, pn>ducing an additional flow of almost 3%. Tests with one pump operating provided additional data to confinn the division of flow resistances between vessel internals (total flow) and vessel nor21cs (loop flow).
Etcam Generator Flow Resistance j
1he steam generator flow resistance was the same as predicted from analysis, so changes in the analysis were not required. The large change in the predicted flow resistance resulting from the change in tubing Reynolds Number and friction factor during plant heatup was also confinned by the now resistance measmements.
Pinion Flow Resistance The reactor coolant piping flow resistance,6% of the total system resistance, was reduced by about 25% to be consistent with measured component flow resistances, accounting for reduced Ap due to close coupling of components and cibows in the piping. Part of an cibow Ap loss occurs as increased turbulence in the downstream piping, but the loss is reduced if a component or another elbow is located at or close to the elbow outlet.
Flow vs Power LEFM measurements at full power indicated that the Prairie Island Unit 2 RCS volumetric flow decreased by about 0.8% as the reactor was brought from rero to full power, This result confirmed the predicted effect of higher velocities in the core, hot leg, and steam generator tubes as these temperatures increase above cold leg temperature. The coolant volumetric How and velocity in these regions increases by 5 to 12%, causing an increase in the total RCS flow resistance applied to the reactor coolant pumps. The decrease in flow as reactor power increases from zero to 100% differs from plant to plant, depending on plant specific coolant
- temperatures, coolant AT (T T,,ia), and component flow resistances.
ho Page 18
South Tem Project Units 1 and 2. ItCS Ilow Me.DatIIment Usine I!! bow Teo Methodolocv 1.icensine Submittal 3.5.3 Additional Prairie Island Tests The flow measurements in later years contributed additional data on system hydraulles perfonnance w hich was used to revise and further validate the,hydraulles analyses, as described in the following paragraphs.
I knoeller Smoothing LEFht and pump input power measurements were obtained at Prairie Island in 1979 and 1980 to recon 0rm RCS flows and hydraulic performance. LEFht data indicated that RCS flows had decreased slightly, by 0.6 to 0.8%, it was also noted that pump input power had decreased by about 2%. After evaluating this data and considering other available information, Westinghouse concluded that the flow decrease was due to impeller " smoothing", where the impeller surface roughness decreases due to wear or crud buildup between high points on the impeller surfaces.1he " smoothing" effect occurs within one or two fuel cycles aller initial plant startup. 'this small flow decrease during the initial cycles has also been measured by elbow tap Dow meters at several other 3 loop and 4 loop plants.
Pumn Imoeller Renlaecment The LEFMs were used at Prairie Island in 1981 to confinn RCS flows aller replacement of a pump impeller. The replacement impeller was predicted to have a higher perfonnance than that of the original impeller, and an increase in loop flow was predicted. The LEFM data confirmed the prediction.
Elbow Tan Flow Comnarison LEFM measurements obtained in 1991 were compared with the 1980 data to confinn that the elbow taps measured the same flow changes over the same period. De comparison indicated that the elbow tap and LEFM flows were in good agreement, with an average difference in flow ofless than 0.3% over 11 years.
3.5,4 System Flow Resistance Analyses Flow resistances are calculated for each component, based on the component hydraulic design data and on hydraulics coeflicients resulting from analyses of test data, such as but not limited to, the Prairie Island hydraulics test program. The component flow resistances are combined to define total system resistance, and then combined with the predicted pump head flow perfonnance to define individual loop and total RCS How. The background and bases for the flow resistance calculations are described below, Reactor Vessel The reactor vessel now resistance is defined in three pants,
- n. The reactor core flow resistance is based on a full size fuel assembly hydraulic test, including the Aps at RCS total now through the inlet and outlet core plates as well as the core.
Page 19
South Texas Ptolect Units 1 and 2 tRCS Ilow Measurement Usine !!! bow 'lan Methodolony 1.lcensinn Submittal b, The vessel internals flow resistance accounts for the Aps with total flow through the downcomer, lower plenum, and upper plenum. The flow resistances are detemiined from hydraulic model test data for each type of reactor vessel, based on Ap measurements within the model,
- c. The vessel nonle now resistances include Aps based on loop flow through the inlet and outlet nonles.
In addition, the overall analysis accounts for small aows that bypass the core through the upper head, hot leg noule gaps, ballle barrel gaps, and control rod guide thimbles.
Steam Generator The steam generator % ridece is defined in five pans: inlet nonle; tube inlet; tubes; tube outlet; and outlet noe. M overall flow resistance was con 0nned by the Prairie Island hydraulics test program (Section 3.$.2). Ihe analysis accounts for the plugged or sleeved tubes in each steam generator, so loep *pecluc flows can be calculated when different numbers of tubes are plugged or sleeved in each loop.
Reactor Coolant Pining The reactor coolant piping How resistance combines the Dow resistances for the hot leg, crossover leg, and cold leg piping. The flow resistance for each section is based on an analysis of the effect of upstream and downstream components on elbow hydraulic loss coefficients, using the results of industry hydraulics tests. The total flow resistance was consistent with the l
measurements from the Prairie Island hydraulics test program (Section 3.5.2).
l 3.5.5 DnGstimate RCS Flow Calculations
'Ihe best estimate RCS flow analysis defines Best Estimate Flow (BEF) and Future Cycle Estimated Flow (FEF) for the elbow tap RCS flow measurement procedure. 'the calculation combines component flow resistances and pump perfonnance predictions based on hydraulle model tests, and defines RCS loop flows at the desired power or temperature with any combination of pumps l
operating, with any fuel assembly design, and with different tube plugging in each steam generator.
l The calculated best estimate flows are in good agreement with calorimetric How measurements from many plants before low leakage loading patterns were implemented, as discussed in Section 3.3.
For the many plants where the comparisons have been made, the calculated best estimate changes in How from cycle to cycle have been in good agreement with changes measured by elbow taps, 3.6 EVALUATION OF SOUTil TEXAS PROJECT RCS FLOW PERFORMANCE RCS elbow tap Dow and calorimetric flow measurements from South Texas Project Units 1 & 2 were evaluated and compared with best estimate flow to determine RCS Dow perfonnance. Elbow tap Dow measurements indicate actual flow changes and are expected to compare well with changes predicted by the best estimate analysis. Calorimetric data from each unit established the baseline now and ident10ed How changes caused by hydraulics changes as well as hot leg temperature streaming biases in later fuel cycles.
The South Texas Project RCS Dow measurement evaluation is described in the following paragraphs.
Page 20 i
m
-.=
South Texas Project Units I nnd 2 RCS riow Measurement Usine Elbow Tap Methodolouv 1,1censine Submittal 3.6.1 Ilest l' stimate I low Predletions t
i South Texas Palect Unit I liest estimate flow analyses defined flows for each of the seven cycles for Unit 1. The Cycle !
initial startup flow was defined to be 407,472 gpm. Ilydraulles changes alrecting subsequent cycle flows defined the following changes in flow, listed on Tabic 3.61.
Impeller Smoothing: As stated in Section 3.5, impeller smoothing is expected to cause a flow a.
decrease of about 0.6% flow aller the Orst cycle. Since the Unit 1 pre-startup tests required longer than nonnal reactor coolant pump operating time, some innpeller smoothing may have occurred before Cycle I startup. For this analysis, the flow decrease due to impeller l
smoothing prior to Cycle 2 was defined to be 0 to.0.6% flow, to allow for smoothing that may have occurred before Cycle 1.
b.
Steam Generator Tube Plugging: The tube plugging at Unit I had a negligible impact on RCS flow until Cycle 6 when the average plugging reached 0.8%, causing an estimated decrease of O.2% flow. Prior to Cycle 7, an additional 0.5% plugging occurred, causing an additional estimated decrease of 0.1% flow.
c.
Fuel Design Changes: Although the fuel design changed over the seven cycles, the best j
estimate analyses determined that the overall impact of the changes on RCS flow was negligible.
4 Considering all of the above, the overall impact of the hydraulle changes was expected to be 0.3 to 0.9% flow over seven cycles of operation, as indicated in Table 3.61. Cycle 1 is defined as the baseline for best estimate flow, and the trend defined on Table 3.61 is plotted on Figure 3.61, with the Cycle I flow speelned as 100%.
llased on the elbow tap flow measurement procedure described in Section 3.4.2, the future cycle j
estimated flow (FEF) is 99.1%, so the estimated flow ratio (R') for Cycle 7 and for future cycles
~
if no hydraulics changes are made is, therefore, 0.991.
South Texas Prolect Unit 2 Ilest estimate flow analyses defined Dows for each of the six fuel cycles for Unit 2. The Cycle I initial startup flow was denned to be 405,756 gpm. liydraulics changes afTecting subsequent cycle flows defined the following changes in now, listed on Table 3.6-1, Impeller Smoothing: As stated in Section 3.5, impeller smoothing is expected to cause a Dow n.
decrease of about 0.6% Cow after the first cycle. Unit 2 pre stanup testing was normal, so the flow decrease for Cycle 2 was defined to be -0.6% Dow, b.
Steam Generator Tube Plugging: The tube plugging at Unit 2 had a negligible impact on RCS Dow until Cycle 6 when the average plugging reached 3.5%, causing a decrease of 0.8% flow, c.
Fuel Design Changes: Although the fuel design changed over the six cycles, the best estimate analyses determined that the overall impact of the changes on RCS flow was negligible.
Page 21
South Texas Protect Units I and 2 RCS Flow Measurement Usine !!Ibow Tan Methodolocy Licensine Submittal Considering all of the above, the overall impact of the hydraulic changes was expected to be 1.4%
How over six cycles of operation, as indicated in Table 3.6 l. Cycle i is defined as the baseline for best estimate flow, and the trend defined on Table 3.61 is plotted on Figure 3.6 2, with the Cycle 1 flow specl0ed as 100%.
Ilased on the elbow tap flow measurement procedure described in Section 3.4.2, the future cycle estimated flow (FEF) is 98.6%, so the estimated flow ratio (R') for Cycle 6 and for future cycles if no hydraulics changes are made is, therefore 0,986.
3.6.2 Evaluation of Elbow Tan Flows South Texas Project Unit 1 Elbow tap Ap measurements were obtained from all 12 transmitters at the beginning of each Unit I fuel cycle. When the op measurements were obtained, Unit I was operating at about 70% power for Cycles I through $ and about 100% power for Cycles 6 and 7. As discussed in Section 3.5, RCS flow decreases as power increases from zero to 100%. 11ased on the Unit I specific parameters, the decrease is 1.2% flow from rero to 100% power, and 0.4% flow from 70 to 100%
power. Considering this flow decrease, the elbow tap Aps were adjusted so all measurements were at a common flow (at full power). The adjusted Aps expressed in inches of water at 100% flow are listed on Table 3.6 2. Another adjustment was made in nonnalizing Dows to the baseline Dow to account for the decrease in cold leg temperature in Cycles 6 and 7, in accordance with the elbow tap flow measurement procedure defined in Section 3.4.2.
The Cycle 1 elbow tap Aps define the baseline for subsequent cibow tap measurements. Table 3.6 2 lists the elbow tap flow comparison of subsequent cycles normalized to the Cycle I flow, expressed as 100% flow. Figure 3.61 shows nonnalized elbow tap flows for the seven cycles, for comparison to best estimate and calorimetric flows.
Sc.uth Texas Project Unit 2 Elbow tap Ap measurements were obtained from all 12 transmitters, as for Unit 1, at the beginning of each Unit 2 fuel cycle. When the Aps were measured, Unit 2 was operating at about 70% power for Cycles i through 4, and about 100% power for Cycles $ and 6. As for Unit 1. Unit 2 elbow tap Aps were adjusted so the measurements were all at a common (full power) flow. The adjusted Aps expressed in inches of water at 100% Cow are listed on Table 3.6 2.
The adjustment was also made in nonnalizing Dows to the baseline now to account for the cold leg temperature decrease in Cycles $ and 6, in accordance with the procedure defined in Section 3.4.2.
The comparison of elbow tap measurements normalized to the Cycle 1 elbow tap baseline Dow is listed on Table 3.6 2 and shown on Figure 3.6 2 for comparison to best estimate and calorimetric Dows.
Page 22
South Texas Project Units 1 and 2. ItCS l' low Mensurement Usine Elbow Tan Methodoloey 1.leensine Submittal 3.6.3 livaluation of Calorimetric Flows Calorimetric flow measurements were obtained from the South Texas Project units at the beginning of each cycle. 'Ihe initial data was obtained at about 70% power for Cycles 15 in Unit I and Cycles 14 in Unit 2, but calorimetric data was also obtained shortly aller full power was attained in these cycles. Since calorimetric flow measurements at full power are more accurate than measurements at reduced power, this evaluation is based on full power measurements. The calorimetric flows are listed in Table 3.6 3 and compared with the best estimate Dows. The definitions of columns on the table are as follows:
MEASURED CAL is the total calorimetric How for the indicated cycle. Listed at the bottom of the column is the average of the cycle Hows, conservatively defined to be the baseline calorimetric Dow for the unit.
% of!!ASE CAL shows the cycle flow difTerences from the baseline calorimetric How defined
(
above.
11EST EST shows the change in the cycle flows from the baseline best estimate Dow, nonnalized to 100% flow as on Table 3.6-1.
ADJUSTED CAL is the measured calorimetric flow adjusted for the known hydraulics changes defined on Table 3.61.
South Texa; Prpiect Unit 1 Total calorimetri: and best estimate Cows for Unit I are listed on Table 3.6 3. The procedure described in Section 3.4.2 would normally be used to define baseline flow. The Cycle 1 How of 404,716 gpm, which is in good agreement with the best estimate flow of 407,472 gpm, would normally be used to define baseline calorimetric How. For additional conservatism, a baseline calorimetric flow based on the average flow for all cycles (404,092 gpm) was defined. The resulting baseline flow is only slightly less than the Cycle i How.
A baseline calorimetric How more representative of actual Cow in the early operating cycles would be based on the average Dow for Cycles I,2 and 4, which is 405,316 gpm, and closer to the best estimate flow. Including all cycles in the baseline flow calculation introduces a conservative flow bias of 0.3% below the b'tseline Dow based on the early fuel cycles.
South Texas Project Unit 2 Total calorimetric and best estimate flows fbr Unit 2 are listed on Table 3.6-3. The procedure described in Section 3.4.2 would nonnally be used to define baseline now. The Cycle i How of 406,944 gpm, which is in good agreement with the best estimate now of 405,756 gpm, would normally be used to define baseline calorimetric How. For additional consenatism, a baseline calorimetric flow based on the average flow for all cycles (402,456 gpm) was defined. The resulting baseline Dow is over 1% less than the Cycle i How. The Dows for Cycles 5 and 6 are well below the best estimate flow and are considered to be afTected by low leakage loading patterns and hot leg streaming.
Page 23 l
South Texas Project Units 1 and 2 RCS Flow Measurement Usine Elbow Tan Methodolony 1 icensine Submittal 3.6.4 Flow Comnarisons South Texas Project Unit 1 Figure 3.61 compares total best estimate, elbow tap and calorimetric flows for Unit 1. liest estimate and cibow tap dows, normalized to the flows for Cycle 1, are in good agreement for the seven cycles, considering that there is some uncenalnty on the time when now decreased due to impeller smoothing, and that less precision was used when averaging eltow tap data during early cycles. Elbow tap and best estimate dows are in very good agreement in the recent cycles.
Figure 3.61 also shows calorimetric flows normalized to the average calorimetric flow for the seven fuel cycles. Although Figure 3.61 shows only small difTerences between calorimetric and best estimate or elbow tap Dows for Cycles 5 through 7. Table 3.6 3 indicates that the difference in adjusted calorimetric How from Cycle i to Cycle 7 would be about 1% flow if impeller smoothing e.~tually occurred before Cycle 1.
South Texas Prolect Unit 2 Figure 3.6 2 compares total best estimate, elbow tap and calorimetric dows for Unit 2. Best estimate and elbow tap flows, normalized to the flows for Cycle 1, are in good agreement in the later cycles. The larger differences in Cycles 2 and 3 have been attributed to the reduced precision used when averaging cibow tap data during these cycles. Elbow tap and best estimate now trends are in very good agreement in the recent cycles.
Figure 3.6 2 also shows calorimetric flows normalized to the average calorimetric How for the six cycles. As noted above, the Cycle 1 flow would have provided a sufTiciently accurate baseline now. If Cycle I had been used to define baseline calorimetric flow, the now difference in Cycles 5 and 6 would be larger and would be a more representative indication of the low leakage loading pattern impact. Ilased on comparisons of adjusted calorimetric flows in Table 3.6-3, the Cycle 6 flow is almost 2% below the Cycle i flow.
3.6.5 Power / Flow Correlation for South Texas Proi g1 t
Westinghouse's review of the radie.1 power distribution and measured calorimetric Dows from South Texas Project Units 1 & 2 indicated that the data, especially from the most recent fuel cycles, was consistent with the power /Dow trend shown in Figure 3.3 4. Figure 3.6 3 plots the apparent now decreases versus the power ditTerence between second row and outer row assemblies for South Texas Project Units 1 & 2. The decreases in RCS flow are based on calorimetric flows adjusted for hydraulics efTects, as listed on Table 3.6-3.
The data from both units defines a similar correlation to that shown on Figure 3.3 4, with the decrease in now approaching 1% at Unit 2. The Unit I data has a similar trend, but with a smaller flow decrease, probably due to the pump impeller smoothing uncertainty discussed earlier. The Dow decreases for recent cycles predicted by the power /Dow correlation are consistent with the conclusion? discussed above. The power / flow correlation thus provides a qualitative confirmation of the hot leg streaming theory and differences between cibow tap and calorimetric flow measurements.
Page 24
South Texas Prolect Units I and 2 RCS Flow Measurement Usine rlbow Tan Methodolony 1icensinn Submittal I
TABLE 3.6-1 i
SOUTil TEXAS PROJECT BEST ESTIMATE FLOW
SUMMARY
l l
l UNIT I i
CYCLE I BEST ESTIMATE FLOW = 407,472 GPM l
J l
CYCLE IlYDRAULICS CilANGE FLOW CilANGE (%)
FLOW (%)
1 N/A 0.0 100,0 2
Impeller Smoothing
-0.6 (*)
_99.4 I
3 N/A 0.0 99.4 4
N/A 0.0 99.4 5
N/A 0.0 99.4 6
S/G Tube Plugging 0.2 99.2 l
7 S/G Tube Plugging 0.1 99.1 l
(')
Impeller smoothing impact = 0 to 0.6%. Only the maximum impact is considered
- here, t
i i
l UNIT 2 CYCLE I BEST ESTIMATE FLOW = 405,756 GPM CYCLE IlYDRAULICS CilANGE FLOW CilANGE (%)
FLOW (%)
l 1
N/A 0.0 100.0 l
i 2
Impeller Smoothing
-0.6 99.4 t
3 N/A 0.0 99.4 4
N/A 0.0 99.4 5
N/A 0.0 99.4 6
S/G Tube Plugging 0.8 98.6 Page 25 l
, _., ~. _.
- ~.., _ - -. - -
South Texas l'rolect Units 1 and 2 RCS Flow Measurement Usine Fibow Tan Methodoloev I.lcensine Submittal TAllLE 3.6 2 SOUTil TEXAS PROJECT ELilOW TAP DIFFERENTIAL PRESSURE
SUMMARY
UNIT l DIFFERENTIAL PRESSURES IN INCllES OF WATER CYCLE LOOP 1 LOOP 2 LOOP 3 LOOP 4 AVERAGE Tm ELBOW TAP *,',
l'F) of IIASELINE 1
496.09 483.96 500.86 460.37 485.32 562.7 100.00 2
494.51 482.62 494.75 457.84 482.43 561.1 99.70 3
497.71 487.92 497.21 460.47 485.83 563.6 100.05 4
492.21 484.82 493.02 458.32 482.09 563.1 99.67 5
497.53 479.29 502.60 458.48 484.48 562.6 99.91 6
489.64 482.46 490.00 455.83 479.48 557.5 99.02 f
7 490.18 487.00 490.49 455.55 480.81 556.5 99.09 ELBOW TAP BASELINE FLOW COEFFICIENT (B) = 10.5455 inches
- fP/#
!! is based on Cycle 1 Average Ap (485.32 psi), Cycle 1 Tcold (562.7'F), and 2250 psia UNIT 2 DIFFERENTIAL PRESSURES IN INCllES OF WATER If g{B El CYCLE LOOP 1 LOOP 2 LOOP 3 LOOP 4 AVERAGE g
1 487.59 454.65 511.04 468.11 480.35 562.7 100.00 2
492.67 456.45 SI1.17 472.33 483.16 562.8 100.30 3
484.38 452.19 511.41 468.39 479.09 563.2 99.87 4
480.17 445.40 505.19 465.09 473.96 563.6 09.33 5
487.14 454.42 513.98 468.82 481.09 556.7 99.64 6
479.67 447.19 505.14 464.99 474.25 556.8 98.94 ELilOW TAP llASELINE FLOW COEFFICIENT (B) = 10.4375 inches
- fe/#
B is based on Cycle 1 Average Ap (480.35 psi), Cycle i Tcold (562.7'F), and 2250 psia Page 26
SaullLTexas 14olect Units 1 and 2 RCS l' low hicasurement Usine I'lbow Tan hiethodolouv l icensinn Submittaj TABLE 3.5 3 SOUTil TEXAS PROJECT CALORlhtETRIC FLOW SUhihtARY UNIT 1 "A
AI' BEST EST ADJUSED CAL CYCLE
% OF llASE CAL pptn gpm gpm 1
404,716 100.2 407,472 404.716 2
406,124 100.5 405,028 400,575 3
411,628 101.9 405,028 414,i12 4
405,104 100.3 405,028 407,549 5
400,544 99.1 405,028 402,962 6
400,880 99.2 404,212 404.113 7
399,656 98.9 403,804 403,286 404,092 100.0
- g UNIT 2
^
Ab BEST EST ADJUSED cal, CYCLE
% OF BASE CAL gpm gpm gpm 4
~
l 406,944 101.1 405,756 406,944 2
406,188 100.9 403,320 408,640 3
402,988 100.1 403,320 405,420 4
404,852 100.6 403,320 407,296 5
399,644 99.3 403,320 402,056 3
6 394,116 97.9 400,076 399,712 A0 402,456 100.0 g 3 Page 27
South Texas Project Units 1 and 2 RCS Flow Measurement Usinn Fibow Tan Methodolouv 1.leensine Submittal l
110URE 3.61 SOUTil TEXAS PROJECT UNIT 1 FLOW COMPARISONS l
SOUTH TEXAS UNIT 1 RCS FLOW HISTORY 1(' t.0 '
======t==a=====---
Colorimetrio l
6 103.0 Best Estimate 102.0 Elbow Tap 101.0
~ = =
= = = = = = = '
2 1 N ---
100.0 #
0 XT Y-N 99.0 -
Mf 9s.0 97.0 ---
96.0 1
2 3
4 6
6
?
Cycle Number SOUTH TEXAS UNIT 1 RCS FLOW HISTORY 400000 T 407000 - - ~-
406000 405000
~
E g 404000 8
-,..** *.,s.
w m
403000 -- - -
g~.*,
- * * * * *. * *.'4, ft 402000 -
J
~~
E Elbow Top using Procedure 401000
- * * ** *
- Raw Elbow Tap twlo BE Flow 400000 Confirmation) 399000 -
Best Estimate 398000
+-
4 1
2 3
4 5
6 7
Cycle Number Page 28
South Texas Prolgt Units 1 and 2. RCS l' low Measurement Usine I'lbow Tan Methodolouv 1.lcensine Submittal l
FIGURE 3.6 2 SOUTil TEXAS PROJECT UNIT 2 FLOW COMPARISONS SOUTH TEXA8 UNIT 2 RCS FLOW HISTORY l
104,0 _
__,___1 g-103.0 Colorimetric Best Estimate l
102.0 Elbow Top 101.0 S
100.0 0 99.0 -
5 98.0 97.0 96.0 1
2 3
4 6
6 Cycle Number SOUTH TEXAS UNIT 2 RCS FLOW HISTORY 408000 m----~~----
407000 Ebow Top udng hocekte 406000
- * * ** *
- Raw Elbow Top (w/o BE Flow Confirmation) 406000 Best Estimate
_h404000-
,,,,.a***..
+v,,
~
403000
~
402000 -
401000 j
400000
=
1 399000 398000 l
1 2
3 4
5 6
Cycle Number Page 29 '
- -.. _. _ - - - _ _ _. = _
l South Texas l'rolect Units 1 and 2 RCS l' low Mensurement Usine I.lbow Tan hiethodolony I.lcensine Submittel FIGURE 3.6 3 FLOW lilAS VS POWER DilTERENCE 4
1 SOUTH TEXAS PROJECT UNIT 1 i
+1%
4 0
o F
0%
L o
O o
o 4
g B
-1%
I 1
A
~
~
S
-2 %
SOUTH TEXAS PROJECT UNIT 2
+1%
o F
0%
c'
'O L
O o
W R
-1%
I o
A
~
~
S O
-2 %
0 2nd ROW - OUTER ROW POWER DIFFERENCE 100 PERCENT POWER Page 30
South Texas Prolect Units I and A RCS 110w Measurement Usinn I!! bow Ten Methodoloev 1leensine Submittal 3,7 ELilOW TAP FLOW MEASUREMENT LICENSING CONSIDERATIONS 3.7.1 Docketcund Plant Tecimical Specifications require that an RCS total flow measurement be perfonned every 18 months to verify that su0icient RCS Dow is available to satisfy t! safety analysis assumptions.
This suncillance is nonnally perfonned at the beginning of each operating cycle. Technical Specifications also require that a qualitative RCS flow verification (l.c., channel check) be l
performed every 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> during Mode 1, lhese surveillances ensure RCS Dow is maintained I
within the assumed safety analysis value, i.e., Minimum Measured Flow (MMF).
l The 18-month RCS flow surveillance is typically satisfied by a secondary power calorimetric based RCS flow measurement and the 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> RCS flow surveillance is satisfied by control board RCS flow indicator or plant process computer readings using inputs from the RCS cibow tap Ap channels. These surveillances and the RCS lew Flow reactor trip are interrelated since the calorimetric RCS flow measurement is used to correlate elbow tap Ap measurements to flow, and the flow at the Ap setpoint for the RCS Low Flow reactor trip is verified to be at or above the now assumed in the safety analysis. The process computer output is nonnallred to the calorimetric flow.
The uncertainty associated with the 18 month precision calorimetric is, therefore, included in the uncertainty calculations for the surveillance criterion and the RCS low Flow trip.
The purpose of this evaluation is to support the use of elbow tap Ap measurements as an alternate method for perfonning the 18 month RCS Dow surveillance. Many plants in recent cycles have experienced apparent decreases in flow rates which have been attributed to variations in hot leg streaming, as discussed in previous sections of this document, llot leg steaming efTects directly impact the hot leg temperatures used in the precision calorimetric, resulting in the calculation of apparently low RCS flow rates, in using the elbow tap op method, the RCS elbow tap measurements are correlated (as described in Section 3.4.2) to precision calorimetric measurements perfonned during earlier cycles when the hot leg streaming effects were decreased.
3.7.2 Sunoortine Calculations in order to implement the elbow tap Ap method of measuring RCS flow, calculations have been perfbrmed to determine the uncertainty associated with the precision RCS Dow calorimetric (s) for
- the baseline cycle (s). These calculations account for the plant instrumentation, test equipment, and procedures that were in place at the time the calorimetric was perfonned.
in addit;on, uncertainty calculations have been perfonned for the indicated RCS Dow (computer and control board indication) and the RCS low flow reactor trip. These calculations reficct the correlation of the elbow taps to the baseline precision RCS How calorimetric (s) noted above.
Additional instrument uncertainties are required to reDect this correlation. Appendix A contains uncertainty calculations that were performed lising South Texas Project speci0c inputs.
Page 31
South Texas Protect t! nits I and 2 RCS Flow Mensurement tisine tilbow Tan Methodolocv I icensine Submittal These uncertainty calculations have confirmed the acceptability of previously perfonned South Texas Project specine safety analyses and associated protection and/or control system setpoints when the periodic surveillance is perfonned via use of Qualified Display Processing System (QDPS) or plant process computer indication on an 18 month basis. In particular, no increase in the RCS total How uncertainty due to the elbow tap Ap method has been determined when using the QDPS or plant process computer indication. ~lhus revision to the Westinghouse Revised Thermal Design Procedure (RTDP) instrumentation uncertainties (currently 2.8% Cow), which are used in deriving i
the Technical Specifications reactor core safety limits and the corresponding DNil limits is not required. if control board indication is utilized, there is a small increase in the instrumentation uncertainty for RCS Dow which does not afTect the RTDP results or the Technical Specifications reactor core safety limits nor the corresponding DNil limits. The low How reactor trip setpoint l
uncertainty has increased somewhat but does not require a change to either the Technical Specifications trip setpoint (91.8% Oow) or to the current Safety Analysis Limit (87% flow) due to the availability of margin in the uncertainty calculation. He revised Technical Specifications including the change to the allowable value is noted in Appendix C.
Technical Specification Table 2.21 columns headed TA, Z, and S are marked as N/A in revised l
Technical Specifications for the Reactor Coolant Flow Low trip. A two column approach (Nominal Trip Setpoint and Allowable Value) is consistent with the NRC's position for the improved Technical Specifications (NUREG 1431) where there is no longer the TA, S and Z columns and is the Westinghouse recommended approach. With the two column approach, the Allowable Value is based on an appropriate determination of channel operability consistent with the uncertainty calculations and the proce:s rack drill allowance. Z and S terms are not applicable to the process racks and therefore are marked N/A. Since TA in Technical Specification Table 2.21 is only used in conjunction with columns S and Z, and is not part of an operability detennination, the TA column is also marked N/A.
3.7.3 Potential Document Imonets The South Texas Project Technical Specifications are affected in four areas:
1)
Specification 2.2.1. Table 2.21, item 12, Reactor Coolant Flow Low (Trip Setpoint, Allowable Value; magnitude changed to reDect uncertainty calculation results);
2)
Specification 3.2.5 (Surveillance Requirement 4.2.5.3 is modified); and 3)
Associated 13ases for this specification (to include a description of the elbow tap op method of flow measurement and to note the indication sources).
Appendix Il contains the 50.92 input for licensing documentation purposes.
Appendix C contains a markup of the South Texas Project Technical Specifications, in the case of the South Texas Project specinc instrument uncertainty analyses shown in Appendix A, the RCS flow uncertainty associated with the elbow tap Ap method (when indication is provided by QDPS or the plant process computer) was less than or equal to the current Technical Specincation value. RCS low Dow reactor trip setpoint uncertainty calculations also verify that the current trip setpoint and Safety Analysis Limit remain valid.
Page 32
South Texas Prolect Units 1 and 2. RCS Flow Measurement Usine Elbow Tao Methodolony Licensinn Submittal k
l l
l APPENDIX A INDICATED RCS FLOW and REACTOR COOLANT FLOW.
LOW REACTOR TRIP INSTRUMENT UNCERTAINTIES Page 33
South Texas l*roject Units 1 and 2 RCS l'Iow Measurement Usine I?ltev Tao Methodolony I leensine Submittal TAllLE A 1 IIASELINE Fl.OW cal.ORlhiETRIC INSTRUhiENTATION UNCERTAINTIES
(! SPAN)
T, P,,
AP,,
P T
T P
m o
mp e3 Sensor SCA -
M&TE-SRA -
l SPE -
STE -
SD -
BIAS-R/E A/D RCA -
M&TE-RTE -
RD D/A RCA -
M&TE-RTE -
RD RDOUT-Noise-CSA -
- INST 1
1 1
1 3
1 USED
'F psia
! AP psia
- f
'F psia INST SPAN 300 1400 120 1400 100' 100*
800 INST UNC, (RANDOM) -
INST UNC.
(BIAS)
~
NDMINAL -
440 1298 1106 622.4 562.3 2250 Pressurizer pressure is not measured, but is assumed based on
.the controller, A conservative uncertainty value is used,
+ T span m
Page 34
South Texas Prolect Units 1 an12 RCS Ilow Measurement Usinc !!! bow Tan Methodolocv 1.lcensinc Submittal TAllL, $.2 FLOW CALORIMIITi(IC S!!NSITIVIT111S FEE 0 WATER FLOW FA TEMPERATURE t/*F MATERIAL DENSITY TEMPERATURE t/'F
=
PRESSURE t/ psi AP
%/*AP FEEDWATER ENTHALP) l TEMPERATURE t/'F l
PRESSURE t/ psi h5 1188 Btu /lbm hF 419.6 Btu /lbm
=
Ah(SG) 768.4 Btu /lbm STEAM ENTHALPY "A
PRESSURE
%/ps1 MOISTURE
%/0.25% Moisture HOT LEG ENTHALPY TEMPERATURE
%/'F PRESSURE
%/ psi hH 646.1 Btu /lbm hC 562.3 Btu /lbm Ah(VESS) 83.8 Btu /lbn Cp(TH) 1.579 Btu /lbm *F COLD LEG ENTHALPY
<e.c TEMPERATURE t/'F PRESSURE
%/ psi Cp(TC) 1.273 Btu /lbm 'F COLD LEG SPECIFIC VOLUME TEMPERATURE t/*F PRESSURE t/ psi Page 35
)
South Te3p Prolect Units 1 and 2 I(CS l' low Mensurement Usinn Elbow Tan Methodoloey I icemine Submittal TAllt.l! A 3 CALORihil?TRIC RCS 1 LOW Mi!ASURiihilINT UNCIIRTAINTillS COMPONENT INSTRUMENT ERROR FLOW UNCERTAINTY FEEDWATER FLOW "3
l VENTURI
%K 1 FLOW THERMAL EXPANSION COEFFICIENT TEMPERATURE
'F l
MATERIAL l
OENSITY (p) l TEMPERATURE
'F PRESSURE psi AP
TEMPERATURE
'F PRESSURE psi STEAM ENTHALPY (h)
PRESSURE psi MOISTURE
- Moisture NET PUMP HEAT ADDITION HOT LEG ENTHALPY (h)
TEMPERATURE
'F STREAMING, RANDOM
'F STREAMING, SYSTEMATIC
'F PRESSURE psi COLD LEG ENTHALPY (h)
TEMPERATURE
'F PRESSURE psi COLD LEG SPECIFIC VOLUME (u)
TEMPERATURE
'F PRESSURE ps1
~ ~
BlAS VALUES FEEDWATER PRESSURE o
h STEAM PRESSURE h
PRESSURIZER PRESSURE h - HOT LEG h
COLD LEG u
COLO LEG FLOW BIAS TOTAL VALUE
~
~
- ** +,++ INDICATE SETS OF DEPENDENT PARAMETERS
~~
SINGLE LOOP UNCERTAINTY (N0 BIAS)
% FLOW N LOOP UNCERTAINTY (N0 BIAS)
% FLOW N LOOP UNCERTAINTY (WITH BIAS)
% FLOW Page 36 i
Smith Texas Project Units 1 and 2 RCS Flow Measurement Usine tilbow Tan Methodotony iIcensine Submittal TAllLil A 4 COLD LiiG IILilOW TAP FLOW UNCliRTAINTY (QDPS/PROCliSS COMPUT!!R)
INSTRUMENT UNCERTAINTIES
% AP SPAN 1 FLOW
_4u PMA -
PEA =
SCA -
M&TE-SRA -
SPE -
STE -
SD -
BIAS-7300 Racks RCA -
M&TE-RTE -
RD -
A/D RCA -
M&TE-RTE -
RD -
FLOW CALORIM BIAS FLOW CALORIMETRIC INSTRUMENT SPAN
- 120.0 % Flow NUMBER TAPS PER LOOP
-3 N LOOP RCS FLOW UNCERTAINTY - 2,6 % FLOW Page 37 m
I South Texas Protect Units I and 2. RCS riow Meaturement thine Fibow Tan Methodolouv I.lcensine Submittal i
TAllLl! A.$ LOW FLOW Rl! ACTOR TRIP i
% AP SPAN
% FLOW SP.A.N 1
PMA1 -
PMA2 -
PEA -
SCA -
M&TE -
SRA -
RTE RD
=
BIAS -
INSTRUMENTRAIGE
- 0 TO 120.0 % FLOW
~
FLOW SPAN
- 120.0 % FLOW SAFETY ANALYSIS LIMIT = 87.0 % FLOW NOMINAL TRIP SETPOINT - 91.E % FLOW TA - 4.0 % FLOW SPAN l
CSA -
% FLOW SPAN MAR -
% FLOW SPAN Page 38 J
South Texas Prolect Units 1 and 2 - RCS Flow Mensitrement Usine Fibow Tan Methodoloey 1.icensine Submittal I
Al'PENDIX 13 NO SIGNIFICANT llAZ ARDS CONSIDERATION Page 39
kuth Texas prolect Units 1 and 2 RCS Flow Measurement ttsine IIlbow Tan Methodolocv I icensine submittal Pursuant to 10CFR50.92 cach application for amendment to an operating license must be reviewed to detennine if the proposed change involves a Significant liarards Consideration. 1he amendment, as defined below, describing the Technical Specification change associated with the change has been reviewed and determined to not involve Significant liarards Considerations. The basis for this detennination follows.
Proposed Change: The current Technical Specification Table 2.21 (page 2 4) "Reactcr Trip System Instrumentation Trip Setpoints," provides the Trip Setpoint and Allowable Value for the RCS Flow lew trip. The Allowable Value will be changed to reflect the increased uncertainty associated with the correlation of the elbow taps to a previous baseline calorimetric. In addition, Technical Specification 3.2.5 (page 3/4.211), " Power Distribution Limits, DNil Parameters", will be changed to allow the RCS total flow to be measured by the elbow tap Ap method. These changes will include the modification of surveillance requirement 4.2.5.3, which currently requires performance of a precision heat balance every 18 months, to allow use of the elbow tap Ap method for RCS flow measurement. Appropriate Technical Specification liases sections will also be revised to reflect use of the elbow tap Ap method for flow measurement and to provide clarification. The revised Technical Specifications are in Appendix C.
llackground: The 18 month total RCS flow surveillance is typically satisfied by a secondary power calorimetric based RCS flow measurement, in recent cycles South Texas Project has experienced apparent decreases in flow rates which have been attributed to variations in hot leg streaming elTects.
These elTects directly impact the hot leg temperatures used in the precision calorimetric, resulting in the calculation of low RCS flow rates. The apparent flow reduction has become more pronounced in fuel cycles which have implemented aggressive low leakage loading patterns. Evidence that the flow reduction was apparent, but not actual, was provided by elbow tap measurements. The results of this evaluation, including a detailed description of the hot leg streaming phenomenon, are documented in Westinghouse report SAE/FSE TGX/filX 0152, "RCS Flow Verification Using Elbow Taps."
South Texas Project intends to begin using an alternate method of measuring RCS flow using the elbow tap Ap measurements. For this alternate method, the RCS elbow tap measurements are orrelated to precision calorimetric measurements performed during earlier cycles which decreased the effects of hot leg streaming.
The purpose of this evaluation is to assess the impact of using the elbow tap Ap measurements as an alternate method for performing the 18 month RCS flow surveillance on the licensing basis and demonstrate that it will not adversely affect the subsequent safe operation of the plant. This evaluation suppods the conclusion that implementation of the elbow tap Ap measurement as an alternate method of determining RCS total flow rate does not represent a significant harards consideration as defined in 10CFR50.92.
Page 40
South Texas Protect Units 1 and 2 ItCS Ilow Measurement Usine 1:Ibow Tan Methodoloev l icensine Submittal 1: valuation: Use of the elbow tap Ap method to detennine itCS total flow requires that the Ap measurements for the present cycle be correlated to the precision calorimetric flow measurement which was performed during the baseline cycle (s). A calculation has been perfonned to determine the uncertainty in the ItCS total flow using this method. This calculation includes the uncertainty associated with the itCS total flow baseline calorimetric measurement, as well as uncertainties associated with Ap transmitters and indication via QDPS or the plant process computer. The uncertainty calculation perfonned for this method of flow measurement is consistent with the methodology recommended by the Nuclear llegulatory Commission (NUllE0/ Cit 3659, PNL 4973,2/85). The only significant difference is the assumption of correlation to a previously perfonned itCS flow calorimetric, llow ever, this has been accounted for by the addition of instrument uncertainties previously considered to be zeroed out by the assumption of nonnallration to a calorimetric perfonned each cycle, liased on these calculations, the uncertainty on the itCS flow measurement using the elbow tap method is 2.6% flow which results in a l
minimum ItCS total ilow of 391,500 gpm and must be measured via indication with QDPS or the plant l
process computer at approximately 100% power.
The specific calculations perfonned were for Precision itCS i: low Calorimetrics for the specified baseline cycles, Indicated itCS Flow (cither QDPS or the plant process computer), and the lleactor Coolant I low -
Low reactor trip. The calculations for Indicated itCS Flow and Itcactor Coolant Flow low reactor trip reflect correlation of the elbow taps to baseline precision itCS Flow Calorimetrics. As discussed above, additional instrument uncertainties were included for this correlation.
The uncertainty associated with the itCS Flow lxw trip increased slightly. It was detennined that due to the availability of margin in the uncertainty calculation, no change was necessary to either the Trip Setpoint (91.8% llow) or to the current Safety Analysis Limit (87% flow) to accommodate this increase.
1he Allowable Value is to be modified to allow for the increased instrument uncertainties associated with the Ap to flow correlation.
Since the flow uncertainty did not increase over the currently analyzed value, no additional evaluations of the reactor core safety limits must be performed, in addition, it was determined that the current hiinimum hicasured Flow (hih1F) assumed in the safety analyses (389,200 gpm) bounds the required hih1F calculated for the elbow tap method (391,500 gpm).
Page 41
South Texas prolect ifnits 1 and 2 1(CS Flow Measurement tIsine Elbow Tan Methodoloev IIcensine Submittal l
liased on these evaluations, the proposed change would not invalidate the conclusions presented in the Ul:SAR.
- l. Does the proposed modification involve a significant increase in the probability or consequences of an accident previously evaluated?
Sullicient margin exists to account for all reasonable instrument uncertaintles; therefore, no changes to installed equipment or hardware in the plant are required, thus the probability of an accident occurring remain unchanged.
'lhe initial conditions for all accident scenarios modeled are the same and the conditions at the time of trip, as modeled in the various safety analyses, are the same. Therefore, the consequences of an accident will be the same as those previously analyzed.
- 2. Does the proposed modification create the possibility of a new or difTerent Lind of accident from any accident previously evaluated?
The proposed change revises the method for RCS flow measurement, and therefore does not introduce any new accident indicators or failure mechanisms.
No new accident scenarios have been identified. Operation of the plant will be consistent with that previously modeled, i.e., the time of reactor trip in the various safety analyses is the same, thus plant response will be the same and will not introduce any different accident scenarios that have not been evaluated.
- 3. Does the proposed modification involve a significant reduction in a margin of safety.
There are no changes to the Safety Analysis assumptions. Therefore, the margin of safety will remain the same.
The proposed change does not impact the results from any accidents analyzed in the safety analysis.
==
Conclusion:==
Dased on the preceding information, it has been determined that this proposed change to allow an alternate RCS total flow measurement based on elbow tap Ap measurements does not involve a Significant liarards Consideration as defined in 10 CFR 50.92(c).
Page 42
fouth Texas Proiect Units I and 2 RCS Flow Measurement Usinn Elbow Tap Methodolocv Licensine Submittal 11
$~
u L
APPENDIX C MARKUPS TO SOUTil TEXAS PP.OJECT i
TECilNIC/', SPECIFICATIONS and DASES l
l Page 43 N'
[
$4 TABLE 2.2-1
{
g REACTOR TRIP SYSTEM INSTRUENTATION TRIP SETPOINTS 3
g M
TOTAL SENSOR E
ALLOWANCE EPROR 3
ITA).
Z 1S1 TRIP SETPOINT ALLOWABLE VALUE A
FUNCTIONAL UN1T E
1.
Manual Reactor Trip N.A.
N.A.
N.A.
N.A.
N.A.
1 a
--s 2.
Power Range. Neutron Flux 8.
a.
High Setpoint 7.5 6.1 0
s109% of RTP**
s110.7% of RTP**
'f
[
]
b.
Low Setpoint 8.3 6.1 0
525% of RTP**
s27.7% of RTP**
3.
Power Range. Neutron Flux.
2.1 0.5 0
55% of RTP** with 56.7% of RTP** with
~2 High Positive Rate a time constant a time constant i
=2 seconds a2 seconds g
a 4.
Deleted.
jj y
a u
ro Ti A
5.
Intermc# ate Range.
16.7 8.4 0
525% of RTP**
531.1% of RTP**
u Neutro', Flux a
e 5
5 6.
Source Range. Neutron Flux 17.0 10.0 0
510 cps s1.4 x 10 cps j
7.
Overtemperature AT 10.7 8.7 1.5 + 1.5#
See Note 1 See Note 2 h
- 8. ' Overpower AT 4.7 2.1 1.5 See Note 3 See Note 4 f
9.
Pressurizer Pressure-Low 5.0 2.3 2.0 a1870 psig
=1860 psig g
- 10. Pressurizer Pressure-High 5.0 2.3 2.0 s2380 psig 52390 psig
{--
- 11. Pressurizer Water Level-High 7.1
'3 2.0 592% of instrument 594.1% of M u sent i
span s n o_.
N/A N/A N/A
.4 o
=91.8% of lao of ko 7-
- 12. Reactor Coolant Flow-Low 4A M
OA design flow
- p design flow
- p 2o r.
- Loop design flow = 95.400 gpm
- RTP = RATED THERMAL POWER E
- 1.5% span for AT: 1.5% span for Pressurizer Pressure p
I IF E.'
South Texas Project Units 1 and 2 RCS Flow Measurement Usine Elbow Tan Methodolony 1,1censine Submittal POWER DISTRIBUTION LIMITS-3/4.2.5 DNB PARAMETERS LIMITING C0lOITION FOR OPERATION 3.2.5 The following DNB.related parameters shall be maintained within the L
limits following:
a.
Reactor Coolant System T,,,, s 598'F b.
Pressurizer Pressure, > 2189 psig*
c.
Reactor Coolant System Flow, = 392,300 gpm**
APPLICABILITY: MODE:1.
ACTION:
With any of the above parameters exceeding its limit. restore the parameter to within its limit within 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> or reduce THERMAL POWER to less than 5% of RATED THERMAL POWER within the next 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />.
SURVEILLANCE REQUIREMENTS 4.2.5.1 Each of the parameters shown above shall be-verified to be within its limits at least once per 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />. The provisions of Specification 4.0.4 are not applicable for verification that RCS flow is within its limit.
4.2.5.2 The RCS flow rate indicators shall be subjected to a channel calibration at least once per 18 months.
4.2.5.3 The RCS total flow rate shall be determined by pees 4s4= 5:t h'e=
measurements at.least once per 18 months. The provisions of Specification 4.0.4 are not applicable.
- Limit not applicable during either a Thermal Power ramp in excess of 5% of RTP per minute or a Thermal Power step in excess of 10% RTP.
- Includes a 2.8% flow measurement uncertainty.
SOUTH TEXAS UNITS 1 & 2 3/4 2 11 Page 45
South Texas Project Units I and 2 - RCS Flow Measurement Usine Fibow Tap Methodolony Licensine Submittal POWER DISTRIBUTION LIMITS BASES lEAT PLUX HOT CHANNEL FACTOR and NUCLEAR ENTHALPY RISE HOT CHANNEL MClQ3(Continued)
When an F measurement is taken, an allowance for both experimental error n
and manufacturing tolerance must be made, An allowance of 5% is appropriate for a full core map taken with the Incore Detector Flux Happing System, and a 3% allowance is appropriate for manufacturing tolerance.
The Radial Peaking Factor, Fxy(Z), is measured periodically to provide assurance that the Hot Channel Factor.RTP)n(Z), remains within its limit.
F The F
limit for RATED THERMAL POWER (F as provided in the Core Operating x
x Limits Report (COLR) per Specificatio#n 6.9.1.6 was determined from ex>ected power control maneuvers over the full range of burnup conditions in tie core.
3/4.2.4 00ADRANT POWER TILT RATIO The QUADRANT POWER TILT RATIO limit assures that the radial power distribu-tion satisfies the design values used in the power capability analysis.
Radial power distribution measurements are made during STARTUP testing and periodically during power operation.
The limit of 1,02, at which corrective action is required, provides DNB and linear heat generation rate protection with x y plane power tilts. A limit of 1.02 was selected to provide an allowance for the uncertainty associated with the indicated power tilt.
The 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> time allowance for operation with a tilt condition greater than 1.02 is provided to allow identification and correction of a dropped or misaligned control rod. In the event such action does not correct the tilt, the margin for uncertainty on F is reinstated by reducing the maximum allowed g
power by 3% for each percent of tilt in excess of 1.
For purposes of monitoring QUADRANT POWER TILT RATIO when one excore detector is inoperable, the moveable incore detectors are used to confirm that the normalized symmetric power distribution is consistent with the OVADRANT POWER TILT RATIO. The incore detector monitoring is done with a full incore flux map or two sets of four symmetric thimbles. The two sets of four symmetric thimbles is a unique set of eight detector locations. These locations are C 8, E 5. E 11. H 3, H 13. L-5, L 11, N 8, 3/4.2.5 DNB PARAMETERS The limits on the DNB related parameters assure that each of the parameters are maintained within the normal steady state envelope of operation assumed in the transient and accident analyses. The limits are consistent with the SOUTH TEXAS UNITS 1 & 2 B 3/4 2 5 Page 46
South Texas Project Units I and 2 RCS Flow Measurement Usine Elbow Tan Methodolocv Licensine Submittal POWER DISTRIBUTION LIMITS BASES 3/4.2.5 DNB PARAMETERS (Continued) initial FSAR assumptions and have been analytically demonstrated adequate to maintain a minimum DNBR of greater than or equal to the design limit throughout each analyzed transient. The T value of 598'F and the pressurizer pressure value of 2189 )sig are analyticM9 values. The readings from four channels will be averaged and twn adjusted to account for measurement uncertainties before comparing with the required limit. The flow requirement (392.300 gpm) includes a measurement uncertainty of 2.8%. FThe RCS flow measurement ^ uncertaintyTof 2;8%
bounds the precision calorimetric measurement method and the elbow: tap lAp measurement method.uThe1 elbow tap Ap measurement uncartainty (2.6%)cpresumesithat elbow tap Ap measurementsfare obtained from. either QDPS or.;the plantLprocess c
computer; The 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> periodic surveillance of these parameters through instrument readout is sufficient to ensure that the para:neters are restored within their limits following load changes and other expected transient operation.
SOUTH TEXAS UNITS 1 & 2 B 3/4 2 6 Page 47
.. _ _ _