TXX-4849, Forwards Response to 850705 Request for Addl Info Re NUREG-0737,Item II.D.1, Performance Testing of Relief & Safety Valves
| ML20199C820 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 06/13/1986 |
| From: | Counsil W TEXAS UTILITIES ELECTRIC CO. (TU ELECTRIC) |
| To: | Noonan V NRC - COMANCHE PEAK PROJECT (TECHNICAL REVIEW TEAM) |
| References | |
| RTR-NUREG-0737, RTR-NUREG-737, TASK-2.D.1, TASK-TM TXX-4849, NUDOCS 8606180261 | |
| Download: ML20199C820 (57) | |
Text
Log # TXX-4849 File # 10010 TEXAS UTILITIES GENERATING COMPANY NKYW/.Y TOWER . 400 N(DNTH 01JVE arrREIrr. I R. 8 B . DA11AB, TEXAS 75301 June 13, 1986
,'"MC L'A."."Oh Mr. Vincent S. Noonan, Director Comanche Peak Project Division of Licensing U. S. Nuclear Regulatory Commission Washington, D.C. 20555
SUBJECT:
COMANCHE PEAK STEAM ELECTRIC STATION (CPSES)
DOCKET NOS. 50-445 AND 50-446 NUREG-0737, ITEM II.D.1 - PERFORMANCE TESTING OF RELIEF AND SAFETY VALVES
Dear Mr. Noonan:
Attached is the response to your request for additional information dated July 5,1985 concerning NUREG-0737, item II.D.1 - Performance Testing of Relief and Safety Valves.
Very truly yours, s V/l6C W. G. Counsil BSD/arh Attachment c - NRC (0 + 40 copies)
A. Vietti-Cook jl@["dObbk
^
f
\
- goqA0 A intVIMitDN 61Y TEK AN t1TitJTIEN EE.Et'i'NIC (TB%tYANY m
M@ M *- $ / >d w -%-
ATTACHMENT 1 RESPONSE TO NRC REQUEST FOR ADDITIONAL INFORMATION (RAI)
DATED JULY 5, 1985 NUREG-0737, ITEM II.D.1 - PERFORMANCE TESTING OF RELIEF AND SAFETY VALVES OUESTIONS RELATED TO SELECTION OF TRANSIENTS AND INLET FLUID CONDITION QUESTION 1
~
The submittal identifies a steam discharge flow condition as a limiting event for the safety valves at this plant (loss of load for maximum pressurizer pressure and locking rotor for maximum pressurization rate). The submittal does not discuss whether single failures that could occur after the initiating event that would result in the dynamic forces on the safety relief valves being maximized were considered. Present a discussion that shows how single failures that would result in maximum dynamic forces on the safety relief valves were considered in selecting the limiting transient.
RESPONSE
The limiting events for dynamic forces on the safety valves were evaluated in EPRI NP-2296-LD, " Valve Inlet Fluid Conditions for Pressurizer Safety and Relief Valves in Westinghouse-Designed Plants," dated March 1982. This report was transmitted to Mr. H.
R. Denton, NRC, from Mr. D. P. Hoffman, Consumers Power Company, on behalf of the participating utilities on Sept. 30, 1982. This l report discusses in detail how the limiting events were selected from licensing FSAR transients, extended high pressure injection events, and cold overpressurization transients and provides information on single failures. The following is a summary of the effects of single failures on determination of the limiting events based on the above report and the Chapter 15 of the CPSES FSAR.
- 1. Licensing Basis Transients--Consistent with regulatory requirements, the FSAR transients are generally conservative and evaluate cingle failures' effects on several parameters including pressurizer pressure and pressurization rate. Peak pressure occurs within a fnw seconds of transient initiation. Single failures within the engineered safeguards system would have little, or no effect, on the pressurization rate or peak pressure observed. For the loss of load and locked rotor event (which bounds other loss of flow events), the peak pressure and pressurization rates are maximized by assuming the following:
PAGE 1
--No Ritctor Trip on Turbins Trip
-No Steam Dump
--Pressurizer spray and heaters are off
,-No Rod Control System operation
-Safety Injection does not operate
-Auxiliary Feedwater does not operate
-control Systems function only if it increases the severity of the event.
For the main feedline rupture event (which also bounds the small steam break event with respect to the peak primary pressure), a sensitivity analysis was performed to determine the break conditions causing the worst case results. In addition, the analysis was performed at Engineered Safety Features design power level and no credit was taken for the following:
-Pressurizer power operated relief valves (PORV's)
-Pressurizer Spray
-Turbine-driven Auxiliary Feedwater Pump
-Protective logic signals from:
-High pressurizer pressure
-Overtemperature N16
-High pressurizer level
-High containment pressure The loss of normal faedwater event with off-site power bounds the same event accompanied by a loss of offsite power (station blsckout). To maximize the dynamic effects on the safety valves, the analysis was performed at Engineered Safety Features design power level and the following were assumed:
-Warct single failure in the auxiliary feedwater system
-Secondary system power operated relief valves and condenser steam dump do not operate
-Pressurizer PORV's operate to decrease margin to water relief through the valves
-Normal reactor control systems do not operate For FSAR transients which result in reactivity and power distribution anomalies, the events are generally terminated or mitigated by the reactor protection system prior to challenging the pressurizer safety valves. Single failures that cause an anticipated transient without trip are beyond the scope of NUREG-0737, Item II.D.l.
- 2. Extended High Pressure Injection Events--Events which could cause an extended high pressure injection event and subsequent passage of steam or water through the safety valves include steam line ruptures'(which bounds an accidental depressurization of the secondary system) and spurious initiation of the high pressure safety injection system (SIS). For the steam line rupture, the SIS is initiated on a low pressurizer pressure which then begins to refill and repressurize the pressurizer. In terms of surge flow and liquid temperature at the valve inlet, this event is bounded PAGE 2
by tha cpurious initiation of high prassure SIS at power.
The spurious initiation of high pressure SIS assumes the following:
-No direct reactor trip on SI signal
-All protective systems unavailable except low ;
pressurizer pressure reactor trip
-No control systems operable
_ . Cold Overpressure Events--The cold overpressure events are described in detail in NP-2296-LD. As discussed in the report, the PORV with the lowest setpoint is assumed to fail. In this condition, the second PORV can adequately mitigate the pressure rise. Therefore, prior to requiring operation of the safety valves, both PORV's must fail.
4 PAGE 3
3 l
OUESTION 2 '
In the PORV operability discussions on low temperature I overpressure transients, variable fluid conditions (steam or water) and temperatures (saturated to subcooled) were identified.
To assure that the relief valves operate under all cold overpressurization events, include the nitrogen bubble case in the discussion. Also, identify the test data that demonstrate operability over the entire range of conditions. Confirm that the high pressure steam tests demonstrate valve operability for the low pressure steam case for both opening and closing of the-relief valve. In addition indicate the set pressure of the PORV for cold overpressure protection.
RESPONSE
EPRI test conditions for the PORV's were chosen based on expected inlet fluid conditions. Comanche Peak's cold overpressure (COP) event, including the nitrogen bubble case, was considered when formulating the enveloping EPRI test conditions. Maximum temperature and pressure conditions that can be achieved at the PORV inlet occur for steam bubble operation. Since pressure is normally maintained below the PORV setpoint, the maximum steam and saturated liquid pressure maintained in the pressurizer during startup and shutdown operations in anticipation of the COP event would occur at the PORV setpoint. PORV setpoints are given below. Tests were limited but designed to confirm operability over a. full range of expected inlet conditions. Steam, steam-to-water, and water flow tests were conducted. Results of these can be found in EPRI Report EPRI NP-2670-LD, Volume 8.
Although the steam tests were conducted only at the higher pressure, it is expected that satisfactory operation would also result at the less severe lower pressures. This can be seen by the successful low pressure, low temperature water tests.
The PORV setpoint are as follows: j PORV A PORV B TEMPERATURE PRESSURE TEMPERATURE PRESSURE DEGREES F PSIG DEGREES F PSIG 0 505 0 445 70 (Note) 505 70 (Note) 445 190 505 150 445 210 520 190 435 230 560 210 440 250 640 230 465 260 670 260 565 280 685 350 570 380 685 380 570 450 2350 460 2350 NOTE: 70 Degrees F is not used as a calibration point.
PAGE 4
OUESTION 3 The EPRI/ Marshall block valve tests were performed with the valves in a horizontal position (valve stems vertical). Identify the orientation of the Comanche Peak-1 block valves. If they are oriented in any direction other than horizontal, provide detailed information on how the EPRI data was extrapolated to assure operability of the block valve in the plant specific orientation.
RESPONSI As required by design drawings and verified by as-built inspection, the block valves are installed horizontally (valve stems vertical).
?
?
PAGE 5
QUESTION 4 The submittal referenced EPRI Safety Valve Test Nos. (929, 931a, 932, 1406, 1411, 1415 and 1419) as being suitable for evaluating the operability of the Comanche Peak-1 valves. The EPRI test blowdowns exceeded the 5% value given in the valve specifications. These increased blowdowns occurred with typical plant ring settings (-71 or -77, -18 relative to the bottom of the ring disk) that are much different from the ring settings specified in the submittal (i.e., -250, -18 relative to the level position). Based on a review of the EPRI test results, the Comanche Peak ring settings will produce larger blowdowns (i.e.,
greater than 10%). The higher blowdowns could cause a rise in pressurizer water level such that water may reach the safety valve inlet line and result in a steam-water flow situation.
Also the pressure might be sufficiently decreased that adequate cooling might not be achieved for decay heat removal. Discuss these consequences of higher blowdowns if increased blowdowns are expected. In order that the suitability of Comanche Peak 1 safety valve ring setting may be evaluated, provide the factory ring settings relative to the bottom of the ring disk or relative to the upper limit of ring travel.
RESPONSE
The previously provided guide ring settings for the safety valves were referenced to the highest locked position rather than relative to the " LEVEL" position (bottom of the disc ring). The following are the ring settings relative to " LEVEL" consistent with the EPRI tested valve settings of -71 to -77 notches (Guide Ring) and -18 (Nozzle Ring):
Valve Valve Nozzle Guide Tag Serial Ring Ring Number Number (Notches) (Notches) 1-8010A N56964-00-0070 -18 -103 1-8010B N56964-00-0069 -18 - 99 1-8010C N56964-00-0006 -18 -100 2-8010A N56964-00-0008 -18 - 82 2-8010B N56964-00-0007 -18 - 94 l 2-8010C N56964-00-0068 -18 -100 The ring settings in the table above are all typical ring j settings for the model of valve, as are the EPRI tested valve ring settings. The difference in individual settings is due to manufacturing dimensional tolerances of the individual valve components. The EPRI valve and the CPSES valves were all tested at Crosby using the same procedure to establish the ring setting to obtain the specified blowdown characteristics. These blowdown characteristics were verified in the EPRI test program.
( Additionally, since the CPSES valve inlet pressure drops are i
equal to that in the EPRI test configuration (See Question 7),
the CPSES valve stability and blowdown are expected to be consistent with that in the test.
l
- PAGE 6 l
l
OUESTIONS RELATED TO VALVE OPERABILITY OUESTION 5 The submittal does not address the expected blowdown and corresponding valve performance for the Comanche Peak 1 Crosby 6M6 safety valves at the plant ring settings. The submittal.
states that the ring setting for the safety valves are -250, -18 relative to the level position. The submittal states that the comparable EPRI tests were Test Nos. 929, 431a, 932, 1406, 1411, 1415, and 1419. The ring settings for these tests were -77 or
-71, -18 relative to the bottom of the ring disk. Explain how the expected values for backpressure and blowdown corresponding to the Comanche Peak 1 ring settings were extrapolated or calculated from test data with such different ring settings, and identify the values for backpressure and blowdown so determined.
RESPONSE
Refer to the Response to Question 4. The expected blowdown and corresponding valve performance for the CPSES safety valves are expected to comparable to the EPRI test.
l l
l l
l l
(
i l PAGE 7 I
OUESTION 6 Thermal expansion of the pressurizer and piping will induce loads on the safety and relief valve piping and on the pressurizer nozzles. This thermal expansion would also induce loading on the inlet flange of a safety or relief valve at the time the valve is required to lift. Evaluate the effects that this loading may have on valve operability.
RESPONSE
The loads induced on the safety and relief valves tested by EPRI exceed the loads for the Comanche Peak safety and relief valves.
The maximum moment tested for the 6M6 valve was during test 908 and was 298.75 in-K. The largest moment predicted for the safety valve inlet at Comanche Peak is 178.175 in-K. This demonstrates functionability for the Comanche Peak safety valves.
Likewise, a bending moment of 43.0 in-K was induced in the inlet of the Copes-Vulcan PORV test valve per EPRI 64-CV-174-2S. The largest moment predicted for the PORV inlet in the Comanche Peak valves is 28.708 in-K. This demonstrates functionability of the Comanche Peak relief valves.
PAGE 8
OUESTION 7 The EPRI guide for application of test results to plant specific evaluations suggests that the inlet piping pressure drops for the ,
Crosby 6M6 EPRI test valves be compared to the calculated Comanche Peak-1 Crosby 6M6 inlet piping pressure drop as a means of assessing valve stability. Provide the pressure drop calculations and the assessment of valve stability. If alternate pressure drop calculations were performed, provide a detailed explanation, and a detailed valve stability assessment.
RESPONSE
The safety valve inlet piping pressure drop calculation has been performed in accordance with the "EPRI PWR Safety and Relief Valve Test Program Guide for Application of Valve Test Program Results to Plant-Specific Evaluations," Interim Report, July 1982. The calculation is available at CPSES for NRC review and the following table provides a summary for the bounding valve piping cenfiguration:
EPRI CPSES TfdT CALCULATED AP OPENING 263 PSI 255 - 269 PSI AP CLOSING 181 PSI 152 - 158 PSI As can be seen in the table, the CPSES pressure difference is comparable to that in the EPRI test. The only value for which the CPSES calculated pressure difference is higher than the EPRI test condition is when a very conservative L/D valve is used in the calculation. This difference is insignificant (<2%) and is within the uncertainties of the methodology, L/D data, and mathematical round-off error.
As stated in the EPRI guide, if the calculated pressure difference is less than the test data, the in-plant valve would be expected to have performance at least as stable as the test valve; therefore, the CPSES valves are expected to have stability comparable to or better than the tested valve.
l l
PAGE 9
OUESTIONS RELATED TO THERMAL HYDRAULIC ANALYSIS OUESTION 8 The adequacy of the thermal hydraulic analysis could not be verified since it is not presented in the submittal. Provide detailed information on the program used so that the methodology for generating fluid parameters can be evaluated. Identify parameters such as timestep, valve flow area, pressure ramp rate, choked flow junction, and node spacing and discuss the rationale for their selection. Provide detailed information on how the program or methodology was verified for this application.
RESPONSE
The ITCHVALVE Computer Program was used for the thermal hydraulic analysis. (See Appendix A)
The adequacy of the thermal-hydraulie analyses can be verified by the comparison of analytical and test results for thermal-hydraulic loadings in safety valve discharge piping for EPRI Tests 908 and 917 presented in Appendix A, Section A.1.2.
In that evaluation, node spacing and time-step size were selected on the basis of stable solutions of the characteristic equations and natching of test data. The safety valve full open flow area of 0.022 ft was used in the model. This area is slightly
- smaller than the Crosby M-orifice area of 0.025 ft for the l tested valve, but results in a good analytical match of the tested fully open valve flow rate. Appropriate water temperatures were used. All pertinent data, including friction factors, loss factors and flow areas were based upon representative calculations and the system layout. Modeling of j the water was conducted with the water seal upstream of the i valves prior to transient intitiation. At time = 0+, the transient was initiated and the water slug position was analytically calculated during.and subsequ_ent to valve opening.
The Comanche Peak Unit 1 plant specific thermal-hydraulic i analysis was conducted based upon the same approach as used for l the comparison to test data. Node spacing and time-step size l were consistent with values utilized in the comparison. Valve l flow areas were selected based upon actual valve data with l appropriate margins applied to account for flow rate uncertainties. All pertinent data, including friction factors, loss factors and flow areas were based upon representative calculations and the system layout. Modeling of the water slug from a temperature profile, considering initial location and movement post-transient initiation, was consistent with the
. comparison study. The pressurizer pressure was held constant through the transient at initial values (see Response 11).
Choked flow is checked internally and automatically every time-step to ensure the proper formulation is applied at every flow path.
, PAGE 10 ,
1 l -. .
.c . . . . .
a
! OUESTION 9 Discuss whether multiple valve actuations were considered in the
, thermal hydraulic analysis. The maximum loading on the piping ,
typically occurs under a multiple valve actuation condition !
during which the valves open in sequence. The experience of EG&G Idaho indicates that the maximum loading occurs when the sequence )
of opening is such that the initial pressure waves from opening of the safety valves reach the common header downstream simultaneously. Additionally, if a PORV is discharging with flow in the common header, the_ piping loads could be significantly affected. Provide justification that sequential opening of the
, valves under multiple valve actuation conditions was considered.
RESPONSE
Two valve opening cases were addressed: 1) the three safety valves opening simultaneously and discharging without PORV flow and 2) the two PORV's opening simultaneously without safety valve flow. The three safety valves are identical and have the same set pressure (i 1 percent). It was, therefore, assumed for the analysis that all three safety valves open simultaneously without PORV flow. Because of similarity, the two PORV's were also assumed to open simultaneously without safety valve flow.
- Maximum common header (area of piping common to both safety and relief valve discharge piping) forces theoretically could be expected when valve sequencing is such that the initial pressure waves from valve opening reach a common downstream junction simultaneously. Based upon engineering judgment:
- 1. The simultaneous opening of the safety valves results in practically simultaneous peak loads at the safety l
valves common branch point. The peak forces occur within approximately 60 milliseconds of each other. As a result, no significant impact in the common header region due to safety valve discharge is expected if the valve sequencing is adjusted such that the peaks of the
~
initial pressure waves reach a common downstream header point simultaneously.
- 2. The total lengths of effective piping between each l valve outlet and the common junction point are not exactly the same. The likelihood of the valve phasing
- being such to compensate for the different lengths is very small; therefore; the peaks of the initial pressure waves from valve opening, either safety or I
relief, would not reach a common downstream junction at exactly the same time.
l 3. There is a significant amount of piping and dynamic
.' supports between the valve outlets and the common
. point. In the unlikely event that increased loadings S
from this common point to the relief tank were to occur, the effects would be limited primarily from near the common point to the relief tank. Significant isolation of the common region from the upstream region PAGE 11
b:cauce of the support configuration exists.
Therefore, the operability and integrity of the valves, the inlet lines to the valves, or the nozzles on the pressurizer would not be jeopardized.
Considerable margin exists between the conservatively calculated maximum stresses and the allowable stresses for the safety valve discharge event. Table A-5 in Appendix A illustrates this for the upstream piping for the faulted load combinations.
Additionally, for the downstream piping for the loading combination (P + DEADWEIGHT + SAFETY VALVE DISCHARGE), the maximum conservatively calculated stress is less than 80% of the allowable value.
l PAGE 12
OUESTION 10 Identify the program or methodology for calculating.the fluid forces for the structural analysis. Discuss the accuracy of the results and the procedures used to qualify the program or methodology.
RESPONSE
See Appendix A, Sections A.1.1 and A.1.2.
PAGE 13
l OUEST10N 11 Identify the initial conditions for the safety and relief valve thermal-hydraulic analyses. Describe the method used for treating valve resistance in the analyses and rdport flow rates corresponding to the resistances used. Because the ASME Code requires derating of the safety valves to 90% of actual flow capacity, the safety valve analysis should be based on a flow rating equal to 13.1% of the flow rate stamped on the valve, unless another flow rate can be justified. Provide further information explaining how derating of the safety valves was handled and describing methods used to establish flow rates for the safety valves and PORVs in the thermal hydraulic analysis.
RESPONSE
The initial conditions for the safety valve water slug discharge case included:
P (Upstream) = 2575 psia h (Steam, Upstream) = 1110 Btu /lb h (Water, Upstream) =Enthalpy based upon a temperature profile consistent with EPRI safety valve dis-change Test 917, i.e., approximately 300* F at the valve inlet and saturation temperature at the steam-water interface.
(An insulation arrangement was implemented that resulted in acceptable heating of the loop seals).
P (Downstream) = 18.0 psia The pressurizer conditions were held constant for the transient at 2575 psia and 1110 Btu /lb.
The initial ccnditions for the relief valve slug discharge case l
included:
P (Upstream) = 2350 psia I
i h (Steam, Upstream) = 1162.4 Btu /lb.
T (Water, Upstream) = 150 F P (Downstream) = 18.0 psia The pressurizer conditions were held constant for the entire transient at 2350 psia and 1162.4 Btu /lb.
l l
l
, PAGE 14 l
Safety cnd rolisf velvss era modaled as two-way junctions. The pressure drop across the valve, provided the system is sub-cooled is given by:
P = 1 joCD v where P = pressure drop CD = discharge. coefficient = f(Cv) ja = fluid density v = velocity through the valve In the case of choking at the valve, the velocity at the valve orifice area is set at the sonic velocity. Upstream and downstream boundary conditions are iteratively set to conserve mass and energy. Choked flow is internally checked to ensure the proper formulation is applied.
The nominal steam flow rating for the CPSES Crosby safety valves (orifice size 6M6) at 2500 psia is 420,000 lb/hr. The minimum analytically determined steam flow through each of the safety valves is greater than 503,500 lb/hr. This is equivalent to a flow of 120 percent of rated. The maximum expected steam flow through the Copes-Vulcan PORV's is 210,000 lb/hr. Values greater than 291,000 lb/hr were analytically calculated. This is a flow of 139 percent of rated.
PAGE 15
OUESTIONS RELATED TO STRUCTURAL ANALYSIS OUESTION 12 The submittal does not present details of the structural analysis. To allow for a complete evaluation of the methods used and results obtained from the structural analysis, please provide reports containing at least the following information:
(a) A detailed description of the methods used to perform the analysis. Identify the computer programs used for the analysis and how these programs were verified.
(b) A description of the method used to apply the fluid forces to the structural model. Since the forces acting on a typical pipe segment are composed of a net, or " wave", force and opposing " blowdown" forces, describe the methods for handling both types of forces.
(c) A description of methods used to model supports, the pressurizer and relief tank connections, and the safety valve bonnet assemblies and PORV actuator.
(d) An identification of the load combinations performed in the analysis together with the allowable stress limits.
Differentiate between load combinations used in the piping upstream and downstream of the valve. Explain the mathematical methods used to perform the load combinations, and identify the governing codes and standards used to determine piping and support adequacy.
(e) An evaluation of the results of the structural analysis, including identification of overstressed locations and a descripticn of modifications, if any.
(f) A sketch of the structural model showing lumped mass locations, pipe sizes, and application points of fluid forces.
(g) A copy of the structural analysis report.
RESPONSE
(a) See Appendix A.
(b) For each piping segment unbalanced or " net" forces were calculated. The hydraulic forcing functions were then simultaneously applied to the appropriate segment of the structural model. The axial extension from the balancing forces (opposing " blowdown" forces) on each end of the structural segments was considered in the evaluation.
However, this effect for this particular application was found to be negligible relative to the net unbalanced forces. Referring to structural analyses comparisons to test results presented in Appendix A for Tests 908 and 917, maximum support and pipe loads compared well PAGE 16 ,
with tset razults. Good comparisons of the maximum displacement values downstream of the safety valve were also seen.
4 (c)~ The structural supports were modeled in sufficient detail to analytically represent the system. .The shock suppressors
- and struts were modeled by inputting a stiffness in series with the piping. Calculated stiffness values were utilized.
All supports were linear and a linear overall system j analysis was conducted.
A' simplified model was used to represent the pressurizer.
i The pressurizer nozzles and pipe connections were i represented with appropriate pipe properties. The
, downstream piping terminated at the relief tank flange where
- the model was anchored.
I The valve bonnet assemblies and the relief valve actuators were modeled as extended masses displaced from the pipe
- centerline. The valve weight and center of gravity were 4 selected from the valve drawings. The stem properties j (diameter and thickness) were then selected to represent the valve frequency.
}
(d) For analysis purposes, the governing code for the piping qualification is the ASME Boiler and Pressure Vessel Code i Section III, 1977 Edition, with Addenda to and including Summer 1979. See Appendix A, Pages A-9 to A-17 for a discussion of the stress analysis.
The governing code for the support qualification is the ASME Boller and Pressure Vessel Code Section III-Division I,
- Subsection NF, 1974 Edition, with addenda to and including Winter 1979.
4 (e) See Appendix A, Pages A-25 and A-26. No overstresses exist i
for either upstream or downstream piping. No modifications are required .
j- (f) Figures A-1 and A-14 in Appendix A illustrate the lumped mass locations and by number the application points of fluid i forces for the safety valve discharge case. The piping l upstream of the safety valves is 6-inch Schedule 160 i (numbers 1-6, 29-34, and 38-43 on the figures). The piping between the safety valve outlets and the common header i (numbers 7-9, 35-37 and 44-46 on the figures) is 6-inch
- Schedule 40. The piping in the header pipework is 12-inch Schedule 80 (numbers 10-27 on the figures).
i' Figures A-2 and A-15 in Appendix A illustrate the lumped mass locations and by number the application points of the
- ~ fluid forces for the relief valve discharge case. A large I
fraction of the piping upstream of the relief valves is 6-inch Schedule 160 (numbers 1-5 on the figures). The i
remainder of the upstream piping and the piping between the valve outlets and the common header is 3-inch Schedule 160 4
PAGE 17
r
. - _ _ . , _ . . - . ~ - . .
piping (numbsrs 6-14 and 34-41 on the figura). Tho piping in the header is 12-inch Schedule 80 (numbers 15-32 on the figures).
(g) The summary report of the structural analysis is included in Appendix A.
PAGE 18
e 9 +M 1' OUESTION 13 F
According to results of EPRI tests, high frequency pressure 1 oscillations of 170-260 Hz typically occur in the piping upstream
)'
of the safety valve while loop seal water passes through the salve. An evaluation of this. phenomena is documented in the Westinghouse report WCAP 10105 and states that the acoustic
- pressures occurring prior to and during safety valve discharge i
are below the maximum permissible pressure. The study discussed i in the Westinghouse report determined'the maximum permissible i
pressure for the inlet piping and established the. maximum allowable bending moments for Level C Service Condition in the inlet piping based on the maximum transient pressure measured or calculated. While the internal pressures are lower than the
- maximum permissible pressure, the pressure oscillations could i
potentially excite high frequency vibration modes'in the piping, creating bending moments in the inlet piping that should be combined with moments from other appropriate mechanical loads.
} Provide one of the following:
i (a) a comparison of the expected peak pressures and banding
, moments-with the allowable values reported in the WCAP i report, or (b) justification for other alternate allowable pressure and i bending moments with a similar comparison with peak j pressures and moments induced in the plant piping.
RESPONSE
! The piping system response for Comanche Peak Unit No. 1 including
! the safety valve loop seal region, is due to frequencies less than 100 HZ. The frequency of the forces and moments in the 170-260 HZ range potentially induced by the pressure oscillations
, is significantly greater than this frequency. The upper limit of j significant frequency content for similar systems is also much less than this (170-260 HZ) range. Industry data indicates that only frequencies of 100 HZ or less are meaningful. The EPRI data confirms this. Consequently, no significant bending moment during the pressure oscillation phase of the transient will j occur.
4 l Pressure stresses based upon a pressure of 2485 psig were included with the bending moments resulting from the deadweight
- and the safety _ valve discharge piping loads. Because of the time phasing of the pressure oscillation (during water slug discharge through the safety valve) and the discharge piping loads (subsequent to water slug discharge through the valve) this term and moment term were not added. They do not occur coincidentally. A comparison of the intensified bending moments from the stress evaluation and the allowable moment presented in WCAP-10105 shows that all values are below the allowables.
} Specifically, the maximum allowable moment from Table 4-7 of l WCAP-10105 for 6-inch Schedule 160 piping for an internal
! pressure of 5000 psi is 516 in-kips. For the combination i pressure plus deadweight plus safety valve water slug discharge, i the maximum stresses occurred at node numbers 3161, 3130, 3130, i
, and 3160 respectively for the CRUN, butt weld, long radius elbow and branch connection components. The moments corresponding to these components are respectively 162.2, 163.5 and 162.2 in-kips.
j PAGE 19
..v. <+-,-----n, - e nv - , , ~ . - -. ,--+w,,-,v,,-----w--wm,--- --- - ,r, --*-,v-ve-~-w- w--we~,--
OUESTIONS RELATING TO PORV CONTROL CIRCUITRY OUESTION 14 NUREG 0737, Item II.D.1 requires that the plant-specific PORV control circuitry be qualified for design-basis transients and accidents. Provide information which demonstrates that this requirement has been fulfilled.
RESPONSE
Both PORV's have been provided with remote manual control capability from the Control Room (CR), and the Hot Shutdown Panel (HSP) in addition to the automatic actuation circuitry described in the CPSES FSAR 5.4.11, 5.4.13, and 7.6.8.
The following components of the PORV control circuitry are qualified for the events shown:
EOUIPMENT EVENTS
- 5. Control Switches & Control Bds. ** , SSE
- 9. Relays ** , SSE
- Although environmentally qualified for LOCA conditions, its operation is not assumed.
The valve is qualified for a seismic event to maintain its pressure retaining function only.
- The control board, control board switches, and relays are located in the control room and are not subject to an adverse environment.
l The documentation to support the qualification of the above equipment is available for inspection at CPSES.
PAGE 20
APPENDIX A 5
PRESSURIZER SAFETY AND RELIEF LINE EVALUATION
SUMMARY
REPORT TEXAS UTILITIES GENERATING COMPANY COMANCHE PEAK - UNIT #1 ,
lT i A.1 PRESSURT7ER SAFETY AND REIT.F LINE ANALYSIS INTDODUCTTON The Presswizer Safety and Relief Valve (PSARV) discharge piping syste for pressurized water reactors, located on the top of the pressurizer, provides overpressure protection for the reactor coolant syst e.
A water seal is l
! maintained upstrem of each pressurizer safety and relief valve to prevent This water seal practically a stem interface at the valve seat.
eliminates the possibility of valve leakage. While this arrangment maximizes the plant availability, the water alug, driven by high syst e pressure upon actuation of the valves, generates severe hydraulic shock locds on the piping and supports.
l l
l t
A-1
h-Under NUREG 0737,Section II.D.1, "Perfomance Testing of BWR 'and NR Kelief ard Safety Valves", all operating plant licensees and applicants are required to conduct testing to qualify the reactor coolant system relief and safety valves under expected operating conditions for design-basis transients and accidents. In addition to the qualification of valves, the functionability and structural integrity of the as-built discharge piping and supports must also be demonstrated or: a plant specific basis.
In response to these requirements, a progra for the ; 3rfomance testing of PWR safety and relief valves was fomulated by the Electric Faer Research Institute (EPRI). The primary objective of the Test Program was to provide full scale test data confiming the functionability of the rea: tor coolant systen power operated relief valves and safety valves for expected operating and accident conditions. The second objective of the program was permit to obtain sufficient piping themal hydraulic load data to confimation of models which may be utilized for plant unique analysis of safety and relief valve discharge piping systans.
1he method of analyses described in the following sections is mnsistent with the findings of the aforementioned EPRI Safety and Relief Valve Test Program.
i A-2
. . 1 i
I l
lv i
i i
. A.l.1 PLANT HYDRAULIC MODE.,
When the pressurizer pressure reaches the set pressure (2,500 psia for a i
safety valve and 2,350 pais for a relief valve) and the valve opens, the high pressure steam in the pressurizer forces the water in the water seal loop through the valve and down the piping system to the pressurizer relief tank. For the pressurizer safety and relief piping system, analytical hydraulic models, as shown in Figures A-1 and A-2 were developed to represent the conditions described above. ,
- The canputer code ITQNALVE was used to perforin the transient hydraulic analysis for the systen. This progran uses the Method of Characteristics approach to generate fluid parameters as a function of time.
One-dimensional fluid flow calculations applying both the implicit and explicit characteristic methods are perforined. Using this approach, the piping network is input as a series of single pipes. The network is generally joined together at one or more places by two or three-way l-junctions. Each of the single pipes has associated with it friction factors, angles of elevation and flow areas.
Conservation equations can be converted to the following characteristic equations:
dz g . Y+c g . . g . c < r . ,,....) a " ;c' n
A-3
h=V-c ..
dP dV 4c 2
- g - oc g . -c(F + ogcose) ah
- E g2 , - sh/so I
an 1 E~7 za variable of length measurement ta time Vs nuid velocity ca sonic velocity Ps pressure oa Guid density F= new resistance ga gravity J e angle off vertical i
Js conversion factor for converting pressure units to equivalent i heat tmits h: enthalpy '
q"'s rate of heat generation per unit pipe length The ca puter program possesses special provisions to allow analysis of
- valve opening and closing situations.
Fluid acceleration inside the pipe generates reaction forces on all segnents of the lines that are bounded at either end by an elbow or bend.
Reaction forces resulting fra nuid pressure and acmentta variations are calculated. These forces can be expressed in terms of the Guid properties
{
available frca the transient hydraulic analysis, perfomed using program I
ITCHVALVE. Tne amentta equation can be expressed in vector fom as:
l A-4
e e '
eV(V
- ndA)
F ey = ch ht ,y eVdv +c f s s From this equation, the total force on the pipe can be derived:
rg (1 - cos eg) ,y r2 (1 - cos a2I aW pipe * { sin og W Bend 1 { sin e2 Bend 2 e
i
+b straight $'* dl Ic pipe A piping flow area ya voltme F: force ra radius of curvature of appropriate elbow a: angle of appropriate elbow Ws mass acceleration g, a gravitational constant All other terms are previously defined.
Unbalanced forces are calculated for each straight segnent of pipe fra the The time-histories of pressurizer to the relief tank using progran FORFUN.
the subsequent structural these forces are stored on tape to be used for analysis of the pressurizer safety and relief lines.
A.1.2 COMPARISON TO EPRI TEST RESULTS Piping load data has been generated fra the tests conducted by EPRI at the Pertinent tests simulating dynamic Cebustion Engineering Test Facility.
opening of the safety valves for representative commercial upstream The resulting downstream piping Icadings envirornents were carried out. !
l A-5
_ I
-,w k
and responses were measured. Upstream envirorments for particular valve opening cases- of importance, which envelope the camercial scenarios, are:
l l
A. cr*1d unter dineharne en11m.d bv sta= - steam between the pressure source and the loop seal - cold loop seal between the stem and the valve.
B. Het water dimehmere rollmed by stRED - staan between the pressure source and the loop seal - hot loop seal between the steam and the valve.
- C. Steam dineharte - steam between the pressure source and the valve.
Specific thermal hydraulic and structural analyses have been cmpleted for the Ca bustion Engineering Test Configuration. Figure A-3 illustrates the placement of force measurement sensors at the test site. Figures A-4 and A-5 illustrate a omparison of the therinal hydraulically calculated results ' using the ITGVALVE and FORFUN ca puter prograns versus experimental results. For' test 908, the cold water discharge followed by
. steam case, Figure A-4 illustrates the force time history of the long vertical run (WE32/WE33) ismediately downstrean of the safety valve.
- For test 917, the het water discharge followed by steam case, Figure A-5 illustrates the thenmal hydraulically amiculated and the experimentally determined force time history for (WE32/WE33) . Although not presented here, ca parisons were also made to the test data available for safety valve discharge without a loop seal (steam discharge).
4 The application of the ITQNALVE and FORFUN ca puter programs for calculating the fluid-induced loads on the piping downstream of the safety
' and relief valves has been demonstrated. Although not presented here, the A-6 l
2 capability has also been shown by direct caparison to the solutions of classical probles.
The application of the structural caput'er progres (discussed in Section i A.1.3 for calculating the system response has also been demonstrated.
Structural models representative of the Ca bustion Engineering Test Configuration were developed. Figure A-6 and A-7 illustrate, respectively, a cm;mrison of the structural analysis results and the experimental results at location (wt32/WE33) for test 908 and test 917, respectively.
I
, l I
A-7
1
}
1 A.l.3 VALVE THRUST ANALYSIS The anfety and relier lines were modeled statically and dynamically (seimnically). The mathanatical model used in the seisnic analysis was modified for the valve thrust analysis to represent the safety and relief valve discharge. The time-history hydraulic forces detemined by FORFUN were applied to the i
piping system lunp mass points. The dynamic solution for the valve thrust was obtained by using a-modified-predictor-corrector-integration technique I
and nomal mode theory.
The time-history solution was found using progran FIXFM3 The input to this program consists of natural frequencies, nomal modes, and applied forces. The natural frequencies and nomal modes for the modified pressurizer safety and relief line dynamic model were detennined with the WESTDYN progran. The time-history displacement response was stored on magnetic tape for later use in cceputing the total systen response due to the valve thrust conditions. The time-history displacements of the FIXFM3 program were used as input to the WESTDYN2 program to detemined the time-history internal forces and deflections at each end of the piping elenents. For this calculation, the displacements were treated as imposed deflections on the pressurizer safety and relief line masses. The solution was stored on tape for later use in the piping stress evaluation and piping support load evaluation.
The time-history internal forces and displacements of the WESTDYN2 program were used as input to the POSDYN2 program to detemine the maximun forces, manents and displacements that exist at each end of the piping elements and
' the maximun loads for piping supports. The results fran progran POSDYN2
)
4 A-8
3 are saved on TAPE 14 for future use in piping stress analysis and support lbad evaluation. .
A.1.4 ETHOD OF STRESS EVALUATION .
In order to evaluate the pressurizer anfety and relief valve piping, appropriate 1 cad combinations and acceptance criteria were developed. The load ccabinations and acceptance criteria are identical to those roccamended by the piping subcannittee of the PWR PSARV test program and are outlined in Table A-1 and A-2 with a definition of load abbreviation p ovided in Table A-3.
A.1. 4.1 PRIMARY STRESS EVALUATION In order to perform a primary stress evaluation in accordance with the rules of the Code , definitions of stress ccabinations are required for the normal, upset, mergency and faulted plant conditions. Tables A-1 and A-2 illustrate the allowable stress intensities for the ap;ropriate canbination. Table A-3 defines all pertinent terms.
A. l . 4 .2 DESIGN CONDITIONS The piping minimum wall thickness, t,, is calculated in accordance with the Code. The actual pipe minimum wall thickness meets the Code requirenent.
The ccabined stresses due to primary loadings of pressure, weight, and design mechanical loads calculated using applicable stress intensity factors must not exceed the allowable l'imit. The resultant manent, Mg, due to loads caused by weight and design mechanical loads is calculated using the follcwing equatien:
I
- A-9
- , - - -- p - , , , _ . - ,m.,we --__.-__--_-_,--____w,--,,,,---_y-,
)
i l
~
MI=
+ }2 #wt . g IDML 2
M*wt M*tML)1 .
l uz
/
,z .
,2 S2 wt .
cMt)i ..
uhere M eMy e M = deadweight moment components x 2 wt wt wt
.M = design mechanical load moment components M
g>My The maximun stresses due to pressure, weight, and DNL in the piping system are reported in Tables A-4 thru A-7.
1 A-10
3' A.1.4.3 UPSET CONDITIONS The ccabined stresses due to the primary loadings of pressure, weight, OBE seinic, and relief valve thrust loadin&s calculated using the applicable stress intensity factors must not exceed the allowables. The resultant shcun aments, ,g, due to loads caused by these 1cadings are calculated as below.
For sei mic and relief valve thrust loading:
2
. ,2 3 yr,32'2 ./,<<.(, sito . ,e2f232 spr
,i . l r,% .(,2wt x j
\, o>
.r, z u2 T2 u2
,s j l
g (,2* e i
l i
A-11 ;
, _; l
. .y where <
- l
"" r .
\
= deadweight moment components
.M .M 4.
M*wt #wt Nt )
= inertial 0BE moment components 3
M*0BE'#0BE' M M 0BE
,M = relief line operation moment components
- M M* SOT g #50TU SOTy
[ A.l.4.4 EMERGENCY CONDITIONS The'canbined stresses due to primary loadings of p essure, weight and safety valve thrust, using applicable stress intensification factors, must the resultant manent, not exceet. the allowable limits. Je magnitude of Mg , is calculated fran the moment ccaponents as shown below:
M + \2 + + M SI / + M %2 1/2 I= I
[M, SOT M*. l[M # SOT #wt l lM3 50T Z
.N wtll N ) N wtlI .
? .
,i..
l where
,e M,w t,Mywt .M2 wt
. deadweight moment components M ,M ,M = safety'line operation moment components 3
- SOT SOTg 50T E E
+
9 A-12
5; A.l.4.5 FAULTED CONDITIONS The ca bined stresses due to primary loadings of pressure, weight, SSE and 5077 , using applicable stress intensification factors must not exceed the allowable limits. For the resultant ament loading, M , the g SSE and SOT y asents are cabined using the sq0are-root-of-the-sm-of-the-squares (SRSS) addition and added absolutely with deadweight for each a ment ca ponent.
(M,, M , M,). The magnitude of the resultant ament, gN , is calculated fra the three ament esconents, as shown below:
g I, /g 2 .
g*SSE 2g 1/2 . g h2
("$OT ) Nt Y
2 2
. [tM . M 2) 1/2 . M i
SOT JSSE / #wt Q(Y Y
'Y 2 N 2 " 1/2
. +
l
((M + M I
SSE 2 } 1/2 + M Z
T
/
\(ISOT' wt') .
where i
l M ,M = deadweight moment components
\t #wt, M*wt M*SSE,M#SSE , M*SSE = inertial SSE moment components M ,M ,M = maximum of 50Tgor SOTE moment components I
- SOTp ISOTy SOTp i
e A-13
l l
3 For the safety and relief piping, the faulted condition load cabination of I yessure, weight, and valve thrust is considered as given in Table A-1 and A-2 and defined in Table A-3 . The pipe break loads (MS/FWPB or LOCA) can be ignored for the PSARV systan. These loads have very little impact on the pressurizer safety and relief systen when empared to the loading conditions discussed in this report.
A.1.4.6 SECONDARY STRESS EVALUATION The cabined stresses due to the secondary loadings of themal, pressure, and deadweight using applicable stress intensification factors must not
- exceed the allowable limit. For the resultant m ment loading, gM , themal aments are combined as shown below:
~
2 2 2 1/2 M
I. /M* MAX - M*MINN + /M -M h
- + [M* MAX - M* MIN
( (yMAX MIN) ( ..
,M ,M 2 = maximum thermal moment considering all thermal cases X # MAX MAX 4
including normal operation
,M 2 = minimum thermal moment considering all thermal cases M* MIN ,M # MIN MIN including normal operation l
This, M, g
is then substituted into the appropriate equations of the applicable code.
A-14
_ _ _ . .__ _ __ _ _ _ _______ ._ _ _ _ _ . . _ _ . . _ _ _ _ . . _ _ . . __ _ .__.__m__. _ , . . , . _ . _ _ _ _ . . . _ _ _ _ _ . .
J TABLE A-1 LOAD QMBINATIONS AND ACCEPTANCE CRITERIA FOR PRESSURIZE AND RETEF VALVE PIPTE AND SUPPORTS . UPSTREAM OF VALVES Piping Allowable Plant /Systa Stream Intensity Oemratina Candition Lnad Cmbination f">wnbi na ti on
- N 1.5 5 ,
1 Nomal Upset N + CBE + SOTg 1.85, 2
' Emergency N + SOTE 2.25 S ,
3 Faulted N + SSE + SoiF 3.0 5, 4
NOTES: for SOT definitions and other load abbreviations.
(I) See Table A-3 Use SRSS for cabining dynanic load responses.
(2)
The bounding nunber of valves (and discharge sequence it setpoints (3) are significantly different) for the applicable systen operating transient defined in Table A-3 should be used.
but (4) Verification of functional capability is not required, allwable loads and accelerations for the safety-relief valves aust be net.
b A-15
TABLE A-2 LOAD CDMBINATIONS AND ACCEPTANCE CRITERIA FOR PRESSURIZER SAFEIT AND RELIEF VALVE PIPING AND SUPPORTS RFTSMICALLY DESIONED DOWNSTREAM PORTION Plant / System Piping Allowable Erwhi nati on Omratina Conditien f nad Cmbination Stress Intensity 1 Normal N 1.0 S 2 upset N . Soro 1.2 S 3
3 up.et N . OBE . SOTu 18%
4 Emergency N + SOT l*03 E h 5 Faulted N + SSE + SCTp 2.4 S h
NOTES: (1) This table is applicable to the sei m ically designed portion of downstream non-Category I piping (and supports) necessary to isolate the Category I portion fran the non-seismically designed piping response, and to assure acceptable valve loading on the discharge nozzle.
(2) See Table A-3 for SOT darinitions and other load abbreviations.
(3) The bounding ntaber of valves (and discharge sequence if setpoints are significantly different) for the applicable syste operating transient defined in Table A-3 should be used.
(4) Verification of functional capability is not required, but l allcwable loads and accelerations for the safety-relief valves j must be met.
l l
(5) Use SRSS for ccabining dynamic lead responses, i
l i
A-16 l
l l
TABLE A-3 .
. DEFINITIONS OF LOAD ABBREVIATIONS
Safety valve discharge transient (1), (2)
SOTE:
SOT F
Max (SOTg ; Mg ); or transition fl w OBE : Operating basis earthquake SSE : Safe shutda n earthquake S a Basic material allwable stress at maxistan (hot) tanperature h
S, s Allwable design stress intensity (1) May also include transition fim, if detemined that required operating procedures could lead to this condition.
(2) Although certain nuclear steam supply design transients (for exanple, loss of load) which are classified as upset conditions may actuate the safety valves,' the extreely lw nunber of actual safety valve actuations in operating pressurized water reactors justifies the mergency condition from the ASME design philosophy 3
and a stress analysis viewpoint.
l l
A-17
J i
9 7
I s
B 6
4 9 5 2 It 10 U
s,s 31 18 32 34
{'
- 36 33 20 33 i 37 22 A 44 -
b 45 15 46 24 '
41 '43 39 I
25 ',
6 ..
38 27 -
1 Figure A-1 SAFETY LINE HYDRAULIC MODEL FROM PRESSURIZER TO RELIEF TANK (STRESSPROBLEM1-53)
4 I
.l I
t-17 18 e 15 21 j 40 4j 1
22
) 39 j4 13 38 12 23 i
19 37 36 x 35 25 b g 4
34 27 8
j f 20 6
i 29 30
.r 1 ;
32 Figure A-2 REllEF LINE llVDRAULIC MODEL FROM PRESSURIZER TO RELIEF TANK (STRESS PROBLEM 1-53)
GO O W O S eD e 66Me e e de 4 $D O OO #O O @ O 9DO@ 9MM e O g4M. $4 awm A
f* h
~
.N
,i M N N b.c \ m p
. . E me \
- = $ N, \ W
, g
\
\\\ \\\\\\\\\\\\ \\g ,
. \ ,
N s ._
\ E N E
\ s
, . \ y tv s r I
=
- }" \
\ $
3 % E w s w W N N$ > k b.
ug O
- # s N w
\
~- ~ g x
l~ e >
as :s
=
a j'
p
- a. +.
e N N
g-4 '
) + rE u
n s - x} c-mm &
rW x s a
[- '
_ .ts 3, s ?
o
) -
{ -
A s s
s I
s $
J L .?
"n'-
- As -
tt- :
+
e=
WW sw xxN :s s
s e
A-20 1
. _ . _ _ _ _ _ _ _ _ . . _ .,__ , _ , _ _ , _ - _ _ _ _ _ _ _ _ . ~ . _ _ . - _ _ _ . - . .
i 1.0E5 <
. l I l
~% l
.8 j
)
p-0
"[
'o I n
l
, l
-1.0E5 i
-2.0E5 l
tests
- ITCEVALVE
-3.0E5 0.1 0.2 0.3 -
0.4 0.5 time (seconds}
Figure A-4 COMPARISON OF THE EPRI FnRCE TIME-HISTORY FOR WE12 AND WE33 FROM TEST 908 WITH THE ITCHVALVE PRE 0!CTED
. FORCE TIME-HISTnRY A-21
2.0E4 lA .
[\ "
1.0E4 ; .
l 3 -
5 l 'li e
ii 7 5 m .- -
si : Anm.
,y v v v v C. O. ,v . ;
e 4 "
> , n >;
10 f 17 Ii l
l 33
-e.
I h s ,e @' ' ;,. x u
W-13 J1 Ip 1
i
.h44_ .
~ k W Mess
- - .. ,. _ q l
,+ ,, ( g
,, ,, .a s
-r II as 1
(= ..
E i
I i
l
f .
O 3
B-
% 1 y
e k 5
g t
I3 i
E W
's i."y r;;se>.... .
h U
. LY - '
Es un e* .s .
W$ sm5.5,a -
3(
.h e $
1 t 7e w e Q i
{E
- e
$ '5 E 3
- i.
- u
\- e ' E = A-35
)}}