ML20207F292
| ML20207F292 | |
| Person / Time | |
|---|---|
| Site: | 05200003 |
| Issue date: | 12/31/1997 |
| From: | Andreycheck T, Chismar S, Delose F WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP. |
| To: | |
| Shared Package | |
| ML20137S565 | List: |
| References | |
| WCAP-14293, WCAP-14293-R02, WCAP-14293-R2, NUDOCS 9906080163 | |
| Download: ML20207F292 (66) | |
Text
v Please follow the following instructions to convert the AP600 tow-Pressure Integral Systems Test at Oregon State University Test Analysis Report, Rev.1 to Rev. 2:
Remove:
l Replace with:
VOLUMEI Rev.1 cover and spine Rev. 2 cover and spine Rev.1 AP600 Cover Sheet Rev. 2 AP600 Cover Sheet-I Rev.1 pages 1.2-1 Rev. 2 pages 1.2-1 Rev.1 pages 1.4-1/2 Rev. 2 pages 1.4-1/2 Rev.1 pages 1.4-5/6 Rev. 2 pages 1.4-5/6 Rev.1 pages 2.1-1/2 Rev. 2 pages 2.1-1/2 Rev.1 pages 4.6-1 to 4.6-8 Rev. 2 pages 4.6-1 to 4.6-8 Rev.1 pages 4.7-1/2 Rev. 2 pages 4.7-1/2 Rev.1 pages 4.7-7/8 Rev. 2 pages 4.7-7/8 Rev.1 pages 4.8-1/2 Rev. 2 pages 4.8-1/2 Rev.1 pages 4.8-5 to 4.8-8 Rev. 2 pages 4.8-5 to 4.8-8 Rev.1 pages 4.9-1/2 Rev. 2 pages 4.9-1/2 Rev.1 pages 4.9-5/6 Rev. 2 pages 4.9-5/6 l
Rev.1 pages 4.13-1/2 Rev. 2 pages 4.13-1/2 Rev.1 pages 4.13/9/10 Rev. 2 pages 4.13-9/10 l
Rev.1 pages 4.15-1/2 Rev. 2 pages 4.15-1/2 Rev.1 pages 4.15-5/6 Rev. 2 pages 4.15-5/6 Rev.1 pages 4.16-1/2 Rev. 2 pages 4.16-1/2 O
Rev.1 pages 4.17-3 to 4.17 6 Rev. 2 pages 4.17-3 to 4.17-6 Rev.1 pages 4.17-11/12 Rev. 2 pages 4.17-11/12 l
Rev.1 pages 4.18-3/4 Rev. 2 pages 4.18-3/4 2
Rev.1 page 5.1-1 Rev. 2 page 5.1-1 Rev.1 pages 5.1.1-1/2 Rev. 2 pages 5.1.1-1/2 Rev.1 pages 5.1.1-5/6 Rev. 2 pages 5.1.1-5/6 Rev.1 page 5.2-1 Rev. 2 page 5.2-1 Rev.1 page 5.4-1 Rev. 2 page 5.4-1 Rev.1 pages 5.4.1-1/2 Rev. 2 pages 5.4.1-1/2 Rev.1 pages 5.4.2-1/2 Rev. 2 pages 5.4.2-1/2 Rev.1 pages 5.4.3-1 to 5.4.3-4 Rev. 2 pages 5.4.3-1 to 5.4.3-4 VOLUME 11 Rev.1 page 5.9-1 Rev. 2 page 5.9-1 Rev.1 page 5.10-1 Rev. 2 page 5.10-1 Rev.1 page 5.11-1 Rev. 2 page 5.11-1 Rev.1 page 5.12-1 Rev. 2 page 5.12-1 l
Rev.1 pages 6.1.3-1/2 Rev. 2 pages 6.1.3-1/2 Rev.1 pages 6.1.3-29/30 Rev. 2 pages 6.1.3-29/30 Rev.1 page 7.1-1 Rev. 2 page 7.1-1 i
Rev.1 pages 7.2.2-1 to 7.2.2-4 Rev. 2 pages 7.2.2-1 to 7.2.2 4 Rev.1 pages 91/2 Rev. 2 pages 9-1/2 i
l 9906090163 990921
~
l PDR ADOCK 05200003 l
A PDR
WESTINGHOUSE NON-PROPRIETARY C1, ASS 3 WCAP-14293 i
Revision 2 AP600 LOW-PRESSURE INTEGRAL SYSTEMS TEST AT OREGON STATE UNIVERSITY TEST ANALYSIS REPORT DECEMBER 1997 Authors:
T. S. Andreycheck R. C. Haberstroh S. A. Chismar L. E. Hochreiter F. Delose W. R. Morrison S. V. Fanto M. Ogrinsh R. L. Fittante F. E. Peters O
C. Frepoli R. F. Wright M. T. Friend H. C. Yeh l
WESTINGHOUSE ELECTRIC COMPANY Energy Systems Business Unit Advanced Technology Business Area P.O. Box 355 Pittsburgh, Pennsylvania 15230-0355 i
C1997 Westinghouse Electric Company All Rights Reserved O
I ntW344w.non:Ib 122297 REVISION: 2
1 1
(v) 1.2 Test Objectives The OSU facility was designed and constructed to specifically examine the LTC performance of the AP600 passive safety-related systems and their interaction with the nonsafety-related active systems.
The range and types of tests investigated in the OSU test facility covered the ranges and phenomena expected for the SBLOCA and LTC transients. Tests were performed to examine the different break sizes and locations to cover the range of phenomena of interest. The data from the tests are used to validate the safety analysis computer codes used to analyze the AP600.
To cover the range of conditions for the LTC transient, various tests were performed. One test modeled as large a break as possible to rapidly depressurize the facility so that decay power would be at a high value when LTC began. Other tests were performed with conditions that would result in a hot IRWST and sump when the primary system transitioned into LTC. Both conditions maximized the production of core steam to be vented through the ADS-4 valves to maintain sump injection.
A detailed scaling analysis was developed for the OSU tests to m! ate the scaled-pressure and reduced-height facility to the AP600 plant. The OSU Facility Scaling Report
- specified the facility dimensions, resulting flow areas for the breaks, and pressure drops needed to preserve the phenomena expected for the AP600 SBLOCA and LTC transients. The scaling study provides the bridge to relate the AP600 SPES-2 Test Analysis Report
- to the similar OSU tests, and to relate the OSU tests to the
[)
i AP600 design. In addition, the AP600 Scaling and PIRT Closure Reportc25> presents an overall 1 discussion of the scaling of the AP600 tets and how they all are related.
The following are the specific test objectives of the OSU program:
To provide data to establish the pedigree of the passive safety-related systems for LTC.
To provide overlap with the full-pressure, full-height SPES-2 tests
- so that an assessment of the
=
scaling effects of the OSU tests could be made. Therefore, similar break locations and sizes (scaled) were examined in both facilities and comparisons were made.
To cover the range of phenomena expected for the AP600 LOCA in addition to the LTC period.
=
n (v) o:umnonA122297 1.2-1 REVISION: 2
7 e
1.4 Test Facility Scaling A detailed component and system scaling analysis was performed for the OSU test facility and is given in the OSU Facility Scaling Report.(2) The results of the scaling analysis were used to specify the design of the facility. De primary objective of the scaling analysis was to design a working scale model capable of producing the same types of flow behavior encountered in the AP600 during the SBLOCA transient, and LTC.
Various scaling techniques can be applied to the design of a small-scale thermal-hydraulic test facility.
The traditional approach has been to use power-to-fluid-volume (PN) scaling, his scaling approach has been successfully applied in various studies such as the FIECHTSEASET Program Final Repon.W The optimum condition for this scaling approach occurs when the scale model implements the same working fluid as the full-scale system, is built at full height using similar materials, and is operated at full pressure. His generally results in constructing a very tall and thin scale model.
Unfortunately, the hydrodynamic behavior in the plenum regions may not be fully represented in the full-height model. A reduced-height, power-to-volume scaled model gives a better representation of multidimensional effects in the plenum and downcomer regions.#)
ne hierarchical two-tiered scaling analysis (H2TS) method has been used to develop the similarity criteria necessary to scale the systems and processes of importance to AP600 integral system and LTC.
De H2TS method, developed by the NRC,is fully described in Appendix D of An Integrated Structure and Scaling Methodologyfor Severe Accident Technicalissue Resolution (') and is referred to as the SASM methodology. Dere are four basic elements of the H2TS analysis method. De first i element consists of system decomposition. Each system can be successively divided as follows: first, I into interacting subsystems; where appropriate, components of those subsystems; then interacting I constituents (materials); and finally, into interacting phases (liquid, vapor, or solid). Each phase can be characterized by one or more geometric configurations, and each geometric configuration can be described by three field equations (mass, energy, and momentum conservation). Each field equation can be characterized by several processes. This is depicted in Figure 1.4-1.
I After identifying the system of interest and decomposing it as in Figure 1.4-1, the second element of I the H2TS method requires identification of the scaling level at which the similarity criteria should be developed. This is determined by the phenomena being considered.
For example, if the phenomenon being considered involves mass, momentum, or energy transport between materials such as water and solid particles, then the scaling analysis should be performed at the constituent level. If the phenomenon ofinterest involves mass, momentum, or energy transport between vapor and liquid, then the scaling analysis should be performed at the phase level. Herefore, identifying the scaling level will depend on the phenomenon being addressed. Table 1.4-1 presents the system hierarchy implemented in the OSU Facility Scaling Report.(2)
O omuw.nonab.12:597 1.4 1 REVISION: 2
Thennal-hydraulic phenomena involving integral system interactions, such as primary system depressurization or loop natural circulation, are examined at the " system" level. Thermal-hydraulic phenomena-such as PRHR decay heat removal, CMT, accumulator, and IRWST passive safety injection, automatic depressurizauon and LCS recirculation cooling-are examined at the subsystem level. Thermal-hydraulic phenomena important to individual components-such as the reactor core, pressurizer, SGs, hot legs, cold legs, coolant pumps, and interconnecting piping-are examined at the module level. Specific interaction between the steam-liquid mixture and the stainless steel structure are examined at the constituent level.
The OSU scaling study presents scaling analysis performed at different levels. The thennal-hydraulic phenomena of interest, the system level at which the analysis was performed, the control volume for the analysis (i.e., the geometric configuration), the applicable balance equations, and the processes imponant to the thermal-hydraulic phenomena of interest are discussed and analyzed for the simulated reactor system as well as the major components in the system.
The third element of the H2TS method requires the performance of a top-down (system) scaling analysis. The top-down scaling analysis examined the synergistic effects on the system caused by complex interactions between the constituents deemed important by the plausible phenomena identification ranking table (PPIRT). This has been modified as discussed ir. Section 1.3, and a revised PIRT is presented in Table 1.3-1. The top-down scaling approach used the conservation equations at a given scaling level to obtain characteristic time ratios and similarity ed'eria, and identified important processes to be addressed in the bottom-up scaling analysis.
The fourth element of the H2TS method required the performance of a bottom-up (process) scaling analysis, which developed the similarity criteria for specific processes such as flow-pattem transitions, and geometry-and flow-dependent heat transfer. The focus of the bottom-up scaling analysis was to develop similarity criteria to scale individual processes of imponance to system behavior identified by the PIRT and develop the design information for the test facility.
The basic objective of the H2TS method was to develop sets of characteristic time ratios for the transfer processes c,f interest. This can be done by writing the control volume balance equations for each constituent, k, as follows:
dV yg = AlQ Vkl IjkmA I 4-I g
k km dt Defining A[Qgyg]:
AlQ Vk} = [QkVk. -[Q Vk}out 1.4-2 k
l k
O c:u3uenon1b.11:097 1.4-2 REVISION: 1 I
F O
sufficient volume and size to correctly model the plant pressure drop and possible three-dimensional flow behavior that could occur in the simulated reactor vessel, plenums, and downcomer. To chose a consistent diameter scale, a simple relationship was derived from the one-dimensional momentum equation to relate the length ratio to the diameter ratio. The choice of the diameter ratio was further verified with a bottom-up scaling approach in which the two-phase flow regimes and transitions between flow regimes were examined using the work of Taitel and Dukler.(8) The possible distortions in the flow regimes and their transitions was also examined for the horizontal piping following the approach of Schwartzbeck and Kocamustafsogullari.(9) 'Ihe flooding review by Bankoff and LeeUO) was also used to verify that the chosen diameter ratio would have minimum surface tension effects if flooding occurred. Using this approach, the facility. dimensions could be specified with confidence that the key parameters and phenomena identified in the PIRT would be preserved in the OSU facility so that the resulting data could be used for AP600 safety analysis code validation.
De OSU tests were designed :0 start in an all-liquid recirculation mode with the simulated RCPs operating with system pressure at.about 400 psia. When a break is initiated, the system begins to depressurize. To preserve the depressurization behavior of the OSU facility, a reference pressure was selected and a scaling rationale developed to relate the lower pressure OSU tests to the higher pressure AP600 transient for the depressurization transients. He results of the scaling approach were used to develop the relationships that led to selection of the OSU facility break areas, ADS valve areas, accumulator gas pressure, and SG secondary-side safety pressures to preserve the scaling relationships between the facility at its reduced pressure and the AP600 plant at its higher pressure. De scaling process used is shown in Figure 1.4-3, in which a top-down scaling approach was used to develop the systems scaling analysis for a simplified control volume of the reactor primary system. A bottom-up
{
approach was then used to develop the fluid property relationships for the depressurization transients.
De approach, originally developed by Kocamustafaogullari and IshiiU 3) and expanded on by Moskal,02) was extended to relate the OSU fluid property conditions to the AP600 plant conditions.
Moskal defines the property relationship:
I M
Y=
1.4-9 pphg, q fg as the key property group to be preserved. His particular grouping also appears in the coefficients for the core velocity from the two-phase natural circulation loop scaling analysis previously described.
The fluid properties and depressurization approach is to select a reference pressure for both the OSU facility and the AP600 that will capture the important parameters identified in the PIRT. Exarmnmg Figure 1.4 2, the AP600 primary system pressure will stabilize, after tlw initial subcooled blowdown, to a near constant value, slightly above the safety valve setpoint for the SG secondary side. The primary pressure will remain at this value for a relatively long period, depending upon the break size and when the ADS activates, which will depressurize the primary system to the containment pressure.
During this time period, the passive safety-related systems of the AP600 will be in operation and the phenomena of importance, which are identified in the PIRT, will be present. Berefore, the cA2344w. mon:Ib-ll1097
},4 5 REVISION: I
secondary-side SG safety valve serpoint pressure was chosen as the reference pressure for the AP600 plant. Note that this ignores the subcooled depressurization ponion of the transient which, for a SBLOCA, is a shon period compared with the total transient length.
j A similar reference pressure can be chosen for the OSU facility where the primary pressure stabilizes above the SG pressure, so that:
I Y
Y I
=
1.4-10 Y
Y o,,
o, where:
I Y ],
OSU reference pressure
=
o i
Y]p AP600 reference pressure
=
o The top-down and bottom-up pressure scaling must also be consistent with the natural circulation scaling, which establishes the facility volume, time, and velocity scaled ratios given the selection of the length and diameter for the facility.
The bottom-up pressure scaling examined the critical flow through the break and the ADS valves, and developed the relationships for the break areas and the valve areas that were consistent with the fluid property scaling given above. Therefore, given a break size in the AP600, a corresponding break size ca. he calculated for the OSU facility that will maintain the time, velocity, and volume scaling for a selected length and diameter scale which was chosen from the two-phase natural circulation scaling relationships.
l l
t t
i e
o m w.wrr16111097 1.4-6 REVISION: 2 i
l
2.1 Overall Facility Description The OSU test facility is a scaled model of the AP600 reactor coolant system (RCS), steam generator system (SGS), passive core cooling system (PXS), automatic depressurization system (ADS), lower containment sump (LCS), chemical and volume control system (CVS), and normal residual heat I removal system (RNS). A complete description of the test facility can be found in Reference 26. In addition, the facility is capable of simulating the AP600 passive containment cooling system (PCS) condensate return process. Figure 2.1-1 is an isometric drawing of the test facility, and Figure 2.1-2 is a simplified flow diagram of the test facility. De facility reflects the scaled AP600 geometry, including the piping routings. All components and piping are fabricated from austenitic stainless steel.
De relative locations of all tanks and vessels-such as the IRWST, CMTs, and acci.nulators were maintained as determined by the scaling approach he facility uses a unique break and ADS measurement system (BAMS) to measure two-phase break and ADS flow.
The RCS is composed of a reactor vessel, which has electrically heated rods to simulate the decay heat in the reactor core, and two primary loops. Each primary loop consists of two cold-leg pipes and one hot-leg pipe connecting a SG to the reactor vessel. A reactor coolant pump (RCP) on each cold leg takes suction from the SG channel head (downstream of the SG U-tubes) and discharges it into the downcomer region of the reactor vessel. A pressurizer with an electric heater is connected to one of -
the two hot legs through surge-line piping. De top of the pressurizer is connected to the ADS 1-3 line. An ADS-4 line is connected to each hot leg.
The reactor vessel contains two DVI nozzles that connect to the DVI lines of the PXS. A flow venturi I
I is incorporated in each DVI nozzle (Reference 26) to limit the loss of inventory from the reactor vessel in the event of a double-ended DVI line break.
His test facility models the primary and secondary side of the SGs with one generator per primary loop. A simulated feedwater line is used for each loop to maintain proper secondary water level. The j
steam produced in each generator is measured and exhausted to the atmosphere through a common -
l diffuser and stack.
l t
The test control logic simulates the response of the AP600 by providing an S signal at a fixed time following a break.
The passive safety injection systems consist of two CMTs, two accumulators, one IRWST, and one passive residual heat removal heat exchanger (PRHR HX). De test facility simulates the AP600 IRWST with a cylindrical tank with scaled water volume and height. The IRWST is located above the reactor core; two injection lines connect to the two DVI lines. Each IRWST injection line also connects to the samp tank with interconnecting piping and isolation valves. De PRHR HX is located inside the IRWST, using IRWST water as the heat sink. The inlet of the PRHR HX is connected to O
the pressurizer side hot leg via a tee at the ADS-4 line, and the outlet is connected to the SG channel O
head at the cold-leg side. Since the inlet is hot and the outlet is cold, water is circulated through this a:um2a.non:ib-11:097 2.1-1 REVISION: 2
system by natural convection. The water volume and elevation of each CMT are scaled and modeled.
They are elevated above the reactor vessel and the DVI lines. A line connecting the top of each CMT to its cold leg provides pressure balance between the RCS and the CMT. Herefore, the CMT injects cooling water by its own elevation head. He accumulators are also modeled with scaled volume and height. However, they are pressurized with nitrogen and, therefore, inject when RCS pressure is below the preselected scaled accumulator pressure.
The AP600 uses four stages of valves to depressurize the RCS. He first three stages of the ADS are provided through connections to the pressurizer. Rese three stages are arranged in parallel, with each stage containing two lines with each line containing an isolation and control valve. The fourth stage of the ADS contains four separate lines.
The OSU test facility uses only one set of valves to model the ADS 1-3 stages for AP600. This is I done using removable flow nozzles (Reference 26) to match the scaled flow characteristics of either one or two lines of valves. The lines of ADS 1-3 split into parallel lines from one connection off the pressurizer in the AP600.
The discharge lines from the ADS 1-3 valves are joined into one line connected to the ADS 1-3 separator. His two-phase flow is : pa"ated using a swirl-vane separator. The liquid and vapor flows are measured to obtain the ADS total flow for mass and energy bahnee analysis. The separated flow streams are then recombined and discharged into the IRWST through a sparger.
He OSU test facility uses one ADS-4 line connected to the top of each hot leg. Each line contains a pneumatically operated, full port ball valve acting as the ADS-4 isolation valve and a flow nozzle I simulating the flow area in the AP600 (Reference 26). Two sets of flow nozzles are used in the test:
one simulates 100-percent flow area and the other simulates 50-percent flow a ea. In the test facility, ADS-4 discharge flows to the ADS-4 separators, where the steam and liquid flows are separated and measured. Steam flow is measured and exhausted to the atmosphere.
He lower containment sump in the AP600 consists of two volumes: normally flooded and normally nonflooded. The normally flooded volume consists of those compartments which collect liquid break flow and ADS flow. The normally flooded volume is modeled in OSU with a cylindrical tank.
identified as the primary sump tank, ne normally nonflooded volume includes those compartments which do not collect any liquid flow. The normally nonflooded volume is modeled in OSU with a cylindrical tank, identified as the secondary sump tank. nese two tanks are connected with a line at a level simulating the curb level in the AP600.
In the AP600, the RNS is used to provide nonsafety-related cooling water injection to the reactor core.
The RNS pump takes suction from the IRWST and discharges it into the DVI lines. The test facility RNS pump takes suction from the IRWST at the scaled location and elevation and discharges equal flow to both DVI lines at scaled locations.
oA2344w-2a non:Ib-ll1097 2,}.2 REVISION: 2
j l
(]
4.6 Automatic Depressurization System 13 Separator V
The ADS provides a means of depressurizing the RCS in a controlled, staged manner through the use of four pairs of valves, with each valve pair sequenced to open at different primary system pressures.
The first three pairs of valves, called ADS stages 1,2, and 3 (ADS 1-3), are located in parallel piping I
paths running from the top of the pressurizer to the IRWST. The liquid drain line loop seal of the I
reparator is filled with water prior to each test to prevent steam from blowing through the line. This portion of the ADS operates independently of the fourth and final stage of the ADS.
The ADS 1-3 flow path was simulated in the OSU test facility by three valves (one valve each having a scaled stage 1, stage 2, and stage 3 flow area), a steam / water separator tank, a vortex (vapor) flow meter, a magnetic (liquid) flow meter, and associated piping. During testing, flow through the ADS 1-3 flow path was measured by separating the vapor and liquid components of the flow, measuring the flow rate of the component flows, recombining the flows, then directing the total metered flow to a sparger located in the IRWST.
Flow through ADS 1-3 may be calculated as:
- ADS 14,f+
ADS 14, g
- 4'6-1 p
dt where the subscripts:
f Liquid phase of water
=
Steam g
=
ADS 1-3
= Stages I through 3 of the ADS SEP Steam water separator tank for ADS 1-3
=
Energy is transported out of the primary system by both the vapor and liquid flows. Also, the stored energy of the ADS 1-3 separator inventory may change due to changes in the amount of liquid and vapor in the separator, changes in temperature and pressure of the liquid or vapor inventory in the separator, or a change in the temperature of the separator tank metal mass. Accounting for these terms, the energy rate equation for ADS 1-3 flow may be expressed as:
d(MADS 14 SEP I ~ (I ) )
QADS 14
- OADS 14, f + OADS 14, g
- CP ref dt I
I
+QADS I4, METAL
- OADS 14, AMB 4'b'2 p.,
O
)
oMhil.non.lb-Illl97 4,6 1 REVtSION: 2
where the subscripts:
METAL Metal mass of the ADS 1-3 separator tank and associated piping
=
AMB Energy loss to ambient environment
=
4.6.1 Automatic Depressurization System 13 Separator Liquid Inventory The ADS 1-3 separator is, in its simplest form, a tank. Liquid inventory in the ADS 1-3 separator is monitored by a level transducer. The functional steps and associated system of equations for operatmg on the output from the ADS 1-3 separator level transducers to calculate liquid mass in the ADS 1-3 steam / water separator tank follows:
Step 1:
Compensate the readings from the ADS 1-3 separator level transducer listed in Table 4.6-1 to account for temperature differences between fluid in the separator tank and fluid in the reference leg of the instrument line. The local pressure and fluid temperature instruments to be used to accomplish the compensation are also identified in Table 4.6-1.
Step 2:
The local pressures and temperatures measured using the instruments identified in Table 4.6-1 are used as inputs to the ASME steam tables to calculate the density of the liquid and vapor in the ADS 1-3 separator:
PADS 14, f " Pf (PT-605, TF-616)
PADS 14. g " Ps (PT-605, TF-617) 4.6-3 where:
PT-605 Data channel ID for local pressure measurement in the ADS 1-3 separator tank
=
TF-616 Data channel ID for local liquid temperature measurement, 'F
=
TF-617 Data channel ID for local vapor temperature measurement, *F
=
As identified previously, both liquid and vapor flow meters have an associated local fluid temperature used to evaluate the thermodynamic properties of the liquid and vapor phases in the ADS 1-3 separator.
Step 3:
Using the compensated liquid level and the ADS 1-3 separator volume as a function of height (see Table 4.6-2), determine the volume of liquid in the ADS 1-3 separator as:
VADS 14 SEP. f " Y(I) ADS 14 SEP x LDP-610COMP oA2344w.11.non:1b-ll1097 4,6 2 REVli10N: 1
r where:
s V(1)
Volume of the ADS 1-3 separator as a function of elevation,in.3/in,
=
LDP-610 COMP = Compensated fluid levels data from level transducer LDP-610, in.
Step 4:
Liquid mass inventory in the ADS 1-3 separator is now calculated as:
MADS l-3 SEP, f " PADS 1-3, f x VADS 1-3 SEP f 4.6-5 Vapor mass in the separator is then calculated as:
MADS 1-3 SEP, g " EADS l-3, g * (YADS 1-3 SEP TOT - VADS 1-3 SEP, f) 4.6-6 where the subscript:
Total volume of ADS 1-3 separator, ft.3 TOT
=
i Step 5:
The rate of change in mass inventory of the ADS 1-3 separator may be approximated by differencing two consecutive calculated values of liquid and vapor mass:
dM AM ADS 1-3 SEP, MADS 1-3 SEP, f, i - MADS 1-3 SEP, f, i-1 ADS 1-3 SEP dt At tj - t.i i
. MADS 1-3 SEP, g, i - MADS 1-3 SEP, g, i-1 4.6-7 ti - t _;
i l
where the subscript:
i Index of data and time arrays
=
l l-O c:u344w-11.non:Ib-11:097 4.6-3 REVISION: 2
l 4.6.2 Steam Flow Rates The density of steam in the ADS 1-3 exhaust line, evaluated using the pressure and temperature inputs from the data channels identified in T ble 4.6-3, is used to calculate the steam mass flow rate through j
a the ADS 1-3 valves as:
bADS 14. g " C x PADS 14, g xW 4.6-8 l
FVM@l where:
Volumetric flow rate of steam, ft.3/ min W=
=
C:
Conversion constant, minutes to seconds
=
and the subscript:
FVM-601 Instrument channel ID for steam vapor flow meter in line between ADS 1-3
=
separator and sparger 4.6.3 Liquid Flow Rates Liquid from ADS 1-3 is ducted from the separator through a magnetic flow meter into a header where it is mixed with steam flow before passing to the sparger in the IRWST. Mass flow from the ADS 13 separator to the sparger is calculated as:
$1ADS 14. f
- C
- PADS 1-3. f XW 4.6-9 2
FMMel where:
W Volumetric flow rate of liquid, gpm
=
2 Conversion constant, gpm to ft.3/sec.
C
=
and the subscript:
FMM-601 Instmment channel ID for liquid flow meter in line between ADS 1-3 separator
=
and sparger The density of the liquid passing through the flow meter is evaluated using data from the pressure and temperature instruments identified in Table 4.6-3.
oA2344w-II.non:lb-llI097 4.6-4 REVISION: 1
J l
l
)-
4.6.4 Total Flow Rate l
r The total liquid and vapor flows through ADS 1-3 are then calculated as:
I QADS 14, f. QFMM401, #ADS 14SEP,1f 4.6-10 l
I-g i
I QADS 14 g. QFVM401 ADS 14SEP, g 4.6-11 At l
The total flow through the ADS is zero when the three ADS 1-3 valves are closed. Thus, the above i
total liquid and vapor flows are used only when at least one valve is open.
The total flow through the ADS (for t 2 valve open time) is then calculated as:
ADS 14. TOTAL
- ADS 14 f
- bADS 14 g For t < valve open time, $g93 3 4 = 0.
The total mass flow rate is integrating step-wise over time to calculate the total mass inventory passed by the ADS 1-3:
MADS 14 "[ (SADS 14. TOTAL x At) 4.6-13
.The flow quality of the ADS 1-3 flow is also calculated as:
X=
ADS 14, g 4.6-14 bADS 14. TOTAL o:\\2344w-11.non:Ib.111197 4.6-5 REVISION: 2
r 4.6.5 Energy Balance Energy flow through ADS 1-3 consists of the following:
Rate of change in stored energy of the ADS 1-3 separator liquid and steam inventory Energy transport rate from the ADS 1-3 separator by exiting steam flow Energy transport rate from the ADS 1-3 separator by exiting liquid flow Rate of change in stored energy of the ADS 1-3 metal components Rate of energy loss from the ADS 1-3 components to the environment The expressions for evaluating these five energy transfer or transport terms are developed in the following sections.
The combination of these five terms results in energy rate equation 4.6-2. Consistent with the influence of valve position on total flow equation 4.6-12, the energy flow rate through the ADS is zero when the three ADS 1-3 valves are closed. Thus:
QADS 1-3 = 0, for the t < valve open time.
4.6.5.1 Rate of Change in Stored Energy of the Automatic Depressurization System 13 Separator Liquid and Steam Inventory The rate of change of energy in the liquid and steam in the ADS 1-3 separator may be expressed as:
d(MADS 1-3 SEP d ~Iref))
I QADS 1-3 " CP dt d(MADS 1-3 SEP, f)
- P. f d(I)f d
-T )
M
- C,f P
f ref dt ADS 1-3 SEP, f dt 4.6-15 d(MADS 1-3 SEP. g) d(I )
+ cP. s d -Tref)
g s
dt dt where:
1 T,,r, Tp,gp = 32*F Expressing the previous equation as a difference and solving for consecutive data, the rate of change of energy of the fluid in the ADS 1-3 separator is calculated:
o:\\2344w II.non:Ib-121297 4.6 6 REVISION: 2
}
i l
(
\\
()
MADS 14 SEP d ~I D ref g
g p bt A(MADS 14 SEP f) d(I )f
" C, f (I ~ Iref)
- C,f M P
f P
ADS 14 SEP, i gg dt 4.6-16 A(MADS I4 SEP. g)
- P. g O(I )
+ C, g (Ig P
~I )
M g
ref At ADS 14 SEP, g At The specific heats of the liquid and vapor phases, cp, f and c,, re;pectively, are evaluated using the p
output of the data channels identified in Table 4.6-2.
1 4.6.5.2 Energy Transport Rate from the Automatic Depressurization System 13 Separator by Exiting Steam Flow The enthalpy of the steam exhaust from the ADS 1-3 separator is determined using the output from the pressure and temperature sensors associated with FVM-601 (Table 4.6-2):
OV hFVM 601, g = h, (PT-605, TF 617) 4.6-17 FVM-601 is the data channel ID for the ADS 1-3 vapor flow meter, and IT-605 and TF-617 denote i
the vapor pressure and temperatures, respectively, associated with that flow meter.
(
i The rate of energy transport of the ADS 1-3 flow due to the steam component, then, is expressed as:
9 DS 14.g " BADS 14,g FV M @ l.
xh 4.6-18 A
4.6.5.3 Energy Transport Rate from the Automatic Depressurization System 1-3 Separator by Exiting Liquid Flow The enthalpy of ADS liquid flow is determined using the output from the pressure and temperature sensors associated with FMM-601 (Table 4.6-2):
h yyer, f = hr (PT-605. TF-616) 4.6-19 y
)
o \\2344w ll.non:lb-121297 4.6 7 REVISION: 2
FMM-601 is the data channel ID of the ADS 1-3 liquid flow meter, and PT-605 and TF-616 denote the data channel ids for the liquid pressure and temperatures, respectively, associated with that flow meter.
The rate of energy transport due to ADS 1-3 liquid flow, then, is expressed as:
Q DS 1-3. f " b xh 4.6-20 A
ADS 1-0. f pygg,oi, f 4.6.5.4 Rate of Change in Stored Energy of the Automatic Depressurizatico System 13 Metal Components The ADS is heat-traced downstream from the ADS valves. Therefore, the change in stored energy of the ADS piping due to energy loss from the fluid is negligible.
4.6.5.5 Rate of Energy Loss from the Automatic Depressurization System 13 Components to the Environment Piping between the ADS 1-3 valves and the separator, and between the separator and the IRWST were provided with heat-tracing. This had the effect of off-setting any energy loss to the ambient environment. Thus, for the purpose of evaluating the rate of energy loss from the fluid to the ambient environment (through piping):
Qgyg n 0.0 4.6-21 O
om44w-limwisi t io97 4.6-8 REVISION: 1
F 4.7 Automatic Depressurization System 4 Separators l
The ADS provides a means of depressurizing the RCS in a controlled, staged manner through the use of four pairs of valves, with each valve pair sequenced so that they will open at different primary system pressures. Mass, flow, and energy calculations associated with the first three pairs of valves, called ADS 13, are described in Section 4.6. A redundant fourth pair of valves, are located on each of the two hot legs and exhaust directly to the containment atmosphere. The ADS-4 valves are used to complete depressurization of the RCS to near containment pressure.
The ADS-4 flow paths were simulated in the OSU test facility by a valve, a steam / water separator tank, a vortex (vapor) flow meter, a magnetic (liquid) flow meter, and associated piping from each of
)
the two hot legs. During testing, flow through the two ADS-4 flow paths was measured by separating the vapor and liquid components of the flow, measuring the flow rate of the component flows, I
recombining the flows, then directing the total metered flow to a simulation of the containment sump.
I The liquid drain line loop seal of the separator is filled with water prior to each test to prevent steam 1 from blowing through the line.
l Flow through each of the ADS-4 flow paths may be calculated as:
MADS 44 ADS 44. f
- BADS 4 4, g +
d i
where the subscripts:
ADS-4 Fourth-stage ADS
=
X
=
Hot leg to which the flow path is connected where:
X 1, for HL-1
=
X 2, for HL-2
=
f Liquid component of ADS-4 flow
=
SEP Steam water separator tank for ADS-4
=
Vapor (steam) component of ADS-4 flow g
=
The total ADS-4 few rate is then calculated as:
- ADS 4-2 Erergy is transponed out of the primary system by the ADS-4 vapor end liquid flows. Also, the I
stored energy of the ADS-4 separatorliquid inventory may change due to a change in the amount of liquid in the separator, a change in temperature of the liquid inventory in the separator, or a change in i
j the temperature of the separator tank metal mass. Accounting for these terms, the energy equation for ADS-4 flow may be expressed as:
otec-4\\2344*-47mn:Ib-012198 4,71 REVISION: 2
O dIMADS 4-X SEP (I 'Iref )
f O DS 4-X " OADS 4-X, f + Q DS 4-X, g + Cp,f A
A dr 4.7-3
+OADS 4-X. METAL
- OADS 4-X. AMB where the subscripts:
METAL Metal mass of the ADS-4 separator tank and associated piping
=
AMB Energy loss to ambient environment
=
The total energy associated with the ADS-4 (accounting for both separators) is calculated by summing I the terms in the previous equation for each of the two separators. The discharged vapor is assumed to I be at saturation temperature and therefore the sensible heat term for the vapor phase is neglected.
4.7.1 Automatic Depressurization System-4 Separator Liquid Inventory The ADS-4 separator is, in its simplest form, a tank. Liquid inventory in the ADS-4 separators is monitored by level transducers. The functional steps and associated system of equations for operating on the output from the ADS-4 separator level transducers to calculate inventory mass in the separator tanks are:
Step 1:
A differential head is calculated to account for the orifice in the liquid line. The inputs to this calculation are: pipe diameter, beta ratio of orifice to pipe restriction
( =1.0 = no orifice), pressure, temperature, and liquid flow. The output differential head is added to the measured LDP value. The readings are compensated (adjusted for orifice effect) from the ADS-4 separator level transducers listed in Table 4.7-1 to account for temperature differences between fluid in the separator tank and fluid in the reference leg of the instrument line.
The local pressure and fluid temperature instruments used to accomplish the compensation are also identified in Table 4.7-1.
Step 2:
The local pressures and temperatures from the instmments identified in Table 4.7-1 are used to calculate the density of the liquid and vapor in the ADS-4 separator:
pf, ADS 4-X Pt (PT-YYY, TF-ZZL) 4.7-4 p, ADS 4-X Pg (PT-YYY, TF-ZZV) g O
otec-4\\2344w-47.non:lb llll97 4.7 2 REVISION: 2
r
)
The entropy of vaporization is calculated as:
}
S, toen = S LOCAL ~ b, LOCR 4 7-19 fg g
f The local flow quality at the location of interest is calculated as:
SSEP ~b.LOCE g
yLOCAL,
S fs 4.7-20 1
i
\\
'Ihe calculation for Xtoeg s performed at four locations: ADS 1-3 separator, ADS 4-1 i
separator, ADS 4-2 separator, and the break separator. The instmments used are identified in Table 4.7-3.
I l
4.7.5 Energy Balance l
\\
l Energy flow through the ADS-4 separator consists of the following:
l 1
l Rate of change in stored eriergy of the ADS-4 separator fluid inventory
=
Energy transport rate from the ADS-4 separator by exiting steam flow Energy transport rate from the ADS-4 separator by exiting liquid flow Rate of change in stored energy of the ADS-4 metal components Rate of energy loss from the ADS-4 components to the environment l
The expressions for evaluating these five energy transfer or transport terms are developed in the following sections.
1 l
1 O
l owc 4us44w 47.non: b-111097 4.77 REVtSION: 1 L
4.7.5.1 Rate of Change in Stored Energy of the Automatic Depressurization System-4 Separator Fluid Inventory The rate of change of energy associated with the fluid (both steam and liquid) inventory of a ADS-4 separator may be expressed as:
I d(M b -I ))
ADS 4-X SEP ref OADS 4 X SEP = cp dt d(MADS 4-X SEP.f )
(T - Tg)
=c f
dt 4.7-21 d(T )
f
- C,f M P
ADS 4-X SEP, f dt d(MADS 4-X SEP. g ) +
d (I )
M g
-TREF)
- C.g g
Expressing the previous equation as a difference:
I d(MADS 4-X SEP d "Iref))
MADS 4-X SEP. f T - Ty
=
dt At
-T )
4-X. *
- C. 8
+ c, g (Ts M
yg P
1 O
0:\\sec42344w.47.non:lb.121297 4,78 REVISION: 2
r 4.8 Break Separator ne purpose of the break separator is to separate break flow into liquid and vapor components and to measure the flow rates of the single-phase flow components. Once measured, vapor flow is exhausted to the ambient environment, and liquid is directed to the primary sump simulation of the test facility.
De break separator consists of,a tank (separator); a vortex (vapor) flow meter, a magnetic (liquid) j flow meter, and associated valves, piping, and instrumentation.
The total break flow may be calculated as:
1 MBREAK "b,BRK + g,BRK +
'8 l f
dt where the subscripts:
BREAK Total break flow
=
I f,BRK Liquid flow from break separator
=
BRK SEP Break separator tank
=
I g,BRK Steam exhaust from break separator
=
To address possible reverse flow from the sump to the break separator, a condition of reverse flow is determined by evaluating the level from LDP-901.
If LDP-9012 75.275 in., then:
~
f f
SEP di i
4.8-2
~ b, ADS 4-2 SEP ~ OVRPLW + b f
SUMP INJ where:
NovRFt.w Liquid mass flow rate through IRWST overflow line measured by FMM-703
=
his new value for A.BRK SEP s then used in Equation 4.9-1 of Section 4.9, to calculate the general i
f mass balance for the sumps.
owc 4u3mw-4s=1b.11to97 4,8 1 REVISION: 2 l
Energy is transported out of the primary system by both the vapor and liquid components of break flow. Additionally, the following occurrences cause the stored energy level of the break separator to vary: a change in the fluid inventory held within the separator; a change in temperature of the fluid inventory held within the separator; a change in the temperature of the metal mass of the separator tank or a loss of energy to ambient. Thus, the energy balance for the break separator, accounting for the change in stored energy of both the liquid inventory of the separator and the separator tank, may be expressed as:
I d(MBRK SEPO OBREAK *O.BRK*O.BRK+CP
+0BRK SEP METAL + OBRK SEP AMB 4'g.3 f
g dt where the subscripts:
METAL Metal mass of break separator tank and associated piping
=
AMB
=
Energy loss to ambient environment 4.8.1 Break Separator Liquid Inventory The break separator is, in its simplest form, a tank. Liquid inventory in the break separator is monitored by a level transducer. For all tests, with the exception of double-ended DVI line breaks, an orifice was in place within the span of the level transducer. Therefore, a correction was made to the level indication to adjust for the pressure drop through the orifice. The functional steps and associated system of equations for operating on the output from the break separator level transducers to calculate i liquid mass in the separator tank follows:
Step 1:
Calculate the pressure drop through the orifice and add this value to the readings from the break separator level transducer listed in Table 4.8-1 to account for the pressure drop through the orifice. The instruments used to measure level, local pressure, and fluid temperature are identified in Table 4.8-1. The flow meters used to calculate the pressure drop are identified in Table 4.8-2.
Step 2:
Compensate the adjusted level indication for the break separator level to account for temperature differences between fluid in the separator tank and fluid in the reference leg of the instrument line. The local pressure and fluid temperatures used to accomplish the compensation are also identified in Table 4.8-1.
O ow4s4 as.nmib-121297 4.g.2 REVISION: 2
7
[
Total break steam flow is calculated as:
J s.BRK
- FVM-905 + FVM906 +
8 g
d 4.8.3 Liquid Flow Rates A single line directs liquid flow from the break separator into the primary sump tank. He mass flow from the break separator to the primary sump tank is calculated as:
$pyy.903 = C x pf x W 4.8-11 i
FMM+05 where:
W
=
Volumetric flow rate of liquid, gpm Conversion constant, gpm to ft.3/sec.
C
=
i and the subscript:
FMM-905 Instrument channel ID for liquid flow meter in line between break separator
=
and primary sump The density of the liquid passing through the flow meters is determined using data frorn the pressure l
and temperature instruments identified in Table 4.8-2.
Total liquid break flow rate passed to the sump simulation is calculated as:
E. BRK SEP f
Qf. BRK. Q
+
4,g.12 FMM405 di l
4.8.4 Total Flow Rate The total break flow rate is then calculated as:
bBRK. TOTAL *
- f. BRK + g.BRK 4.8-13 he flow quality of the break flow is also calculated as:
N X gg =
4.8 14 3
5 BRK. TOTAL l
o.\\sec4\\2344w-48.am:lbil1097 4.g.3 D'ISION: 1 l
l L
4.8.5 Energy Balance The energy flow through the break separator consists of the following:
Rate of change in stored energy of the break separator fluid inventory Energy transport rate from the break separator by steam exhaust flow Energy transpon rate from the break separator by liquid flow to the sump Rate of change in stored energy of the metal of the break separator tank
=
Rate of energy loss to the environment The expressions evaluating these energy transfer or transpon terms are developed in the following subsections:
4.8.5.1 Rate of Change in Stored Energy of the Break Separator Fluid Inventory The rate of change of energy in the break separator fluid may be expressed as:
I d(M (T -T,,f) d(M. BRK SEP) +
d(I )
r f
f I
OBRK SEP " C.f
- I I
"I P. f M. BRK SEP P
f dt dt dt where:
1 I
T.T ref REF Reference temperature,32 F
=
Tf Measured liquid temperature, *F
=
Expressing the previous equation as a difference:
4.8-16 d( M BRK SEP (I ~Iref))
f f
f dUnf)
+ P. f M BRK SEP C.f P.f P
f f
dt At At Expanding the terms on the right-hand side of the equation:
- f. BRK SEP cp, f (T - T,,f)
= cp, f (T - T )
f f
rer At ti - 1; i l
4.8 17 AT T.i - T.1-1 f
f f
M. BRK SEP P.f N. BRK SEP. i C.f P
f f
at tj-Q O
ch4Qh48.non:Ib-121297 4,8 6 REVISION: 2
where:
i Index of data and time array
=
The specific heat of the liquid, cp, f, is evaluated using the pressures and temperatures identified in Table 4.81.
The rate of change of energy in the steam in the break separator may be expressed as:
I d( M. BRK SEP (I ~ I ))
d( M, BRK SEP )
d(I )
g g
ref g
C.g P.gN U )I g
P
+ P. g M, BRK SEP dt g
ref g
dt g
dt 4.8-18 Expressing the previous equation as a difference:
I d(M, BRK SEP (Ig ' I ))
S ref S
g C,g "C.g U d )T gg M, BRK SEP P
P g
ref I
At P, g g
At j
i 4,8 19 Expanding the two terms on the right-hand side of the equation:
-T ) T 8, BRK SEP M, BRK SEP, i g, BRK SEP, i-l cp,, (Tg g
ref g
, ep, g (7, 7,,f) g ep,, M, BRK SEP M, BRK SEP, i "C,g P
g g
g ti
_g The specific heat of the steam, cp, f,is evaluated using the pressures and temperatures measured by the instruments identified in Table 4.8-1.
4.8.5.2 Energy Transport Rate from the Break Separator by Steam Exhaust Flow The enthalpy of the steam exhaust from the break separator is calculated using the output from the pressure and temperature sensors associated with FVM-905 and FVM-906 (Table 4.8-2):
h, pyy_m = h (PT-YYY, TF-ZZZ) 4.8-21 g
g O
owc 4u344w-48mn:lb 121297 4.8 7 REVISION: 2
FVM-XXX denotes a specific flow meter, and PT-YYY and TF-ZZZ denote the vapor pressure and temperatures associated with that flow meter. He rate of energy transpon from the break separator due to exhaust steam, then, is expressed as:
Q VM-XXX " FVM-XXX X h. FVM-XXX
' ~
F g
where:
Qpyy xxx Rate of energy transport due to steam flow through flow meter FVM-XXX
=
Re total energy transport rate, then, is calculated as:
)
l Q
BRK, g " O VM-905 + O VMeo6
.8-3 F
F f
4.8.5.3 Energy Transport Rate from the Break Separator by Liquid Flow to the Sump He enthalpy of liquid flow into the break separator is calculated using the output from the pressure and temperature sensors associated with FMM-905 (Table 4.8-2):
hr. FMM+os = h (PT-905, TF-912) 4.8-24 r
FMM-905 denotes the flow meter to which the enthalpy is applicable, and PT-905 and TF-912 denote the liquid pressure and temperatures associated with that flow meter. He rate of energy transport from the break separator due to liquid overflow into the sump, then, is expressed as:
Q3gg,f = $ gg, f x h, FMM 905 4.8-25 3
f 4.8.5.4 Rate of Change in Stored Energy of the Metal of the Break Separator Tank The break separator tank and associated piping is heat-traced. Herefore the change of stored energy of the metal components due to energy loss from the fluid is negligible.
4.8.5.5 Rate of Energy Loss to the Environment The rate of energy loss from the break separator and its associated piping to the environment is zero because the separator and the steam exhaust lines are heat-traced. Thus, the separator and steam lines are treated as an adiabatic boundary. Heat loss to the environment from the liquid drain line running from the break separator to the sump is neglected since the run of pipe is small.
owc 4ch-48.nonib.11:097 4.8-8 REVISION: 1 J
r 4.9 Sumps f
In the AP600, the sump collects all liquid released from the pnmary system and serves as the source of post-accident LTC water inventory. In the OSU test facility, the sump is modeled by two tanks: a primary sump tank and a smaller secondary sump tank. Dese tanks have associated with them piping, vapor and liquid flow meters, and other pressure and temperature measurement instrumentation.
Accounting for all possible flow paths associated with the sump, the general mass balance on that component may be expressed as:
- STM XHST -bSUMP INJ where the subscripts are defined as:
=
Both primary and secondary sump tanks BRK SEP Liquid flow from break separator through FMM-905
=
Liquid flow from the ADS 4-1 steam / water separator through FMM-603 ADS 4-2 SEP =
Liquid flow from the ADS 4-2 steam / water separator through FMM-602 IRWST Liquid overflow from IRWST into sump through FMM-703
=
STM XHST Steam exhaust from sump through FVM 903
=
SUMP INJ Liquid flow from sump to DVIline through FMM-901 and FMM-902
=
Energy is transported into and out of the sump by both the vapor and liquid components of flow associated with the sump. In addition, the stored energy associated with the sump may change due to a change in the liquid inventory held within the sump, a change in temperature of the liquid inventory held within the sump, or a change in the temperature of the metal mass of the sump tanks, or heat loss to the ambient. Dus, the energy balance for the sump, accounting for the change in stored energy of both the liquid inventory of the sump and the sump tank (s) may be expressed as:
I 8"
cp
=QBRK SEP + OADS 4-1 SEP + OADS 4-2 SEP + kIRWST.9-2 4
- OSTM XHST
- OSUMP INJ
- OSUMP METAL
- OSUMP AMB where the subscripts are defined as:
METAL Metal mass of sump tanks and associated piping
=
AMB
=.
Energy loss to ambient environment oMec4\\2344w 49.non:Ib-ll1097 4.9-1 REVISION: 2
4.9.1 Sump Liquid Inventory For convenience, liquid mass in the primary and secondary sump tanks are calculated separately, then summed to yield the total liquid mass in the sump. Liquid inventory in the sump tanks are measured by a level transducer. Load cells installed on the two sump tanks are not used for inventory calculations.
After reviewing the primary-and secondary-sump fluid temperatures for all tests (using thermocouples in Table 4.9-1, and including TF-907 which is located at the top of the primary-sump tank), it was determined that the sump temperatures were always less than the saturation temperature evaluated at the total pressure. It was concluded that 1) the water region clearly remained subcooled in the sump tanks, and 2)it was not possible for a steam region to be supported above the water level in the sump tanks (as occurs in some internal system components, such as the CMTs). The region almve the water level is a cool mixture of air and some steam, and an accurate calculation of its composition is not possible since the partial pressure of the steam in the mixture is unknown.
Therefore, steam inventory is not modeled in the sump tanks and the region above the water level is not included in sump inventory.
4.9.1.1 Use of Level Measurement for Mass Calculation O
The functional steps and associated system of equations for operating on the output of the sump level transducers to calculate liquid mass in the sump tanks follows:
Step 1:
Compensate all readings from the primary and secondary sump level transducers to account for temperature differences between fluid in the tanks and fluid in the reference legs of the instrument lines. The two channels of level data to be compensated are identified in Table 4.9-1. 'Ihe instruments used to measure local pressure and fluid temperatures to be used to accomplish the compensation are also identified in Table 4.9-1.
Step 2:
The local pressures and temperatures as recorded from the instrument channels identified in Table 4.9-1 are used to calculate the density of the liquid in the sump tanks:
Pr, j = Pr, j(P, T) 4.9-3 where the subscripts are def' ed as:
m f
Liquid phase of water
=
j Either sump tank
=
o*du3d4w-49 non:1b-111097 4.9-2 REVISION: 1
l De total steam mass flow rate exhausted by the primary sump and the break separator is calculated as:
bSTM XHST FVM-903 + b 4*9~9 FVM406 De total steam mass flow is calculated by integrating the mass flow rate step-wise over time:
MSTM XHST *1 (bSTM XHST x At) 4.9-10 4.9.3 Sump Injection Two lines provide for liquid to flow from the primary sump into the DVI lines and the reactor pressure vessel simulation. Mass flow from the sump through either of the two sump injection lines is calculated as:
M
= C x pFMM-XXX.x Wpyy_xxx 4.9-11 FMM-XXX 2
where:
W
=
Volumetric flow rate of liquid, gpm C
Conversion constant gpm to ft.3/sec.
=
and the subscript:
FMM-XXX Instrument channel ID where:
=
XXX
= 901 (piping run to DVI-1)
XXX 902 (piping run to DVI-2)
=
ne density of the liquid passing through the flow meters is determined using data from the pressure and temperature instruments identified in Table 4.9-2 as input to the ASME steam tables. He total mass flow rate to the DVI lines from the sump is calculated as:
MSUMP INJ FMM 901 + b 4.9-12 FMM402 he total mass injected by each injection line into the DVI line is calculated by integrating the product of the measured mass flow rates and the time interval over which the measurement is taken:
Mygg_xxx = {( brMM-XXX, i x At; )
4.9-13 obsc4\\2344w-49.norrib.ll1097 4,9-5 REVISION: 1
The tctal liquid mass injected from the sump is calculated as:
i l
bi
" biFMM401 + bIFMM 402 49-I4
{
SUMP INJ 4.9.4 Total Flow Rate Out of the Sump Total mass flow rate out of the sump is then calculated as:
4.9-15 SUMP
- SUMP INJ +
STM XHST Total mass flow out of the sump is calculated by integrating the mass flow rate step-wise over time:
biSUMP. TOTAL
- 1 SUMP x At 4.9-16 4.9.5 Energy Balance Energy into the sump from the break separator, the two ADS-4 separators, and the IRWST overflow lines are calculated in and obtained from their respective modules. Thus, the calculation of an energy balance on the sump requires that the following parameters be evaluated:
Rate of change in stored energy of the sump liquid inventory Energy removal rate from the sump by steam exhaust flow Energy removal rate from the sump by injection flow supplied to the DVI lines Rate of change in stored energy of the metal of the sump tanks Rate of energy loss to the environment The expressions for evaluating these five energy transfer or transpolt terms are developed in the following sections.
4.9.5.1 Rate of Change in Stored Energy of the Sump Liquid Inventory The rate of change of energy in the liquid in the sump may be expressed as:
d(M. SUMP (T -T ))
E ~Iref) d( M. SUMP ) + C, f f
d( T )
r f ref g
M,Scyp 4 S,17 C,f
- C,f P
P f
P g
dt dt dt Expressing the previous equation as a difference:
d( M, semp Ur-T,,g))
l f
g 4'9,} g E -Iref)
C,f "C,f N. SUMP
- C,f P
P f
P f
dt at At obec4\\2344w-49.non:lb 121297 4.9 6 REVISION: 2
i 1
4.13 Steam Generator Primary Side The AP600 includes two SGs, which under normal operating conditions provide the heat sink for reactor core heat removal and steam source to power the turbine. The OSU test facility incorporates two simulated SGs. For the purposes of developing mass and energy balance equations to represent the thermal hydraulic performance of the RCS primary side, the primary side of the simulated SGs is considered as three interconnected sections:
Inlet plenum Tubes (both uphill and downhill sides)
Outlet plenum
'Ihe mass and energy balance equations are developed by considering the equations for each of these sections.
4.13.1 Inlet Plenum 1
A general mass balance equation for the inlet plenums tnay be written as:
IP Mg x - M -TUBE
=
IP l
where the subscripts:
IIL X Hot leg
=
X = 1 (SG-1)
X = 2 (SG-2)
IP SG inlet plenum
=
IP-TUBE Interface between inlet plenum and tube bundle of SG
=
Similarly, the energy equation for the SG inlet plenum is written as:
I d(M (T-T ))so x g
OSG = cp h -TUBE MIX EX E MIX ~ IP-TUBE SO X lP di
~ OIP SG X METAL ~ OIP SG X AMB
(
\\
owus44-413.non:it 111m7 4.13-1 REVISION: 2
O where the subscripts:
=
X = 1 (SG-1)
X = 2 (SG-2)
AMB Ambient conditions associated with inlet plenum
=
METAL Metal mass of inlet plenum
=
MIX Fluid mixture conditions
=
4.13.1.1 Mass Balance ne mass stored in the steam generator inlet plenums is calculated from level measurements in the plenums. The level measurements must be compensated for temperature differences between the fluid in the plenums and that in the sense lines of the instruments. The data channel ids of the level, pressure, and temperature instmments 10 be used in calculating fluid mass in the inlet plenums are listed in Table 4.13-1. The following approach was used in performing the temperature compensation of the level instrument output and calculating the fluid mass in the SG inlet plenums.
Step 1:
First, compensate the readings of the inlet plenum level transducers to account for temperature differences between fluid in the plenum and fluid in the reference legs of the instrument lines. As noted above, the channel ids of the two level transducers to be compensated, one for each SG, are identified in Table 4.13-1. The instruments used to measure local pressure and fluid temperatures to be used to accomplish the compensation are also identified in Table 4.13-1.
The LDPs, which provide the SG plenum levels, provide incorrect readings when the pumps are running.
1)
The plenum level is to be based on the density-corrected LDP, no pump-flow cornetions to be included in the level calculations.
2) he plenum liquid volume and mass is to be based on two attematives:
a)
Plenums assumed to be liquid solid during pump flow.
b) Liquid volume and mass to be based on density-corrected LDP after pump flow stops.
3)
The time period of pump flow to be defined as:
a)
Starting at the beginning of the test.
eAsec42344. 413.non:Ib inio97 4.13-2 REVtSION: 1
Step 6:
De rate of change in mass inventory may be approximated by differencing two consecutive calculated values of liquid mass:
dM AM (M IBES i ~ MTUBES, i-1)
TUBES TUBES Tt 4,}3,37 dt At t; - tg where the subscript:
i Index of the data and time arrays
=
he total mass flow rate from the inlet plenum into the tubes of a given SG is calculated from Equation 4.13-1. With the change in mass storage of the tubes of a given SG calculated from Equation 4.13-26, the terms of the general mass balance on the SG tube bundle, Equation 4.13-18, may be rearranged to solve for the mass flow from the tubes into the outlet plenum of interest.
4.13.2.2 Energy Balance Equation 4.13-15 defines the general form of an energy balance for the SG tubes. The tubes may contain both vapor and liquid at the same time. The energy balance must account for both phases of the working fluid. Thus, the left-hand side of Equation 4.13-15 may be expanded as:
I d(M (T-T,,r)) TUBES P. f Fr. TUBES - T ) d (M. TUBES)
- e. r Mr. TUBES d(I, TUBES) f f
c p
rer dt g
dt dt d
+ c. 5 Es. TUBES - T,,f) (M. TUBES) d(T. TUBES) g g
dt P, g
- g. WBES di P
4.13-28 where:
1 T,,f, TREF Reference temperature; 32*F
=
i l
i l
owc 4\\2344w 413.non: b.12:297 4.13 9 REVISION: 2 i
Equation 4.13-28 may be further expanded to specifically address the liquid volume on the hot leg
(" uphill" side) and cold leg (" downhill" side) of the SG tubes, and written in its difference form to operate on the data:
I d(M (T-T )) TUBES OI. TUBES U, TUBES ref f
f M. TUBES P. f f~
ref)
Cp dt P. I f
At At OI, TUBES E, TUBES g
g
+C.g M. TUBES E' 8 (T T'*I}
+
p g
At 8
At 4.13-29 The data channel ids of the instmments to be used to evaluate the thermal transport properties are listed in Table 4.13-1. Note that the value for T will be taken as the minimum of the average g
measured temperature and saturation temperature. The value for T, will be taken as the maximum of the average measured temperature and saturation temperature.
The equation for the rate of change in internal energy of the metal of a tube bundle is written as:
dTTUBE SG X METAL 4.13-30 OTUBE SG X METAL = MTUBE SG X METAL
- C METAL
- p di Representing the rate of change in energy stored by the metal of the tubes of each SG in difference form:
ATTUDE SG X METAL 4.13-31 OTUBE, SG X METAL =MTUBE, SG X METAL
- C, METAL
- p At The data channel ids to be used for the metal energy storage calculations are listed in Table 4.13-1.
The quality in the hot-leg side of the tubes is calculated as:
M TUBE OP S
4.13-32 X=
M, TUBE OP + M. TUBE OP E
f For use in the overall system energy balance calculations, the fluid stored energy in the SG tubes is given by the following:
UTUBE SG X = U, TUBE SG X + U. TUBE SG X 4.13-33 r
g o:\\sec4\\2344w-413.non Ib-121297 4.13-10 REVISION: 2
I' 4.15 Pressurizer The pressurizer is a tank like structure through which mass flows from the primary system to the first
{
three stages of the ADS valves under accident mitigation. At steady-state, the pressurizer has an initial liquid and steam volume. During a transient, the initial pressurizer and primary-side inventory are vented by the ADS through a pipe running from the top of the pressurizer to the IRWST.
l l
Subscript notation in this section is as follows:
l f
Liquid phase of water
=
g
= Steam PRZR Pressurizer
=
SL-PRZR = Connection between the pressurizer and the surge line ADS 13 Connection between the pressurizer and ADS 1-3
=
AMB Ambient environment
=
METAL Pressurizer metal l
=
(
l 4.15.1 Inputs and Assumptions
/]
A general mass balance for the pressurizer may be expressed as:
U 1
dM PRZR
, gSMRZR _g 4,}$.]
l dt ADS 13 l
Similarly, a general energy balance on the pressurizer may be expressed as:
l l
l d (M (Tpgza -T,f) )
4.15-2 cp SWRZR - Q DSl3 SEA!. SAMB dt A
1 1
The pertinent data channels associated with the pressurizer are shown in Table 4.15-1. In addition, the following values are provided by the ADS 1-3 module:
Energy rate from pressurizer by way of ADS i 3 (Btu /sec.)
QADS13
=
b 8
^
'}
ADSt3 The geometry of the pressurizer is indirectly defined by Table 4.15-2, showing volume as a functiot of fluid height. In the equations below, this function is expressed as V(level).
f l
owmis.non:iw 11097 4.15-1 REVISION: 2 l
lo
I Table 4.15-3 provides the data channels and metal structure informa: ion needed for the pressurizer metal segments.
In the calculation of the heat loss to the metal, the heat capacity of a segment is defined as a function of the metal temperature (heat cap (T g))). Table 4.15-4 provides this function.
w 4.15.2 Mass Balance Calculation The fluid level reading from LDP-601 needs to be compensated to account for temperature differences between fluid in the pressurizer and fluid in the reference leg of the instmment line. Since two temperature sensors are defined for the pressurizer, the temperatures are averaged for the fluid temperature:
T = T1 + n 4.15-3 g
2.0 The density of the fluid in the pressurizer is obtained from an ASME steam table specific volume call:
I' 4.15-4 Pr = VCL(Pr, T )
g The mass of the fluid is then calculated:
Mr
- Pf x V(level ) x C 4.15-5 f
where:
Conversion constant, in.3 to ft.3 C =
The fluid specific heat, c,g, is defined from a steam table call:
p c,r= CPL ( P, T )
4.15-6 p
g g
O c:vec4c344*-415.non:tb-li no97 4.15-2 REVISION: 1
A 4.15.3 Energy Balance i
The energy flow associated with the pressurizer consists of the following components:
1 Rate of change in stored energy of the pressurizer inven:ory Rate of energy transport to the pressurizer from the surge line Rate of energy transport from the pressurizer by way of ADS 1-3 Rate of change in stored energy of the pressurizer metal Rate of energy loss to the environment e
1 Expressions for each of these components is developed in the following subsections.
l 4.15.3.1 Rate of Change of Energy in the Pressurizer Fluid Inventory
)
I The stored energy of the pressurizer inventory must account for both the liquid and vapor phases of the inventory:
d (M (T-T ))pg7, d W Ur-T gD yf f
g PRZR S
8 MI PRZR
~
cp P' I dt dt P, g
. dt The equation for energy change associated with the liquid phase of the pressurizer inventory may be expressed:
I d( Mr (T -Tur))PRZR f
C,f P
di 4.15-20 d( M.PRZR }
d( I )
l f
f E -Tag)
P. f M,pgzg C,f P
f f
dl dt where:
1 Tyg,Tggy Reference temperature,32'F
=
i i
1 l
l c:\\sec4\\2344w-415.non Ib 121297 4.15-5 REVISION: 2
Similarly, the energy change associated with the vapor phase of the pressurizer inventory may be expressed as:
I d( Mg (T -Tref) )
g C,g P
dt 4.15-21 d( M, )
d( T )
M g
C,g d -Tref)
Pg s
dt P
g dt Expressing the preceding two equations as differences:
I d( M, pgzg Ur -Trer))
f C,i P
dt 4.15-22 E
AT f'PRZR f
E ~Iref)
+ P, f M,pgzg C,f P
f f
At and:
I d( M, pgza U -T,,f) )
l g
s C,g P
dt 4.15-23 cp, g U -T,,f)
+ cp,, M, pgzg g
g g
t Expanding the two terms on the right-hand side of the preceding two equations, the liquid phase expression becomes:
I I
cp, f (T -T,,f)
- ' f (T i~T'}
f t
t - _i 4.15-24 ATr T, ; - T, ; _i f
f M,pazg
=c,f M, pgza, ;
C,f f
p P
f O
os344.srev21:344w.4:5 non:ib.12:297 4.15-6 REVISION: 1
N 4.16 Pressurizer Surge Line The pressurizer surge line is piping that connects the pressurizer to the primary system. During a transient simulation where the ADS is actuated, the surge line becomes part of the relief path from the primary system to the pressurizer, ADS, and containment.
Subscript notation in this section is as follows:
f
= Liquid phase of water g-
= Steam SL
= Surge line HL-SL
= Junction between the hot leg and the surge line SL-PRZR = Connection between the pressurizer and the surge line ADS 13
= ADS 1-3 AMB
= Ambient environment METAL
= Pressurizer surge line metal A general mass balance on the pressurizer surge line may be expressed as:
s 8'
=M
- M -PRZR gt4t SL i
Similarly, a general energy balance on the pressurizer surge line may be written as:
I d (M(TSL-T ))
g C
P Hl.41. ~
~
dt 4.16.1 Inputs and Assumptions De pertinent data channels associated with the pressurizer surge line are shown in Table 4.16-1. In addition, the following values are provided by the pressurizer calculations:
l NISL-PRZR Mass rate of junction between surge line and pressurizer (Ibm /sec.)
=
hSL-PRZR Enthalpy of fluid in the junction between surge line and pressurizer (Bru/lbm)
=
g The geometry of the pressurizer surge line is indirectly defined by Table 4.16-2. showing volume as a function of fluid height. In the equations below, this function is expressed as V(level). The highest f
O volume value in this table is referenced in subsequent equations as V o:Wc4\\2344-416 son:1b-llll97 4,16-1 REVISION: 2
Table 4.16-3 provides the data channels and metal stnicture information needed for the pressurizer surge line metal segments. Only one metal segment was used to model the pressurizer surge line.
In the calculation of the heat loss to the metal, the heat capacity of a segment is defined as a function of the metal temperature (heat _ cap (Tsurf0))). Table 4.16-4 provides this function.
4.16.2 Mass Balance The fluid level reading requires compensation because the sensor is outside the actual component.
Since only one temperature sensor is defined for the pressurizer surge line, no temperature averaging is required for the LDP compensation.
He density of the fluid in the pressurizer surge line is obtained from a standard steam table specific volume call:
1.0 4.16-3 pg = VCL(P, T )
g g
He mass of the water in the pressurizer surge line is then calculated:
Mr = pg x V(level ) x C 4.16-4 f
where:
Conversion constant, in.3 to ft.3 C
=
The specific volume (v ) and the heat capacity (c.8) f the gas are both obtained from a steam table g
P call:
c. g' Vs = CPV ( P,, T )
4.16-5 p
g He density of the gas in the pressurizer surge line is:
j l
l l
Pg = 1.0 4.16-6 v,
(
O l
ow4u3uwa16 non: b-ii1o97 4.16 2 REVISION: 1 l
l
c.
l 1 ';
1 Flow for tests with inoperative flow meters may.be inferred from the cold-leg differential pressure -
cells. Test SB01 has functional flow meters and differential pressure cells and can be used to calibrate this method. He resultant single-phase liquid flow rate is calculated as:
i AP g
g2 FMM-XXX 4.17-6 FMM-XXX,
REF-XXX APREF-XXX where:
REF-XXX Average flow in Ibm /sec through FMM-XXX during the
=
calibration period for Test SB01. The calibration period is the period of full power operation prior to the start of the transient.
APpyy_xxx Differential pressure corresponding to FMM-XXX per Table 4.17-1
=
AP Average value of APpyy.xxx during the calibration period for
=
REF-XXX Test SB01 Once flow in the cold legs becomes two-phase, the magnitude of the output from the flow meters is not indicative of the actual flow in those components. He liquid mass in CL-1 and CL-3 is then calculated using local level transducers as follows:
Step 1:
Compensate the readings from the downcomer level transducer listed in Table 4.17-2 to account for temperature differences between fluid in the downcomer and fluid in the reference leg of the instrument line. He local pressure and temperature transducers to be used to accomplish the compensation are also identified in Table 4.17-2.
Step 2:
Tables of level versus volume were developed for the cold legs. From the collapsed liquid level calculated from Step 1, calculate the liquid volume in the
{
cold leg using linear interpolation:
j V, a x = f(LDP-YYYCOMP) f where the subscript:
CL X CL-1 or CL-3
-O
=
G l
l.
o:us44wnanwv:ns.c4c344 4 7.non:1b o!169s.
4.17 3 REVISION: I
The elevation of the cold leg relative to the downcomer LDP span extends from [
]a.b.c in. Thus, the table of volume versus elevation for the cold legs is mapped onto the downcomer levels readings so that:
Downcomer-Comtensated Water Level Water Level in CL-1 and CL-3 (in.)
(in.)
[
Ja,b.c
(
3a.b,c
[
ja.b.c
[
ya.b.c Step 3:
The local pressures and temperatures measured using the instruments identified in Table 4.17-2 are used as inputs to the ASME steam tables to calculate the density I
of the liquid and vapor in the cold legs:
Pf,ct x = p(Pyr-xxx, T7p_xxx) 4.17-8 Pg.ct x = p(PFr-XXX, TTF-ZZZ) where the subscripts:
PT-XXX Channel ID for local pressure measurement in downcomer
=
TF-XXX Channel ID for local temperature measurement of liquid temperatures
=
TF-ZZZ
=
Channel ID for local temperature measurement of vapor temperatures Step 4:
Using the local thermodynamic properties of water as determined from local pressure and fluid temperature measurements, the liquid and vapor mass in the cold leg is calculated as:
1 M,ctx
- Pt.ctx x V, et x f
f 4.17-9 M,CL x
- Pg. CL x x (VToT. CL X - V. ct x) g f
where the subscript:
TOT Total volume of component
=
O o:u344wnonwv2wc4cu4*-417.non:ib-012098 4.17-4 REVISION: 2 i
Step 5:
The rate of change in mass inventory of the cold leg may be approximated by differencing two consecutive calculated values of the liquid and vapor masses:
dM, et x AM, et x _
Mr. ct x.1 - M, et x, 3 3 f
f f
i dt at ti-tig 4.17 10 dM, et x N, et x M. ct x, i - M. CL x, i-1 g
g s
g dt At t; - 1_
3 where the subscript:
i A specific value in the time and corresponding data array
=
The total rate of change of the mass inventory of CL-1 or CL-3 is then calculated as:
dM AM et w BL CL w BL dt at 4.17-11 E, CL 1 + E. CL 1 + M, et 3 + AM, et 3 l
t g
f g
4,17,1,2 Energy Terms The fluid in these two cold legs may, depending on the time and nature of the test, be in either a liquid or a vapor phase. Therefore, the change of energy for either CL-1 or CL-3 may be written as:
I d( Met x (T-T,,f))
d( T )
d( Mct x, f )
g M
C P P.f ct x. f
+ P.f (I -Iref) dt dt f
dt 4.17-12 d( T,)
d( Mct x g )
+C,g Mct x. g
+ P. g (T -T )
P dt g ref dt l
l l
l l
\\
l l
o A2344wnonvev2h4\\2344 w-417.non:lt> 121297 4,17 5 REVISION: 2 J
Writing the previous equation in its difference form:
l 4.17-13 d( MCL x (T-T,,r) )
AT AM f
CL x. f M
C P P, f CL x. f P. f d E )
f nf dt At At AT AMetx
- C g Met x' 8 sM P
.s t
At The data channel ids for the instmments to be used to define the thermal transpon propenies are identified in Table 4.17-2.
The rate of energy change of the metal mass for CL-1 and CL-3 is calculated as:
OCL w EL METAL *OCL 1 METAL
- OCL 3 METAL expanding, QCL X METAL = M X cp X "U
where the subscript:
TFM-20X = Temperature-sensing element to be used for this calculation:
X=1 CL-1 X = 3 CL-3 Similarly, the heat flux from the surface of the cold leg is calculated as:
OCL w BL AMB
- OCL I AMB + 0CL 3 AMB
~
The cold legs and CMT balance lines are similar in that they are predominantly horizontal pipes. The cold-leg surface heat flux calculations and equations are, therefore, identical to those of the CMT balance line. The balance line equations are provided in Subsection 4.4.6 and the associated cold-leg instrumentation is listed in Table 4.17-2.
O o \\2344wnon\\rev2\\sec4\\2344w.417.norrib-121297 4.17-6 REVISION: 2
1
(~N)
Step 5:
The rate of change in mass inventory of the cold leg may be approximated by
(_/
differencing two consecutive calculated values of the liquid and vapor masses:
dM, cg x M. CL x M. ct x. i - M, CL x, i-1 f
r r
f dt at ti - t _3 i
4.17-26 dM,ctx AM. CL x M. ct x, i - M CL x, i-1 g
g s
g dt At ti -t ;
i where the subscript:
i A specific value in the time and corresponding data array
=
The total rate of change of the mass inventory of CL-2 or CL-4 is then calculated as:
dM E
CL w/o BL CL w/o BL dt at 4.17-27 AM, CL 2 + AM, CL 2 + AM, ct 4 + AM, CL 4 f
g f
g 4.17.2.2 Energy Terms The fluid in these two cold legs may, depending on the time and nature of the test, be in either a liquid or a vapor phase. The change of energy for either CL-2 or CL-4 may be written as:
I d( Met x (T - T,,f))
d( T )
d( Mct x, f )
f C P L
P, f (T - Trer) r dt dt dt 4.17 28 d( T,)
d( M
+ cp,, Mct x, g
+ P, g (I -I )
CL x, s I dl g
ref dt 1
(^
O omuwnonwv2sece44. 4:7.non::b.121297 4.17-11 REVISION: 2
Wnting the previous equation in its difference form:
I d( MCL x (T - T,,g)
AT AM M
r
- C. f (I ~ I )
CL x, f cp P, f et x' I P
f ref dt t
At 4.17-29 AT AM et x
+ cP.g M
-M CL x.g At
.s s
at The data channel ids for the instruments to be used to define the thermal. msport properties are identified in Table 4.17-4.
The rate of energy change of the metal mass for CL-2 and CL-4 is calculated as:
OCL w/o BL METAL *OCL 2 METAL
- OCL 4 METAL expanding, E]20x 4,i7 31 Qct x yg7xt - M x cp x O
where the subscript:
TFM-20X = Temperature-sensing e!: ment to be used for this calculation:
X = 2 CL - 2 X=4 CL-4 Similarly, the heat flux from the surface of the cold leg is calculated as:
1 1
OCL w/o BL AMB "OCL 2 AMB
- OCL 4 AMB 4.17-32 The cold legs and CMT balance lines are similar in that they are predominantly horizontal pipes. The l
cold-leg surface heat flux calculations and equations are, therefore, identical to those of the CMT l
balance lines. The balance line equations are provided in Subsection 4.4.6 and the associated cold-leg l
instrumentation is listed in Table 4.17-4.
1 O
c:u3awnonvevn=4c344.-4 7.non:tb-i21297 4.17-12 REVISION: 2
where the subscript:
Vapor conditions at the local hot-leg pressures and temperatures g
=
TOT Total, or liquid + vapor
=
- The total fluid mass in the hot leg is calculated as:
Mgt x _ = M, at x + M, gt x 4.184 g
g
- Step 5:
The rate of change in mass inventory of the hot leg may be approximated by differencing two consecutive calculated values of liquid mass:
dMut x AM go x Mgt x,i - Mgt x, 33 p
4,} g,g dt At tg - t;.g where the subscript:
i
=
Index of the data and time arrays 4.18.2 Energy Terms
'Ihe fluid in the two hot legs may, depending upon the time and nature of the test, be either in a liquid or a vapor phase. Thus, the change of energy for the fluid in either HL-1 or HL-2 may be written as:
d( Man x (T -T,,f))
d( Tg) d( Mgt x, f )
cp P. f Mgt x' I P, f Ef nf) dt di dt 4.18-9 d( T )
d( MHL X.
)
g
+ cP, s M P. g (T - T,,f)
HL X, a dt g
dt i
where the subscript:
Steam g
=
f Liquid phase of water
=
HL X X = 1 for HL 1
=
X = 2 for HL 2 k
e:u344wnonw,2wce2344 4 s.non:1b.i21597 4.18-3 REVISloN: 2
Writing the above equation in its difference form:
I d( Mgt x U - T,,fD AT E
f gt x, f M
C P P, f dt HL x, r
- I
~ "
At At 4.18-10 AT AMan x C,g N P
HL X,g P, g @g - I )
+
ref At At The data channel ids for the instruments to be used to define the thermal transport properties are identified in Table 4.18-1.
The rate of energy change of the metal mass for HL-1 and HL-2 is calculated as:
ATTFM-20Y O
"M HL X METAL METAL X cp x 4, } g.] }
where the subscript:
TFM 20Y Designates the temperature sensing element to be used for this calculation:
where Y = 5 => HL-1 6 => HL-2 The hot legs and CMT balance lines are similar in that they are predominantly horizontal pipes. The hot-leg surface heat flux calculations and equations are, therefore, identical to those of the CMT balance line. The balance line equations are provided in Subsection 4.4.6 and the associated hot-leg instrumentation is provided in Table 4.18-1.
O c:C344wnonWv2\\sec4C344 -418.non: I b-121297 4,}g.4 REVISION: 2
r.
)
f 5.1 Analysis of Matrix Test SB01 Matrix Test SB01 (OSU Test U0001) modeled a 2 in. cold-leg break SBLOCA with LTC and without operation of the nonsafety-related systems. De break was located at the' bottom of CL-3 on the core I makeup tank side of the facility. De test included a simulated failure of one of the ADS-4 lines with I no vacuum break in the ADS discharge line. Since SB01 is the first LOCA test, more detail is l
' included in Section 5.1 than in subsequer.t sections.
He analysis of Matrix Test SB01 is di ided into three sections:
t l
l General facility performance (Subsection 5.1.1) describes the overall response of the system throughout the test. De prformance of the facility is characterized by the figures listed in l
Table 5.1.1 1.
l SBLOCA (Subsection 5.1.2) provides a discussion of the system behavior from the start of the
=
test, through system depressurization, to approximately [
Ja.b.c seconds into the transient, and includes the initial system blowdown, the establishment of natural circulation, and the initial portion of the IRWST injection cooling (Figure 5.2 2).
LTC (Subsection 5.1.3) discusses the behavior of the remainder of the test after [
]a.b.c j
i
. seconds and includes the completion of IRWST injection and the establishment of sump injection, i
The refill and subsequent recirculation of the CMT is considered as a separate discussion within Subsection 6.1.1. The period between SBLOCA and LTC is not discussed specifically since the
. system was behaving in a stable manner, l
l l
1 i
i I
~
I onwwnanvev2um.-50.non:itut11097 5.1-1 REVISION: 2 i
I i
l (9
5.1.1 Facility Performance Q,)
The simulated break was located on the bottom of cold leg 3. A flow nozzle simulating one ADS-4 valve was installed in the ADS 4-1 line (HL-1 to the ADS 4-1 separator) to provide the single-failure simulation. A flow nozzle simulating two valves was installed in the ADS 4-2 line (HL-2 to the ADS 4-2 separator). Additionally, flow nozzles simulating two ADS 1-3 pairs of valves were installed in the ADS 1-3 inlet lines.
The reactor heater control decay algorithm maintained maximum reactor heater power output for
[
]a.b.c seconds, and then the power was programmed to begin to decay, simulating the total pos trip energy input of the AP600 nuclear fuel. This test was performed with reactor heater rod HTR-C2-317 removed and replaced with a dummy rod.
Facility performance is divided into separate discussions of the five phases of the test:
Blowdown Natural circulation ADS em l
I
'V IRWST injection Sump injection j
l The overall performance of the 30,000-second (8-hour) test is shown in Figures 5.1.1-1 to 5.1.1-4.
Figure 5.1.1-1 shows the pressurizer pressure throughout the test with the various phases and operating components indicated for each phase. For clarity, the time scale is cut between 2000 and 12,000 seconds since there was no change in the operating mode during this period. Figure 5.1.1-2 shows the total injection flow rates into the DVI line from the various systems as a function of time.
I Figure 5.1.1-3 shows the calculated steam generation rate in the core throughout the test.
Figure 5.1.1-4 shows the variation in average measured core outlet temperature and peak clad temperature relative to the core outlet saturation temperature.
Figures 5.1.1-1 and 5.1.1-2 shows that a continuous flow of water into the reactor vessel was provided by the passive safety-related systems as the primary system was depressurizing. The operation of the passive safety injection systems overlapped so that, as one system drained or emptied, another provided flow into the simulated reactor vessel for continuous core cooling.
Sufficient flow to the core was maintained so that the measured average core outlet temperature was O
saturated or subcooled for significant ponions of the transient, and the core steam flow was less than
(.)
the passive safety system injection flow (Figures 5.1.1-2 and 5.1.1-3). As the system transitioned into o:cm.newv2cww-50.non ib-i2:597 5.1.1-1 REVtSION: 2 o
LTC, the water injected from the sump was hot since it originated in the primary system. The hotter sump water combined with the lower driving head during the sump injection resulted in continuous steam generation in the heater rod bundle (Figure 5.1.1-3) after [
]a.b.c seconds. This steam was then vented, primarily through the ADS-4 valves.
5.1.1.1 Blowdown Phase The blowdown phase corresponds to the first [ Ja.b.c seconds of Matrix Test SB01 (Figure 5.1.1-1).
De test was initiated (0 time) by opening the break valve located at the bottom of CL-3 and the blowdown phase was completed when steam pressure reached the steam generator (SG) safety valve setpoint. 'Ihe hot leg of the reactor coolant system (RCS) was at 420 F and 370 psig prior to test initiation. The simulated S signal was generated at 0.5 seconds after the break signal and initiated the following actions.
In the first [
]a.b.c seconds, the SG safety relief setpoint pressure was raised to [
]a.b.c psig, and the reactor shifted to power (kW) control mode with a programmed power de nand for 600-kW total power. The main feedwater pump tripped and the feedwater was isolated at [ ]a.b.c seconds. The passive residual heat removal heat exchanger (PRHR HX) outlet valve and CMT discharge valves opened at [ ]*** seconds, and the reactor coolant pumps (RCPs) tripped at [ ]a.b.c seconds after the break signal.
Forced flow continued through the PRHR HX and the CMTs until the RCPs stopped at about
[ ]a.b.c seconds, at which time PRHR HX flow changed to natural circulation. As the RCS depressurized and coolant escaped through the break, the pressurizer level decreased rapidly and steam formation began in the reactor vessel upper head. At about [ ]*AC seconds, the level in the reactor vessel indicated the vessel was beginning to lose inventory as the vessel drained and some liquid flashed to steam. The upper plenum volume began to show a collapsed level decrease indicating that there was steam collecting in this volume at about [
]*A* seconds.
As primary system pressure fell to near a steady state condition, the system transitioned into the natural circulation phase once the pumps coasted down and the system reached the SG pressure relief setpoint at 335 psig at about [
]a.b.c seconds. During this period, there was initially liquid solid natural circulation in the PRHR and CMT systems. The CMTs provided recirculating flow to the reactor vessel, while the PRHR removed energy from the primary system.
5.1.1.2 Natural Circulation Phase ne upper head drained to two-thirds empty after [
]aAc seconds. The mass loss through the break caused a rapid decrease in pressurizer level and emptied the pressurizer at about [
]a.b.c seconds.
The pressurizer surge line was completely emptied at about [
la.b.c seconds. The primary system was initially at a pressure above the SG secondary-side pressure, therefore the SGs continued to remove energy from the primary system. As the system continued to drain, the SG tubes started to c:\\2344wnon\\rev2\\2M4w.50.non: I b-121597 5.1,12 REVtSION: 2
[9 an excess of mass in the system to be vented through ADS-4 before IRWST injection could occur.
kl CMT-1 and CMT-2 were completely empty at [
]"' seconds, respectively.
He collapsed reactor vessel level reached a minimum value of about [
]"' in. at [
]'6' seconds.
Although this level is below the top of the heater rod heated length, the actual level of the top of the two-phase mixture is much higher, as shown in the behavior of the core outlet thermocouples (Figure 5.1.1-4), which do not exceed saturation temperatures.
At about [
]'6* seconds, the RCS system pressure decreased to about [ ]"' psig, which was sufficiently low that the IRWST static head was greater than RCS pressure, and IRWST injection began.
5.1.1.4 In-Containment Refueling Water. Storage Tank Injection Phase IRWST injection was split between the two DVI lines beginning at [
]"' seconds and continually diminished (Figure 5.1.1-2) as the differential head between the IRWST and the RCS decreased with draining of the IRWST. IRWST injection was sufficient for the primary system to refill. He pressurizer and pressurizer surge line emptied a second time at about [
]"' seconds, respectively.
I When the pressurizer had a liquid level, a partial vacuum of approximately 12.3 psia was created in the 1 ADS 1-3 separator and sparger pressure. The partial vacuum was maintained from [
]'**
(,)
I seconds as steam in the ADS lines was condensed in the IRWST. The partial vacuum was broken as the level in the IRWST decreased below the sparger nozzles. No vacuum breaker was installed on the sparger line inside the IRWST for this test. A vacuum breaker is included in the AP600 plant design, thus this OSU test response may not be typical of the AP600. The surge line then began to reflood almost immediately at [2220]"' seconds and the pressurizer at about [3226]"' seconds. The reflood was caused by RCS levels increasing above the reactor vessel nozzles because IRWST injection exceeded the inventory losses, and by the condensation in the ADS 1-3 lines. The maximum pressurizer level attained was about [47]'*' in. at [5984]'** seconds, but it immediately began to decrease he pressurizer was empty at [6080]"' seconds and remained empty for the remainder of the test. He surge line stayed full until [6080]"" seconds when the level decreased to about [25]"# in. and remained there until [9524]'** seconds. The level again decreased to about [15]'** in. at [9400]'** seconds and oscillated between about [10 to 20]'** in. for the remainder of the test.
Both CMT balance lines began to refill at about [
]"" seconds when the IRWST injection increased j
the reactor vessel level sufficiently to begin covering and refilling the cold legs. At about [
]'6' seconds, when the CMT-2 balance line had completely refilled CMT-2 began to rapidly refill and reached the [
]"' in. level (about two-thirds full) at about [
]'** seconds. De CMT refi!1 is
)
discussed in more detail in Subsection 6.1.1. After the CMTs were partially refilled, there was no injection flow from the CMTs because the higher static head of the IRWST held the CMT discharge line I
check valves closed.
r i
LJ o:u344wnonvev2u344w-50mdb 122297
$,1,1-$
REVtSION: 2 i
Steam generation staned again at about [
]"' seconds and continued for the remainder of the test (Figure 5.1.1-3). CMT-1 and CMT-2 remained at essentially constant levels for several thousand seconds and then began slow draindowns at about [
]'6* seconds, respectively. The draindown for both CMTs was slow and did not occur until the IRWST relative level was [
]'6" in. below that of the CMTs. Data indicate that the CMTs drained for a while, and then the differential head between the IRWST and the CMTs again closed the CMT discharge check valves, terminating draining until the differential shifted the other way and draining recommenced. Both CMTs were completely empty at about [
]"" seconds, which coincides closely with the primary sump injection valve opening at [
]"* seconds. A possible correlation is that when the primary sump valve opened, the IRWST had just reached its minimum level of about [
]"" in., which is about [ ]""
i in. below the instrumented level for the CMTs, and that there was still a slight panial vacuum remaining in the CMTs. Also, when the primary sump injection valves opened, there was a short period in which the IRWST and primary sump levels equalized, causing a decrease in RCS fluid levels and resulting in a rapid drop in CMT levels from about [
]""in.
Staning at about [
]'6" seconds, there was a series of pressure, level, and flow oscillations that occurred throughout the components of the facility lasting until about [
] seconds. These oscillations will be discussed in detail in Subsection 6.1.3.
At[
]"* seconds, the PRHR HX inlet fluid temperature instantly increased from [
]'6' 'F to saturation temperature. The temperature increased at about [
] seconds after pressure, level, and flow oscillations began in the facility and was possibly caused by the inlet line burping, which once again allowed the line to fill with saturated steam. Following the burp, all of the PRHR HX temperatures began to slowly approach saturation.
'Ihe break separator level began to increase at the same rate as the primary sump at about
[
]"* seconds. This increase occurred when the sump level reached the height of the break separator loop seal. As a result of this increase, the break separator level reached the height of the break in CL-3, causing break flow to reverse and flow from the break separator into the RCS through the break at about
[
]"* seconds. Break flow then remained essentially zero or slightly negative throughout the rest of the test.
5.1.1.S Sump Injection Phase
~ Primary sump injection (Figure 5.1.1-2) began through the check valves around the sump injection valves at about [
]"* seconds, when primary sump and IRWST levels were essentially equal. At
[
]'6' seconds, the primary sump injection valves automatically opened when the IRWST reached its low-low level setpoint of [
]"' in.
In the LTC mode of operation, primary system inventory was lost through the ADS 4-1 and ADS 4-2 valves to the primary sump. System inventory was made up through pnmary sump and IRWST injection through the DVI lines and some small flow from the primary sump through the break a:umwnonwv2cm..so.non.ib-122297 5.1.1 -6 REVISION: 2
{J T
5.2 Analysis of Matrix Test SB18
-1 Matrix Test SB18 is considered the reference test since it was configured in the same manner with an I ADS vacuum break as the remaining tests. Matrix Test SB18 (OSU Test U0018) simulated a 2-in.
cold-leg (CL) break SBLOCA with LTC and without operation of nonsafety-related systems. The break was located at the bottom of CL 3 with a simulated failure of one of the ADS-4 lines. CL-3 is on the CMT side of the facility.
The analysis of Matrix Test SB18 is divided into three sections:
General facility performance (Subsection 5.2.1) describes the overall response of the system throughout the test. The performance of the facility is characterized by the figures listed in Table 5.1.1-1.
SBLOCA (Subsection 5.2.2) provides a discussion of the system behavior from the start of the test, through system depressurization, to approximately (
]a.b.c seconds into the transient, and includes the initial system blowdown, the establishment of natural circulation, and the initial portion of the IRWST injection cooling (Figure 5.2-2).
LTC (Subsection 5.2.3) discusses the behavior of the remainder of the test and includes the completion of IRWST injection and the establishment of sump injection.
,C The refill and subsequent recirculation of the CMT is considered as a separate discussion within Subsection 6.1.1. The period between SBLOCA and LTC is not discussed specifically since the system was behaving in a stable manner.
I Matrix Test SB18 was a duplication of the break conditions of Matrix Test SB01. The purpose of perfonning Matrix Test SB18 was to confirm the ability of the facility to replicate its response to a SBLOCA, with the same configuration from the beginning to the end of the test program.
l The differences between SB01 and SB18 are as follows:
In SB18, a vacuum breaker was installed on the ADS 1-3 sparger line inside the IRWST to I
eliminate partial vacuums in the pressurizer and ADS 1-3 separator.
l In SB18, pressurizer heater logic was changed so that the PLC initiated a signal to open the pressurizer heater SCR contactor at [
]a.b.c seconds after S signal actuation, thereby ensuring de-energization of the heaters.
In SB18, to prevent heating at the top of the CMTs prior to break valve opening, CMT p
balance line isolation valves were closed and opened by the operator 1 minute after the TEST pushbutton was pressed.
o:C344wnonWv20344w 52.non:Ib-122397 5.2-1 REVISION: 2 i
5.4 Analysis of Matrix Test SB09 Matrix Test SB09 (OSU Test U0009) simulated a 2 in. LOCA in the CL-3 to CMT-1 balance line l.
with LTC and without the operation of the nonsafety-related systems. 'Ihe break was located on the horizontal section, of the balance line before the venical rise to CMT-1. Except for the break location, I this test was similar to SB01 and SB18, including the simulated failure of one of the ADS 4 lines.
Changes to the OSU facility since the performance of SB01 are noted in the Final Data Repon.(D ne analysis of Matrix Test SB09 is divided into three subsections as follows:
Facility performance is discussed in Subsection 5.4.1. It provides a brief outline of the response c1 the test facility; funher details are available in the Final Data Report.(D 1
The short-term transient for SB09 encompassed the stan of the simulation up to [
]a.b.c l
e seconds. This period included the blowdown, natural circulation, ADS, and initial IRWST stages of the transient.
l ne analysis of the long-term transient for SB09 encompassed the time frame from [
]a.b.c e
seconds to the end of the test. This phase of the transient includes the IRWST injection and covers the transition to sump injection. The long-term transient actually started at IRWST injection, which is discussed as part of the short-term transient. Between the end of the shon-term transient and [
]a.b.c seconds, the system remained relatively inactive with the exception of the CMT-2 refill. At [
]a.b.c seconds, CMT-2 began to refill: CMT-1 [
]a.b.c during the SB09 transient. CMT refill phenomena are discussed further in i
l Subsection 6.1.1, and the discussion of the long-term transient provided here begins at
[
]a.b.c seconds.
De discussion of the short-and long-term phase of the transient focuses on imponant thermal-hydraulic phenomena identified,in the PIRT (Table 1.3-1). Key indicators of the quality of the analysis on which this discussion is based are the mass and energy balance results. Dese are discussed in detail in Subsections 6.2.2 and 6.2.3.
l l
1 O
L l
l o:umnonwv2cm-54.non:lt li1097 5.4-1 REVISION: 2 l
i 5.4.1 Facility Performance The performance of the OSU test facility during Matrix Test SB09 in reference to the five transient phases is outlined in the following:
Blowdown Natural circulation
- ADS IRWST injection Sump injection The overall performance of the facility during the transient is shown in Figures 5.4.1-1 to 5.4.1-4. Figure 5.4.1-1 shows the pressurizer pressure throughout the test with various phases and operating components delineated on the figure. The time scale was reduced for clarity since there were only small changes in I system pressure during the long-term phase of the transient. Figure 5.4.1-2 shows the DVI line injection I flows from the CMT, accumulator, IRWST and sump at each time in the transient. Figure 5.4.1-3 shows the calculated core steam generation rate throughout the test. Figure 5.4.1-4 shows the variation in average measured core outlet temperature and peak clad temperature relative to the core outlet saturation temperature.
Figures 5.4.1-1 and 5.4.12 show that there was a continuous flow of water to the core from the passive safety-related systems throughout the transient. Once initiated, the ADS lines rapidly depressurized the primary system, enhancing the CMT and accumulator injection flow rates. Ultimately, the ADS-4 valves reduced the system pressure sufficiently to start gravity-driven IRWST injection. He passive injection systems operation overlapped so that as one source of water drained the next was available to continue the cooling process. De level of steam generation in the core and the response of the average measured j
core outlet fluid temperatures and maximum clad temperatures are shown in Figures 5.4.1-3 and 5.4.1-4.
Dese figures show that the cooling flow prevented core heatup, and the core remained covered. He core remained subcooled for large periods of the transient. When steam was produced, the rate of generation remained well below the rate at which water was delivered to the core.
5.4.1.1 Blowdown Phase ne blowdown phase began at time zero when the break was initiated and continued until the primary 1
I circuit pressure is in equilibrium with the secondary-side pressure at around [
]'6# seconds. During this phase of the transient, cooling flow was provided from the intact CMT, while the CMT with a broken balance line injected very little mass. CMT-2 remained in the recirculation mode until almost the end of this phase, and heat was removed from the primary circuit via the SGs. The pressurizer and surge line completely drained at [
l'6" and [
]'6' seconds, respectively.
o:u344wnnnwv2c344. 54.non:ib-090498 5.4.1-1 REVISION: 2
Y 5.4.1.2 Natural Circulation Phase In this LOCA simulation, the single-and two-phase natural circulation phase initially continued the gradual reduction in system pressure characteristic of blowdown; later on in this phase of the transient, the rate of depressurization increased significantly once all the SG tubes had drained at about [
]"'
seconds. The tubes in SG-2 in the nonbalance line loop completed draining almost [
]"' seconds later than those in SG-1. After [
]'6' seconds, heat removal from the primary circuit continued via the PRHR and the break. In response to the enhanced depressurization rate, CMT-2 transitioned into a rapid draindown at [
] seconds, and the falling CMT level reached the ADS low-level setpoint so that the ADS-1 valve began to open at [
]"' seconds. By [
]'6' seconds, both accumulators began to inject.
5.4.1.3 Automatic Depressurization System Phase ADS-1 actuation was followed by ADS-2 and ADS-3 [
]'6' and [
]'6" seconds later. An increased accumulator injection rate was observed with initiation of ADS-2. The influx of cold water combined with increased venting via the ADS led to an even more rapid depressurization of the primary system.
Actuation of ADS-4 at [
]'b' seconds completed the depressurization to the extent that the IRWST began injecting at [
]"' seconds via DVI-2 and [
]'6' seconds via DVI-1. During the rapid accumulator injection, increased flow path resistance reduced flow out of CMT-2 and stopped flow out of CMT-1. CMT flow resumed at rates approaching [ ] Ibm /sec. as the accumulators drained. Because the CMT-1 balance line contained the break, CMT-1 injection was delayed and overlapped with IRWST-2 injection for more than [
]'6' seconds. CMT-1 flow continued concurrently with in [
]'b' seconds of IRWST-1 injection. He minimum RPV mass inventory of [
]'** lbm occurred just before IRWST injection began. Actuation of ADS-1 and ADS-2 rapidly refilled the pressurizer as water and steam flowed out of the ADS. He pressurizer gradually drained by [
]'** seconds.
5.4.1.4 In Containment Refueling Water Storage Tank Injection IRWST injection signals the transition from the short-to long-term phase of the transient. Initially, IRWST injection was delivered solely through the IRWST-2 DVI line with flow gradually increasing as the driving head between the IRWST and the RCS increased. The pressure differential increased because RCS pressure decreased as the core steam generation decreased from [
]'** lbm/sec. at IRWST-2 initiation to zero at [
]"' seconds. The maximum IRWST flow was established shortly thereafter, then it gradually decreased with the decrease in pressure differential as the IRWST continued to drain. He influx of water from the IRWST was enough to keep the core subcooled until [
]"' seconds. Steam was subsequently generated in the core for the remainder of the transient. Following the restart of core steam generation, IRWST injection between [
]"' seconds was marked by oscillations in pressure and level throughout the primary system. These oscillations were also observed in the ADS-4 liquid flow rates.
5.4.1.5 Sump Injection on%nonwam.54 nonab o90498 5.4.1 -2 REVISION: 1
5.4.2 Short Term Transient For the 2-in. balance line break LOCA simulation, Matrix Test SB09, the short-term transient l
encompassed the time frame up to 3000 seconds. As shown in Figure 5.4.1-1, this period included
)
full depressurization of the facility through all four stages of the ADS, together with CMT and accumulator injection plus the initial stages of IRWST injection. The variation in mass, energy, pressure, and temperature throughout this stage of the transient are illustrated in the plot package outlined in Table 5.4.2-1. The plots concentra:e on the primary system, including the accumulators, CMTs, IRWST, primary sump and flows from the primary system via the ADS, break, and IRWST overflow.
l There were two principal parameters to be examined for the short-term transient:
Adequate flow from the passive systems to the reactor vessel must be maintained.
Adequate flow into the core must be maintained to ensure that decay heat was removed from
=
the simulated fuel rods, without a temperature excursion.
1 '
'Ihese parameters are addressed in the following discussion.
5.4.2.1 Maintenance of Core Cooling Mass Injected to the Primary Systems 1 Figures 5.4.2-5 and 5.4.2-6 show the combined effect of the injection flows for the short-term phase of the transient. Separate plots of the individual contributions to the total flow can be located by consulting the plot package index given in Table 5.4.2-1.
Figures 5.4.2-5 and 5.4.2-6 show how the CMTs, accumulators, and IRWST supplied a continuous flow of water to the core. During the first [
)*AC seconds, cooling flow was provided primarily by j
CMT-2, since CMT-1 flow was severely limited by the break in its balance line. By [
]*AC
{
seconds, CMT flow was supplemented by flow from both accumulators. The rate of flow from the CMTs was reduced or stopped by accumulator injection. Accumulator flow produced maximum DVI injection rates during the entire transient with values of [ ]aAc lbm/sec. and above in each DVI line.
Following the end of accumulator injection, the CMTs again provided cooling flow. Because CMT-1 was still almost full at the end of accumulator injection, it provided greater flow than CMT-2 through the remainder of the CMT drain period. While CMT-2 emptied [
]*** seconds before the flow from IRWST-2 began, CMT-1 was still injecting at a rate of (
)*AC lbm/sec, so continuous injection was maintained. IRWST-1 injection began [
]a b.c seconds before the CMT-1 draindown completed.
1 Since continuous IRWST injection through both DVI lines began before CMT-1 had fully drained, there was no period of the short-term transient when the passive safety systerns failed to provide flow j
to the RPV.
a.u344.nonwv2s2344w.f 4.non 1b.121597 5.4.2-1 REVISION: 2
[
Reactor Pressure Vessel and Downcomer Behavior The effect of water flow on the average measured core inlet / outlet temperatures and peak clad temperatures during the short-term phase of the transient is shown in Figures 5.4.2-3 and 5.4.2-57.
The core outlet temperature first reached the saturation point at [ ]a.b.c seconds. The core outlet fluid temperature became subcooled again retuming to the saturation level for the period between [
]a.bx seconds, after which, the influx of water from the accumulators kept the core subcooled until
[
Ja.b.c secs. At [
Ja.b.c seconds, the influx of water from the IRWST was sufficient to subcool the core again. The core then remained subcooled until the end of the short-term transient.
Figure 5.4.2-57 shows that there were no significant excursions in heated rod temperatures throughout the short-term transient, therefore, sufficient core inventon and flow were maintained through this phase of the transient to remove the decay heat generated. For significant portions of the transient, a two-phase mixture was present in the core and upper plenum regions, with core boiling kept at a low level.
The following discussion tracks the variation in water level and mass throughout the reactor vessel and downcomer. The mass and level for the core region are shown in Figurea 5.4.2-44 and 5.4.2-45. The collapsed liquid level in the core indicates that the heated rods remained covered with a single or two-phase mixture throughout the short-term transient. The minimum core inventon of [
]a.b.c lbm occurred at about [
Ja.b.c seconds into the transient before the initial accumulator injection was fully established. As shown in Figure 5.4.2-45, the collapsed liquid level dropped to [ Ja.b.c n. below the top of the heated rod length during this phase of the transient. The average void fraction of the core two-phase mixture may be estimated by dividing the measured core collapsed liquid level by the
[
]a.b.c n. heated rod length. In this test, the minimum collapsed liquid level corresponded to a core void fraction of [
].a.b.C By the end of the short-term transient, the effect of IRWST injection ended core boiling (Figure 5.4.2-55), and the core was again water-solid.
The collapsed liquid levei in the upper plenum region and the associated fluid mass are shown in Figures 5.4.2-49 and f.4 2-48. Figures 5.4.2-50 and 5.4.2-51 show that the upper head had only partially drained while accumulator injection and ADS-1 actuation occurred; then it rapidly refilled by about [ Ja.b.c in. during Matrix Test SB09. The upper head resumed draining and refilled later in the short-term transient during maximum IRWST injection.
The mass of fluid and collapsed liquid level in the RPV downcomer are shown in Figures 5.4.2-41 and 5.4.2-42. The downcomer collapsed liquid level fell to the bottom of the cold-leg piping during the first [
Ja.be seconds. IRWST injection maintained the collapsed liquid level within the cold-leg pipe perimeter after [
]a.b.c seconds in the transient.
O or3mmowv2uws4mn:ib.12 297 5.4.2-2 REVtSION: 1
O 5.4.3 Long Term Transient The long-term transient started with initiation of IRWST injection, covered the transition from IRWST to sump injection, and provided information on the LTC response of the AP600. For the 2-in. cold-leg balance line break, Matrix Test SB09, the long-term transient analyzed begins at [
]a.b.c seconds and extends to the end of the test at'[
]a.b.c seconds. The behavior of the test facility during this period of the transient is discussed in this subsection using the plot package detailed in Table 5.4.3-1. This I
analysis concentrates on the components of the primary system that remained active during the LTC phase, that is, the RPV, the hot legs, ADS-4, the sumps, and the IRWST.
Thermal-hydraulic phenomena of interest for the long-term transient are:
Maintenance of core cooling and removal of energy from the primary system.
Level oscillations (from [
]a.b.c seconds. There were system wide level and pressure oscillations, which are discussed further in Subsection 6.1.3).
5.4.3.1 Maintenance of Core Cooling Mass Injected into Primary System Total DVI line flow, CMT flow, and IRWST flows are shown in Figures 5.4.3-6 and 5.4.3-7. Flow from the primary sump is shown in Figure 5.4.3-19. From around [
]a.b.c seconds, there was a contribution to the DVI flow from the CMT-2 as the previously refilled CMT-2 drained. CMT-1 did not ref?ll during test SB09.
During the pre-sump injection phase of the transient, IRWST flow proceeded at a gradually decreasing rate with the effect of the primary system oscillations superimposed. At [
Ja.b.c seconds, flow from the primary sump began through the main injection valves, which opened as the IRWST has reached the low-low level set-point. This resulted in a reversal of flow through the IRWST injection line-1 almost equal to the IRWST flow into DVI-2. The net result was that the IRWST level decreased less than [ ]a.b.c in.
between the inception of sump injection flow through the primary injection valves and the end of the transient. The initial sump injection through the check valves around the main injection valves at [
]a b.c seconds decreased the IRWST rate by about [ Ja.b,e percent in each of the DVI lines.
Reactor Pressure Vessel and Downcomer Response The effect of water inflow on the average measured downcomer fluid temperatures, core inlet and core l
outlet temperatures, and heater rod temperatures during the long-term phase of the transient is shown in Figures 5.4.3-4,5.4.3-5, and 5.4.3-38. Figure 5.4.3-4 shows that there was a general increase in average downcomer fluid temperatures during teh long-term transient. By the end of the test, this o:uu4.nonwv2cu4w.54.non:1b-121297 5.4.3-1 REVISION: 1
average temperature reached a value about [ Ja.b.c F below saturation. Figure 5.4.3-5 shows that the core exit temperature remained at or near saturation for the majority of the long-term transient after
[
]a.be secs. Figures 5.4.3-34 to 5.4.3-36 show that the DVI line flow method described in Section 4.11 indicates that a small level of boiling was maintained after [
]a.b.c seconds into the transient. Nevertheless, the level of boiling was small, and the test results showed that the inflow from the IRWST and sumps was sufficient to maintain cooling.
Figure 5.4.3-38 shows that there were no significant excursions in heated rod temperatures throughout the long-term transient therefore, sufficient core inventory and flow was maintained through this phase of the transient to remove the decay heat generated. For significant portions of the transient, a two-phase mixture was present in the core and upper plenum regions.
The following discussion tracks the variation in water level and mass throughout the reactor vessel and downcomer. The mass and level for the core region are shown in Figures 5.4.3-28 and 5.4.349. The collapsed liquid level in the core indicated that the heated rods, were always covered with a single-or two-phase mixture. During the sump injection stage of the transient (beyond [
Ja.be seconds),
the collapsed liquid level remained just below the top of the heated rods, and the core void fraction was [
].a.be The reduction in the core collapsed liquid level following the start of sump injection produced no marked impact on core cooling; in this test the sump water was relatively cold (Figures 5.4.3-4 and 5.4.3-5). During sump injection the calculated steam generation rate was at a maximum of about [
]a.be Ibm /sec. (Figure 5.4.3-36).
I Prior to sump injection, the collapsed liquid level in the downcomer (Figure 5.4.3-32) region remained I between the mid-level of the hot leg and cold legs. After initiation, the collasped liquid level in the I downcomer region remained between the mid-level of the DVI lines and the hot legs. Figure 5.4.3-33 shows the mass of water in the upper head, which remained below [ Ja b.c lbm from the inception of sump injection until the end of the test.
The mass of water in the RPV is shown in Figure 5.4.3-25. For the sump injection portion of the long-term transient, the reactor vessel water mass reached an equilibrium value of about [
Ja.b# Ibm, which is
[
]a.b.c percent of the initial vessel water inventory. From [
la.b.c seconds, oscillations in vessel inventory were observed. Figures 5.4.3-51 to 5.4.3-56 illustrate these oscillations using plots on a restricted time frame from [
]a.be seconds. These oscillations are observed in primary system measurements from the upper plenum to the ADS-4 flows. The SB09 oscillations have a less uniform period during this time interval than do the oscillations observed in other tests. The oscillations in ADS flow lagged behind those in the upper head pressure by around [ Ja b.c seconds. These oscillations and possible mechanisms for their production are discussed further in Subsection 6.1.3.
The mass of fluid and collapsed liquid level in the RPV downcomer are shown in Figures 5.4.3-26 and 5.4.3-27. The collapsed liquid level remained above the cold leg midplane until [
]a.b.c seconds when it started to fall to an elevation below the center of the hot legs. This was close to the time that o:umme2umw.s(non: b.121297 5.4.3-2 REVISION: 2 t
CMT-2 completed draindown and corresponds to the time the cold legs began to drain (Figure 5.4.3-41)
I and the ADS-4 valve liquid flow started to decrease (Figure 5.4.3-44). Once sump injection was I established, the downcomer inventory became stable at about mid-elevation of the hot legs.
5.4.3.2 Energy Transport from the Primary System During the long-term transient, energy continued to be deposited in the primary system from the heated rods, metal, and fluid flowing from the primary sump. The SGs and PRHR remained inactive throughout this phase of the transient, and the principal path for energy out of the primary system was via the ADS-4 valves.
Integrated mass flow from the primary system via the ADS and the break is shown in Figure 5.4.3-43.
During the LTC phase of the transient, the only significant outflow was through the ADS-4 valves. After
[
. ]'6' seconds, there was some reverse flow through the break as is indicated by the reduction in the integrated flow shown in Figure 5.4.3-43; the break flow integral returned to its pre-[
]'6' m nd value prior to sump injection. During the sump injection phase of the transient, outflow from the ADS-4 valves was liquid. By the end of the test, liquid flowed out through these valves at a combined average rate of [
]'6' lbm/sec.
Figure 5.4.3-36 shows the calculated steam generation rate as determined by the DVI line flow method.
During the sump injection phase of the transient, steam was generated at up to [
]'6' lbm/sec., although the steam vortex meters indicate little or no flow out of the ADS-4 valves. However, there are two indications that steam is leaving the primary system by this route.
Figure 5.4.3-46 shows total measured system fluid inventory. During this phase of the transient after the start of primary sump injection (from [
] seconds, that is, when core steam generation was most significant), the total system inventory fell by about [
]'6' lbm.
This amount corresponds to a steam flow rate of [
]'6' lbm/sec., which would not have been detected by the vortex meters.
Examination of the fluid thermocouples on the outlet of the ADS-4 valves indicates that temperatures remained at or above saturation temperature following the start of sump injection.
It was not possible for all the steam generated in the core to flow from the upper head to the downcomer via the bypass holes (Subsection 6.1.3). 'Iherefore, steam was leaving the primary system via ADS-4.
Figure 5.4.3 50 shows all the components to the system energy balance. Further discussion of steam loss from the primary circuit is provided in the mass and energy balance discussions of Section 6.2.
i
\\
om44w onvev2uws4mid-o9049s 5.4.3-3 REVISION: 2
TABLE 5.4.31 OSU TEST ANALYSIS STANDARD PLOT PACKAGE FOR SUBSECTION 5.4.3 LONG TERM TRANSIENT Plot No.
Component Variables Units Description 1
RPV RPVPWR kW Core power 2
Primary sump TSMPII, TSMPI2
'F Sump injection line temperatures 3
DVI TDVIL1, TDVIL2
'F DVI line temperatures 4
RPV TOIDC, T02DC, T03DC,
'F Water and saturation temperatures in ST01DC downcomer 5
RPV TOIRPV, T08RPV.
'F Core inlet / outlet temperature, ST08RPV saturation temperature 6
DVI-l
- WWTDVIL1, Ibm /sec.
Individual components and total flow
- WWTIRWII, in DVl-1 WWTIRWI3 7
DVI-2
- WWTDVIL2, Ibm /sec.
Individual cemponents and total flow
- WWTIRWI2, in DVI-2 WWTIRWI4 8
CMT CLDP-502, CLDP-507 in.
Collapsed liquid level in CMTs 9
CMT CLDP-509. CLDPSIO in.
Level CL-CMT balance lines 10 IRWST IRWST lbm Mass of fluid in IRWST 11 IRWST CLDP-701 in.
Collapsed liquid level in IRWST 12 IRWST UIRWST Btu Fluid energy in IRWST 13 Primary sump AMPSMP lbm Primary sump fluid mass 14 Primary sump CLDP-901 in.
Primary sump level 15 Primary sump UPSMP Btu Primary sump fluid energy 16 Secondary sump AMSSMP lbm Secondary sump fluid mass 17 Secondary sump CLDP-902 in.
Secondary sump level 18 Secondary sump USSMP Btu Secondary sump fluid energy 19 Primary sump WSTSMPET, WWTSMPIT Ibm /sec.
Primary sump steam and liquid injection rate 20 Primary sump MISMPil, MISMPI2, Ibm Integrated primary sump and IRWST MISMPIT, MIIRWT flows 21 SG MSSGIPI, MSSGIP2, Ibm Mass of fluid in SG side inlet / outlet MSSGOPI, MSSGOP2 plena 22 Surge line PLM lbm Fluid mass in surge line 23 Surge line CLDP-602 in.
Collapsed liquid level in surge line 24 Surge line UPSL Btu Fluid energy in surge line 25 '
RPV MWRPV lbm Total fluid mass in reactor vessel e
o.uu4wnonvn2ch54.non itw>o498 5.4.3 4 REVISION: 1
I l
i l
/'N 5.9 Analysis of Matrix Test SB15 V
i Matrix test SB15 (OSU Test U0015) simulated a 2-in. hot-leg break LOCA with LTC and without the operation of the nonsafety-related systems. The break was at the bottom of HL-2 and except for the l break location, this test was identical to SB18, including the simulated failure of one of the ADS-4 lines.
Analysis of Matrix Test SB15 is divided into three sections as follows:
Facility performance is discussed in Subsection 5.9.1. It provides a brief outline of the response of the test facility; further details are available in the Final Data Report.(3) l The short-term transient for SB15 encompassed the stan of the simulation up to [
]a.be seconds. This period includes blowdown, natural circulation, ADS and initial IRWST stages of the transient.
Analysis of the long-term transient, SB15, encompassed the time frame from [
]a.b.c seconds to the end of the test. This phase of the transient concluded with the IRWST injection phase to the initiation of sump injection. The long-term transient actually started at IRWST injection, which is discussed as part of the short-term transient. Between the end of the shon-term transient to [
]a.b.c seconds, the system remained relatively inactive with the exception l
s of the CMT refill. At [
]a.b.c seconds, CMT-1 began to refill and CMT-2 followed
[
la.b.c seconds later. CMT refill phenomena is discussed further in Section 6.1.1 and the discussion of the long-term transient provided here begins at [
Ja.b.c seconds.
The discussion of the shon-and long-term phase of the transient focuses on important thermal.
hydraulic phenomena identified in the PIRT (Table 1.3-1). The mass and energy balance results are key indicators of the quality of the analysis on which this discussion is based. These are discussed in detail in Subsections 6.2.2 and 6.2.3.
I i
l C
on*wanwulmw-59.non:Ib.ll1097 5,9 1 REVIS'!ON: 2 l
(
5,10 Analysis of Matrix Test SB19 Matrix Test SB19 (OSU Test U0019) simulated a 2-in. break LOCA with LTC and without the l
operation of the nonsafety-related systems. By automatically controlling the BAMS header pressure, the effect of containment backpressure was simulated. The break was located at the bottom of CL-3 and except for the simulation of backpressure, this test was similar to SB18 and SB01, including the simulated failure of one of the ADS-4 lines. Changes to the OSU facility between SB18 and SB19 are identified in the Final Data Report.W The analysis of Matrix Test SB19 is divided into three sections, as follows:
Facility perfonnance is discussed in Subsection 5.10.1. It provides a brief outline of the response of the test facility; further details are available in the Final Data Report.m ne short-term transient for SB19 encompassed the start of the simulation up to e
[
]a.b.c seconds. This period includes blowdown, natural circulation, ADS, and initial IRWST stages of the transient. For this test, the CMT refill occurred during the shon. term transient. At [
]a b.c seconds, CMT-1 began to refill, and CMT-2 followed [
Ja.b.c j
seconds later. This CMT refill phenomena is discussed further in Subsection 6.1.1 and is excluded from this discussion of the short-term transient.
O The analysis of the long-term transient for SB19 encompassed the time frame from e
[
Ja.b.c seconds to the end of the test. This phase of the transient included IRWST injection and covered the transition to sump injection. The long-term transient actually started with IRWST injection, which is discussed as part of the short-tenn transient. Between the end of the short-term transient and [
Ja.b.c seconds, the system remained relatively inactive, so this discussion begins at [
]a.b.c seconds.
The discussion of the short-and long-term phase of the transient focuses on important thermal-hydraulic phenomena identified in the PIRT (Table 1.3-1). The mass and energy balance results are key indicators of the quality of the analysis on which this discussion is based. These are discussed in detail in Subsections 6.2.2 and 6.2.3.
l l
O a:umnonwv2nwsi-ib.mo97 5.10-1 REVISION: 2 l
y 5.11 Analysis of Matrix Test SB21
(
Matrix Test SB21 (OSU Test U0021) simulated a double 4-in. cold-leg break LOCA with LTC and without the operation of the nonsafety-related systems. De breaks were located on the top and bottom of CL-3 and except for the break size, this test was similar to SB18 and SB01, including the simulated failure of one of the ADS-4 lines. Changes to the OSU facility since the performance of SB18 are noted in the Final Data Repon.0)
De analysis of Matrix Test SB21 is divided into three sections as follows:
Be facility performance is discussed in Subsection 5.11.1, which provides a brief outline of the response of the test facility; funher details are available in the Final Data Repon.(3)
The short-term transient for SB21 encompassed the start of the simulation up to
[
]a.b.c seconds. This period included the blowdown, natural circulation, ADS, and initial IRWST stages of the transient.
i ne analysis of the long-term transient for SB21 encompassed the time frame from e
[
Ja.b.c seconds to the end of the test. His phase of the transient includes IRWST injection and covered the transition to sump injection. De long-term transient actually started with IRWST injection, which is discussed as part of the short-term transient. Between the end of the short-term transient and [
]a.b.c seconds, the system remained relatively inactive with the exception of the CMT refill. At [
]a.b.c seconds, CMT-1 began to refill, and CMT-2 followed [
]a b.c seconds later. CMT refill phenomena is discussed further in Subsection 6.1.1, and the discussion of the long-term transient provided here begins at
[
]a b.e seconds.
De discussion of the short-and long-term phase of the transient focuses on important thermal-hydraulic phenomena identified in the PIRT (Table 1.3-1). He mass and energy balance results are key indicators of the quality of the analysis on which this discussion is based. These are discussed in detail in Subsections 6.2.2 and 6.2.3.
O a
a:umnonwv2um-51.non: b-111097 5.11-1 REVISION: 2
5.12 Analysis of Matrix Test SB23 Matrix Test SB23 (OSU Test UOO23) simulated an 0.5-in. break LOCA with LTC and without the operation of the nonsafety-related systems. De break was located at the bottom of CL-3 and except for the break size, this test was identical to SBl8 and similar to SB01, including the simulated failure of one of the ADS-4 lines. As noted in Section 1.5, the original scaling methodology used indicates tnat the selected OSU break area was larger than necessary for a true simulation of a 0.5-in. break.
However, the results are acceptable to validate codes since the calculations account for this variation when predicting test results.
De analysis of Matrix Test SB23 is divided into three sections, as follows:
ne facility performance is discussed in Subsection 5.12.1. It provides a brief outline of the e
response of the test facility; funher details are available in the Final Data Repon.W The short-term transient for SB23 encompassed the stan of the simulation up to
=
[
Ja.b.c seconds. This period includes the blowdown, natural circulation, ADS, and initial IRWST stages of the transient.
ne analysis of the long-term transient for SB23 encompassed the time frame from e
[
]a.b.c seconds to the end of the test. This phase of the transient includes IRWST injection and covered the transition to sump injection. The long-term transient actually started with IRWST injection, which is discussed as part of the shon-term transient. Between the end of the short-term transient and [
]a.b.c seconds, the system remained relatively inactive with the exception of the CMT refill. At [
]a.b.c seconds, CMT-1 began to refill, and CMT-2 followed [ ]a b.c seconds later. CMT refill phenomena is discussed further in Subsection 6.1.1, so the discussion of the long-term transient presented begins at [
]a.b.c seconds.
De discussion of the short-and long-tenn phase of the transient focuses on important thermal-hydraulic phenomena identified in the PIRT (Table 1.3-1). De mass and energy balance results are key indicators of the quality of the analysis on which this discussion is based. These are discussed in detail in Subsections 6.2.2 and 6.2.3.
1 i
oA2mwnonwv2s2Wst.nostb-tim 5.12-1 REVISION: 2
c-I'b) 6.1.3 Flow Oscillations During Long-Term Cooling Cyclic flow, pressure, level, and temperature oscillations were observed in about 65 percent of the OSU tests. These oscillations occurred during the latter stage of the IRWST injection phase of SBLOCA simulations. Several possible causes of these oscillations, evaluation of test data, and the extrapolated effect in the AP600 plant are identified in this section. The most likely cause of the oscillations is also presented in detail.
6.1.3.1 Introduction ne flow paths in the reactor vessel during IRWST injection are schematically illustrated in Figure 6.1.3-1. Liquid from the IRWST entered the downcomer through the two DVI lines. which are located 180 degrees apart, flowed down through the downcomer and up through the core, where heat addition from the electrical heater rods generated steam. A two-phase water / steam mixture exited the core. Liquid (and possibly some steam) was released from the reactor vessel through the hot legs and ADS-4 lines, which are open during this portion of the test. Steam was separated from the liquid in the upper head and flowed into the downcomer through a series of ten small holes in the downcomer top plate. This steam condensed on the surface of the cooler water in the downcomer.
He key features of the oscillations are summarized in Table 6.1.3-1. He oscillations exhibit several
/O common characteristics:
b The oscillations begin after net steam was generated in the core during IRWST injection.
=
He oscillations were regular with a period between 110 and 135 seconds.
The oscillations started gradually and end gradually (Figure 6.1.3-2).
For all the tests, the measured upper plenum collapsed liquid level oscillated around the top of the hot leg nozzles, and in all tests the measured downcomer collapsed liquid Ievel covered the cold-leg nozzles.
The difference in the oscillation start and stop time and the oscillation period is related to the break size and location. These two parameters affect the amount of the steam generated in the core and eventually, the amount of steam condensed in the downcomer.
Five mechanisms have been postulated to explain these osciPations. Each hypothesis is described in detailin Subsection 6.1.3.2. A detailed analysis of the oscillations for SB01 is presented in Subsection 6.1.3.3. He oscillations observed in the other tests are also briefly discussed in Subsection 6.1.3.3 in I comparison with the transient SB01.
7 l
o:us44wnonwv2\\2344w44 non.Ib.12:297 6.1.3 1 REVISION: 2
)
l i-
6.1.3.2 Proposed Hypotheses Flow oscillations were observed to occur during FLECHT-SET gravity-feed reflood testsO8)and have been analyzed. The oscillations were found to be natural oscillations, which are U-tube (manometer) oscillations due to the gravity force alone, with an oscillation period of three seconds. The vessel in the OSU test is shorter than that in the FLECHT-SET test. Since the period of natural oscillation is proportional to the height of the vessel, the period of natural oscillations in the OSU test is shorter than the three second period observed in the FLECHT-SET test. However, the observed oscillation period in the OSU test was [
]a.b.c seconds. Therefore, the oscillations in the OSU test are not natural oscillations, but are forced oscillations due to force induced by the steam generation and condensation.
In the OSU test, the driving force of the oscillations was determined to be the steam generated in the vessel, which caused the pressure in the vessel to increase. To have oscillations with repetitive cycles, this pressure build-up in the vessel was relieved by venting the steam. The possible vent paths are: (a) through the hot legs and the ADS-4 lines, (b) through the upper head bypass holes and the cold legs, and (c) through the upper head bypass holes with condensation at the top of the downcomer, with the injected DVI line flow as shown in Figure 6.1.3-1.
Five candidate hypotheses for the oscillations observed during LTC were investigated and are described and evaluated in this section. The candidate hypotheses are:
1.
Level fluctuations in the upper plenum opened and closed the steam vent path at the hot-leg nozzle. Level fluctuations were driven by pressure changes resulting from altemately covering and uncovering the hot-leg nozzle. No condensation cccurs in the downcomer.
2.
Slug flow in ADS-4 lines caused pressure surges when the steam slugs discharged into the separator by changing the two-phase flow regime and pressure drop in these lines.
- 3. He observed pressure fluctuations were driven by changes in condensation rates for steam flowing from the upper head and condensing in the downcomer.
4.
The observed pressure and level fluctuations were due to alternately covering and uncovering of the hot-leg nozzle and steam condensation in the downcomer.
- 5. De observed pressure and level fluctuations were due to alternately covering and uncovering of the upper head bypass holes.
The following sections describe assessment of the data and will show only Hypothesis 4 can be the probable cause of the oscillations.
O o:u344wnonwv2u344wanon: t b-l i i o97 6.1.3-2 REVISION: 1
That is, the oscillation periods in the AP600 will be 5.7 times longer than those in the OSU tests, i.e.,
about [' ]**'* minutes, relatively slow and stable.
'Ihe oscillations will not become unstable in the AP600, because the pressure in the vessel cannot increase beyond the value where the hot leg becomes uncovered and the level oscillation amplitude is limited by the distance between the hot leg and the highest elevation the mixture level can attain.
Furthermore, since the oscillations are the forced oscillations with oscillation period of order of 100 seconds, which is much larger than natural oscillation period of order of 2 seconds, the oscillations will not be amplified by the resonance. The core remains covered during the oscillations for the following reasons. When the mixture level dmps below the hot leg, steam vents through the hot leg and the vessel depressurizes. As the vessel depressunzes, the DVI injection rate increases and the hot-leg flow rate decreases. When the injection rate t-We larger than the hot-leg flow rate, the mixture level in the upper plenum rises and the hot leg will be covered. Therefore, the mixture level oscillates around the hot leg and keeps the cold leg covered during the oscillations.
6.1.3J Cooclusions Pressure, level and flow rate periodic oscillations have been observed in several OSU test during the i
l last period of the IRWST injection phase. Five mechamsms have been postulated to explain the oscillation process and are reported in Subsecuon 6.1.3.2. The fourth mechanism is consistent with all the tests. In Mechanism 4, the onset of the oscillations requires the following plant conditions:
- 1. Steam generation in the core -
- 2. Steam condensation in downcomer
- 3. Upper plenum liquid level covering the hot-leg nozzle 1
l 4.
Downcomer liquid level covering the cold-leg nozzles.
l As soon as the oscillations are established, the driving force is the plugging and unplugging of the steam vent path, which is typically ADS-4. In this process, the liquid mass injected in the system, is at first accumulated when the steam vent path to the ADS-4 is open, then discharged when the steam vent path is closed. The product of the oscillation amplitude (in terms of mass) and oscillation fr g-y is equal to the injection flow rate. 'Ihe greater the steam condensation in the downcomer condensation, the lower the frequency and the higher the amplitude.
i i
'Ihe SB01 oscillations have been analyzed in detail and the postulated Mechanism 4 is supported by I the calculations. All other tests have been analyzed briefly in comparison with transient SB01. 'Ihe oscillation period is about [
]** seconds in all the tests with the exception of SB01 where it is
[
J'M seconds and SB15 where it is [
]*M seconds. 'Ihe steam generation rate at the onset of the oscilianons is about [
]' lbm/sec. for most of the tests (SB01, SB14, SB15, SB18 and SB23).
oA2344woomkv2u34ew.64.nas:Ib.Illor7 6.1.3-29 REVISION: 2 1
l l
In SB19, it is lower ([
Ja,b,e Ibm /sec.) due to the break backpressure, which is holding the pressure in the downcomer while in SB09 and SBIO it is higher ([
].b.c lbm/sec.) because some steam is vented through the break at the top of the cold leg and this is reducing the pressure in downcomer. In SBl3, a very high value ([ Ja b,e Ibm /sec.) of steam genention rate is evaluated before the onset of the oscillations. This can be explained only if some steam is vented by the break in the DVI line or from the CMTs through the balance but there is no evidence of this in the SB13 data.
1 O
9
'"*'mamwammit>.11 ton 6.1.3-30 REVISION: 1 n
r i
/~";
7.1 Variations in Break N.
\\_,]
I The 2-in. break at the bottom of CL-3 is modeled by Test SB01 and SB18 in the OSU test matrix. In all tests simulating design basis LOCA events, the passive safety systems successfully reduced the RPV pressure to containment pressure, and stable IRWST injection wu achieved. Furthermore, LTC via sump injection was efficient in the tests which were carried out for that length of time.
Due to the low-pressure nature of the OSU facility, actuation of the passive safety-related systems based on pressurizer pressure, as would occur in a SBLOCA event for the AP600, was impractical.
Consequently, CMT and PRHR isolation valves were opened automatically early in each test at about the same time. Likewise, the OSU primary system pumps were tripped early, independent of break size, and the reactor decay power profile was very similar in all cases. Therefore, CMT and PRHR system behavior is similar at the beginning of all the SBLOCA break transients. The effect of break size on individual component performance for selected tests is discussed in the following subsections.
p l
[N tb o:\\2344wnon\\rev2\\2344-714.non: l b-121297 7,}.1 REVISION: 2 L
f' l
7.2.2 Core Makeup Tank Drain / Refill Behavior l
Refilling of the CMTs was commonly observed to occur in the OSU tests at times on the order of
[
]a.b.c of seconds into IRWST injection. However, during SB19, the test performed with a i
simulated containment backpressure of [ ]a.b.c psig, CMT refill began at about the time of [
3a.b.c, he behavior of the CMT refill following initial draindown, and the mechanism driving this behavior, l
was discussed in Subsection 6.1.1. His section presents a discussion about the test conditions observed during CMT reflood.
1 i
Break Size Effects Data from the four cold-leg break tests served as the basis for this comparison; Test SB23,0.5-in. cold-leg break I
Test SB01,2-in. cold-leg break Test SBM,4-in. cold-leg break Test SB21,4-in. top and bottom cold-leg breaks
=
The CMT-1 water level history for each of the preceding four tests is shown in Figure 7.2.2-1.~ From b
this figure, the following are observed; All four tests exhibited draindown and refill behavior.
De 0.5-in. break took the longest time to drain and did not initiate refilling until much later in the transient than the larger breaks. De CMT also refilled to a much higher level for the 0.5-in. break than did the other break sizes.
Both of the 4-in. break tests perfonned similarly with respect to time. De CMT did not drain to quite as low a level and refilled sooner and to a slightly higher level for the 4-in. top and bottom break than for the 4-in. bottom break.
ne draindown for the 2-in. break behaved similarly to that of the 4-in. breaks. However, the e
reflood, while occurring about [
ja.b.c seconds earlier in the transient and refilling to about l
6-in. lower in the CMT, was similar to that of the 0.5-in. break.
Similar draindown and refill behavior is noted for CMT-2 in each of these tests, as shown in
' Figure 7.2.2-2.
Figures 7.2.2-3 through 7.2.2-6 show the time histories of the water level in CMT-1, the CL-3/CMT-1 balance line, and the downcomer, as well as the pressure in the CMT for each of the four tests oA2344wnonWy2C344-722.non: l b. I l l l97 7,2,2-1 REVISION: I f
j c
l i
I identified above. These plots support the reflood mechanism described in Subsection 6.1.1; that liquid from a filled cold-leg balance line spilled into the CMT, causing steam to condense and pressure to drop, resulting in more liquid being drawn into the CMT. The refilling process was observed to terminate when the CMT pressure reached approximately [
]a.b.c ps a or the liquid level in the cold leg dropped below approximately [ ]a.b.c nches.
Hot Leg versus Cold-Leg Break The liquid level history of CMT-1 for the hot-leg break simulation, Test SB15, was compared to that I of the 2-in, cold-leg break. Test SB01, in Figure 7.2.2-7. The plot shows that the draindown behavior of both breaks was similar, with the hot-leg break refilling about [
]a.b.c seconds earlier than cold-leg break. Both transients resulted in the CMT being reflooded to about the same elevation.
Comparing the history of the liquid level in CMT-1, CL-3/CMT-1 balance line, and the downcomer with the CMT-1 pressure history as shown in Figure 7.2.2-8, it is again noted that there was a pressure drop in the CMT consistent with high water levels in the downcomer and the balance line, so that after the pressure reached a minimum, the water level in CMT-1 was observed to begin increasing.
DVI Line Breaks The CMT draindown and refill behavior was noted to be dependent upon the size of the DVI line break. As shown in Figure 7.2.2-9, for the 2-in. DVI line break, Test SBl3, CMT-1 drained in a manner similar to that observed for Test SB01 and reflooded at about the same time, but to a level approximately [ Ja.b.c-n. higher than was observed in Test SB01. For the 2-in. DVI line break, CMT-I completed the second draindown by about [
Ja.b.c seconds into the test, or about [
ja.b.c seconds before CMT-1 drained for the second time in the reference 2-in. cold-leg break test. For Test SB12, the double ended guillotine DVI line break, CMT-1 drained quickly and did not reflood during the test.
As shown in Figure 7.2.2-10, CMT-2 was observed to drain fastest for the double ended guillotine DVI line break (Test SB12), with Tests SB13 and SB01 exhibiting similar draindown behavior.
Similar to the broken loop CMT-1, the intact loop CMT-2 did not refill in Test SB12. However, in Test SBl3, CMT-2 exhibited a refill behavior similar to that observed in SB01, although delayed by about [
Ja.b.c seconds. Also, for SB13 CMT-2 redrained much more quickly than it did in Test SB01, completing its second draindown by about [
Ja.b.c seconds into the test.
Figure 7.2.2-11 shows that the liquid levels in both the downcomer and the cold-leg balance line were high at the time CMT-1 reflood initiated during Test SB13. As CMT-1 was observed to refill, the local pressure in the CMT remained low, about [ ]a.b.c psia. However, for Test SB12, the double-ended guillotme (DEG) cold-leg balance line break, Figure 7.2.2-12 shows the water level in the cold-leg balance line was too low (about [ ]a.b.c in.) throughout the test to be drawn into CMT-1 and the CMT did not refill.
on23awnonwvn23u.722.non ib.1i n97 7.2.2-2 REVISION: 2
L l
l
' ' p This information, combined with the preceding observations for CMT-2 performance for both DVI line V
break tests, suggests that the mechanism for CMT reflood was water from the cold-leg balance line being drawn into the CMT, resulting in condensation that reduced the pressure causing additional water to be drawn into the CMT, resulting in refilling of the CMT as described in Subsection 6.1.1.
Cold Leg Balance Line Breaks As expected, the draindown and refill behavior of the CMTs was observed to be dependent upon their location with respect to the break location. As shown in Figure 7.2.2-13, the response of broken loop CMT-1 to a 2-in cold-leg balance line break was similar to that of the reference 2-in. cold-leg break, ahhough the initial draindown was delayed by about [
Ja.b.c seconds, or until about [
]a.b.c seconds into the test. During SB01, CMT-1 refilled to an elevation of about [
]a.b.c in., while in j
Test SB09,it was observed to not refill. For the DEG birak, the CMT draindown was not started until about [
]a.b.c seconds after initiation of the test, and was a gradual process that continued until about [
]a.b.c seconds into the test; in Test SB10, CMT-1 was not observed to refill.
From Figure 7.2.214, however, CMT-2 was observed to reflood for all three tests with the reflood occurring earliest for the DEG cold-leg balance line break. The DEG cold-leg balance line break was also observed to refill and hold a water level that was about [ ]a.b.c in. grea*er than either of the other two breaks. It was also noted that, for both of the balance line breaks, the second draindown was completed at about the same time, which was about [
]a.b.c seconds before that for base case f
\\
Test SB01.
l From Figure 7.2.215, it was noted that the liquid level in the downcomer for Test SB09 remained high over the time period in which CMT refill might be expected; from approximately [
]a b.c seconds. However, since the primary system break was in the CL-3/CMT-1 balance line, it I
could not fill and promote the reflooding of CMT-1. Similarly, the break in CL-3/CMT-1 balance line for Test SB10 precluded CMT reflood.
However, the intact cold-leg balance line did promote refill of CMT-2 Figure 7.2.2-16. The requisite l
conditions were present in both Tests SB09 and SB10 to support drawing of water from the cold-leg into CMT-2; high downcomer level (> [ Ja.b.c n. of water), high cold-leg balance line level, and low l
CMT pressure. These observations further support the mechanism of CMT reflood as described in Subsection 6.1.1.
Inadvertent Automatic Depressurization System l
l The inadvertent ADS event was noted to produce an initial draindown of CMT-1 that was similar to that observed in the reference 2-in. cold-leg break, Figure 7.2.2-17. 'Ihe CMT refill occurred at about the same time in both tests, but was observed to result in only about [ Ja.b.c in. of water being drawn into the CMT compared to the approximately [ ).b.c in. for the reference test. Comparing water O
levels in the cold-leg balance line, the downcomer, and the CMT, Figure 7.2.218, it was noted that l
o:u344wnonvn2u3u.722_non:ib.li 197 7.2.2-3 REVISION: 1
the downcomer water level, and therefore, the cold-leg water level decreased below about [
]a.b.c-n.
at about the time CMT-1 refilling was terminated. Also at that time, the CMT pressure recovered to a nominal value of about [
]a.b.c psia. Thus, although at a reduced pressure, the CMT could not draw fluid from the downcomer as its liquid level had dropped below the cold leg.
He level instrument for CMT-2 was not operable for Test SB14. However, data from channels LDP-504, LDP-506, and LDP-508, which collectively spanned the same elevation as the overall level measurement of LDP-502, indicated the CMT-2 ref;ll behavior was similar to that of CMT-1; CMT-2 drained within about the first [
Ja.b.c seconds of the test, refilling began at about [
]a b.c seconds into the test with a maximum refill level of about [ Ja b.c nches of water reached at about
[
]a.b.c seconds, and the second drain of the tank completed at about [
]a b.* seconds into the test.
Containment Backpressure Effects Containment backpressure, as simulated in Test SB19, was noted to have some affect on the CMT I refill behavior compared to the behavior noted in Test SB01. Specifically, for Test SB19, both CMTs were observed to stop their initial draining before becoming completely empty, and refilled quickly (Figures 7.2.2.-19 and 7.2.2-20). This suggests the higher system pressure, and therefore, higher steam temperatures in the CMTs, promoted higher condensation rates on the inside surfaces of the CMTs early in Test SB19, which, in turn, resulted in lower CMT pressures relative to the cold leg earlier in the test. Combined with the earlier IRWST injection, also a consequence of increased containment backpressure, which provided for high downcomer water levels earlier than occurred in reference Test SB01, the low CMT pressures early in Test SB19 allowed the CMTs to draw water from the cold-leg I balance lines earlier than was observed in Test SB01. Thus, the results of Test SB19 demonstrate that l
the early IRWST injection, promoted by the increased containment backpressure, provided for higher I water levels in the downcomer and cold legs earlier than was observed in Test SB01.
Both CMTs were also observed to complete their second draindown by about [
l Ja.b.c seconds before the same event occurred in Test SB01. Comparing the liquid level in the downcomer and the cold-leg balance line for bot!' CMT-1 and CMT-2, Figures 7.2.2-21 and 7.2.2-22 respectively, it is noted that the downcomer fluid level was above [ ]a.b.c nches until the CMTs completed their second draindown. Also, as the CMTs approached complete draindown, the fluid level in the cold-leg balance lines dropped to about [ Ja.b.c in. and the CMT pressure increased. This suggests that the drop in water level in the downcomer and cold-leg precluded the inflow of water into the CMT, resulting in complete draining of the tank.
O 1
oA2344wnonsrev2\\2344 722.non:Ib-111197 7.2.2 4 REVISION: 2
/^
9.0 REFERENCES
1.
Dumsday, C.L., et al, AP600 Low-Pressure Integral Systems Test at Oregon State University Final Data Report, WCAP-14252, May 1995.
2.
Reyes, J.N., Hochreiter, L.E., Lau, L.K., and A.Y. Lofi, Low Pressure Integral Systems Test Facility Scaling Report, WCAP-14270, January 1995.
3.
Cunningham, J.P., et al., AP600 SPES-2 Test Analysis Report. WCAP-14254, May 1995.
4.
Kemper, R.M., Applicability of the NOTRUMP Computer Code to the AP600 SSAR Small-Break LOCA Analysis, WCAP-14206, November 1994.
5.
Hochreiter, L.E., FLECHT SEASET Program Final Report, NUREG/CR-4167. EPRI NP-4112, Prepared for the U.S. Nuclear Regulator Commission, November 1985.
6.
Modro, S.M., et al., Evaluation of Scaled Integral Test Facility Conceptsfor the AP600, Idaho National Engineering Laboratory, SMM-27-91, Transmitted to U.S. NRC, July 1991.
7.
U.S. Nuclear Regulatory Commission, An Integrated Structure and Scaling Methodologyfor
("
Severe Accident TechnicalIssue Resolution, NUREG-CR-5809, (l991).
8.
Taitel, Y. and A.E. Dukler, A Modelfor Predicting Flow regime Transitions in Horizontal and Near Horizontal Gas Liquid Flow, AIChE Journal, Vol. 22, No.1, pp. 47-54, January 1976.
9.
Schwartzbeck, R.K., and G. Kocamustafaogullari, Similarity Requirementsfor Two-Phase Flow-Pattern Transitions, Nuclear Engineering and Design,116, pp.135-1457,1989.
10.
Bankoff, S.G., and S.C. Lee, A Critical Review of Flooding Literature, NUREG-CR-3060, (1983).
11.
Kocamustafaogullari, G. and M. Ishii, Scaling Criteriafor Two-Phase Flow Natural and Forced Convection Loop and Their Application to Conceptual 2x4 Simulation Loop Design, Argonne National Laboratory, ANL-83-61, NUREG/CR-3420, May 1983.
12.
Moskal, *Ihomas E., Examination ofScaling Criteriafor Nuclear Reactor Thermal-Hydraulic Test Facilities, Ph.D. Dissertation, Camegie-Mellon University, Pittsburgh, PA, March 1987.
13.
Meyer, P.E., NOTRUMP, A Nodal Transient Small-Break and General Network Code, WCAP-10079-P-A, (1985).
I a:us44wnonwv2c344w.non:ib-11:197 9-1 REVISION: 1
)
14.
Cunningham, J.P., et al., AP600 Core Makeup Tank Test Analysis, WCAP-14215, 1
December 1994.
15.
Baker, O., Design of Pipe Linesfor Simultaneous Flow of Oil and Gas, Oil and Gas Joumal, July 26,1954.
16.
McAdams, W.H., Heat Transmission, 3rd ed., McGraw-Hill Book Company, New York,1954.
17.
Chamber, Cleveland, Kleiner, and Tukey, Graphical Methodsfor Data Analysis, Wadswonh, 1983.
18.
Yeh, H.C. and Hochreiter, L.E., An Analysis of Oscillations in Simulated Reflood Experiments.
American Nuclear Society, Trans. Vol. 22, p. 490,1975.
19.
Holman, J. P., Heat Transfer, McGraw-Hill,1976.
20.
Andreychek T. S., et al., AP600 low-Pressure Integral Systems Test at Oregon State i
University Test Analysis Report, WCAP-14292, Rev. 0, July 1995.
21.
Westinghouse Electric Corp., }XCOBRA/ TRAC OSU Long-Term Cooling Preliminary Validation Report, LTCT-GSR-003, August 1995.
22.
Griffith, P., Notes on Two-Phase Pressure Drop Through Fittings and Valves, MIT Two-Phase Flow course,1972.
23.
Beck, J. V., Blackwell, B., and St. Clair, C. R. Jr., inverse Heat Conduction-lll Posed Problems, Wiley-Interscience Publications,1985.
24.
Idelchik, I. E., Handbook of Hydraulic Resistance, Third Edition,1994.
I 25.
Brown, W. L., et al., AP600 Scaling and P/RT Closure Report, WCAP-14727. Rev.1, 1
July 1997.
I l 26.
Hochreiter, L.E., et al., Low Pressure Integral Systems Test Facility Description Report, I
WCAP-14124, Rev. O, July 1994.
O c:ChnonWv2Chnon:Ib.llll97 92 REVISION: 2