ML14315A087

NRR E-mail Capture - STP Slides for 11/6/14 Public Phone Call

ML14315A087
Person / Time
Site:	South Texas
Issue date:	11/04/2014
From:	Harrison A South Texas
To:	Balwant Singal Division of Operating Reactor Licensing
References
Download: ML14315A087 (30)

Text

NRR-PMDAPEm Resource From: Harrison Albon [awharrison@STPEGS.COM]

Sent: Tuesday, November 04, 2014 4:45 PM To: Singal, Balwant; Stang, John

Subject:

STP Slides for 11/6/14 Public Phone Call Attachments: STP Head-Loss Testing for Addressing GSI-191 - 110614 - Final.pdf 1

Hearing Identifier: NRR_PMDA Email Number: 1681 Mail Envelope Properties (8C918BCF8596FB49BD20A610FA5920CF02182452)

Subject:

STP Slides for 11/6/14 Public Phone Call Sent Date: 11/4/2014 4:44:54 PM Received Date: 11/4/2014 4:46:33 PM From: Harrison Albon Created By: awharrison@STPEGS.COM Recipients:

"Singal, Balwant" <Balwant.Singal@nrc.gov>

Tracking Status: None "Stang, John" <John.Stang@nrc.gov>

Tracking Status: None Post Office: CEXMBX04.CORP.STPEGS.NET Files Size Date & Time MESSAGE 3 11/4/2014 4:46:33 PM STP Head-Loss Testing for Addressing GSI-191 - 110614 - Final.pdf 668820 Options Priority: Standard Return Notification: No Reply Requested: No Sensitivity: Normal Expiration Date:

Recipients Received:

STP Strainer Testing to Address GSI-191 South Texas Project Nuclear Operating Company Alion Science and Technology University of New Mexico Thursday, November 6, 2014 1

Desired Outcome

• Establish common understanding of STP strainer tests

- Purpose and scope

- Procedures

- Facility and diagnostics

• Identify and understand NRC staff issues with test plan

• Focus continued discussion for resolution of issues in follow-on public meeting 2

Review of STP RIR Head Loss Approach

• Current Licensing Amendment Request

- Modified NUREG/CR-6224 correlation for conventional P

• Challenged on bases for: inconsistent derivation, inconsistent prediction, uncertainty assessment (x5)

- Exponential factors for chemical P

• Challenged on multiplicative effect and bases for S,M,L magnitude

• Requests for Additional Information

- HTVL data used for initial calibration of VISTA correlation

• Reynolds scaling illustrates wider relevance of available data

• Alternate quantification of realistic response for model uncertainty

- Strainer performance data used for initial L* (L-star)

• Additive description of chemical head loss

• New interpretation of limited data

• Audit

- STP committed to develop and apply L* additive chemical approach

- NRC emphasized need for head-loss data under representative performance conditions 3

Background

• LAR calculates Total Head Loss (THL) head loss as follows:

THL = 5 X H X M; where:

- H is from NUREG/CR-6224 using theoretical compression

- M is a chemical effects multiplier conditioned on break size, bed thickness, and sump temperature

• Revised approach THL = NCHL + CHL; where:

- NCHL is calculated using VISTA head loss

- CHL is chemical head loss based on L* head loss 4

High-Level Plan for Test and Analysis

• New horizontal flume at UNM to test full-scale strainer module (January shakedown)

• Four, multi-day tests (finished end of March)

• Data refine and validate L* additive chemical P

• Data calibrate VISTA correlation of strainer response

• Schedule supports current resolution plan 5

Application of Test Results

• STP closure path is based on analytic evaluation of full accident spectrum with quantitative understanding of uncertainty

• VISTA correlation of HTVL data will be compared to nonchemical strainer response

- Porosity controlled by compaction is the only factor lacking direct measurement (dominant uncertainty)

• Refined L* for chemical head-loss increment will be added to VISTA predictions to obtain total head loss

• Tests provide calibration of L* and validation of VISTA to STP strainer response

• Tests provide validation of composite analytic evaluation to actual STP strainer response 6

Available Test Data

• Existing STP HTVL tests (Alion)

- fiber and particulates only

• Existing STP flume tests (ARL)

- fiber, particulates and WCAP surrogate

• Existing and planned Vogtle HTVL tests (UNM)

- fiber, coatings and WCAP surrogate

• Existing Sv characterization (UNM)

• Planned STP strainer tests (UNM) 7

Available Data Assessment

• Previous strainer qualification testing

- Some tests meet debris transport goals

- No tests with revised mass ratios

- All tests with 120F tap water

- All tests with WCAP-16530 surrogate chemical

• Alion HTVL tests with diverse loading

- Used for initial calibration of VISTA correlation

- Concerns about nonprototypic bed formation

• No chemical products of concern were identified during prototypical testing 8

Motivation for Testing

• Additional data needed to support L*, additive chemical head loss method

- Most likely, low chemical loads not well characterized

- Additional debris + chemical combinations needed

• Demonstrate physical connection between Reynolds scaling of VISTA and plant flow conditions

- Measure Re of debris-loaded strainer (with uncertainty)

- Identify any gaps in HTVL database

• Quantify magnitude and causes of uncertainty between analytic head-loss evaluation and strainer module performance

• Obtain performance data at the transition to potential strainer failure (most effect on risk) 9

Current L* Additive Chem Head Loss 10 9

8 7

6 CHL (ft) 5 4

3 2

STP strainer test data 1 Correlation 0

0 500 1000 1500 2000 2500 3000 L* (g/m2)

• Illustrates analytic methodology for data interpretation

• Prior tests useful for uncertainty quantification

• Not based on current debris loads

• Not designed to support L* quantification 10

Generate Additional L* Data 10 9

Hypothetical Trends 8

7 6

CHL (ft) 5 4

3 2

STP strainer test data 1 Correlation 0

0 500 1000 1500 2000 2500 3000 L* (g/m2)

• Focus batch resolution on low concentrations

• Check L* envelope with alternate conventional beds 11

Purpose of VISTA

• Reynolds scaling provides common basis for comparing all available data

• Common basis essential to select, define, defend desired tests

• Common basis essential to demonstrate testing sufficiency

• Rigorous treatment of residual uncertainty (compression)

- Calibrate factor of 5 to actual percentiles of variation

• Provides correlation to strainer data under representative performance conditions

• Addresses recognized analytic deficiencies of NUREG/CR-6224 Provides alternate view of real behavior to judge model uncertainty

• Correlation is essential to diagnose and evaluate subtle interactions in accident space

• Correlation is essential to accurately associate infinite debris combinations

• Final application to risk quantification will fully acknowledge uncertainties 12

VISTA Uncertainty Quantification

• Model for conventional debris head loss

• Familiar scaling for a range of flow conditions

• In a flume test:

- Constant particle-to-fiber mass ratio

- Water properties known from in-situ characterization

- Velocity known by pump flow measurement

- ONLY average bed porosity and bed thickness across strainer are UNKNOWN

• Formal uncertainty propagation can be used to calibrate the uncertainty factor applied to all predictions (like defining the 95th percentile) 13

Planned FIESTA Tests

• FTA-000 (Shakedown)

• FTA-100 (Clean Strainer)

• FTA-200 (1/4th inch contiguous bed)

• FTA-300 (1/16th inch thin bed)

• FTA-400 (DBA load)

• Exact loads to be determined by CASA Grande spectrum analysis

• Intermediate batches in FTA-200 will help define threshold for FTA-300 14

High-Level Test Description

• Strainer module testing using protocols similar to those used in the STP 2008 flume tests

• 4 tests with an STP module in a flume facility

- 1 clean strainer + 3 debris/chem series

• Target steady state test conditions emphasize maximum plant vulnerability (minimum NPSHmargin)

- Maximum flow rate and temperature similar to time of recirculation

• Two Steps: (1) conventional debris + (2) chemical product

- Temperature and velocity sweep before chemical addition

• Debris tests incorporate prototypical amounts of particulate (e.g. paint from STP reduced inventory)

• Debris tests use incremental additions of WCAP chemicals based on 30 days of precipitate formation

- L* head loss per area per chemical added across full conc. range 15

Typical Flow Sweep

• Rates based on flume heat exchange

• Min/max P variation limited to 25% of steady

• Add more debris if bed does not return to prior P 16

WCAP-16530 Surrogate

• Standard external preparation will be adopted for consistency with industry practice

- Standard external preparation is not likely to produce realistic chemical products or quantities - difficult to judge uncertainty

• Standard room temperature settling/

storage/mixing procedures

• Introduction to higher temperature flume

- Bench tests will quantify possible redissolution

- Compensate mass inventory to preserve total solid 17

Chemical Corrosion Inventory

• Chemical batches added to pre-established fiber and particulate bed

• Batches provide incremental response independent of total inventory added

• Finer batch resolution will capture:

- First 2-3 days of potential passivation (UNM) day inventory (UNM corrosion)

• Coarse batch resolution will capture:

day inventory (x5 UNM corrosion) day inventory (WCAP corrosion) 18

External Chemical Preparation 19

FIESTA Facility Description

• Use of STP strainer module

• Polycarbonate channel custom-built to achieve representative approach velocity

• Procedures to encourage near 100% transport STP Strainer Module from Alden Test (2008) 20

FIESTA Facility Description

• Dimensions: 32 (L) x 6 (H) x 4 (W) STRAINER POLYCARBONATE CHANNEL WALLS

• Polycarbonate walls with steel structure and support system

• Inline heating

• Multiple drains DRAINS on flume floor to track debris transport UNM FIESTA Facility 21

FIESTA Facility Description

• Temperature Range: 25 C to 85 C (nominal test operation at 55 C)

- Temperature cycling possible

• Prepared chemical precipitates added via inline system from auxiliary tank

• Debris added using industry standard methods

• Intent to achieve near 100% debris transport 22

FIESTA Primary Instrumentation Diagnostic Purpose Flume Sampling Frequency Mode Volumetric flow rate Face velocity on the test 0.1 to 0.02 Hz (every 10 to 50s) Online module Differential Pressure Hydraulic loss through the matched with flow samples Online Array debris bed. Clean strainer response along the module.

Static pool pressure Fluid density in combination matched with flow samples Online with level Liquid Temperature Fluid properties, pH correction matched with flow samples Online Room Temperature Differential pressure correction matched with flow samples Online Atmospheric Pressure Pump NPSH Once per 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> or as specified in test Spot Read plan Liquid Level Fluid density, chemical conc, 4 times per hour or as specified in test Online or Spot water make up. plan Read Total pipe volume Chemical conc. NA once pH Chemical debris preparation Once per 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> after first chemical Bench Reading and flume test conditions add or as specified in test plan ICP Flume concentration for mass Once per 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> after first chemical Grab Sample balance add, or as specified in test plan Mass balance Milligram to kg accuracy for NA As needed surrogate chemical preparation Particle sizing Characterize surrogate NA As needed chemical product Light table Verify debris preparation NA Each debris 23 batch

FTA-000 (Shakedown)

• Instrument operation/calibration

• Perfect and train debris preparation/storage

• Perfect and train chem preparation/storage

• Perfect and train cleaning procedures

• Demonstrate near 100% debris transport

- Narrow flume

- Mild agitation

- No bed disturbance

- Post test fiber recovery for transport calibration 24

FTA-100 (Clean Strainer)

• Objectives:

- VISTA correlation for clean strainer using different fluids

• Sequential phases of one test:

- DI water

- DI water + baseline

- DI water + baseline + chem surrogate

• Procedure:

- Establish fluid condition

- Vary temperature and velocity

• Benefit:

- Reliable interpretation of composite head loss

- Possible direct measurement of fluid viscosity and density 25

FTA-200 (1/4th in. thin bed)

• Objectives:

- Strainer performance near lower limit of contiguous bed (most likely condition)

- Assess filtration transition during loading (use in FTA-300)

- Establish L* for alternate bed configuration

• Sequential phases of one test:

- Batch addition of premixed fiber + particulate

- Batch addition of prepared chemical surrogate

• General Procedure:

- Establish fluid condition

- Incremental debris addition

- Flow sweep

- Incremental chemical product addition 26

FTA-300 (1/16th in. thin bed)

• Objectives:

- Dominant risk condition if adverse response

- Establish L* for alternate bed configuration

• Sequential phases of one test:

- Batch addition of premixed fiber + particulate

- Batch addition of prepared chemical surrogate

• General Procedure:

- Establish fluid condition

- Incremental debris addition

- Flow sweep

- Incremental chemical product addition 27

FTA-400 (DBA Load)

• Objectives:

- Compare results to previous ARL test under similar conditions

- Assess degree of uncertainty in L* for extreme debris conditions

• Sequential phases of one test:

- Batch addition of premixed fiber + particulate

- Batch addition of prepared chemical surrogate

• General Procedure:

- Establish fluid condition

- Incremental debris addition

- Flow sweep

- Incremental chemical product addition 28