ML18213A091
| ML18213A091 | |
| Person / Time | |
|---|---|
| Issue date: | 07/31/2018 |
| From: | Tammie Rivera NRC/RES/DRA/FRB, Electric Power Research Institute |
| To: | |
| Shared Package | |
| ML18213A072 | List: |
| References | |
| Download: ML18213A091 (27) | |
Text
Module III - Fire Analysis Incipient Detection Joint EPRI/NRC-RES Fire PRA Workshop August 6-10, 2018
Objectives Understand
- what constitutes a very early warning fire detection system
- different types of smoke detection technologies
- current regulatory guidance Present overview of
- completed research
- quantitative risk-scoping study 2
Outline Background on smoke detection principles and technology What constitutes very early warning fire detection system Current regulatory guidance on the use of Incipient Fire Detection Research testing program and accompanying risk-scoping study 3
Smoke Detection Principles Fire produces a variety of changes in the ambient conditions
- fire signatures (aerosol, energy release, gas, transport, etc.)
Smoke detection technologies generate an electrical signal When selecting a smoke detector, the highest signal to noise ratio in the earliest period of fire development is preferred Smoke detectors monitor the signal and alarm when a setpoint is reached.
- Setpoint is typically reported as percent per foot obscuration (%/ft obsc.)
Modern smoke detectors are capable of having the set point configured from a fire alarm control panel.
4
Smoke Detector Technologies Photoelectric
- Light obscuration
- Light-scattering Ionization Cloud Chamber Video Image Photos Ref. NFPA Fire Protection Handbook Section 14, Chapter 2, Twentieth Edition 5
Spot Type vs Air Sampling Detectors (ASD) 6
Very Early Warning Fire Detection (VEWFD)
System NFPA 76, Fire Protection of Telecommunications Facilities, defines VEWFD systems as, Systems that detect low-energy fires before the fire conditions threaten telecommunications services.
NFPA 76 specifies, in part, a VEWFD shall
- alert sensitivity of 0.20%/ft obsc. / alarm sensitivity of 1.0%/ft. obsc.
Both setpoints are above ambient and at each sampling point (port/detector).
- 200 ft2 coverage for area wide / 4 ft2 coverage for air return grill 7
Use in U.S. NPP Fire PRA Use in nuclear power plants (NPPs) is to provide advanced warning
- Provide more time for operator response
- More time = higher success of suppressing fires Some plants have used ASD since mid-1990s as risk reduction measure
- Robinson IPEEE Some plants use ASD as an enhanced detect system without quantifying performance via risk assessment
- TMI, exemption for thermo-lag 330 performance issue.
Interest in crediting VEWFDS in fire PRA 8
FAQ 08-0046, Incipient Fire Detection (2009)
Provided an interim staff position regarding the use of VEWFD systems in fire PRA Determines the probability of non-suppression (Pns)
- Event tree used to quantify
- Structure and several parameter estimates based on EPRI 1016735 Several limitations on use
- Only electrical enclosure less than 250V
- Only in-cabinet detection (not applicable to area-wide)
- Fast acting components are ratio out FAQ based on NFPA standard objectives and system performance expectations provided by vendors.
- DATA NEEDED 9
FAQ 08-0046 Simplified Event Tree Detector System Fire Initiating Successful Operator Availability and Fire Suppressed End Point Event Response to Alert Reliability (1-) (1-) (1-1) No Fire Damage to Targets OK Outside Cabinet 1 Fire Damage NS (1-2) No Fire Damge to Targets OK Outside Cabinet 2
Method results in Fire Damage NS Pns of 0.02 or less (1-2) No Fire Damage to Targets OK Outside Cabinet 2 Fire Damage NS 10
NRC/RES Confirmatory Research Program Initiated to provide data Evaluates
- Operating experience (USA/Canada, Nuclear/Non-Nuclear)
- Literature (vendor, journal, standard, listings, codes of practice)
- Operator performance
- System performance via testing Provides
- Test results
- Risk scoping study Parameter estimates based on operating experience (fire events database), testing, fire PRA methodology Documented in NUREG-2180, DELORES-VEWFIRE 11
Experimental Approach Detectors
- 5 ASD VEWFD systems
- 1 VEWFD spot-type
- 2 conventional Scales of testing
- Laboratory 1 small cabinet, 2 full size cabinets from Bellefonte
- Full scale small room 2 banks of up to 5 fully interconnected cabinets Area-wide - ceiling & air return
- Full scale 2 banks of 5 cabinets with up to 3 partially interconnected Area-wide - ceiling & air return 12
Laboratory Scale Testing Small Cabinet 13
Laboratory Scale Testing Large Cabinets 14
Full Scale -
Small Room 15
Full Scale - Large Room 16
Smoke Source Had to be developed to mimic a prolonged overheating incipient stage condition
- Cartridge heater in copper bus bar with material attached to exterior
- Material is elevated to piloted ignition temperatures
- Modified cable bundle used in some tests
- 3 heating ramp periods (HRPs), 15-, 60-, 240-minutes 17
Generic Results 18
Risk Scoping Study Quantify probability of non-suppression (Pns) using event tree model Event tree parameters estimated using available operating experience, test data, human reliability engineering tools, and non-suppression curves.
Two Event Trees
- In-Cabinet
- Area-wide 19
In-Cabinet Event Tree Detector System System Effective Successful Field Operator Conventional Fraction of Fires that Successful MCR Fire Avaliability, Detecting Incipient Response Enhanced Suppression Detection / End State have an Incipient Stage Response Reliability Stage (Fire Watch Posted) Suppression i x SF 1- 1- 1- 1- 1- 1-1 OK Cabinet Damage 1 1-3 OK Cabinet Damage 3
NS Fire Damage Outside Cabinet 1-2 OK Cabinet Damage 2
NS Fire Damage Outside Cabinet 1-1 OK Cabinet Damage 1
NS Fire Damage Outside Cabinet 1-2 OK Cabinet Damage 2
NS Fire Damage Outside Cabinet 1-2 OK Cabinet Damage 2
NS Fire Damage Outside Cabinet 1-1 OK Cabinet Damage 1
NS Fire Damage Outside Cabinet 20
Area-Wide Event Tree Detector System Fraction of fires that System Effective Successful Field Conventional Successful MCR Enhanced Fire Availability, have an incipient Detecting Incipient Operator Response Detection / End State Response Suppression Reliability stage Stage (Fire Watch Posted) Suppression i x SF 1- 1- 1- 1- 1- 1-2 OK Cabinet Damage 2 1-3 OK Cabinet Damage 3
NS Fire Damage Outside Cabinet 1-2 OK Cabinet Damage 2
NS Fire Damage Outside Cabinet 1-1 OK Cabinet Damage 1
NS Fire Damage Outside Cabinet 1-2 OK Cabinet Damage 2
NS Fire Damage Outside Cabinet 1-2 OK Cabinet Damage 2
NS Fire Damage Outside Cabinet 1-1 OK Cabinet Damage 1
NS Fire Damage Outside Cabinet 21
Conclusions For area-wide applications, the ASD VEWFD systems outperformed conventional spot-type detectors For in-cabinet applications, the ASD VEWFD systems performed inline with in-cabinet ionization detectors Risk scoping study provides a more detailed evaluation of the performance of smoke detection systems used to protect electrical enclosures The Pns estimated by the risk scoping study is more scenario specific and time dependent, rather than a one size fits all approach (FAQ 08-0046) 22
Research Insights
%/ft. obscuration may not be the best metric to evaluate the generic performance of VEWFD systems
- Smoke characteristics vary by material, degradation rate and mode
- Light scattering may perform better detecting large particles (e.g., CSPE), while cloud chamber/ion may perform better detecting a large number of small particles (e.g., ETFE)
Cloud chamber difficult to verify system meets NFPA 76 sensitivity settings 23
Spreadsheet Tools Available 24
Format Two Sheets (in-cabinet; area-wide)
Follows NUREG-1805 approach
- Inputs
- Results
- Solved Event Trees Estimates the Pns for damage outside cabinet Application specific damage states not addressed 25
Appendix H User guide for VEWFD event tree non-suppression probability calculation tool Provides a step-by-step procedure for using the excel spreadsheets.
26
Reminder NUREG-reports are a technical document NUREG-reports do not constitute regulatory guidance, unless endorsed via a RG or other means (SER, etc.)
27