ML20038C402
| ML20038C402 | |
| Person / Time | |
|---|---|
| Issue date: | 11/24/1981 |
| From: | Murley T Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20038C400 | List: |
| References | |
| REF-10CFR9.7 NUDOCS 8112100573 | |
| Download: ML20038C402 (23) | |
Text
_
i AAB
.o m,,
4 "55 98 "O
L" E
PRESSURIZED THERMAL SH0CK OF PRESSURE VESSELS o
E*
COMMISSION BRIEFING
=
NOVEMBER 24,1981 1
\\
i
i PRESSURIZED THERMAL SHOCK OF PRESSURE VESSELS COMMISSION BRIEFING
- 1. Current Status ll. Recent Information 111. Staff Actions IV. Staff Conclusions T. E. Murley November 24,1981
\\
l
-4 9
1 I
N D
<C 1-W
&ZLU M
D O
m 4
PRESSURIZED THERMAL SHOCK t
There Is No Clear Cut Time When a PWR Vessel Changes from Safe to Unsafe It is Largely a Matter of Judgment, Which Depends on a Good Understanding of the Risks involved
-- Frequency of Overcooling Transient
-- Severity of Overcooling Transients
-- Probability of Repressurizing the Primary System
-- Existence of a Flaw of Sufficient Size to.Propogate at the Location of High Thermal Stress
-- Material Properties of Plate and Weld Metal
-- Irradiation History of Vessel (Neutron Embrittlement) i 1
\\
.~
SCHEMATIC CURVE OF OVERC00 LING TRANSIENT PROBABILITY Rancho Seco Transient PROBABILITY OF OMC00 LING Severe Transient Which Would TRANSIENT ause Vessel Cracking SEVERITY OF TRANSIENT
\\
l
SOURCES OF MARGIN IN FRACTURE MECHANICS ANALYSES
-- Assumed Presence of Flaw in Critical Weld (Highest Fluence)
-- Actual Shift in RT Will Probably Be Less Than RG l.99 NDT
-- Code Allowable Toughness Curves (K IA) Are Lower Bound Curves IC
-- Code Allowable Toughness Curves Are Assumed to Shift by the Same Amount as the Charpy Curves. This is Believed to Be Conservative
-- Real Crack Stress Intensity Will Be Lower Than Assumed Because Real Cracks Will Be Irregular and Not Perfectly Sharp
-- Warm Prestressing May Be Effective in Some Transients
-- Cladding Could Reduce Stress Intensity for Under-Clad Cracks
\\
I
POTENTI AL SOURCES OF NONCONSERVATISM IN ANALYSIS
-- Some Overcooling Transients Can Be More Severe Than the Rancho Seco Transient Used as a Benchmark
-- Fluence Calculations May Be in Error More Than Expected Uncertainty 4
-- Copper Content May Be Locally Higher Than Expected
-- RT Shift May Be Higher Than RG l.99 NDT
-- Fracture Toughness Properties May Be Lower Than Code Allowable
--Thru-Cladding Cracks increase Stress intensity
-- Local Vessel Temperatures May Be Lower Than Buik Coolant Temperatures Due to Imperfect Mixing
\\
l l
D G
W Z
_O x
O u_
Z Zw Ow
SUMMARY
OF OWNERS GROUP MEETINGS WESTINGHOUSE - OG Evaluating Remedial Actions Generic Report Due 12/31/81 At least 3 EFPY Remaining Before Concern with any Plant B&W-OG
- - Oconee-1 Detailed Report Due 12/31/81
- 32 EFPY (Full Service Life) Remaining if Credit is Taken for Operator Action CE - OG Detailed PlantSpecific Analyses Due 12/31/81 At least 5 EFPY Remaining for All Plants if Credit is Taken for Warm Prestressing
\\
l
DISCUSSIONS WITH FEDERAL REPUBLIC 0F GERMANY
-- Action Has Been Taken on Two Older PWRs (Stade and Obrigheim)to Remove Outer Fuel Elements and Replace with Dummy Elements
-- For Newer Plants, a Limit of I x 10' nyt Has Been Placed on the Allowable Fast Neutron Fluence.
-- Overcooling Transient at Neckarwestheim Plant on September 21,1971
-- Plant in Hot Standby Condition
-- Safety Valve Opened and Stuck Open
-- Cold Leg Temperature in One Loop Dropped from 550 F to 320 F in 5-10 Minutes
-- Pressure Decreased from 2200 psi to 1400 psi in 5-10 Minutes and Then increased to 2130 psi in 45 Minutes
\\
SUMMARY
OF 60-DAY RESPONSES ON RTNDT NSSS Vessel Weld Plant Vendor Fabricator Location RTNDT( F)
NRC Staff Licensee OconeeI
'B&W B&W Long.
170 160 TMI-1 B&W B&W Long.
160 170 Ft. Calhoun CE CE Long.
280 255 Maine Yankee CE CE Long.
240 180 Calvert Cliffs I CE CE Long.,
230 178 H. B. Robinson W
CE Long.
290 242 San Onofre 1 W
CE Long.
270 229 Turkey Pt. 4 W
B&W Circum.
290 193
\\
1 l
PRELIMINARY CONCLUSIONS FROM 60-DAY RESPONSES i
RT May Be Lower Than Staff Estimates NDT Earlier Staff Estimates for Low Nickel Welds Are Probably Substantially Conservative I
Further Staff Analysis Is Needed to Understand the Basis for Licensee Estimates 1
i
\\
SUMMARY
OF 60-DAY RESPONSES ON RT LINT 4-NDT i
Licensees Generally Do Not Consider it Appropriate to Define a Single Value of RT as an Upper Limit for Continued Operation NDT
-- No Alternate Limits for Continued Operation Were Proposed i
i 9
4 I
\\
~
i 1
l
SUMMARY
OF 60-DAY RESPONSES ON OPERATING PROCEDURES All Eight Plants Provide HPSI Termination Criteria All Eight Plants Provide Instructions for Terminating or Throttling l
Feedwater Flow Generally, Operating Procedures Place More Emphasis on Maintaining ECC Flow Than on Preventing Pressure Vessel Overcooling CONCLUSION Staff Will Require Industry to Take Action to improve Operating Procedures and Operator Training to Mitigate Overcooling Transie nts
\\
t
SUMMARY
OF ORNL ANALYSES A Range of Overcooling Tra.nsients Was Used to Evaluate Oconee I 1
Vessel Integrity 1
The Most Seve're Transient that Has Occurred (Rancho Seco Event) Will Not Pose a Threat to the Oconee i Vessel for About 15 More EFPY Using Conservative Material Properties and Fracture Mechanics Analyses, Very Severe Overcooling Transients Were Calculated to Pose a Threat to the Oconee l Vessel Today e
\\
i NRC STAFF REVIEW OF ORNL REPORT Useful Summary of the Background and Present Status of the Pressurized Thermal Shock issue i
Pressure and Temperature vs. Time Calculations Were Very Conservative (Some Aspects were Unrealistic)
-- Material Properties and Fracture Mechanics Calculations Are Believed to Be Conservative Probability of Occurrence of Severe Pressurized Overcooling Transients is Sufficiently Low that immediate Corrective Action is Not Warranted
\\
DISCUSSIONS WITH PROFESSOR G. C. SlH OF LEHIGH UNIVERS11Y
\\
Professor Sih Expressed Views that Curr'ent Fracture Mechanics Methodology Might Not Re Adequate for Predicting Vessel Failure Correctly.
The Staff is Evaluating His Comments.
\\
1 I
.w.e.-
y g-
?
I i
i a
e b
6 t-
~
4..
i =
O a
f M
- )
f
.e o
4 &
f q
s y
7-na n.
r 1 b' 4
s s
f a
)
f I
Y
..g 9
h s'
.s i
s,f'.
yg.
'oe
/
4,
,p J
)
sl j s
~
J's m
7
.O
- d sq w a
/
<:C
/
u_
LL.
+
n
. F-M O
M M
/
,,k j
/
A s
dY s
O 4
y 'E
/
4 e
k A
f g
oU a
w d
f
/
6 ed" 4 e
/
s 9
- -- -,~.,., - - -,.
a----
+
i
< ),
ps 4
, *\\
8
~
y r
. g SCHEDifLE OF FUTURE ACTIONS i
1 s
s December 2,1981 EPRI Report on Feasibility of in-Place Annealing of Vessels December 8,1981 ACRS Subcommittee Meeting with Staff and EPRI 4
December 31,1981 Owners Groups' Reports Scheduled January 18,1982 150-Day Responses from Eight Licensees Summer 1982 Staff Position on Pressurized Thermal Shock issued Continuing Confirmatory NRC & Industry Research Programs and NRR Technical Assistance Program (USI)
\\
i
/
f~
j 1
~
ISSUES TO BE ADDRESSED
)
h Wh~at Should Be the Regulatory Limit?
f i
i c 2'.
What Credit Should Be Allowed for Operator Action?
i
~
What Should Be the Evaluation Model?
3.
i
- Neutron Fluence Uncertainties
- Credit for Low Nickel Weld Metal
[
- Credit for Warm Prestressing 4
- Assumed Crack Shape
~
y.
- Effect of Stainless Steel Cladding l D vs. 2-D Fracture Mechanics
- Coolant Mixing in the Downcomer 4.
What is the Residual Risk with the Proposed Regulatory Limit?
- Probability of Overcooling Transients
- Probabilistic Fracture Mechanics Analysis 5.
What Are the Benefits and Costs of Corrective Actions?
6.
What Should Be the Schedule for Implementing Fixes?
1 There Will Probably Be Substantial Technical Disagreement Between the Staff and industry on These issues.
l 1
The Work Needed'to Technically Resolve the issues Cannot Be Fully Completed by Summer 1982.
\\
A.
1 eP g
N Z
~
O D
J O
ZO O
u._
<C F-C.A i
)
i i
r i
l 1
I l
STAFF CONCLUSIONS
-- Pressurized Thermal Shock is a Safety Concern for Older PWR Vessels Having High Copper Content Material
-- Immediate Corrective Action is Not Necessary, But Substantial Work is Needed by the Industry
-- Some Corrective Actions Will Likely Be Required for Some Older Plants
-- Bases for Staff Conclusions That No Corrective Actions Are Required Now:
-- Chances of a Severe Overcooling Transient in Older PWR Plant During Coming Year is About I in 100
-- Even if a Transient as Severe as Rancho Seco Were to Occur, Vessel Failure Would Not Be Predicted Using Nominal Material Properties
-- Analyses Include a Number of Conservatisms
\\
l
!~
O PROGNOSIS Staff Can Have a Regulatory Position by Summer 1982 Future Analyses Will Probably Show Current Staff Analyses to Be Conservative Some Older Plants May Have to Take Corrective Actions Before tiv End of Their Full Service Life
\\
- )
!(}(((
(_ ( ( ( I
'l l
(
- ( ( ( ( ( ( ( ( { ( ( '( ( ( ( ( ( ( I( ( ( ( ( J [ [ f ( (
<=>
TRANSMITTAL TO:
Document Control Desk, 016 Phillips
.b P53 ADVANCED COPY TO:
C The Public Document Roomk g
DATE:
}
3O
[}
M
/
/
ch
- Attached -are ' copies _of_ a C. omission meeting ~ Thef-6 transcript /s/ and related meeting docunie'nt/s/.
are being forwarded fox entry on the Daily Accession 3:>
List and placement in the Public Document Room.
No 50 other distribution is requested or required.
Existing DCS identification numbers are listed on the individual b
documents wherever known._
~
~'
pc=>
MEE';ING TITIFE:
+4A >
e
()
/
Q ll /M Y./[)
Open C
MEETING DATE:
Closed DCS COPIES:
/
Copies (1 of each Checked)
ITEM DESCRIPTION:
Advanced May To PDR:
- Original be Duplicate
[hf 1.
- Document Dup
- Copy *e f
F 5
f W.
VW
/
/
gg 2.
\\
' *Q
?
)
3.
4, jggik i
.g DEC1 j g u.5. %mC" g
c -;
4.
- f 9
h
., a N'f 5.
p.
e b
- 5=
of each SECY paper.)
P.-
@ MTP0)o?Nay0VP&&&Mff#F h