ML19305A488
| ML19305A488 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 06/26/1979 |
| From: | Mattson R Office of Nuclear Reactor Regulation |
| To: | Rogovin M NRC - NRC THREE MILE ISLAND TASK FORCE |
| References | |
| NUDOCS 7908210397 | |
| Download: ML19305A488 (1) | |
Text
_-
k v
Distribution:
C nneket File )M ruuvWMYYEYI NRR Rdg.
DSS Rdg.
R.J.Mattson W.Minners Rdg.
26 1979 Docket No. 50-320 MEMORANDUM FOR: Mitchum Rogovin, Director, NRC/TMI Special Investigation Group FROM: Roger J. Mattson, Director, Division of Systens Safety
SUBJECT:
TMI-2 STEAM GENERATOR B Enclosed is a current evaluation of the cause of the primary to secondary leak that occurred in Steam Generator B at TMI-2 during the accident.
Original signed by Roger J. Mattson Roger J. Mattson, Director Division of Systens Safety
Enclosure:
As Stated cc:
- 11. Denton, NRR
' ' ~ ' ' -~
" " "" " d ' d -- -
i 79080 V
J039 7 a
i f
M DSS UMinners':cj RMattson l
06/25/79 06/'-(/79 1
NRCFORM 318 (9 76) NRCM 0240 W u. s. oovannuent painrine orrect sets-eas.eaa
49 o
UNITED STATES
~g 8
NUCLEAR REGULATORY COMMISSION n
a wasmwovon, o. c. rosse o
s
~~~
a u an NOTE TC R. J. Mattson',M FROM:
R. J. Bosnak
SUBJECT:
TMI-2 STEAM GENERATOR B
Reference:
Note to R. J. Bosnak from R. J. Mattson of June 1,1979 on TMI-2 Steam Generators A baseline eddy current inspection of steam generator B at THI-2 during Decenber 1977 revealed approximately 400 tube minor dents to 95% through wall penetrations.g with defec.ts ranging from Thirty five tubes with defects greater than 40% of the wall thickness required plugging. A majority ei the eddy current signals (ECT) were indicative of a dimple (ding) 1.r.., reduction of inside diameter without de?.ectable reduction in wall thickness. These defects presumably occurred daring the fabrication process. Other ECT signals were indicative of scab type defects.
Circumferential cracks can initiate at such locations under conditions of high cycle fatigue. It is surmized that excessive flow induced vibrations may have caused high cycle fatigue failures at other B&W plants notably Oconee 1, 2, and 3.
Concern about excessive tube vibration at TMI had been raised after the baseline inspection in Decenber 1977.2 A test was designed to investigate the reduction in alternating stress by installation of tube sleeves at two locations of concern and addition of intennediate supports at two different locations at the upper most tube span.
It is possible that during the period between the Decenber 1977 baseline inspection and the accident at TMI-2 in March 1979, circumferential cracks may have been initiated at locations of high flow induced vibrations.
It is estimated that a circumferential crack with a depth which had progressed to greater thare seventy percent through wall would be unable to withstand the pressure and thennal loads imposed during the March 1979 transients. Such a crack would then be expected to pop through the remaining wall resulting in primary to secondary leakage.
Contact:
J. R. Rajan, DSS:MEB, X27538/72 l
l0006 79071cc on;s-Cr=
n.
\\
.N JUN 141979 R. J. Mattson The bending and thermal stre'sses in the tubes during pressure and thermal transients similar to those that ccurred during the TMI-2 accident were evaluated by B&W recently.3 In the analysis of the postulated transient, primary system pressure builds to the maximum value associated with the safety valve setpoint. Since primary flow is unavailable, a steam environment exists on the primary side of the tihes. On the secondary side, the steam generator is completely depressurized, boiled dry. A temperature differential of approximately 450 F may exist across the steam generator tube walls under such conuitions.
Initiation of auxiliary feedwater flow to the steam generator at this stage results in a rapid cooling of the tubes which in turn produces significant tensile loads in the axial direction. The bending stress on ghe tube outer wall due to a temperature differential, AT, of 450 F across the tube wall can be as high as 80 ksi tension which will cause plastic deformation of a portion of the tube.
A circumferential crack of depth seventy percent through wall or more located in this region is likely to penetrate the tube wall and the crack opening is likely to increase resulting in a primary to secondary leak. During a decrease in AT across the tube wall and a consequent reduction in bending stress, the circumferential crack would tend to close up, resulting in either a decrease or complete stoppage of the primary to secondary leak, depending on the size of the crack.
Other possible sources of primary to secondary leakage were also examined. These include:
a.
Leakage due to failure of the welds, attachments or other modifications made in the TMI Unit 2 steam generators to instal 1 instrumentation for monitoring vibration flow and pressure data.4 b.
Leakage as a result of other design modifications made in the steam generators for example:
(1) Tube sleeving modifications, (2)
Lane flow blockers (3) Auxiliary feedwater nozzle modifications, and (4) Secondary side, lane tube stiffeners.
l l
I l
i
a I4 E R. J. Mattson tc.
Failure of steam generator tube plugs installed after the init.lal baseline inspection of 1977.
d.
Leakage of the tubes due to other types of damage viz. wear, stress-corrosion cracking or erosion / pitting.
While the possibility of leakage due to these mechanisms cannot be ruled out, the most probable cause appears to 'oe high cycle fatigue cracking discussed earlier.
yL O R D osnak, Chief Mechanical Engineering Branch
)
Division of Systems Safety cc:
F. Schroeder, DSS J. Knight, DSS W. Minners, DSS J. Rajan, DSS H. Silver, DPM F. Cherny, DSS
.6
~
Da#crence::
1.
Letter from H. Silver to Metropolitan Edison Company - Summary of Meeting on St_am Generator Tube
- Inspection June 30, 1978.
2.
Letter from Metropolitan Edison Company - Steam Generator Tube Sleeve Qualification Program, Dec. 22, 1977 to S. A. Varga, NRC.
3.
B&W Report of May 7, 1979, " Evaluation of Transient Behavior and Small Reactor Coolant System Breaks in the 177 Fuel Assembly Plant Appendix 2 (Steam Generator Tube Thermal Stress Evaluation) 4.
B&W Report of Dec. 22, 1979 "Once Through Steam Generator Instrumentation Program for Three Mile Island #2."
Report No. 773570139 e
d 1