ML20237J631
| ML20237J631 | |
| Person / Time | |
|---|---|
| Site: | Trojan File:Portland General Electric icon.png |
| Issue date: | 08/13/1987 |
| From: | Wu P Office of Nuclear Reactor Regulation |
| To: | Mccracken C Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML20237J050 | List: |
| References | |
| TAC-65727, NUDOCS 8708180264 | |
| Download: ML20237J631 (8) | |
Text
-.
s e,.
ENC'L0SURE l-
[g* ascw*'o, UNITED ST ATEs NUCLE AR REGULATORY COMMISSION y V, c [j
[M w AssincTow. o. c. 20sss q.y t.
fa AUG 131987 MEMCPANDUM FOR:
Conrad E. McCracken, Acting Chief
.fr Chemical Engineering Branch Division of Engineering & Systems Technology FROM:
Paul Wu, Alternate Task Leader Pipe Well Thinning Task Force Chemical Engineering Branch Division of Engineering & Systems Technology
SUBJECT:
TROJAN PLANT SITE VISIT TRIP REPORT
~
The Pi,.a Wall Ti. inning Task Force ude a site visit at the Trojan Plant or.
July 22-23, 1987. The main purposes of the visit were to:
1)togather first-hand information regarding the main feedwater line pipe wall thinning problem, 2) to review the licensee's pipe wall thinning monitoring program,
- 3) to evaluate the results of licensee's failure analysis, and 4) to select representative samples from the degraded piping sections for independent analysis and verification.
The task force has accomplished 3 out of the 4 missions it set out to achieve.
Regarding the failure analysis, the licensee is still conducting its analysis and nothing was available to the task force. However, based on the task force observation on the damaged components and subsequent infonnation provided to the task force on July 30, 1987, we detennined that reasontble assurance has been provided by the licensee for safe operation of the Trojan feedwater systems through the end of 1988 operating cycle. Detailed staff analysis is described in the attached trip report. The task force will review results from the licensee's failure analysis and those from the staff / consultants independent verification to determine the long-term operability of the Trojan feedwater systems.
Ill{
Paul Wu, Alternate Task Leader Pipe Wall Thinning Task Force Chemical Engineering Branch Division of Engineering & Systems Technology
Enclosures:
As stated cc:
R. Starostecki P. Kuo L. Shao T. Chan J. Richardson G. Arlotto G. Knighton R. Bosnak C. Cheng C. Serpan T. Marsh
l EhCLOSURE 1 TRIP rep 0RT TROJAN NUCLEAR PLANT MAlh FEEDWATER SYSTEMS PIPE WALL THINNING EVENT INTRODUCTION:
Mein feedwater systems, as well as other power conversion systems, are important to safe operation. Failures of active components in these systems, for l
example, valves or pumps, or of passive components such as piping, can resu t in undesirable challenges to plant safety systems required for safe shutdown and accident mitigation.
~
Failure of high-energy piping, such as feedwater system piping, can result in complex challenges to operating staff and the plant because of potential systems interactions of high-energy steam and water with other systems, such as electrical distribution, fire protection, and security systems.
During the 1987 refueling outage at Trojan, the licensee discovered severe cases of wall thinning in the feedwater systems. Expanded inspection revealed that not only the nonsafety-portion of the feedwater lines experienced severe wa11' thinning, similar degradation had occurred in the safety-portion of these lines.
In addition, the licensee identified pipe wall thinning in 2 straight-run sections of the safety-portion of the feedwater system.
An NRC task force consisting of P. Wu (Task Leader. ECEB), Warren Hazelton (MTEB), Jim Brammer (ILRB), Mel Cowgill (Consultant, BNL), and Spencer Bush (Consultant, PNL) was dispatched to the site on July 22-23, 1987 to review the event and to evaluate the corrective actions and other related activities The task force attended briefings, interviewed taken by the licensee.
licensee staff, and made visual examinations of selected piping sections.and In addition, piping samples were components removed from the feedwater lines.
selected for staff / consultant..idependent verification of licensee's The task force concluded its review activities in the afternoon of analysis.
July 23, 1987.
BACKGROUND:
Trojan has experienced erosion / corrosion of secondary system piping including ext'raction steam piping, heater drain piping before the recent discovery of similar problem in the main feedwater system.
l l
l 1.
4 During the 1987 refueling outage at Trojan, wall thickr.ess discrepancies were observed by the licensee in the main steam lines at the outlet of the steam flow venturi. The cause of the discrepancies was machining of the pipe in order to install the venturi.
Records from the venturi vendor were reviewed by the licensee and it was confirmed that machining had bee,n perfonned.
The same vendor also installed the venturis in the four main feedwater lines.
Therefore, examinations were also performed by the licensee on the spool pieces containing venturis.
As a result of this examination, the licensee discovered a total of 19 elbows and 2 Straight runs in the safety-related portion of the feedwater line to have wall thickness below or will be below the code allowable value before the end of the next fuel cycle.
In the non-saft ty related portion of the feedwater line, over 35 items including elbows and pipe sections also have wall thickness below the code allowable value. At the present time, the licensee has replaced all elbows and pipe sections having wall thickness below the code allowable. The licensee is also conducting a failure analysis on the degraded items from the feedwater line.
However, no conclusion has been reached regarding the mechanism or causes of the wall thinning problem.
Since this is the first time that wall thinning has been identified in the safety-related portion of the feedwater line, the staff has prepared an Information Notice No. 87-36, Feedwater Line Break, to alert the industry on the wall thinning problems especially involving the safety-related portion of the feedwater line at the Trojan plant.
On July 9,1987, the licensee made a preser;tation to the staff in Bethesda regarding its action plant to resolve the pipe thinning issue. On July 10, 1987, the licensee provided additional information and pipe replacement statistics tg the staff.
REVIEti 0F LICENSEE'S EROSION / CORROSION MONITORING PROGRAM Prior to 1985. Trojan's monitoring program was confined to high pressure-extraction steam lines and limited sections of the heater drain system where
)
the possibility of flashing was a concern.
Following the 1985 failure of Trojan's heater drain pump discharging elbow, Trojan expanded its monitoring During the current inspection, primary emphasis was placed on past problems / failures and past inspection results. OT inspection with O' program.
longitudinal wave and 45' supplemented by 60' shear wave techniques were ISI program is performed in accordance with the 1974 edition of employed.
Section XI with addenda through sumer 1975 which has no requirement to wd! thickness. As a result of the Surry failure, main feedwater enmine suction piping was examined and the scope of heater drain pump discharge Initially, a 4" x 4" grid was used in the piping inspection was increased.
inspection.
If wall thickness discrepancy was discovered, a l' x 1" grid will be used for taking measurements. We find that although the licensee's monitoring program is consistent with the industry practice, more definitive guidance on the sample selection and sample size expansion should be provided.
4
SUMMARY
OF INSPECTION RESULTS:
All 45 and 90 degree elbows on Seismic Category I feedwater piping were inspected at 4-inch intervals (measured at the extrados) at 12 points equally spaced on the circumference. All straight piping left in service was scanned for the thinnest point on the circumference at 12-inch intervals.
Results of those inspections and information on piping replacement are shown in Table 1.
Visual inspections conducted by the task force members on removed piping indicated that erosion / corrosion was occurring at pipe welds and in the weld heat-affected zones. During our site visit, the task force members also visually examined 4 replaced elbows including a 60 degree elbow removed from the feedwater pump discharge side. This elbow had suffered severe localized erosion attack in addition to the uniform thinning caused by erosion-corrosion. As shown in Figure'1, the attack and the severe wall thinning occurred at the intrados of the elbow. We also visually inspected 2 reducers and 4 straight-run piping sections. They did not reveal any noticeable degree of wall thinning or localized erosion attack. However, the licensee was unable to locate the 2 straight-run sections which were removed from the safety-portion of the feedwater systems, and showed tiger-stripping type of damage morphology.
FAILURE ANALYSIS AND DAMAGE MECHANISM (SJ:
The licensee is still conducting its failure analysis to identify the cause and mechanism (s) which led to the o' served piping degradation. Based on its o
visual observation and preliminary analysis, the licensee suggested that the current piping degradation is a form of erosion / corrosion.
Based on the evidence shown in Figure 1 and other visual examinations made by the task force members, surface morphology of damaged elbows indicated that severe localized erosion damage had occurred in addition to the uniform-type of wall thinning caused by erosion / corrosion.
Piping samples were selected by the task force and they will be shipped by the licensee to the BNL for staff / consultant independent verification.
RESPONSES TO TASK FORCE QUESTIONS:
During the exit interview, the task force requested that the licensee provide the following additional information:
(1) a description of all safety-related piping or fittings replaced during the 1987 outage due to wall thinning, (2) a description of the technical basis for assumed erosion rates used to determine that portions of the piping did not have to be replaced, and (3) justifications for safe operation of the feedwater systems through the next This information is essential for the task force operating cycle at Trojan.
to determine the safe operation of the Class 2 piping of the feedwater systems. By "05000344/LER-1987-018, :on 870509,during Local Leak Rate Testing Containment Spray & RHR Recirculation Suction Valves Outside Containment Exhibited [[Topic" contains a listed "[" character as part of the property label and has therefore been classified as invalid.s.Caused by Normal Packing Degradation.Valve Packings Tightened|letter dated July 31, 1987]], the licensee provided information addressing the above questions.
EROSION / CORROSION RATE:
In the "05000344/LER-1987-018, :on 870509,during Local Leak Rate Testing Containment Spray & RHR Recirculation Suction Valves Outside Containment Exhibited [[Topic" contains a listed "[" character as part of the property label and has therefore been classified as invalid.s.Caused by Normal Packing Degradation.Valve Packings Tightened|July 31, 1987 letter]], the licensee stated that the erosion / corrosion rates are estimated using the following formula:
..=w..
L.'
' s Nominal Original Minimum Measured Wall Thickness Erosion / Corrosion Rate = Wall Thickness Years In Service The equation assumes that the rate has been constant throughout the life of the plant. This implies that the major factors:
(1) water temperature, (2) water velocity..'(3) piping geometry, (4) water pH, and (5) water dissolved oxygen content have been essentially kept unchanged.
In response to the task l
force question, the licensee stated that the Trojan is a base-loaded plant operating at 100 percent power. ' Water-temperature downstream of the last feedwater heater has remained constant at about 440*F and the water velocity has remained at about 22.6 fps. Piping geometry has not changed. Feedwater pH downstream of the last feedwater heater has varied between 8.7 to 9.0 over the plant life with an average value of 8.82.
DissoJved oxygen in the Class 2 feedwater piping has remained constantly below 3 ppb. Based on this information, the task force determined that it is reasonable to assume a constant danage rate at least for the Class 2 feedwater piping system.
i Typical Values for erosion / corrosion rates calculated on the basis of the current inspection results'on Category I feedwater piping are:
Observed Description Erosion / Corrosion Rate, inches / year 0.005 to 0.014 Straight Piping 0.015 to 0.038 90 Degree Elbows Closely Spaced 0.015 to 0.036 90 Degree Elbows ESTIMATION OF REMAINING PIPE LIFE The licensee indicated in its "05000344/LER-1987-018, :on 870509,during Local Leak Rate Testing Containment Spray & RHR Recirculation Suction Valves Outside Containment Exhibited [[Topic" contains a listed "[" character as part of the property label and has therefore been classified as invalid.s.Caused by Normal Packing Degradation.Valve Packings Tightened|July 31, 1987 letter]] that the following equation was used to determine remaining life of the Seismic Category I feedwater piping:
Measured Minimum Wall - Minimuu Allowable Wall 2 x (Observed Erosion / Corrosion Rate)
Remaining Life
=
where Pressure x Pipe Outside Diameter 2 x (AHowable Stress + 0.4 x (Pressure))
Minimum Allowable Wall
=
The equation above for minimum allowable wall thickness is in confonnance with ASME Section III NC/ND 3641.1.
CORRECTIVE ACTIONS:
The licensee conducted 100 percent inspection of the Seismic Category I It has replaced piping in the secondary system which feedwater piping system.
had wall thickness below minimum code allowable and piping predicted to be at i
_____-___-____________-_______________-__--____m
.' or below minimum code allowable thickness by the 1988 refueling outage. The replacement piping and elbows have the same specifications as those of theIn existing piping material i.e., ASTM A-106 Grade B and ASTM -234 Grade B.
addition, the licensee plans to slowly increase feedwater pH over the current level to about 9.?.
The task force detemined that these corrective actions and planned changes will reduce the propensity of erosion / corrosion induced i
wall thinning of the feedwater system while at the sare time protecting the i
l steam generator tubes.
CONCLUSIONS:
1.
Pipe wall thickness measurements were taken using ultrasonic techniques (UT) which is the current industry practice. The licensee's have also shown that UT measurements and those taken by micrometers agree within 0.005 inches or less. We determined that this accuracy is adequate for pipe wi.ll thinning monitoring purpose.
2.
The licensee indicated that the plant operating conditions, i.e.,
temperature, pressure, flow rate, pH, and dissolved oxygen in the feedwater downstream of the last feedwater heater have remained essentially constant throughout the plant life in the last 10 years.
Therefore, we determine that it is reasonable for the licensee to use 10 years as the basis to derive an estimated average pipe wall thinning rate.
3.
As the licensee stated and subsequently verified by the task force, there were small diameter pockets of erosion that exceeded uniform erosion / corrosion areas.
It is possible that these pockets could have been missed using a standard 4" x 4" inspection grid. Since these conditions are considered anormalies caused by very severe flow conditions, created by unique piping geometry which is not representative in the Seismic Category 1 feedwater piping, we determine that the methodology used by the licensee to estimate remaining pipe life to be adequate and it was a conservative approach for the licensee to calculate the remaining pipe life by doubling the observed erosion / corrosion rate.
The licensee has replaced all piping components and straight sections 4.
which had wall thickness below the allowable code minimum or will reach this value before the end of the 1988 operating cycle.
In addition, the licensee stated that the feedwater pH will be increased to about 9.2 over the next operating cycle. We determine that these corrective actions are
. adequate to ensure the safe operating of the Seismic Category 1 feedwater piping throughout the 1988 operating cycle.
The licensee is continuing its failure analysis to detennine the exact 5.
cause and mechanism (s) which led to the observed piping degradation.
Results from the licensee's analysis and those from the staff / consultant independent verification will be used as the basis to evaluate the long term operability of the feedwater systems at the Trojan Plant.
D E
T A
4 L
1 S8 0
E S8 R
0 E9 Y
N1 T
E K
R CE EF 3
8 8
5 1
A 0
5 7
3 7
A I H HT 1
S 2
1, S
T N
W F
O 1
O LO N
B L
L AD E
WNE E
0 LE S
9 BH M
AT E
D W
T N
OE S
A LR Y
LO S
G AF N
E R
I EB E
P D
T I
OE A
P CU W
L D
T MA E
H UV E
G M
F I
IE A
NL N
R I B 1
A T
MA J
S W
e O
WO l
R
}
E OL b
T I
H LL
(
T EA a
T E
B H
D R
E T
E O
ED T
F VO F
A AC O
L S
H M E
E.
'7 8
TYOU G
R
'6 L
9 I
Y 3
6 L
1 9
ALTM N
7 1
RE I
N T
3 A
VDN N
E NIEI I
F OTI M H
A ICF D
T S
SEI EE OPTHC L
RSNTA L
REE L
A ORDHPCE W
C
,I AR
/NRSE OATRN IEN E
SYEOE O/NTB RSO EEPDE HMEV DCOTA ENCCH TI E
A GPE M8NXL I 3I EC Y
S T 0.P
)
S I EC 2
R E0PR
(
S I
S G
A AL D
G E
N G
D N
G P
EDL E
T I
N S
E I
N E
HNLRU I
R TAAOF C
T I
D C
T E
T P
L A
T P
P I
I E
L I
I D
S F
P W
P F
P L
)
)
E E
1 2
N R
W
(
(
I l
l
W,
..* o e
f 4
(
t.
w g
~
., s FIGURE 1(a) Severe Wall Thinning at the Intrados (upper right corner) of a 60* Elbow i
r
- 4, e
e e.
s
(
. ;. u.;.
.h
~..-Q 1
FIGURE 1(b) Severe Localized Erosion Attack in the Intrados Region of the Elbow I