ML20059N622
| ML20059N622 | |
| Person / Time | |
|---|---|
| Site: | Sequoyah  |
| Issue date: | 10/04/1990 |
| From: | Wallace E TENNESSEE VALLEY AUTHORITY |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9010170030 | |
| Download: ML20059N622 (6) | |
Text
-
- 1. ',., e
- +
TENNESSEE VALLEY' AUTHORITY CHATTANOOGA. TENNESSEE 374o1 SN 1578 Lookout Place 08T 041990 I
U.S.' Nuclear Regulatory Commission
.i ATTN: Document Control Desk LWashington, D.C.
20555.
Gentlemen
)
In the Matter of
)
Docket Nos. 50-327 l
Tennessee Valley Authority
)
50-328 SEQUOYAH NUCLEAR PLANT (SQN) UNITS 1 AND 2 - MICR0 BIOLOGICALLY INDUCED f
LCORROSION-(MIC) - STATUS UPDATE
Reference:
TVA letter to NRC dated January 20 '1988, "Sequoyah Nuclear Plant'(SQN) Units 1 and 2 - Microbiological 1y Induced Corrosion.(MIC) Program" t
This letter provides a status report for SQN's MIC inspection and monitoring program at the. request of NRC staff.
The enclosed report includes TVA's 4
evaluation of radiography data performed on a sample of six stainless steel' butt-welds over an 18-month period. Recent test information concerning the mechanism for MIC has also been included.-
As discussed -in Enclosure 1 (Item VI) of the referenced letter, TVA's MIC program for SQN included provisions for monitoring'MIC growth on a sample of welds with. existing MIC' indications and identification of colonies on new
- elds.
w The'results of.TVA's monito-ing program indicate:
1.
Leakage associated with the MIC welds with throughwall leaks has been minuscule..
- 2.. Maximum growth rate for welds with throughwall leakage and welds.with 10 years.'of' service is 0.57 inch'per year.
3.: Additional: radiography data will be needed to establish the ef fectiveness of SQN's blocide treatment-system.
-4.
The' original growth rate criteria (5/8-inch-per-year circumferential growth rate) was reasonable and conservative.
y The enclosed report provides further details regarding these results.
1 Ak Q h[
kib
)
7 e
. M An Equal Opportunity Employ r
7 - -
wv.
i u
+.
+
_2_
l' OCT 641990:
U.S. Nuclear; Regulatory Comission Please direct questions concerning this 1ssue to Don V. Goodin at (615) 843-7734.
Very truly yours, TENNESSEE VALLEY AUTHORITY I
~
E. G. Wallace, Manager
~ Nuclear Licensing and.
Regulatory Affairs 1
Enclosure cc (Enclosure):
Ms. S. C. Black,' Deputy Director
- Project Directorate II-4 U.S. Nuclear Regulatory Comission
. One White Flint,~ North-11555 Rockville Pike Rockville; Maryland. 20852 l
. Mrs J. N. Donohew-Project Manager U.S. Nuclear Regulatory Comission x
-One White. Flint, North 1
11555-Rockville Pike Rockville, Maryland -20852 i
NRC Resident' Inspector-
- l
-Sequoyah: Nuclear Plant j
2600 Igou Ferry. Road Soddy Daisy,' Tennessee 37379 Mr. B.-A.
Wilson, Project Chief U.S. Nuclear ' Regulatory Comission j
- Region II 101 Mariettu Street, NW, Suite 2900 Atlanta.-Georgia 30323
- {
d i
~
[d,
.c I ,
y,
.a.
r
,3 -
,. g g f'
ENCLOSURE MICROBIOLGICALLY INDUCED CORROSION (MIC) LEAK INSPECTION j
AND MONITORING PROGRAM r
1.0 INTRODUCTION
l This report provides a current summary of the leak inspections and monitoring of MIC on SQN's essential raw cooling water (ERCW) stainless steel welds.
The results of 18 months of monitoring corrosion and growth in stainless steel butt welds are presented.
_j
2.0 BACKGROUND
Pitting damage resulting from microbiological growth in the wolded areas of the stainless steel headers has been found on ERCW piping. MIC damage has been confined to the butt welded configuration. To address the presence of
.g this condition in SQN's ERCW system, TVA has developed an inspection and replacement program. This program includes visual inspection for leakage and radiography of. leakers ~and.nonleakers.
3.0 CORROSION MECHANISM The MIC leaks in SQN's ERCW system occur at or near the stainless. steel butt welds. MIC damage is observed as small pits, which produce voids in the weld
- )
after entering the weld metal.- MIC damage initiates as a small opening on the inside diameter (ID) of the pipe. A MIC growth nodule begins to grow over this opening on the inside of the pipe. These nodules are easily seen on l
radiography. The cavity in.the pipe opens up under the-ID entrance hole. The cavity-contains.a porous corrosion product lef t behind by the selective corrosion of the austenite phase of the weld.
This leaves the' dendritic (ferritic)' structure of the weld as delta fer-ite. A small hole on the outside diameter of the pipe!is where the microbes exit-the pipe. Testing by-TVA has shown that ferric chloride is produced in the' cavity. Using American
. Society for Testing and Materials (ASIM) G-48, " Ferric Chloride Test,"
selective: attack of the austenitic dendritic structure can be reproduced in the laboratory.
In a leaking weld, the porous dendritic structure traps or screens out sediment in the river water, and the'1eak slowly decreases until it stops. Although the leak' stops, the evidence of the leak is still visible from staining and corrosion product buildup on the pipe.'
i 4'. 0 '
SUMMARY
OF EXPERIENCE There have been 52 leaks identified since the discovery of MIC at SQN in October 1986.
Inspection of removed welds adjacent to leakers, random radiography of nonleakers, and inspection within the piping-when the system is
~
}
open have shown that nearly all welds inspected have had damage or buildup of biological nodules on the weld ID.
n p
l1 i
L 1
n
,Y at i
W;
_a;
'. c o
' ' 0 4,
t In September 1987, all leaking welds were removed.
Since that time, the
-incidence of.new leakers has dropped significantly. The drop in the discovery rate is attributed to the smaller population of "old" vintage welds (1978-1979) and the' slow progression of pitting damage.
Radiography has shown:that.new welds made with both 316L and ER NicrMo-3 filler material indicate evidence of MIC attack, although none of the welds have shown throughwall leakage. The ER NicrMo-3 material was not attacked, but pitting damage was found on the heat affected zone of the weld adjacent to one of the two_ welds inspected.
In June 1988, a monthly leak rate determination program for identified lenkers was begun. This inspection was performed on leaking welds that were identified during two walkdowns (March and June 1988). The initial leakage determination showed that only one of the 17 leaks inspected indicated active dripping (four drops per minute), with four additional leaks still damp to the touch. Th'e July 1988 inspection found that all the welds were dry, except the
?
dripping weld, which slowed to two drops per minute. By September 1988, the dripping had stopped; however, this weld remained wet to the touch until it was replaced in February 1989. This is the only weld that continued to drip for such a long period of time after discovery.
Since this initial' group of 17, there.bave only been two additional leaks identified. Neither of these leaks continued to drip after the first month, although one weld that was identified in December 1989 remained damp until it was replaced in May 1990.
The identification of leaks usually requires _ prodding the stain or corrosion -
product until visible leakage can be observed. This results in a much larger leakage rate than was initially observed, but is still within the " drops per minute" range. With time, the leak begins to seal itself. The sealing of the leak is because of occlusion of the pit with river sediment.
These pits have
-a " tunneled" morphology with selective attack of either the delta' ferrite
~
phase or the austenite phase'in the weld. material. This results in a complex leakage path that is easily blocked by the particulate matter in the raw water, 5.0 RADIOGRAPH DATA COLLECTION AND RESULTS SQN's monitoring program is intended to provide additional information on the growth rates associated with the MIC damage on the new and existing welds and to evaluate the performance of new alloys used in the system.
The radiograph data used to make the growth rate determination was collected, lj under.SQN's Preventative Maintenance Program (PMs 2241-067 and 2242-067).
In cach case, the results were evaluated according to Technical Instruction (TI)-109,."Non-Destructive Testing of Stainless Steel Butt-Welds To Assess
. Damage Resulting From Microbiological 1y Induced Corrosion." The evaluation is based on the total linear circumferential length of the indications found in the weld.
l
)
a-h y..
y F
The dansge' that is shown by the radiography is evaluated to demonstrate-seismic integrity of the welded joint. The radiograph' data is'also trended to study the growth rates of the identified damage.
For the purpose of repeatable trending, all the film is read by a single Level JII radiographer.
J This information is used to evaluate the growth potential of the identified damage to ensure that it does not compromise the structural integrity of the piping system in the time period between the-initial assessment of the damage and the repair or next scheduled radiography exam..TI-109 currently requires 1
that a 5/8-inch-per-year circumferential growth rate be'used to evaluate the growth potential for these welds.-
Weld radiograph data has been collected on welds that were:- (1) identified as leaking by walkdowns, and (2) selected as a representative' sample of nonleakers. This sample contains six welds that include original stainless welds (installed in 1978-1979) and repair welds that have been installed since 1987.
The naterial of construction of the af fected piping is 316 stainlese teel.
i The original welds that were made in 1978 and 1979 used 308 stafaless steel as
-a filler material.- This material does not have the molybdenum content of the 316 stainless steel and is more susceptible to chloride pitting, which is the corrosion mechanism beneath the biological growth on this materir.l. Repair t
welds that were made in October and November 1987 used 316L ae a filler material. This material matches the base material in alloy content, but will have some segregation of the alloy elements during solidification, which
^
lowers its-corrosion resistance relative to the base material.
Since radiography showed the initiation of MIC attack on the replacement 316L weld material, the repair welds that were made during the Unit 2 Cycle 3 outage (February 1989) used ER-NICrMo-3 as a filler material.
This weld material is the equivalent of Inconel 625. This material has a high resistance to chloride pitting; however, the potential for corrosion of the weld heat affected zone increases since Inconel 625.is galvanically cathodic to.the base material..In most cases, galvanic attack in-the raw water systems has been minimal, primarily-because of the relatively low conductivity of water in the Tennessee River.
'l The radiograph data shows that the growth rate of MIC damage is very high as a weld begins to exhibit the formation of nodules and voids.
The growth rate in older established MIC colonies is much less, and in most cases, does not show any change'in linear length. The largest growth rate that was observed in an established MIC colony was 0.57 inch per year.
The growth did not continue beyond a single 6-month monitoring period. Based on the observed growth rate.
TVA's current growth rate estimate of 0.625 inch per year is acceptable for determining the growth potential in leaking and nonleaking MIC welds.
y l
e
't; y,,s4-
]
c.
_4 i
6.0 BIOCIDE RESULTS SQN implemented its chlorine / bromine biocide treatment program in LJanuary 1989. Since that time, no new MIC indications were found to have initiated in the sample of nonleaking welds. A MIC-type indication was found l
in'one of two additional welds that were radiographed in September 1989.
These two welds were repair welds ride in February 1989, after the initiation of blocide. These welds were added to the sample to evaluate tha effects of the blocide treatment and the performance of the new weld material (Inconel 625).
The radiograph data is as yet inconclusive regarding the effectiveness of the l
new blocide system.
In the original sample of nonleakers, no new indications were found after the initiation of the new biocide. However, since the -
j corrosion attack was well established in these welds by the time of the implementation of the blocide program, new indications of attack svuld not normally be expected.
7.0 CONCLUSION
S
'TVA's experience with SQN's MIC program has-shown that throughwall damage has.
i been slow to develop.
- The leakage associated with the.throughwall leaks, evaluated to date, has been very minor. 'In most or all cases, leakage stops as the pit becomes blocked by the particulate matter in the raw water.
The maximum growth rate found ir. leaking welds and welds that had been in service for ten years was 0.57 inch per year. TVA's original growth rate criteria of 5/8 inch per year is considered to be reasonable and conservative.
The radiograph data is as yet inconclusive regarding the effectiveness of'the new blocide system. Additional radiograph data will be needed to establish the effectiveness of SQN's biocide treatment. system.
4 l.
l l
f I
H