ML19309H174
| ML19309H174 | |
| Person / Time | |
|---|---|
| Site: | Point Beach  |
| Issue date: | 04/04/1980 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML19309H173 | List: |
| References | |
| NUDOCS 8005090299 | |
| Download: ML19309H174 (17) | |
Text
.
8005090 M Q O
SAFETY EVALUATION REPORT RELATED TO POINT DEACH UNIT 1 STEAli GENERATOR TUBE DEGRADATiNDUETODEEPCREVICECORROSION April 4, 1980 O
4
INTRODUCTION In accordance with the Confirmatory Order dated November 30, 1979, Point Beach Unit I was shutdown on February 29, 1980 for steam generator hydrostatic testing and eddy current inspection after having completed the authorized operating period of sixty (60) effective full power days (EFPD's) since the restart subsequent to the October 1979 steam generator inspection. The evaluation herein provides an update of the SER issued in support of the Confirmatory Order to reflect the operating experience at Unit 1 since the Order was issued, and the results of the steam generator inspection obtained during the February 29, 1979 outage.
The background information and results of two consecutive inspections (August and October,1979) as discussed in the November 30, 1979 SER are incorporated into this evaluation by reference.
BACKGROUND CONFIRMATORY ORDER DATED NOVEMBER 30, 1979 Inservice inspections of the Point Beach Unit 1 steam generators performed during i.ne August and October 1979 outages indicated extensive general intergranular attack (IGA) and stress corrosion cracking on the external surfaces of the steam generator tubes within the thickness of the tubesheet (generally referred to as " deep crevice corrosion").
In view pf these findings and of the apparent high rate at which this corrosion phenomenon was developing, the licensee agreed to certain conditions to assure safe operation of Unit 1 for a period of sixty (60) effective full power days.
This commitment was formalized by a Confirmatory Order dated November 30, 1979, amending the Operating License to include, in part, the following conditions:
1.
a) Hydrostatic testing to be performed within 30 EFPD's.
b) Hydrostatic testing and eddy current inspection within 60 EFPD's.
Submittal of the proposed eddy current inspection program for NRC staff review.
Eddy current inspection results also to be submitted, with no resumption of power until the Director, Office of Nuclear Reactor Regulation determines in writing that the results are accept-able.
2.
More restrictive limits on primary to secondary steam generator leakage.
3.
More restrictive limits on primary coolant activity.
4.
Unit I not to be operated with more than 18% of tubes plugged in either of the steam generators.
While not covered under terms of the Confinnatory Order, the licensee implemented additional measures in an attempt to retard further tube degradation. These measures included 1) a crevice flushing program to remove harmful chemicals from the tubesheet crevices, 2) reduced operating temperature and pressure, 3) continued close surveillance of feedwater chemistry and condenser tube leakage, and 4) sludge lancing to be performed within 12 months of the return to power.
hv
. DEFECTS AT OR ABOVE TUBESHEET The Safety Evaluation issued in support of the November 30, 1979 Confirmatory Order reflected thestaff's understanding that the extensive degradation observed during the August and October 1979 inspections involved general intergranular attack and cracking within the tubesheet crevices, exclusively. Subsequent to the Confirmatory Order, however, the staff became aware of five (5) tubes with defect indications at or above the tubesheet which had not been addressed in the November 30 SER.
In response to our request, the licensee submitted by letter dated December 21, 1979 additional details regarding the defects in these five tubes andan evaluation of their significance. The licensee reviewed the single frequency eddy current test results since 1975 for the subject five tubes and compared the signals of these past inspections to the same frequency signal obtained during the multi-frequency inspection in October 1979.
This comparison showed that the signals have not changed through three or four inspections since 1975. On the basis of this review the licensee concluded that the defects observed in October 1979 at or above the tubesheet have remained essentially unchanged since at least 1975 and occurred as a result of earlier thinning or cracking rather than to the intergranular attack phenomenon currently being experienced in the tubesheet crevice area and which' was only first observed in November,1977.
In response to our request, the licensee submitted by letter dated December 21, 1979 additional details regarding the defects in these five tubes and an evaluation of their significance.
Based upon our review of this submittal and a subsequent conference call with the licensee on December 22, 1979, we concluded that (1) the eddy current indications at or above the tubesheet, which were observed during the October 1979 inspection, are old defects, possibly due to wastage or stress corrosion cracking, which were active mechanisms in 1975 and earlier, (2) these indications are not related to the active phenomenon of general intergranular attack and cracking currently being experienced in the tubesheet crevices, and (3) the staff conclusions set forth in the November 30, 1979 SER remained valid and that the unit could continue to be safety operated under terms of the Confirmatory Order. Nonetheless, we have continued our investigation into the significance of'the defects found at or above the tubesheet, particularly with regards to eddy current capabilities to detect these defects and their safety significance. This matter is addressed in further detail in 'this evaluation.
OPERATING EXPERIENCE SUbIEQUENT TO THE CONFIRMATORY ORDER Following the issuance of the Confirmatory Order, Point Beach Unit I was returned to power on December 1,1979. On December 11,1979, Unit 1 experienced a rapid increase in primary to secondary leak rate, to 26n gpJ, and was forced to shutdown under terms of the Confirmatory Order.
The source of the leak was identified as one leaking tube and two leaking plugs in steam generator B.
Although not required by either the Technical Specifications or the Confirmatory Order, the licensee performed multifrequency eddy current examinations in both the A and B steam generators. A total of approximately 1900 tubes were inspected. The inspection bounded all areas of previously observed deep crevice corrosion by at least one row and column of tubes.
The inspection boundaries were expanded whm new indications were observed l
. nea t9s boundary. A set of randomly selected tubes outside the boundaries were alsc 11soected.
Representatives from the NRC staff and consultants were at the site c. Je: ember 16, 1979 to observe the inspection in progress.
As a result of this ins 3ection, twenty (20) tubes were plugged in steam generator A and fif teen (15) tutes were plugged in steam generator B.
None of the observed indications occurred at or above the top of the tubesheet. The inspection program and results were formally documented in Licensee Event Report 79-021 SIT-0 dated December 22, 1979.
Pricr to resuming power operation, 2000 psid primary to secondary and 800 psid secondary to primary hydrostatic tests were performed. No tube feilures or addi-tiora: leakage resulted from these tests.
Based upon our review of the December 11 tube leak occurrence and the inspection results we concluded that the conclusions reached in the November 30, 1979, SER remair.ed valid and that the operating restrictions imposed by the Confirmatory Order continued to provide adequate assurance of safe operation.
Point Beach Unit 1 was returned to power on December 22, 1979 and operated to the completion of its authorized 60 EFPD operating period (on February 24,1980) with only a very minor, but equivalent to a consthnt 30 gpd primary to secondary leak.
This v.as within the trace amount of equivalent leakage normally experienced at this : nit.
MARCH 1930 INSPECTION RESULTS FIELD EDY CURRENT TESTING The eddy current testing (ECT) program implemented during the March 1980 steam generator inspection was submitted for NRC staff review by letter dated February 26, 1930.
This program was modified to incorporate NRC staff comments.
ECT "
1001 of the tubes in regions of previously observed deep crevice corrosion ac
/ity (including the kidney shaped central bundle region) was performed within boundaries bounding previously observed defects by at least one tube row and column. Where defects were observed to occur at the boundary, the inspection was expanded to bourd these defectives by one tube row and column. An additional 3% random sample was inspected on tne cold leg side and also among tubes on the hot leg side in areas not being 100% inspected. Representatives of the NRC staff were on site during the inspection to monitor the inspection as it proceeded, and to facilitate timely decisions from NRC/NRR regarding the need for additional inspection or tube pullir.] for laboratory examination.
Multifrequency eddy current testing (ECT) conducted in accordance with the approved program revealed 18 defect indications on the hot leg side in steam generator A and 24 defect indications on the hot leg side in steam generator B.
In addition,3 tubes in S.G. 8 and 6 tubes in S.G. A were found with undefinable indications within the tubesheet.
On Mcrch 31, a hydrostatic test conducted after the ECT inspection revealed two tubes leaking ai approximately 2 drips / minute and two wet plugs in S.G. E.
Following plugging of these tubes and repair of the wet plugs a second hydrotest revealed another lea <ing tube in S.G. B which was plugged.
Table I l
sum arizes the ECT indicattd defect depths in the two steam generators.
Table II l
sum arizes the elevation of the defect indications above the lower, primary surface of the tubesheet.which is about 23 inches thick.
Some defects affected several inches of tube length and one tube had indications running from the tube expansion at the :rimary surface of the tubesheet to approximately one inch below the upper, secor.dar/ tubesheet surface.
The elevations indicated in Table II are the highest elevati:r.s mached by each defect.-
(
4 4
4 i j
TAB'_E I ECT INDICATED DEFECT DEPTHS i
DEFECT DEPTH IN NUMBER OF TUBES PERCENT OF TUBE WALL S.G. A S.G. B 93 to 100 5
3 80 to 89 7
7 70 to 79 2
7 60 to 69 3
3-50 to 59 2
40 to 49 1
2 TABLE II ELEVATION OF ECT DEFECT INDICATIONS DISTANCE ABOVE THE NUMBER OF TUBES PRIMARY TUBESHEET SURFACE (INCHES)
S.G. A S.G. B 0-4 1
j 2
5-9 10-14' 2
2 15-19 8
6 20-21 8
12-
-l 1/2" AB0VE SECONDARY T.S. SURFACE e
. No defe:tive tubes were discovered outside of the central bundle region on the hot leg site cor anywhere on the cold leg side of either steam generator.
Tables I and 11 in Appendix I provide a tube by tube evaluation of ECT indicated defect deaths and elevations and results of re-evaluations of ECT tapes from previous inspectio.s for each defective tube.
Study of these tables reveals that 15 tubes in stean generator A and 4 tubes in steam generator B had the same ECT indications but were overlooked in either the December or the December and October 1979 inspections. All of the tubes with defect indications were plugged except those that v.ere removed for laboratory examination. All the ECT indications were of small amplitude and indicate very small volume defects.
TUBE DULLING AND LABORATORY EXAMINATIONS In their February 26, 1980 submittal the licensee committed to remove a tube from the Unit I steam generators if one was found with an eddy current testing indicated defect at or above the top of the tubesheet, such as were observed in five tubes during the October 1979 inspection. The primary interest in removing this type of tube was two fold:
(1) to determine if the intergranular attack occurring within the tubesneet crevices is resulting in tube degradation at or above the upper secondary surface of the tubesheet and (2) to correlate field ECT with laboratory examination of the defects. As indicated in Table II one tube was discovered in steam generator B with an indication approximately 1/2" above the top of the tubesheet.
This was tube R19-C37 and the indication was 58% deep.
In accordance with their commit-ment, this tube was removed from the steam generator for laboratory examination.
In addition, the NRC (after a review of the ECT results) required removal of two other tubes for laboratory examination. These were tubes R30-C41 which had a 47%
indication approximately 21" above the primary face of the tubesheet and tube R26-C53 which had a 86 indication approximately 18" above the primary face of the tubesheet.
Removal of these tubes was intended to provide additional data regarding the extent and magnitude of IGA and the accuracy of ECT. The tube removal procedures extended the outage time approximately six days and resulted in approximately an additional 155 manrem exposure.
LABORATORY RADIOGRAPHY AND EDDY-CURRENT TESTING Radiography and ECT were performed on all three of the removed tube specimens by Westinghouse at their Pittsburgh R&D facility.
As a result of the pulling process the original 22.1/2" length of tube R30-C41 within the tubesheet was elongated to approximately 24-3/4". This measurement was based on the ring lef t on the tube at the top of the tubesheet.
Radiography of the removed tube revealed many defect indications in the region up to 23-1/4" from the tube end.
Many ECT indications existed up to 23-1/2" from the tube end.
No radiographic or ECT incfications existed at or above the ring marking the top of the tubesheet.
The latoratory ECT examination indicated an approximately 70 to 80% defect based on evaluation of the single frequency (400 XHZ) signal, located 23-1/2" from the tube end.
Based on the elongation caused in the tube removal process, 23-1/2" corresponds to approximately 21.3" from the tube end in the unstrained tube.
= - -..
=
4 1
4 Th e e' : I".~ indicated a 475 defect at 400 KHZ approximately 21" from the tube -
erd.
~'s : evaluation of ~tne defect based on the multi-frequency signal estimated.
i the ce#e:t cepth in the ssee 70L to 80% range as obtained in the laboratory (at 400 KHZ) t- -Pe tbsence of tLoesheet interference effects. Defect depths are reported based or t.e single frequency signal when possible since it is the technique currer ~., 3:: roved by the ASME Code.
The p.1:f r.; of tube R2E-C53 elongated the original 22.5" of tube in the tubesheet 4
crevice tc coproximately 25-7/16". Radiography of the removed tube revealed many defe:t ~1:i:stions in the region up to approximately 19.8" from the tube end as well i
as a sir;ie defect 25" above the tube end. Eddy current testing revealed many defect j
indicat'ar.s up to 19.8" from the tube end. Eddy current testing also revealed two 90% de#s:ts located apprcximately 7/16" and 2-7/16" below the tubesheet ring.
No j
radio;-aaric or ECT indications existed at or above the ring marking the top of the tubesteet.
I None cf the above laboratory ECT indications for tube R26-C53 were specifically i der.t' f' c: 'n the field.
Some of the indicated defects may have been introduced 4
or ra:e crse during the tube pulling operation.
" Squirrel" indications (minor i
disturtences in the ECT signal of underterminable origin) were observed in the' field i
over -ht 'uil length of ube within the. tubesheet.
It was not possible to verify
~
thrcu;h lab:ratory ECT tne 86% ECT indication observed in the field 18" above the tube er.:, since this corresponded to one of the locations where the tube broke.
durin; u: ling.
However, this. field ECT indication will be compared with the results of :he #rac:ography analysis of the fracture surface as 'part of a detailed report wht:h t e l'censee has comitted to submit by April 30, 1980.
Tube Eli-C37 was of particular interest because of the field ECT indication of i
a SE defect located approximately 1/2" above the tubesheet.
Unfortunately, when-the tube v.as examined there was no ring clearly indicating the top of the tubesheet as there was on the other two tubes which were removed. Since the section of tube withir :1e tubesheet experiencesa different load and elongation during the removal process thar the section of tube above the tubesheet', the exact-location of the l
top of le tubesheet relative to the tube cannot be directly quantified.
Radiograohj and ECT of the removed tube revealed many defect indications in the region to to 23.75" from the tube end.. Radiography also showed crack like indica-tions a. sroximately 24-3/8" above the tube' end and ECT indicated an approximate i
605 defe:t 2'-1/2" above the tube end. No ECT indications were observed.above the 60t ir di:a tion.
i-Althou;i tne 605 laboratcry ECT indication corresponds well with the 58% field ECT l
indica:i:n, its elevation cannot be directly. correlated to the field indications -
l because the location of the-top of the L tubesheet is not identifiable.. Calculations-l based :r strains in the cther tubes which were removed indicate that-this -defect L
would
.1.'s been inside the tubesheet. Nonetheless, it is the defect with the highest -
i eleva:H-ir. the tube, Lits depth corresponds well-to the field ECT depth and it-l
- ould :s es defect of ir.terest given the non-unifonn straining of the tubes during re :va!.
Metalk ree ic Examinaticr.s Metal'.:; a:-H exaninatis c consisted.primarily of photomicrographs (PM) to determine.
at ct; e'evation IGA existed in the tubes.
l L
l
, for tube R30-C41 PMs were prepared for sections centered on the top of the tubesheet and approximately 0.35" below and 0.45" above the top of the tubesheet.
In each of thest regio" D"-.f 50 and 200 power magnification were made.
The 200 power PMs were cen :G s.. the region in the 50 power photomicrographs indicating the greatest surface irregularities.
For the,section of tube below the top of the tubesheet the PMs showed shallow grain boundary separation on the order of 0.0025" maximum. At the top of the tubesheet, shallow surface separation was observed affecting grain boundaries to just over 0.001" in depth.
Similarly above the top of the tubesheet surface separation of the grain boundaries was observed to a depth of approximately 0.001 inches.
Extensive general IGA as is occurring deeper in the tubesheet crevice was not observed in any of these regions.
Photomicrographs were also prepared for tube R26-C53. Again the PMs were centered about the top of the tubesheet and approximately 0.4" below and 0.2" above the top of the tubesheet.
The section below the top of the tubesheet showed shallow grain boundary separation penetrating approximately 0.002" maximum.
The region centered about the top of the tubesheet showed no grain boundary separation although some surface irregularities penetrating less than 0.001" existed. Above the top of the tubesheet some areas of grain boundary separation penetrating approximately 0.003' were observed.
Extensive general IGA as is occurring deeper in the tubesheet crevice was not observed in any of these regions.
Five photomicrographs were made of tube R15-C39. One was centered on the 60%
defect described earlier while the other four were centered approximately 1-5/8" and 3/4" below and 1" and 1-3/4" above the defect.
The two sections below the defect showed IGA penetrating to depths of nearly 0.004".
Photographs of the tube surface at the defect show a crack running less than approximately 1/2" longitudinally then turning and running less than approximately 1/4" circumferentially.
Photo-micrographs of a section made through the defect show a crack penetrating approximately 0.017" surrounded by localized IGA. The longitudinal section made for the PM may not have included the deepest section of the crack.
Section 0 above the defect-indicates one localized area of grain boundary separation approximately 0.001" deep and section E above the defect shows no grain boundary separation but some shallow surface irregularities less than 0.001" in depth.
PROPOSED CONDITIONS FOR CONTINUED OPERATION The licensee has proposed the following conditions to allow continued operation of Point Beach Unit 1.
l.
Within 90 EfPD, a 2,000 psid primary-to-secondary hydrostatic test and -a 800 psid secondary-to-primary hydrostatic test will be perfonned. An eddy current examination consisting of about 1,000 tubes in the central region of the hot leg in each steam generator and 3% of the remaining tubes outside this area will be performed.
2.
Primary coolant activity for Point Beach Unit I will be limited in accordance with the provisions of Sections 3.4.8 and 4.4.8 of the Standard Technical-Specifications for Westinghouse Pressurized Water Reactors, Revision 2, July 1979, rather than Technical Specification 15.3.1.C.
3.
Close surveillance of primary-to-secondary leakage will be continued and the reactor will be shutdown for tube plugging on confirmation of any of the following conditions:
. 3rimary-to-secon:sry leakage of 150 gpd (0.1 gpm) in either steam generator; e.
b.
Any primary-to-secondary leakage in excess of 250 gpd (0.17 gpm) in either
+
stea: generator; or An upward trend (average over a three-day period) in primary-to-secondary c.
lealage in either steam generator in excess of 15 gpd (0.01 gpm) per day, when ceasured primary-to-secondary leakage is above 150 gpd in that steam generator.
4.
Tne reactor will be shutdown, any leaking steam generator tubes plugged, and a, eddy current examination as described in Item 1., above, will be performed if leakage due to crevice corrosion in either steam generator exceeds the limits stated in Technical Specifications 15.3.1.D.
5.
Unit I will be operated at a reactor coolant pressure of 2,000 psia with the associated parameters (i.e., overtemperature aT and low pressurizer pressure trip; Dint) with the linits indicated in the Safety Evaluation Report appended to your leth r of January 3,1980.
On return to power operation, the licensee proposes to continue the following program to assist in retardng further tube degradation:
Unit I will be oprated at a reduced reactor coolant system hot leg temperature.
a.
b.
Continue close surveillance of feedwater' chemistry conditions and condenser tube leakage.
}
Perform sludge lancing within nine months of returning to power.
c.
EF LCION ECT PRC3RA!4, RESULTS, AND CAPABILITIES Mertbers of the NRC staff and their consultant from Oak Ridge National Laboratory were on site during the inspection to review the testing and evaluation techniques.
Edd/ current testing examinations were conducted in accordance with the' program prososed in the licensee's February 26, 1980 submittal and approved, with comment, by *.he SRC.
This progra 1 bounded the areas where deep crevice corrosion was pre-viously observed and was expanded in any areas where new indications were found.
Tne randon inspection of peripheral hot leg tubes and cold leg tubes revealed no deep crevice corrosion.
Therefore, the inspection performed is ' adequate to ensure that the great majority of tubes with deep crevice corrosion have been removed from servi:e by plugging.
Tne Mer:h 1980 ECT resul s show a marked reduction in the number of tubes with in-dicatd defects compared to the August and October'1979 inspections.
In addition, fifteen of the 24 ECT indicated defects in steam generator B and 6 of the 18 ECT indicated defects in stea generator A were shown to exist previously through re-en-ir.a, ion of the ECT ta es from previous inspections.
Thus, the number of new :
defe: t discovered in this inspection is smaller than the raw data indicates.
The ir.s:s: tion results suggest that some of the remedial actions taken by the licensee fcM:..".; the October 1979 inspection, particularly the lower temperature operation, it s;;ceeding in retarding the rate of further deep crevice corrosion, especially m!
sir:s: r.s time of' the Dece-ber 1979 outage.
l
, As :is:ussed in our November 30, 1979 SER the accuracy of the eddy current technique is it,.. hat diminished ir the tubesheet region ar,J cann:. be. fully relied upon tu dits:: svery tube degraded by deep crevice corrosion.
Tnis appears to be particularly true f:r tubes subject to general IGA, but which do not contain cracks.
Partially thra;n aall cracks of significance are generally detectable, even in the tubesheet regW., wi th ECT. As experience has shown, however, very small volume defects which in turn produce very small amplitude ECT signals may be easily overlooked (as was the case with the 19 tubes above). Our evaluation of the safety significance of IGA ar.d stress corrosion cracking cccurring within the thickness of the tubesheet is discussed in our November 30, 1979 SER which is incorporated into this SER by reference.
Witt regard to the tubes observed during the October and March inspections to contain defects at or slightly above the top of the tubesheet,we have concluded that multifre-quer.cy ECT can detect defects of a significant size to threaten tube integrity during norral or postulated accident conditions. All of the defects discovered at or above the top of the tubesheet are small amplitude, small volume defects. Assuming the defects at or above the tubesheet to be wall thinning (wastage related), rough estimates of the size of the defects were made by the staff based on comparison with the ECT sign 3tJres from the ASME Code calibration standard. These estimates show that if these defects are wastage related, the volumes of these defects are very small compared to wia: is necessary to burst or collapse the tube under postulated accident conditions, as determined by independent tests sponsored by NRC (NUREG/CR-0718).
In the case of tube R19-C37 which exhibited a field ECT indication of 58% approximately 1/2 inch above the tubesheet, the laboratory examination indicates that the defect indication observed in the field is most likely a crack.
NRC sponsored burst and collapse tests (NUREG/CR-0718) have been perfonned on free standing tubes with EDM notches (simulating a crack) of up to 85-90% (through wall) in depth.
The results indi:a e the lower bound burst strength to exceed the maximum primary to secondary
+
pressure differentials during normal operation or postulated accidents for notches (cra:ks) ranging to about 1 inch in length.
It should be noted that the burst strength of a tube containing a crack defect slightly above or below the top of the tubesheet is considerably higher than for free standing tubes, because of the re-straint against radial expansion of the tube provided by the tubesheet. The above tests indicated a collapse failure to be a much less limiting failure mode than a burs failure mode for free standing tubes during postulated accidents.
Cracks of sufficient size to cause a burst or collapse failure under postulated accidents are considered by the staff to be well within the detectable capability of the multi-frequency eddy current techninue, regardless' of the location of the crack relative to tt too of the tubeshset.
Tube ReTcval and Laboratory Exam Laboratory radiography and ECT confirm the position taken by the staff that general
,e IG;.a not be detectable in the crevice of the tubesheet until it is severe enough for preferential crack growth to occur. Detection of defects below the top of the tubesheet by laboratory examinations is due partly to increased capability of ECT without the influence of 'the. tubesheet and partly to the creation of new or the openir. of old defects during the removal process.
Laboratory radiography and i CT confir a the absence of defects above the tubesheet in tubes R30-C41 and R7f.-053.
I UMor*a ately the top of the tubesheet could not be identified on tube R19-C3/.
, Howecar, assuming that the upper most defect detected in the tube is the defect which was identified by field ECT, there is a good correlation between the laboratory and field ECT. More importantly, the defect which was detected was small enough so as net to j npardize tube integrity.
Primary-to-secondary and secondary-to-primary hydrostatic tests conducted on March 6 revealed one tube (R23-C44) which exhibited a sligh leak at a rate of 3 drips per minute, and'one wet plug in a previously d
plugged tube (R23-C50) both in S.G. B.
No tube ruptures occurred. The defect M
four.d by ECT just above the tubesheet in tube R19-C37 in S.G. B withstood the simulated accident pressure differentials. This provides additional support to our
.2[i3(
previously stated conclusion that multifrecuency ECT can detect defects at or above the top surface of the tubesheet which 'would eopardize tube integrity during nor. mal operating or postulated accident condi ions, obs The staff wants to emphasize that as inspection techniques with increased capabilities, m
such as cultifrequency ECT, are developed, that many small volume defects which ect previously went undetected will now be found. These defects must be evaluated in
- f the context of the magnitude of defects which jeopardize tube integrity during normal tr(
or postulated accident conditions. As inspection techniques become more capable, M
correspondingly more discriminate criteria must be established.
Many plants which V
have not been inspected with multifrequency ECT are going to show new defects when "C
multifrequency inspections are performed.
These results must be dealt with re(Mially
]
and requirements for tube inspection, plugging, and removal must be carefully upplied, nt METALL0 GRAPHIC EXAMINATIONS
- l Members of the NRC staff and their consultant from Brookhaven National Laboratory met
'+ t with representatives from WEPC0 and their Westinghouse consultants in Pittsburgh f
on March 28, 1980 to review results of the metallographic examinations.
Review of 91 the photonicrographs described earlier revealed no general IGA similar to that occurring 1) within the tubesheet crevice above the top of the tubesheet in tubes R26-C53 or R30-gr C41.
Shallow grain boundary separation on the order of two grains or less existed on all photomicrographs of these tubes. Shallow grain boundary dissolution of this i }
nature can result from several mechanisms including previous operating environments nnin or tube pickling during manufacturing.
This grain boundary separation is much less
/ "I severe than that occurring within the tubesheet. The staff has concluded that the en shallow grain boundary dissolution at and above the top of the tubesheet is not I
significant in terms of tube integrity. Metallographic examination of tube R19-C37 revealed stress corrosion cracking and shallow IGA of the tube near the top of the b
tubesheet. Re-evaluation of past ECT tapes showed that this defect existed as far b
back as 1976 but was overlooked using single frequency ECT. The nature of the crack in is similar to that of stress corrosion cracks which occurred during previous operating periods. The staff believes that this is an old defect which has not significantly changed since 1976.
1)
CONCLUSI0."S E-Based on the information presented above the staff has reached the following co)n-in it' clusions:
- am he
- 1) The inspection 'and' tube plugging performed has been adequate to ensure the H n' great majority of defective tubes have been removed from service.
Mfi t h'
- 2) Multiple frequency eddy current testing used to perform the inspection is capable of detecting defects near the tubesheet and tube support plate interfaces which I
would jeopardize integrity of the tube during nonnal operation or postulated l
accide'it condi tions.
, 3) Hydrostatic tests sirJ13 ting postulated accident conditions performed prior tc returning to operation will identify any significant defects overlooked during ECT examination.
- 4) Ir.tergranular attack at and above the top of the tubesheet as observed in the re.cVed tube samples is extremely shallow and poses no threat to tube integrity at or above the top of the tubesheet.
- 5) Based on the number cf new defects, the rate of deep crevice corrosion a; pears to have decreased.
- 6) A maximum 90 effective full power day operating period, prior to the next ECT inspection as proposed by the licensee, will provide adequate assurance that a large number of tubes will not simultaneously reach a point of incipient failure.
- 7) Remedial actions proposed by the licensee will continue to mitigate the effects of postulated accidents and retard the rate of corrosion.
The staff has determined that the following conditions should be required for continued operation:
- 1) Within 90 effecting full power days from the date of this order, a 2,000 psid primary-to-secondary hydrostatic test and 800 psid secondary-to-primary hydrostatic test shall be perforrec. Also during this plant outage, an eddy current examina-tion shall be perforred on tubes in each steam generator.
The program shall require such examinations of about 1000 tubes in the central region of the hot leg, three (3) percent of all hot leg tubes outside this central region and 3%
of the cold leg tubes. The Central region shall encompass all areas where deep crevice corrosion has previously been observed.
- 2) Prirary coolant activity for Point Beach Nuclear Plant Unit 1 will be limited in accordance with the provisions of Sections 3.4.8 and 4.4.8 of the Standard Technical Specifications for Westinghouse Pressurized Water Reactors Revision 2, July 1979, rather than Technical Specification 15.3.1.C appended to License 0FR-24.
- 3) Close surveillance of primary to secondary leakage will be continued and the reactor will be shut down for tube plugging on detection and confirmation of any of the following conditions:
a) Sudden primary to secondary leakage of 150 gpd (0.1 gpm) in either steam generator; b) Any primary to secondary leakage in excess of 250 gpd (0.17 gpm) in either steam generator; or c) An upward trend in primary to secondary leakage in excess of 15 gpd (0,01 gpm) per day, when measured primary to secondary leakage is above 150 gpd.
l l
i 4 The reactor will be shut down, any leaking steam generator tubes plugged, and an eddy current exanination performed if any of the following conditiens l
are present:
a) Confirmation of primary to secondary leakage in either steam generator in excess of 500 gpd (0.35 gpm); or, b) Any two identified leaking tubes in any 20 calendar day period.
This eddy current program will be as described in item 1.
l 5.
The NRC Staff will be provided with a summary of the results of the eddy current examination performed under items 1 and 4 above. This summary will include a photograph of the tubesheet of each steam generator which will verify the location of tubes which have been plugged.
6.
The licensee will r.ot resume operation after the eddy current examinations required to be performed in accordance with condition 1 or 4 until the Director Office of Nuclear Reactor Regulation has determined in writing that the results of such tests are acceptable.
These conditions are similar to those in the November 30, 1979 Order except that' the approved operating period.has been lengthened from 60 to 90 effective full power days. and no shutdown to perform hydrostatic tests are being required prior to.the end of the 90 day period.
These conditions differ from the licensees proposal in that the primary.to secondary leak rate limits and requirements for ECT examination are more conservative.
On the basis of our review and evaluation, we conclude that continued ' safe operation of Point Beach Unit 1 may be permitted within the stated terms of the Confirmatory 1
Order.
~
I k
I i
J E
APPENDIX I -
l TABLE I POINT BEACH #1 'A' S/G M.F.
M.F.
Tube
- Dec.
Oct.
R C
1980 1979 1979 12 19 80%
SAME SAME 19-21" ATE F(2,51 No R651 N.C.
na 7
22 295/96%
SAME NDD/SAME 12" ATE /17" ATE R251 N.C.
12" ATE /17" ATE R551 18 22 66%
SAME NDD 12-17" ATE R251 N.C.
R551 10 23 41%
NDD 20" ATE R251 R551 7
24 83%
MAYBE(?)
NDD 17"-20" ATE NDD R551 R251 8
24 79%
MAYBE(?)
NDD 17"-21" ATE NDD R551 R251 25 45 69%
Squirrels NDD 12"-20" ATE R351 R851
- 20 48 85%
SAME SAME 21" ATE R251 N.C.
R851 l
9 49 90%
NDD 21" ATE R251 17 50 85%
NDD 19" ATE R251 19 50 97%
NDD 11" ATE R251 20 50 97%
NDD 11" ATE R251 12 59 87%
MAYBE(?)
NDD 21" ATE NDD R951 R151 12 61 83%
NDD i
17" ATE R151 14 63 83%
MAYSE(?)
19" ATE Squirrels i
R151 l-
j POINT BEACH #1 'A' S/G M.F.
M.F.
Tube
- Dec.
Oct.
R C
1980 1973 1979 15 66 60S I
18" ATE i
8 27 Squirrels SAME 15-20" ATE R251 N.C.
15 28 Squirrels No 21" ATE Squirrels R251 '
28 34 Squirrels SAME 18-21" ATE R251 N.C.
I 28 35 Squirrels SAME l
17" ATE R251 N.C.
! 20 41 91 %
NDD l
19" ATE R351 25 43 73%
SNiE Very S.V.
17" ATE N.D.D.
R351 R751 11 46 Squirrels SAME 12"-21" ATE R351 29 52 Squirrels SAME 14" ATE R151 N.C.
i.
t I
s I'
f l
l_ _
~
APPENDIX i TABLE 11 B S/G INLET POINT BEACH *1 M.F.
M.F.
S.F.
Tube =
n Dec.
Oct.
Aug.
R C
1980 1979 1979 1979 18 26 75%
SAME Changed NDD 18" ATE R151 No R651 R551 change 13 26 73%
SAME SAME NDD 21" ATE R151 N.C.
R651 N.C.
R551 13 33 71%
SAME Changed NDD 20" ATE R151 N.C.
R651 R552 6
24 91%
SAME SAME NDD 11" ATE R151 N.C.
R651 N.C.
R552 20 35 68%
SAME Changed NDD 21" ATE R151 N.C.
R351 R552 8
37 89%
NDD 5" ATE R151 19 37 58%
SAME SAME NDD 1/2" ATS 53%
R351 N.C.
R552 R151 N.C.
10 41 70%
SAME NDD 21" ATE R251 N.C.
R751 R651 30 41 47%
SAME Some Change NDD 21" ATE R251 N.C.
R751 R651/R151 30 42 48%
SAME Changed NDD 21" ATE R251 N.C.
R751 R151 22 46 76%
SAME NDD 15" ATE R251 N.C.
R351 l 24 48 84%
Changed NDD 12" ATE R251' R351 R652
' 30 48 85%
SAME SAME NDD 21" ATE R251 N.C.
R951 N.C.
R652 25 49 84%
Changed NDD 5" ATE R251 R351 R652 l20 51 99%(?)
SAME NDD l
16" ATE R251 N.C.
R351 R652 23 54" 86%
Full length some are new R351 R251*
l l--_-_.
l l
B S/G INLET POINT BEACH 01 M.F.
M.F.
S.F.
Tube #
Dec.
Oct; Aug.
R C
1980 1979 1979 1979 I
23 57 56%
NDD 17" ATE R251
' 21 58 I 83; SAME 21" ATE R251 14 59 755 NDD 21" ATE R251 21 63 62%
SAME NDD 21" ATE R351 R1051 12 67 665 NDD 21" ATE R351 R1051 2
72 92%
SAME NDD Top of Roll R351 N.C.
R1051 26 53 86% (New) 18" ATE 3D 43 Squirrels SAME SAME 21" ATE R251 R751 26 53 Squirrels NDD.
Full T.S.
R251 25 55 Squirrels NDD Full T.S.
R251 22 63 Squirrels SAME SAME 21" ATE R251,
R1051 22 64 Squirrels SAME '
No 20" ATE R351 Squirrels R1051-j 25 55 74%(New) 15" 1.TE I
i l
j