IR 05000341/1987022
| ML20236C438 | |
| Person / Time | |
|---|---|
| Site: | Fermi  |
| Issue date: | 07/23/1987 |
| From: | Harper J, Phillips M, Wohld P NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML20236C397 | List: |
| References | |
| 50-341-87-22, IEB-85-003, IEB-85-3, NUDOCS 8707300067 | |
| Download: ML20236C438 (8) | |
Text
-
_.
__ _.
_
_
,
j
.
.
.
!
!
l U.S. NUCLEAR REGULATORY COMMISSION
REGION III
-i i
Report No. 50-341/87022(DRS)
j a
Docket No. 50-341 License No. NPF-43 j
Licensee: Detroit Edison Company l
2000 Second Avenue
Detroit, MI 48224 Facility Name:
Enrico Fermi Nuclear Power Plant, Unit 2 l
Inspection At:
Enrico Fermi 2 Site, Monroe, Michigan Inspection Conducted: March 30 through July 9, 1987
[f/f/
I 23,/987 Inspector:
P. R. Wohld Date'
'
b 7 3-O f
F J. C. Harper Vendor Inspection ranch, NRR Date Approved By: M P
ps, C ief 28/8~7 Operational Programs Section Date Inspection Summary Inspection on March 30 through July 9, 1987 (Report No. 50-341/022(DRS))
Areas Inspected: Special safety inspection of the licensee activities with respect to main steam isolation valve spring failures, Jamesbury butterfly valve failures, and IE Bulletin 85-03 regarding motor-operated valve switch settings.
Results: Of the three areas inspected, one violation was identified (Paragraph 3).
i I
'
..
8707300067 870723-PDR ADOCK 05000341 Q
!
l
I i
--
.
- .
,
.,
.
L l
DETAILS
'
,
h lL 1:
>
' 1..
Persons Contacted
'a.
Detroit Edison Company
+F. Agosti, Vice' President,. Nuclear Operations
- +S. Cashell, Licensing
- +K.. Earle, Technical Engineer,' Nuclear Production
+L. Esau, Maintenance Engineer
- S. Frost, Licensing
+C. Gelletly, Nuclear Projects and Plant Engineering
+J. Green, ' Systems Engineering.
+A. Lim,. Systems Engineering
+W. Orser, Vice President, Nuclear Engineering.and Services-
+T. Randazzo,. Director,' Regulatory Affairs
+B. Sheffel, Nuclear Production, Tecnical Engineering and ISI
- +J.' Thorpe, Systems' Engineering
.+G. Trahey, Director, Quality Assurance b.
U.S. Nuclear Regulatory Commission
- J. Harper, Inspector, Vendor Inspection Branch, Office of NRR
.#P. Pelke,< Project Inspector, Region III
+ Denotes those. attending the. exit meetingLheld June 5, 1987, at the.
,
Fermi Site.
!
- Denotes those_ participating in the phone conference. held July 9, 1987..
'
Additiona'l plant technical and administrative personnel were
,
contacted by the inspectors during the course of the inspection.
2.
Main Steam Isolation Valve Spring Failures (LER 86011)
During April 1986, when investigating valve leakage problems, the licensee discovered the failure of four Main Steam Isolation Valve (MSIV)
-springs on two valves, three on valve F0-28B and one on valve F0-28A.
The springs were part of a sixteen spring system for each valve used to assist valve closure. The licensee has evaluated the safety implications i
of the spring failures and determined that the impact is minimal. This
'
is based on the large number of springs available to assist closure and the fact that a safety-related air supply is availabe for MSIV closure.
i
?
'
'
.
__
l (
.
.
.
The inspector reviewed the licensee's followup activities during the week of May 19, 1986, including physical inspection of the valves, and considered the initial activities by the licensee to be acceptable.
Since the initial inspection, however, the material problem with the
springs still remains to be fully resolved because important information
from both old and replacement spring metallurgical evaluations was not adequately considered.
]
i The spring failure event was reported on May 28, 1986, in LER 86-11-00.
In that LER the licensee indicated that quench cracking and embrittlement due to improper heat treatment and tempering were the cause of the inner spring failures. Additionally, the licensee offered that the larger outer springs were manufactured with a different material, implying that this justified their exclusion from any metallurgical evaluation.
There were two problems with the licensee's activities subsequent to this report:
(1) the licensee failed to consider the material embrittlement properly when later endorsing a 105% compression test for the valves; and (2) the inner and cuter springs have since been determined to be made of the same material, yet, no metallurgical testing was done or proposed on a sample representative of the outer spring population installed.
Subsequent to compression testing and reinstalling the old inner springs, the licensee discovered two more failed inner springs on March 21, 1987.
Mr. J. C. Harper from the Vender Inspection Branch of NRR conducted an inspection, between March 30 and April 1, 1987, and found that the licensee had discovered on May 11, 1986, that four inner external MSIV springs'(17/32" Dia x 4-5/8" 00) on the outboard valves were broken.
These had been supplied to the valve manufacturer (Atwood & Morrill) by Duer Spring. Of the springs that broke, three had one break approximately 2/3 the distance from the bottom of the springs and the remaining spring broke in several places. According to Atwood & Morrill Company, Inc.
(A&M), the outer springs were also supplied from Duer Spring on the same purchase order. A&M claims that the outer springs were from a different heat number.
i Three broken springs were found on the upper units and the remaining broken spring was found on the lower unit. As stated previously, four inner external MSIV springs of two outboard valves were discovered broken (17/32" dia. x 4-5/8 00).
Valve B2100F028A contained one broken spring and valve B2100F028B contained three broken springs. These springs were AISI 5160 material, purchased with valves from A&M, Part No. 33134-824-3219.
The springs were purchased by A&M from Duer Spring (A&M Purchase Order No. AM5243 Heat No. 8067703, Oven No. 20224-R5, test report dated January 6, 1972).
Two of the broken springs and three non-failed springs were submitted to the Detroit Edison Engineering Research Division (DER)
for metallurgical examination. The outer springs (6-1/4 inch 0.0.) were
'
not examined.
From their investigation, DER concluded that the root cause of the spring failure (Heat No. 8067703) was three fold in nature:
(3) temper embrittlement; (2) quench cracking; and (3) surface imperfections.
e
. _ _ _ _. _ _ _ _. _
__.._______w
..
.
L
{
i
!
According to DER [in a 6-24-86 report], "The metallurgical results revealed that failure was attributable to pre-existing seams and quench
!
cracks (induced during heat treatment) which propagated as brittle
fracture due to temper embrittlement of-the spring material." The three non-failed springs were examined and quench cracks were found in two of
,
the three springs; all three springs exhibited temper embrittlement.
J l
Based on these findings, it was recommended that all inner springs be replaced.
From independent review of the DER metallurgical report, the NRC inspectors concur with the DER conclusions. The materiaT, AI5f-5160 used in the
failed springs, is highly susceptible to 500 F embrittlement (or temper embrittlement) because the alloy contains substantial amounts of chromium, phosphorus, and manganese. Temper embrittlement is caused by tempering a susceptible' alloy in the 400-700 F range, thereby allowing manganese (Mn)
and phosphorus (P) to' segregate at the grain boundaries and subsequently
,
render the material brittle. As reported by DER, Scanning Electron
Microscopy (SEM) of the fracture surfaces revealed predominate intergranular type fracture and numerous grain boundary cracks (this was reported to have been observed on the failed spring sample away from the quench cracks).
The fracture surface SEM micrograph reported by DER is good evidence that-the spring material was temper embrittled during heat treatment (HT). The DER investigation stated that the non-failed inner springs examined were also temper embrittled. As a result of destructive impact testing it was
determined that the non-failed springs had a similar grain boundary l
fracture as the failed springs.
During forming, the bar stock was heated to 1700-1750 F and the springs coiled. The springs were then quenched in oil to 400oF.
Subsequently, the springs were tempered in a furnace at 850 F for 1.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />.
According to A&M during the time frame when the subject springs were manufactured, j
Duer's heat treating processes were being carried out manually. Manual
heat treating is subject to many more errors in process control than
.
automated, continuous heat treating processes.
It is conceivable that a i
loss of control during manual heat treating caused problems with the i
failed MSIV springs (quench cracking and temper embrittlement).
In this i
situation, overheating during austenizing and/or using an improper i
quenching rate and medium (water instead of oil) are likely causes of
'
quench cracking.
However, it is more probable that time delays between the austenizing/ quenching process and the tempering process caused the quench cracking, since a manual HT process transfers materials from one HT station to the next in a noncontinuous manner.
It is known that time delays before tempering and tempering a material that is below recommended tempering temperatures promotes quench cracking. At present, Duer employ-an automated, continuous HT process.
_ - _ _ _ _ _ _ _ - _ _ _ _ _ - _ _ _ _ _ _ -
<
?!
.
. Detroit Edison' contracted independent consultant, Failure' Analysis;
'
2 Associates (FAA) to determine whether. fatigue.or stress corrosions
~
cracking.were probable causes of failure for the springs. 'According sto FAA', in its position.as advisor to Fermi 2 Nuclear Production, FAA did not perform an independent metallurgical evaluation of the' failed springs. However,,FAA did.revie' some of the metallurgical evidence prepared by. Detroit Edison DER.. FAA did perform a fracture' mechanics analysis and based on those results concurred with General Electric's
'
(GE'.s) position that a 105%' overload test on the springs would be a suitable means of separating defective spr.ings.
For BWRs in operation, the GE Service Information Letter,' dated July.18,1986, titled
" Inspection of Atwood & Morrill MSIV External Springs" recommended replacement of any cracked or broken springs fo.llowing a five percent
,
overload test performed with springs installed or as a bench test. The-l springs were to be stroked ten times to 105% of the normal ~ stroke at a compression rate 'ess than one inch per second. According to GE, "A-spring which passes this (the 105% of load) test is expected to ' provide satisfactory service at normal load." The FAA justification of GE's recommendation was based on their fracture mechanics analysis.
- However, it. should be. noted that the FAA fracture mechanics analysis was in contradiction to the conclusions reached in the DER fracture mechanics study. DER concluded from their fracture mechanics analysis that the 105% overload test would not be an adequate means of sorting out-i defective springs.
The FAA fracture mechanics analysis did not take into ~
account that the inner springs had been temper embrittled, therefore their.
analysis was carried out using literature fracture toughness' parameters of properly heat treated AISI 4340 spring material instead of a comparable material that was temper embrittled. On the other hand DER's fracture mechanics analysis utilized fracture toughness values for the embrittled springs (30-40 ksi-inh) in order to calculate a critical crack size of
.03 inches (using the A&M reported service stress of-.87.6 ksi -' valve in
'
open' position). The actual critical crack sizes measured on the tested material by DER varied from.044.077 inches.
DER estimated that in a five percent overlaod condition a critical crack size.of.027" would result, as opposed to a critical crack size of.030 inches in the non-overloaded condition (assucing embrittled material).
Based on these
calculations it was correctly cencluded'that a proof load test would not be an effective means of. sorting out defective springs.
On June 2, 1986 a Part 21 evaluation was completed by Fermi 2.
The evaluation was limited to the inner-external springs and did not address temper-embrittlement as being a controlling factor in the root cause of
the failure. According to the evaluation, "... All non-failed springs (
- .
L, in all eight of the MISV's were proof tested according to a plan outlined by Detroit Edison consultants, FAA and concurred with by GE and Detroit Edison Engineering." [ DER stated three weeks later, however, in their i
June 24, 1986 report (and as noted above) that they did not concur with the proof testing adequacy.]
l
)
I
!
i I
i k
.
m
.,
.
- The. basis for the 105% compression test was unacceptable' based both on Lthe' June 24,1986'. input from DER and the additional spring failures discovered March 21,.1987. The Fermi. staff failed to take'any,further corrective action _ after they learned of-the material embrittlement problem -
!
and test: inadequacy until after the second spring failure was inadvertently
<
. discovered... Also, on. learning that the outer springs are'made of the same
!
material, and'after a request from Mr. Harper,.there is still a question
'
on the material condition of the installed outer springs. When asked for i
a test report on.the replacements for'the inner springs, the licensee provided a copy of DER Report 87C63-11, dated April 23, 1987.
This i
report indicates quality problems with these springs as well, based on cracks identified in a three spring sample and recommended magnetic particle testing 'on the remaining springs.
Because of questions regarding the licensee's handling of the MSIV springs, as noted above, a telephone conference was held on July 9.
During the conference the licensee's staff indicated that they intended to replace all. currently installed valve springs in early 1988 during a shutdown-for valve. leak rate te' sting. 'The licensee's staff also stated that all
.
replacement springs will be:
(1) Tested for spring constant.
.i
.(2) Magnetic particle tested after heat treatment and before.. painting.
(3) Compression teste'd to the solid condition.
Also, 13 spare' inner springs and 13 outer springs will be ordered with the replacement springs and will be compression cycled 5,000 times each,
.over ten times the design l number, to' assure the quality of the springs.
'
The completion of these proposed activities. appears fully adequate to conclude any question with respect to the valve springs.
I While a violation was considered with respect to the licensee's handling of correcting the MSIV spring problem, considerations with respect to the safety significance, the low probability of unacceptable spring failure combinations, and the fully adequate corrective actions now proposed indicate that a violation was not appropriate.
LER 86-011-01 addressing the MSIV problems will remain open pending the
)
successful completion of corrective actions.
3.
Jamesbury Butterfly Valve Failures (LER 86027)
{
l Approximately 32 safety-related, Jamesbury butterfly valves are used at
]
Fermi. Three valves have failed within that many years (T23-F409 in September 1984, T46-F401 in November 1985, and T23-F410 in August 1986).
-The shafts were frozen in position by corrosion products from the carbon
!
steel shaft binding with epoxy impregnated fiberglass bushings used-for I
shaft bearings. The vendor design no longer incorporates carbon steel i
shafts and replacement parts are now provided in stainless.
j
6 L
_
_
. - - - - - - - - - _ _ _ _ _ _ _ _ _ - - _ _ _
_- - --
_ _-__
/
c.
.
A review of the licensee's corrective action to all three ' failures showed that it: lacked death and timeliness.
In the opinion of the inspector, the des.ign weakrjess of the materials used should have-been better appreciated after the first. failure tand more positive measures taken to l
'
. prevent.the_two subsequent failures.,Even after the last failure,.there have been no appropriate preventive maintenance measures put in place and increased stroke time' evaluations were being limited to'an inappropriately restricted subset of the remaining valves. An evaluation was done to limit
.the' valves receiving increased attention but it did not account for degrading conditions.all the valves could have seen during storage, installation, and the long construct'on period.
The 25% valve stroke time increase used by the licensee to detect valve problems allowed the latest failure to occur ijradually. A review of stroke time trend. data for other Jamesburg valves indicated erratic behavior suggesting the need for closer evaluation of the physical condition of additional valves.
The plant staff was aware of the inspector's concerns and was reviewing their program for additional measures that could be taken.. The failure to take prompt and effective corrective action to the initial failure
!
resulted in subsequent failures from the same mechanism and, even after the third failure, measures taken appeared less than fully adequate to prevent additional failures.
This is considered to be a violation of Criterion XVI, Corrective Action, in Appendix B of 10 CFR Part 50 (461/87022-01(DRS)).
4.
IE Bulletin Followup (0 pen)-IE8 85-03: Motor-0perated Valve (MOV) Common Mode Failure During Plant Transients Due To Improper Switch Settings.
~
The bulletin was discussed briefly with the licensee to determine the status of their progress in addressing the issues raised and a number of drawings and procedures were received for. review.
The level of involvement in addressing the issues expressed in the bulletin was less than desired and one significant issue was identified.
The licensee has adopted a valve control logic that deliberately backseats-globe and gate valves using the Limitorque motor-operator torque switch. This was pointed out as highly undesirable in the bulletin; however, the licensee's staff believed there was no problem with it. At the request of the inspector, Detroit Edison was reviewing the acceptability of their approach with Powell and Rockwell, the valve
.
manufacturers.
Resolution of this concern is considered an unresolved
!
item (341/87022-02(DRS)).
J
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ -
- - _ _ -
-
_
- -
_
a.;. ;
The inspector expressed concern for MOVs at the Fermi site for several
- reasons which, in'~ addition to the backseating ' issue and Jamesbury valve failures,- included the failure of two MOVs when tested during the j
inspection' period. The resident inspectors are also concerned about the
,
reliability of the site's MOVs.
In response to this, the licensee.is.
'
' gearing up its Bulletin 85-03 effort to issue contracts, as' necessary, and' schedule the required effort.
No violations were identified.
!
i 5 '.
Unresolved Items i-Unresolved items are matters about which information is' required to ascertain whether they are acceptable items,Jviolations, or deviations.
'!
One unresolved item identified during the inspection is discussed in j
Paragraph 4.
'
6'.
Exit Meetings.
.The inspector met with the licensee representatives'(denoted in
'I Paragraph' 1).on June 5,1987, and. held a phone conference on July 9, 1987, to discuss the scope and findings of the inspection.
The licensee acknowledged.the statements made by the inspector with respect to items
i discussed in.the report.
The inspector also discussed the.likely informational content of the inspection report with regard to documents or processes reviewed by i
the inspector during the inspection. The applicant did not identify any such documents / processes as proprietary.
<
l l
l l
________________________a