ML20198S574

From kanterella
Jump to navigation Jump to search
Requests NRC Review of Proposed Second non-ASME Pipe Repair to Previously Repaired Spool Piece in SSW Sys.Nrc Originally Granted Approval During 971118 Telcon
ML20198S574
Person / Time
Site: Pilgrim
Issue date: 01/15/1998
From: Desmond N
BOSTON EDISON CO.
To:
NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
BECO-2.98.004, GL-90-05, GL-90-5, NUDOCS 9801260223
Download: ML20198S574 (5)


Text

___-- - --

0 10 CFR 50.55a i

Generic Letter 90-05 Boston Edison Pilgrim Nuclear Power Statiori hocky Mll Road Plymouth, Massachusetts 02360-5599 January 15,1998 BECo Ltr. L98.004 Nancy L Desmond Regulatory Relations Group Manager U. S. Nuclear Regulatory Commission Attn.: Document Control Desk Washington DC 20555 Docket W. 50-293 License No. DPR-35 Beauest for NRC Review of a Proposed Non ASME Code Pipe Repair Purpose This letter requests NRC review of a proposed non-ASME code pipe repair to a previously repaired spool place in the salt service water (SSW) system. Based on a verbal discussion between the NRC and Pilgrim of 110 foilowing evaluation, the NRC granted approval for the repair via telephone on November 9,1997. Shnrtly following the repair, the affected spool was replaced during a forced outage that commenced on November 23,1997. Although the spool p!ece has subsequently been replaced, this letter provides a mechanism for formally requesting and receiving NRC approval of the second norr-ASME code repair of the spool.

Backaround By letter of July 7,1997, Boston Edison Company reported degradation of a spool piece associated with Pilgrim Nuclear Power Station's (PNPS) salt service water (SSW) system.

This system provides the ultimate heat sink for containment heat removal.

In addition, the July 7,1997, letter requested NRC permission to perform a temporary repair in accordance with Generic Letter (GL) 90-05 and, in part, ASME Code Case N-562 using carbon steel plates.

I O Relief to perform this repair was requested from the NRC under the purview of 10CFR50.55a (g)'S)(i). Verbal permission was granted, the repair was made with carbon steel plates, and the NRC safety evaluation was received in an NRC letter dated October 1,1997.

The piping immediately downstream of the MO-3806 butterfly valve had through-wall leaks due to localized delsmination of the ruuber lining and subcequent erosion and corrosion of the carbon steel pipe prior to the repair. The leaks were adjacent to the pipe slip-on flange that mates with the valve. This location is downstream of the reactor building closed cooling water (RBCCW) heat exchanger.

9901260223 900115

  • f'W ADOCK 05000293 1 of 5 PDR

., s e d

Aft:r th3 r:prir, we performed insp;ctions to monitor tha condition of tha r;prir cnd th3 erosion / corrosion rate. The pipe in the area of the repair was found to be eroding at a rate greater than that projected; therefore, we intended to perform further repairs using 300 series (316 or 304) stainless steel plates in place of the carbon steel plates used in our first repair

[

ef, fort.

Evaluation Boston Edison Company (BECo) performed a structural integrity evaluation of the affected piping using ultrasonic testing (UT) data for wall thickness in the vicinity of the original leaks.

1 The pipe evaluation is in accordance with the guidance provided in Generic Letter 90-05 for a W

through wall flaw in American Society of Mechanical Engineers (ASME) Safety Class 3 piping. The method evaluates the stress intensity factor "K" in the pipe with the limiting i

ircumferential length removed based on the pipe stresses from existing PNPS piping

=

analysis of record for combined loads. including seismic (SSE). The maximum allowable flaw length was calculated using the GL90-05 fracture toughness criteria of Ke = 35 (ksi)(in)".

There were three discrete through wall flaws; all are within the stress criteria allowabie flaw size.

GL90-05 requires that the flaw size be limited to the lesser of 3 inches or 15% of the length of the circumference. Based on the measured flaw sizes, including the total length that is below tm,n adjacent to the through-wall area, the flaws are within the criteria. The GL90-05 proximity requirement that the adjacent through wall flaws be spaced at greater than twice tmin was also considered. Therefore, the piping was structurally sound and capable of performing 5

its design function.

Based on the known operating history, it'spection, maintenance and tost requirements for the SSW system as validated through interviews with the design engineers, system engineer, j

and QC/ISI inspectors, the preliminary root cause of the through-wall leaks has been attributed to delamination of the aging rubber pipe lining. The erosion / corrosion rate is further exacerbated by localized high flow velocities resulting from throttling of the butterfly valve immediately upstream. Rubber lined steel piping flaws experience accelerated resion and corrosion where tho rubber lining has delaminated. Where the lining remains intact the pipe remains at its nominal full wall thickness (tonm). Hence, the wall thinning is local to the areas where lining has delaminated, while elsewhere there is no effect. Therefore, the through-wall leaks in this piping were due to a small area delamination of the lining resulting in localized erosion and corrosion.

Prior to the July 7,1997, request, PNPS performed an analysis using a hydrat:lic model for the SSW system to evaluate the actLal pressura at the subject location in the SSW piping.

This analysis showed the pressure at this location is usually slightly negative except at the highest yearly tides (above +11 ft). At the highest tides, this location has a slight positive l

pressure, resulting in service water leakage. No safety related components are within the proximity of the piping flaw location that would be directly affected by this leakage. The leakage would be accommodated by the design of the auxiliary bay.

There is usually a small vacuum in the pipe at this location related to the changing tides. Air in-leakage has a negligible effect on the flow rate through the RBCCW heat exchcnger.

Conclusion of EvaluatiorJ The above discussion and associated calculations / operability evaluation demonstrated that the pipe structuralintegrity was acceptable. The effect from SSW leakage into the auxiliary equipment bay and/or air in-leakage into the flow stream (i.e., when the pressure is negative 20f5 4

.~

. ~ = -. -. -. - - --. _. - -

3-'

^

o l

. ct this location) were acceptablD.- Thtreforo, the systim tssocint:d with the degraded spool '

piece was capable of performing its safety function; hence, it was operable.

The? preliminary root cause determination indicated-the flaws can.be attributed to U

delaminationL of the aging: rubber ~ pipe-lining. The erosion / corrosion rate was further 9

' exacerbated by localized high flow velocities resulting from throttling of the butterfly' valve immediately upstream.

4:

Monitorina Measures

Immediate compensatory measures were not required to assure system operability or. safe operation because the piping was structurally sound and leakage did not adversely impact system operability.

Ongoing pipe monitoring using UT was being perfctmed periodically to ensure that the pipe condition. did not deteriorate beyond acceptabis limits, in addition, operator tours were

. performed once per shift to monitor for changes to the leakage rate.

4 In addition, GL90 05 requires that a minimum of 5 locations be subject to augmented-j -

inspections to evaluate other system locations for similar degradation. As stated in the July

'7,1997, request, auxiliary bay SSW piping is inspected on a programmatic basis. Therefore, I.-

the only locations that required immediate inspection were similar locations downstream of
the other RBCCW and TBCCW heat exchanger outlet valves.

(

To address 'this, 5 locations described in our July 7,1997, request were inspected in accordance with GL 90-05 guidance prior to the initial repair. All augmented inspection results at these locations found values greater than the manufacturer's ts Reason for Non-Code Temporary Repair j.

As provided in our July 7,1997, request, the impact a code repair would have ori plant operation was assessed. Also assessed was the impact of 9 r: umber of non-code repair methods.'The code repair methods require removing one loop of the SSW system from service and cross tying the RBCCW systems during power operation, placing Pilgrim in a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> limiting condition for operation (LCO) under Technical Specification section 3.5.B.3.

The code repair we considered viable (spool replacement) requires removing a loop' from

- service for greater than the LCO's-24 hours, resulting in a plant shutdown. Hence, we requested relief in accordance with the guidance of GL 90-05 for a non-code repair that could

' be executed with the loop in-service; Description of Proposed Temporary Repair-

.A revised temporary non-code repair in the area previously repaired was proposed to stop potential leakage caused by erosion / corrosion occurring at a faster rate than projected at the time of; developing.the ' original' repair. This proposed stainless steel repair' maintained.

-structural integrity until-the piping spool was replaced during the November 23,1997, forced

- outage.4 The temporary repair was a stainles: steel cover plate welded to the pipe at the leak i location.~

~ ' '

~

- The repair added stainless. steel cover plates to the earlier repair. One cover plate was fillet twelded to the existing cover plate in the flange area near the valve (MO-3806). The second -

E

~ e cover plate was welded to the pipe downstream of the existing cover plate. This was done

because the carbon l steel cover plate originally installed was locally eroding.and the pipe downstream of the original cover plate was also locally eroding. The purpose of the' original.

13 of 5 J

--.-?---'

A M

sa.

4-i'-

+-

t---ig'h w-t' syyip H d ege --

p.i.g 9-w og g.

  • gmf-g ry g9 y yyrp.-y,t-tr--b--

9 r yi, y

y j,i r-P g --

r-

(July 7,1997) t:mporcry modification was to stop th3 lxk in tha SSW syst:m (Spool JF 29-8-2) and maintain structural integrity until replacement. The leak was in the service water outlet piping from RBCCW heat exchanger E-209B, downstream of valve MO-3806.

ABME Code Case N-562, although written as guidance for a weld overlay repair method, was used as a technical guide to attach the cover plate. The cover plate method was selected as the preferred temporary repair instead of the overlay method for the following reasons:

The cover plate repair method would stop the leak with lest risk of enlarging the flaws than the overlay method. All other guidance of N-562 was followed as applicable.

The cover plate was acceptable for 100 psi, although the pressure at the leak's location ranges from a slight vacuum to a slight positive pressure. (it is dependent on tide level because the line dischar0es to the sea.) The line's 100 psi design specification was selected at Pilgrim's construction to make it uniform to other parts of the system that are subjected to higher pressures; therefore,100 psi was a conservative value for the repair.

The pipe's stress is low (4 ksi) as shown by BECo's calculation M747 (which was provided with our July 7,1997, request). If it is intensified by a factor of 2.1, as prescribed by N-562, it would be within the allowable limit of 18 ksi.

The cover plate method was less intrusive to the structural integrity of the pipe because it exposes the pipe to less heat from the welding process. Qualified welding procedures were used.

The cover plate method a:d not affect plant operations.

The rate of material removal due to direct water impingement was not accurately predictable prior to the original repair. The carbon steel material chosen for that repair is prone to direct impingement erosion, but the estimated rate was such as to allow the selection of carbon steel. Later (post-repair) data indicated that it was preferable to use a 3/8" rolled 300 series stainless plate material for the cover plates which wN e welded with E308 wire.

A cover plate had been welded to the 18 inch SSW pipe and flange where flow-assisted corrosion / erosion had occurred. The cover plate was eroding and the pipe downstream of the cover plate was also eroding. Two additional cover plates were installed to address the erosion problem. One stainless steel cover plate was welded to the existing cover plate and one was welded to the pipe down stream of the existing cover plate. ASME Code Case N-562 was used as a guide for performing the temporary modification. The cover plate was treated like an overlay weld, and the same N-562 rules were applied for overlap of the thinned area and for the NDE required. N-562 allows welding to the pipe with water inside with proper procedure qualification.

The minimum width of the new cover plates oeyond the required repair area is given in the Code Case as:

Sm = 0.75 x (Rxt )"

where R = outer radious of pipe; t

= nominal pipe wall thickness For this case, R = 9 in, and the nominal wall thickness is 0.312 in. This results in Sm = 1.26 in. A distance of 2 inches was used.

4 of 5

4 l

Repair's Safety impact Pilgrim performed a safety impact evaluation of the repair that determined the following:

.' The safety-related functions of tne SSW system remained qualified for plant design bases loads after completion of the temporary repair.

The temporary repair did not increase the probability of occurrence or consequences of an accident or malfunction of equipment important to safetv. The possibility of creating an accident or malfunction other than those evaluated in the UFSAR was not increased because the temporary modification did not introduce any interaction with other safety-related systems.

This temporary repair did not increase the probability of occurrence or consequences of failure of equipment important to safety because no new failure mechanisms were introduced.

Commitments Our July 7,1997, letter contained commitments that would have remained in force with the new repair; however, spool piece replacment made these compensatory measures unnecessary.

Should you require further information on this issue, please contact P.M.Kahler at (508) 830-7939.

k 0%

N.L. Desmond PMK/dmc/pipea ec: Mr. Alan B. Wang, Project Manager Project Directorate 1-3 Office of Nuclear Reactor Regulation Mail Stop: OWF 1482 U. S. Nucient Regulatory Commission 1 White Flint North 11555 Rockville Pike Rockville, MD 20852 U.S. Nuclear Regulatory Commission Region i 475 Al'endale Road King of Prussia, PA 19406 Senior Resident inspector Filgrim Nuclear Power Station Mr. Peter LaPorte, Director Massachusetts Energy Management Agency 400 Worcester Road P.O. Box 1496 Framingham, MA 01701-0313 Attn: Mr. James Muckerheide 5 of 5

_ _ _ - _ _ _ _ _ _ _ _ _ _ __ __- _ ___ _ _