ML20212G607

From kanterella
Jump to navigation Jump to search
Power Reactor EVENTS.May-June 1986
ML20212G607
Person / Time
Issue date: 12/31/1986
From: Massaro S
NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD)
To:
References
NUREG-BR-0051, NUREG-BR-0051-V08-N3, NUREG-BR-51, NUREG-BR-51-V8-N3, NUDOCS 8701210084
Download: ML20212G607 (43)


Text

- - - - - - - - - - - - - - - - - - - - - .

NUREG/BR-0051 Vol. 8, No. 3

, [T

\ /

POWER United States Nuclear Regulatory Commission REACTClR EVENTS Date Published: DECEMBER 1986

Power Reactor Events is a bi-monthly newsletter that compiles operating experience information about commercial nuclear power plants. This includes summaries of noteworthy events and listings and/or abstracts of USNRC and other documents that discuss safety.related or possible generic issues. It is intended to feed back some of the lessons learned from operational experience to the various plant personnel, i.e., managers, licensed reactor operators, training coor-dinators, and support personnel. Referenced documents are available from the USNRC Public Document Room at 1717 H Street, Washington, D.C. 205ss for a copying fee. Subscriptions of Power Reactor Events may be requested from the Superintendent of Documents, U.S. Covernment Printing Office, Washington, D.C. 20402, or on (202) 783-3238.

Table of Contents Page 1.0 SUMMA RIES OF EVENTS... I 1.1 Fuel Rod Damage Due to Baffle Jetting at McGuire Unit 1.. . . I 1.2 Failure of Emergency Service Water Pump Due to Cavitation at Susquehanna Units 1 and 2.. 5 1.3 Failures of Residual Heat Removal Pump Wear Rings Due to Intergranular Stress Corrosion Cracking at Browns Ferry Unit 2.. 7 1.4 Fire Protection System Actuation Results in Emergency Core Cooling System Actuation and Flooding at Browns Feny Units 1,2, and 3. 9 1.5 Urs sie on inadvertent Actuations of Deluge Spray System at River Bend = . . . 14 1.6 References.= 1G 2.0 EXCERPTS OF SELECTED LICENSEE EVENT REPOR TS.;  :.- 17 3.0 ABSTRACTS / LISTINGS OF OTHER NRC OPERA TING EXPERIENCE DOCUMENTS 29 3.1 Abnormal Occurrence Reports (NUREGD090) .- . . . . . 29 3.2 Bulletins and Information Notices.= 31 3.3 Case Studies and Engineering Evaluations. _ . 34 3.4 Generic Letters.-- . . . 39 h 3.5 Operating Reactor Event Memoranda. .. . . . . . . . . 40 3.6 NRC Documentation Compilations..... 41 Editor: Sheryl A. Massaro s

' Office for Analysis and Evaluation of Operational Data j'l U.S. Nuclear Regulatory Commission Period Covered: May-June 1986 Washington, D.C. 20555 8701210084 861231 p PDR NUREG PDR BR-OOS1 R C--- --_

1.0 SUMMARIES OF EVENTS 1.1 Fuel Rod Damage Due to Baffle Jetting at McGuire Unit 1 On June 26, 1986, debris was discovered on the McGuire Unit 1* reactor core baffle plate during performance of the reactor core verification procedure following reactor core reload. Video inspections of the baffle plate revealed what was suspected to be loose fuel pellets. Westinghouse (the nuclear steam supply system vendor) performed a video inspection of the reactor core and confirmed the presence of four whole fuel pellets and some pieces of fuel pellets near core location P-3. The damaged fuel was in a low power region of 3

the core, reducing the potential for fission products in the coolant. It was determined that the damage was due to baffle jetting. The core was unloaded, the baffle gaps were measured, and several relevant fuel assemblies were inspected; no further damage was found.

The driving force for baffle jetting is a pressure differential across the baffle joints due to reactor coolant flowing downward on the outer surface and flowing upward on the core side of the baffle plates. Two types of baffle gaps related to fuel failures are (1) center-injection joints, where the direction of the impinging flow is perpendicular to the outer row of fuel pins and (2) corner-injection joints, where water flows parallel to the outer row of pins adjacent to the baffle plate. Fuel assembly damage from baffle jet impingement also has been identified at Trojan, Farley, Point Beach, and at other reactors both within the United States and overseas. The fuel damage at McGuire occurred in a corner-injection location. The events leading to the debris discovery and corrective actions are detailed below.

On May 16, 1986, McGuire Unit 1 completed fuel cycle no. 3. Personnel began unloading the reactor core, and completed this activity on June 12. Following the reactor core unloading, an examination of selected fuel assembles was performed. On June 17, it was discovered that fuel assembly D-08 had grid strap assembly damage. On June 18, a determination was made to use fuel assembly D-08 during the next fuel cycle because the damage to the grid strap assembly was insignificant.

On June 21, personnel began loading the reactor core, and on June 26 the loading was completed. Personnel began performing the reactor core verifica-tion procedure, and observed debris on the core baffle plate near reactor core location P-3. (See Figure 1.) A video inspection of the core baffle plate was performed, and what was suspected to be fuel pellet debris was observed near reactor core location P-3. The NRC was notified of the possibility of fuel damage in the reactor core.

During the morning of June 27, Westinghouse personnel performed another video inspection of the core baffle area and confirmed the presence of four whole fuel pellets and some pieces of fuel pellets near reactor core location P-3.

4 McGuire management then established a task force to investigate and address the possible fuel damage.

  • McGuire Unit 1 is a 1150 MWe (net) MDC Westinghouse PWR located 17 miles north of Charlotte, North Carolina, and is operated by Duke Power Company.

Center-injection Joint Corner Injection Joint a

i I

P-03 I I

~

1 3 Damaged Pins Core Baffles i

Jet Flow - II / /// / / / / / / / / / / /1

['7gggggo

/ 0000

/

/ /

Defective Joint  !  !

/ r k

Fuel Bundle s

i Figure 1 McGuire Unit 1 Cycle 3 Baffle Jetting Damage 2

i -

On June 28, a vacuum system was set up inside the reactor building in upper containment for the removal of the loose fuel pellets. On June 29, Westinghouse personnel vacuumed the loose fuel pellets from near reactor core location P-3. The fuel pellets were placed inside a container that was suspended in the refueling cavity for temporary storage prior to being removed to the spent fuel pool. The NRC was notified of a possible loss of fission product barrier, per the spent fuel damage procedure. Westinghouse personnel vacuumed the upender area of the reactor building refueling cavity for any

, possible pieces of fuel pellets. Health physics personnel utilized an

)

underwater radiation monitoring instrument with a maximum range of 1000 R/hr to i measure the radiation level of the fuel pellets. The radiation level was greater than 1000 R/hr. Westinghouse and licensee personnel conducted a video inspection of the spent fuel pool for loose fuel pellets; no fuel pellets were found.

Personnel then began removing three fuel assemblies from the reactor core that were next to fuel assembly D-03. Fucl assembly D-03 was suspect because it had been in reactor core location P-3 during the last video inspection. At this time, fuel assembly D-03 was residing in reactor core location L-5. The video inspection confirmed damage to D-03. More video inspections of the assembly were taken to determine clearances involved, prior to removing the assembly from the reactor core, to ensure there would be no interference with the fuel handling mast. Licensee and Westinghouse personnel reviewed the videotape of fuel assembly D-03 and decided to move the fuel assembly to the spent fuel pool for further inspections. A container was fabricated to position under the fuel assembly to catch loose fuel pellets if any fell during transport.

On July 1, fuel assembly D-03 was taken to the spent fuel pool. Fuel assembly E-47, which was in reactor core location L-4 in contact with the damaged face of fuel assembly D-03, also was removed from the reactor core and transported to the spent fuel pool. Following the fuel assembly inspection, fuel assembly E-47 showed no indications of damage. Fuel assembly D-03 revealed damage in the area of fuel rods 15,16, and 17 on face 1 of the assembly. The licensee then decided to unload the reactor core for further inspections. Video inspections performed on fuel assembly C-56, which was located in reactor core location P-3 during core cycles 1 and 1A, and fuel assembly D-05, located in reactor location P-3 during core cycle 2, revealed no damage.

On July 2, personnel began unloading the reactor core, and on July 4 the unloading was completed. Licensee and Westinghouse personnel began inspecting 16 fuel assemblies in the spent fuel pool that had been in corner baffle locations during the last fuel cycle. No fuel damage was found.

On July 5, a video inspection of the refueling cavity upender and lower internal areas was completed and showed no debris or fuel pellets. The corner injection baffles at reactor core location P-3 were measured. Baffle gaps up to 0.007 inches were noted.

i On July 6, the reactor core plate inspection was completed. Three pieces of debris were found, including a fuel rod spring and pieces of fuel rod cladding.

The remaining corner baffle gaps were measured (16 total). Westinghouse and licensee personnel completed video inspections of the 16 fuel assemblies that had been located in corner baffle locations. Fuel assembly D-08 had a small sharp edge on grid strap assembly 2 of face 4. Previously, during the post 3

irradiation examination, fuel assembly D-08 was determined to have a slight botton deformation in the corner between faces 1 and 3. This problem was unrelated to the problem with fuel assembly D-03. Fuel assembly inspections of eight fuel assemblies that had resided in center baffle locations during the I fuel cycle 3 revealed no damage.

On July 7, a video inspection of the area under the reactor core support plate was performed, and some debris was found. The debris and fuel pellets were transported to the spent fuel pool for storage. A video inspection of the spent fuel pool transfer canal and upender areas was performed and no debris or h fuel pellets were found.

For the next fuel cycle, licensee management decided to redesign the reactor core without assembly D-03. The revised reactor core design called for the use of fuel assemblies ZV-1 and D-36. During the time period from July 19 to July 25, these two fuel assemblies were modified by McGuire and Westinghouse personnel. This modification involved the installation of stainless steel rods to replace the existing fuel rods on the faces of the fuel assemblies that would be in the direct flow path of the baffle gap at reactor core location P-3. On July 27, personnel began loading the reactor core, and on July 31 the loading was completed.

Throughout this event, health physics personnel routinely surveyed the spent fuel pool cooling system piping, filters, and associated components for ur. usual radiation levels that might be attributed to fuel pellets in the system. No abnormal radiation levels were detected.

Based on indications of iodine-134 activity in the Unit I reactor coolant system, it was concluded that the fuel damage to fuel assembly D-03 probably had occurred just after the beginning of the core cycle 3, followed by gradual further degradation throughout the remainder of the core cycle. This conclusion was further supported by an increase in neptunium-239 activity from the end of core cycle 2 into core cycle 3.

Based on this conclusion, there were an estimated five leaking fuel rods present during core cycle 3. These consisted of the three damaged fuel rods in fuel assembl power level)y D-03, with welland developed two additional defects,fuel mostrods (based likely carri donover assumed core average from core cycle 2.

During core cycle 3, fuel assembly D-03 was located in reactor core location P-3. A subsequent inspection of corner baffle gaps at reactor core location P-3 revealed four baffle gap measurements of 0.005 inches, one baffle gap measurement of 0.003 inches, and three baffle gap reasurements of 0.005 inches.

Reactor core location A-5 had one baffle gap measurement of approximately 0.006 inches. Reactor core location E-1 had one baffle gap measurement of 0.003 '

inches and another of 0.005 inches. All other baffle gap measurements were less than 0.003 inches. The initial installation baffle gaps were less than 0.003 inches per Westinghouse specifications. $

Fuel assembly D-03, located in reactor core location P-3, during core cycle 3, '

was damaged on the face of the fuel assembly parallel to the baffle plate. The damage found indicated an enlarged corner baffle gap. The resulting increased j 4

)

i jet of water flowed parallel to the face of the fuel assembly between the fuel l assembly and the adjacent baffle plate. This type of flow also caused fuel rod

) swirling and vibration to occur at the rod locations that were damaced on fuel assembly D-03. The baffle jetting induced rod motion causes fuel rod fretting, resulting in abnormal clad wear against the Inconnel grid strap assemblies.

Baffle jetting parallel to the fuel rods ultimately results in fuel rod failure to adjacent fuel rods.

The licensee's planned corrective actiqns include the following:

(1) All 16 corr.er baffle gaps will be measured during the 1987 refueling outages.

(2) The affected faces of fuel assemblies located in corner baffle locations will be visually inspected durina the next refueling outages.

(3) The primary chemistry at the station will be monitored for 30 days following the unit startup; the data will be reviewed, and any course of action required at that point will be addressed.

(4) Neptunium-239 and iodine-134 activity in the reactor coo'lant system will be monitored to better identify activity changes that might indicate fuel damage, beginning with Unit 1 core cycle 4 The actual time of occurrence of the fuel rod failures in the event is unknown.

, Although fuel handling was not the cause of the damage, it is known that fuel handling aggravated the damage and caused further spread of uncontained fuel pellets. The damaged fuel was in a low power region of the core reducing the potential for fission products in the coolant. The coolant activity remained within Technical Specification limits the entire cycle with one post-trip exception. The observed fuel failures had minimal effect on public health and safety, and did not result in radioactivity levels or effluent releases in excess of those allowed by the operating license. (Refs.1 and 2.)

1.2 Failure of Emergency Service Water Pump Due to Cavitation at Susquehanna Units 1 and 2 On May 24, 1986, Susquehanna Units I and 2* were shut down after all four emergency service water (ESW) pumps were delcared inoperable. Unit I has been operating at 100% power, and Unit 2 at 91%.

The pumps were declared inoperable following the failure of the C ESW pump on May 22, caused by severe cavitation damage, and the discovery on May 24 of cavitation damage on the A ESF pump. Following the shutdown of both units, the B and D ESW pumps were inspected and also found to have cavitation damage.

However, the danage was less severe and the B and D ESW pumps were declared operable on May 28, 1986. Due to their similar design, three of four residual f

  • Susquehanna Units 1 and 2 are each 1032 MWe (net) MDC General Electric BWRs located 7 miles northeast of Berwick, Pennsylvania, and are operated by Pennsylvania Power and Light.

5

heat removal service water (RHRSW) pumps were inspected. Both Unit 1 RHRSW pumps were found to have minor cavitation damage. The Unit 2 RHRSV pump inspected had no damage. The four ESW pumps and two RHRSW pumps were repaired i and were operable as of June 10, 1986. The cavitation damage was caused by impeller suction recirculation cavitation, which resulted from operating the pumps at flows significantly less than design. Periodic inspections will be conducted to ensure pump integrity. An engineering review of the design and operating configurations of the system is being performed by the licensee to determine what changes can be made to prevent recirculation cavitation.

The ESW system at Susquehanna is comprised of two redundant loops with two 50%

capacity pumps in each loop. All four ESF pumps are sing 1c-stage, vertical circular pumps, Byron Jackson Type 24BXF. The A ESW loop contains the A and C pumps and the B ESF loop contains the B and D pumps. The ESW system provides

- ling to the diesel generators, emergency core cooling and reactor core ation cooling systen room coolers, RHR pump motor oil and seal coolers, and l tne control structure and switchgear chillers. The sequence of events which led to declaring the ESW system inoperable follows.

On May 22, 1986, an overcurrent alarm for the C ESW pump was received in the control room. Investigation revealed the pump motor to be running at low amperage, and the pump discharge check valve was closed. A sheared pump shaft was suspected. The C ESW pump was declar'ed inoperable, and a limiting con-dition for operation (LCO) was entered at 1:00 a.m. On May 23, the C ESW pump was disassembled. The bottom portion of the pump suction bell had separated from the pump body and had fallen into the pump pit. Inspection of the damaged parts revealed the suction bell Pd been penetrated around its entire circumference by what appeared t h e cavitation. On May 24, inspection by an underwater diver revealed similar but less severe damage to the A ESW pump suction bell. The A ESW pump was declared inoperable at 9:45 a.m. on May 24, 1986. Since the condition of the remaining B and D ESW pumps was not known, they were declared inoperable at 12:00 a.m. on May 24, and a controlled shutdown of both units was begun. As a precautionary measure, an Alert (emergency classification) was declared, and the Technical Support Center was activated. The Alert was ended at 10:00 p.m. on May 24, 1986, a few hours after both units re' ached hot rhutdown. Although the A, B, and D ESW pumps were declared inoperable, they were ' functional and continued to perform at or near their design capability during and following the unit shutdowns.

Visual inspection of the B ar,d 0 ESW pumps were subsequently made. Both pumps were found to have cavitation damage similar to the A and C pumps. However, the cavitation damage did not penetrate the suction bell wall of either pump.

The B and D ESW pumps were declared operable at 9:15 a.m. on May 28, 1986. Due to a lack of spare parts, temporary repairs were made to the C ESW pump to return it to a functional condition until the needed parts could be obtained.

On May 28, the C ESW pump was declared functional, thus restoring the A ESW loop to a functional status.

Following receipt of the required spare parts, the A and C ESW pumps were repaired, retested and declared operable on June 6, 1986. The B and D ESW pumps were then removed from service for repairs, retested and declared I operable on June 10, 1986. The impellers and suction bells of all four ESW pumps, as well as other miscellaneous parts, were replaced during the repairs.

6

, Due to their similar design, the decision was made to inspect the RHRSW pumps.

The RHRSW pumps at Susquehanna are two-stage vertical circulator pumps, Byron Jackson type 28KXL. Cavitation damage was found on the impeller liners on the Unit 1 A and B RHRSW pumps and the liners were replaced. Since no damage was found on the Unit 2 A RPPSW pump, the Unit 2 B RHRSW pump was not inspected.

The damage to the ESW and RHRSW pumps was determined to be caused by impeller suction recirculation cavitation. The cavitation is caused by operating the pumps significantly below their design flow rates. Current ESW and RHRSW operating configurations limit the ability to operate those systems in a manner that would prevent recirculation cavitation. A review of the design and method of operation for both systems is currently being performed by the licensee to determine what changes are required to avoid recirculation cavitation. In the interim, each pump will be inspected periodically. The inspection freouency will ensure that the integrity of the pump is sufficient to support 30 days of continuous operation following a design basis accident. Based on the results of the pump inspections, a conservative replacement frequency for the affected pump components will be determined. (Refs. 3 and 4.)

1.3 Failures of Residual Heat Removal Pump Wear Rings Due to Intergranular Stress Corrosion Cracking at Browns Ferry Unit 2 NRC Information Notice 86-39, " Failures of RHR Pump Motors and Pump Internals,"

issued May 20, 1986, discussed the failure of a residual heat removal (RHR)

I pump at Peach Bottom Unit 3 due to impeller wear ring failure. After receiving this notification, the licensee for Browns Ferry

  • inspected that facility's RHR pumps, since they are the same model and manufacturer (Bingham-Willamette single-stage centrifugal pumps, model 18x24x28 CVIC), with service times comparable to those at Peach Bottom. (See Figure 2.) The findings, reported to the NRC in May 1986, are described below.

The 2D RHR pump was found to have linear indications in the pump impeller lower wear ring; the upper wear ring had no indications. Subsequent metallurgical examination on the lower wear ring showed the indication to be due to intergranular stress corrosion cracking (IGSCC), as at Peach Bottom. The RHR 2A pump was then inspected, and also showed pump impeller lower wear ring indications. The interin solution is the replacement of all lower impeller wear rings for the Unit 2 RHR pumps, including cross-tie RHR pumps to Unit 2.

This is planned prior to restart of Unit 2. The remaining Unit 1 and 3 pumps will be modified once a long-term solution is determined by the licensee, the pump supplier, and the pump manufacturer. The inspection findings are detailed below, Units 1 and 2 were in refueling outages at the time of the inspection, and Unit 3 was in an extended maintenance outage with the unit in cold shutdown.

The condition described below was discovered in Unit 2, although the potential exists for similar conditions on Units 1 and 3.

  • Browns Ferry Units 1, 2, and 3 are each 1065 MWe (net) General Electric BWRs located 10 miles northwest of Decatur, Alabama, and are operated by Tennessee Valley Authority.

7

l

)

N7h ^*  %\\N W

7 ." ...

T ,

' T.' f

/F .k .

8 l "55" s p i 8 1

'  ! ff'Tl' l l 1.' .

L  : 8 8

l I l p

,  !, t -  ?-

I i a g

f . p

_[ f 'k i % J' f gdd w y-~

'/

Suction ,

/ ,

[ ' ,- ~ ~"

fl u

f r

f,g iv, Dis, charge e i

' - ~ k p 'i' I

, ///_ ')

II ,pd 7 l

- Y -

f O

--.__ l 2 gy /

r is 1 1 1 1 -T-$-

Casing Wear Lower Impeller [

Ring Wear Ring \

Figure 2 Typical CVIC RHR Pump Sectional Assembly I

i 8

The RHR 2A and 20 pumps were being inspected using dye penetrant tests. Linear indications were identified in the lower wear rings on both pump impellers.

The upper wear rings of these pump impellers revealed no linear indications.

Metallographic analysis on the RHR pump 20 lower impeller wear ring indicated the linear indications resulted from IGSCC. This determination was based on several factors: the pump impeller service environment, wear ring material, 4

and metallographic examination. The examination showed brittle fracture surfaces with no evidence of decreased ductility or fatigue. The wear ring material is of American Iron and Steel Institute (AISI) type 410 stainless steel, with hardness values ranging frcm 33 to 37.5 Rockwell C (within the vendor specified range).

Immediate corrective action has been initiated to inspect and replace the lower '

wear rings on all Unit 2 pumps and the Unit 2 cross-tie RHR pumps. This will be accomplished prior to restart of Unit 2. Also, current evaluations are under consideration by the licensee, General Electric (the nuclear steam system surrly vendor) and Bingham-Willamette for long-term solutions to this problem.

Alternatives under consideration are: the use of wear rings of lower hardness values, the use of a different wear ring material, and the use of an impeller of a configuration that does not utilize wear rings.

At present, the licensee is of the opinion that there are no short-term safety concerns. In concert with General Electric, it has been determined that: (1) the wear ring cracking has a relatively low probability of leading to catastrophic pump failure, and (2) redundant RhR pu.nps are available even if the one pump fails due to ring cracking.

Browns Ferry utilizes four RHR pumps on each unit. The RHR pumps are relatively low head, high capacity pumps. Several operating modes of the RHR system are utilized to support various plant ccnditions. During accident situations, the low pressure coolant injection mode provides cooling water to the reactor vessel. Other important RHR modes are the containment spray and shutdown cooling supply. (Ref. 5.)

1.4 Fire Protection System Actuation Results in Emergency Core Cooling System Actuation and Flooding at Browns Ferry Units 1, 2, and 3 On April 30, 1986, during work on a fire header in the coolino tower area of Browns Ferry Unit 1,* a spurious actuation of the fire protection system occurred, spraying some cable trays in Units 1 and 3. These trays were washed down, but were not checked out further. On May 3, 1986, all eight emernency diesel generators (EDGs) and the emergency equipment cooling water (EECO) system started automatically at 1:34 a.m. and again at 2:25 a.m. Since Unit I was defueled, no licensed operators were in the control room at the time. On the second actuation, it was realized that the core spray and keep-fill valves were open. About 60,000 gallans of water were pumped to the reactor vessel.

  • Browns Ferry Units 1, 2, and 3 are each 1065 MWe (net) MDC General Electric BWRs located 10 miles northwest of Decatur, Alabama, and are operated by j Tennessee Valley Authority.

9


j

The vessel overflowed, causing a high level alam in the refueling pool area  ;

(Units 1 and 2 were defueled, with the vessel and drywell open) and spilling at least 28,000 gallons into the ventilation system. The event is detailed below.

On April 30, 1986, at 1:55 p.m., hydrant No. 20 and fire protection isolation valve HCV-0-26-1021 separated from the fire header causing a serious leakage from the fire protection high pressure water system. On April 28, an equipment tagout request had been processed to replace fire hydrant No. 2 at cooling tower No. 1. The reouest identified valve HCV-0-26-1021-1AT as the boundary valve for the system maintenance. The work description was to replace hydrant No. 20. On April 30 at 10:00 a.m., hydrant No. 20 and HCV-0-26-1021 had been excavated, and the concrete thrust block removed to allow removal of the hydrant. At 1:45 p.m., fire pump C was started to perform Surveillance Instruction (SI) 4.11. A.1.b. This SI was required to prove operability afttr maintenance on the fire pump breaker. Pressure values recorded during the SI were slightly higher than normal. Typical pressures are 175 psi at the pump and 160 psi in the plant on the fire header; SI pressures of 183 psi at the pump and 179 psi on the header were recorded.

At 1:55 p.m., the Unit 1coperator was notified of a fire header leak at cooling tower No.1; he checked fire header pressure at 179 psi. (Valve HCV-0-26-1021-1AT had separated from the header at this time.) Ten minutes la ter, the operator removed fire pump C from service and began monitoring the fire header pressure, per shift engineer request. At 2:10 p.m., fire header pressure had decreased to 30 psi. The operator had the fire pump C discharge valve throttled to allow slow charging of the header if necessary. By 2:15 p.m., fire header pressure had decreased to psi. Five minutes later, isolation valve HCV-0-26-1019 was closed by the Unit 2 Assistant Shift Engineering (PCV-0-26-1019 isolates all high pressure fire protection on cooling tower No. 1), and the leakage was stopped.

When pressure was then restored by the raw service water (RSW) charging system, the following fixed spray systems initiated, wetting down the associated areas:

(1) In the Unit I reactor building, several cable trays were sprayed; raw cooling water booster pumps IA and IB were sprayed; and panel 25-6A, which contains reactor protection system instrumentation, was sprayed.

In the Unit 3 reactor building, cabla trays were sprayed, raw cooling (2) water booster pump instrumentation panel 25-185 was sprayed, and raw .

cooling water booster pump 3A was sprayed.

At 2:25 p.m., fire header pressure was 80 psi. The shift engineering established fire watches in the reactor building, turbine building and cooling tower. Welding and burning was stopped in the reactor and turbine buildings.

A 4:20 p.m., a 1-hour report was made to the NRC on the fireheader incident. By 5:15 p.m., the Unit 1 fixed spray systems were returned to normal; by 6:00 p.m., the Unit 3 fixed spray system was returned to nomal. All fire systems '

were normal except cooling tower No.1, which remained isolated. Reactor and turbine building fire watches were released.

l 10

The licensee determined that the events resulted from the following:

(1) Fire system construction using mechanical joints requires extended bound-aries to avoid pressurizing mechanical joints when the joints' structural support is removed; the clearance was not extended, nor was the system construction properly reviewed to ensure adequate clearance boundaries j with supports removed.

(2) Rods which can be installed and fastened to a collar on the pipe adjoining mechanical joints were not installed. These support rods are installed on many of the plant fire hydrants, but not the hydrants at the cooling towers.

(3) To identify mechanical joints which are not structurally sound with

! supports removed, some must be excavated. The fire protection system is installed by typical drawings in part, and the underground piping confituration is unknown. Fire protection procedures did not include an inspection to assess the consequences of removing the thrust block after excavation.

l (4) The three spray systems which actuated should have remained inactive; leakage past check valves could have caused the actuation. Two check valves in the water supply which holds the valves closed would have to leak to cause the valve to open en low water supply pressure.

The three inadvertent starts of all eight EDGs on May 3,1986 occurred due to grounding of the high drywell pressure switches during the spraying of portions of the reactor building on April 30. During the first initiation, the core spray injection valves opened, resulting in overflowing of the reactor cavity into the ventilation system. An estimated 28,000 to 33,000 gallons of water were spilled. The EDG starts are as follows:

(1) On the first automatic start of all eight EDGs, at 1:24 a.m., C3 and 03 EECW pumps started. After attempting to shut down the EDGs and receiving

" auto start lockout " the operator discovered the high drywell (DW) pressure initiation lights were on for core spray logic divisions I and II. Also, the high pressure coolant injection (HPCI) auto initiation

light was on. No other annunciators were energized on Unit 1, and the sequential events recorders was inoperative.
The operator depressed the core spray initiation logic reset button and cleared the high drywell pressure initiation signal and secured the EDGs
core spray. At this time, it was not noticed that the injection valves were open. The core spray testable check valve had not indicated the disc

~

being open. Charging pressure on the loop was normal, and all core spray residual heat removal (RHR) pump motor breakers had previously been disabled to prevent inddvertent starting and injection into the vessel.

It was suspected at this time that the search for a Unit 1 battery board l

ground had caused the initiations. Due to the core spray and RHR pump l motor breakers being disabled, no emergency core cooling system (ECCS) i pumps started, and no annunciations were energized in the control rooms.

Control room personnel were engrossed in searching for the cause of engineered safety features (ESF) actuation, and thus were not aware the core spray inboard in,iection valves were open.

11

i The " fuel pool system abnormal" alarm was received in the control room.

An operator was dispatched to the refuel floor to determine levels. Upon notification that water from the reactor cavity was spilling into one l ventilation duct beside the reactor cavity, an operator imediately opened reactor water cleanup system blowdown to maximum. The source was identified to be from the condensate charging system through the core spray injection and discharge valves. Valves 75-25 and 75-53 were closed.

Water from the reactor cavity also entered into the fuel pool skimmer surge tank through a gate. This had caused the surge tank level to increase and give the alarm.

All elevations in the Unit I reactor building were checked for water leakage. There was water coming from the vent duct near the primary containment purge filter, and there were 6 to 8 inches of water and sludge standing in the floor in the southwest quadrant of the Unit I reactor building. Health physics proceeded to rope off all contaminated areas.

Rough calculations later showed that approxinately 28,000 to 33,000 gallons of water had been spilled. Reactor cavity level was determined to be about 2 inches below the vent ducts. The cause of this indication was isolated to pressure switches 64-58 A-D.

(2) A second automatic start was received on the eight DGs at 3:45 a.m. on May 3. EECW pumps C3 and D3 started, the Unit 1 signal caused initiation core spray high drywell pressure divisions I and II, HPCI auto water initiation logic activated, and the core spray and RHR discharge valve loops I and II received a signal to open. The operator imediately closed .

the core spray injection valves before they reached full open. The RHR outboard valves had been tagged closed previously for other maintenance; thus, no injection occurred from RHR.

Power was removed from the core spray injection valves to prevent any further openings. Attempts were made to reset initiation logic, but it would not reset. Upon investigation it was discovered that pressure switches 64-58 A and C were picked up, causing the high pressure initiations. Pressure switch 64-58C was reset, thus clearing the initiation signal, allowing reset of high drywell pressure initiation logic and shutdown of EDGs and EECW pumps C3 and D3. Pressure switch 64-58 A was still picked up at this time.

After consultation with offsite personnel it was considered likely that the pressure switches became grounded due to water getting into their when the high pressure fire protection system initiated and sprayed in this area on April 30, 1986.

On the next shift, after instrument mechanics had removed covers from junction box (JB) NN-PP on panel 25-68, pressure switches 65-58 A and C picked up again, causing the third ESF actuation as previously described ,

with the exception that the injection valves on core spray had been closed '

with power renoved.

The signals were reset and the EDGs secured aoain. Plant personnel inspected and dried out terminal boxes and instruments on panel 25-6A and 6B on Unit I. All JBs were inspected on Unit I with water found only in boxes on panel 25-6A and B. No water was found in the Unit 3 switches or JBs.

12

h k

The ESF actuations were determined to have been caused by moisture inside pressure switches64-58A and C causing an electrical short. It is suspected that this moisture came from the fire protection actuation on April 30, 1986.

A contributing factor was the delay in action to inspect and evaluate equipment

, operability subjected to the fire protection spray. It is also suspected that ,

l inadequate sealing and waterproofing of conduits, junction boxes, and instruments in the areas subjected to fire protection sprays contributed to the event.

The cause of induction of water into the Unit 1 ventilation system ducts was the operator's failure to recognize that injection valves were open on the system as a result of the ESF actuation. Water flowed from the 10,000 gallon condensate transf er system head tank located on top of the reactor building via

! a 4-inch line in the keep fill system to the core spray system. The pressure on the core spray tank is maintained by two 1000 gpm condensate transfer pumps.

1 The licensee stated that no personnel contaminations resulted from the water f spillage.

On May 11, 1986, spurious actuation of the high pressure fire suppression system occurred again in the Unit 3 reactor building at the same elevation that had been sprayed with water during the actuation on April 30, 1986. The actuation was caused by an apparent operator error in restoring the fire protection system to normal after completion of fire fighting control efforts I that had taken place at the cooling towers on May 10.

Af ter securing a high pressure fire pump at 10:10 p.m., the RSW head tank isolation valve was not reopened until the PSW low pressure alarm was received at 11:00 p.m. When a fire pump is started, the head tank is automatically isolated to prevent overfilling but must be manually reopened. The fire pumps operate in parallel with the RSW pumps. After the alarm, the RSW pumps started charging the fire protection header; however, the fire zones 8 and C opened,

( spraying portions of the reactor building. Each fire zone is actuated by I operation of a deluge valve. The valve is opened by system pressure whenever the pilot chamber pressure is diminished any significant amount below the

header pressure. The check valves in the line charging the pilot chamber leaked when the system pressure was low. Consequently, when the system was l

repressurized, the pressure beneath the deluge clapper exceeded the pressure in

, the pilot chamber and the valve actuated. The valves were reset and an inspec-tion made of the various panels sprayed with water.

{

l During the overflow of the reactor cavity event and the second fire protection

spray initiation on May 11, 1986, the operators in the Unit 1 control rooms l were not licensed by the NRC. These operators were completing the unit operator logbook and signing the logout with a shift engineer routinely reviewing the logout. In the present plant condition with Unit 1 and 2 i

defueled, an NRC licensed operator is not required in the control room but only l reouired to be on site. The significance of not having licensed operators in

the control room during these type events is beinn reviewed by the NRC.

l t

(Ref.6.)

13 l

t

. . - . . . - . ., . -n--. - - - - . - - - - - - - , - - . , . - - . - - - , - - - - - - - - - - - - , - . . - . - - - - - - - - - , . - , - -

i 1.5 Update on inadvertent Actuations of Deluge Spray System at River Bend

\

i Power Reactor Events, Vol. 8, No.1, described on pp. 6-7 a January 1986 event i at River Bend

  • in which an unlabeled switch resulted in the inadvertent I actuation of the deluge spray system. On May 19, 1986, the fire protection deluge system was inadvertently actuated again, from an indeterminate cause.

This actuation deluged the main turbine bearings Nos.1, 2, and 3. Main turbine operation, including vibration, was monitored closely. The actuation was secured in about 15 minutes. Later the sane day, with the unit at 73%

power, the main turbine tripped on a high bearing vibration signal, causing closure of the turbine stop valves and a subsequent reactor scram.

Investigation revealed water accumulation in the No. 3 bearing vibration probe {

cable connector, which caused a false trip signal. There was no actual high vibration otherwise indicated. The event is detailed below.

At 6:54 a.m. on May 19, 1986, the main turbine bearing fire protection deluge system inadvertently actuated. This actuation deluged the main turbine bear-ings Nos. 1, 2, and 3. Main turbine operation, including vibration, was monitored closely, with no abnormalities noted. The actuation was secured at approxinately 7:09 a.m. Investigation of the area and monitoring of the tur-bine operation revealed no unusual effects from the deluge. Normal operation j continued until 2:20 p.m., with the unit at 73% power, when the main turbine -

tripped on No. 3 bearing high vibration, resulting in a reactor scram. All systems responded normally to the turbine trip; i.e., the control and stop valves closed, the turbine bypass valve opened, and the reactor scrammed as designed. The reactor water level was properly maintained by the nomal feed-water system.

The cause of the deluge nas not been determined. The system is manually actu-l ated only. Once actuated, the deluge continues until manually isolated. The switch used to initiate the systen was found in the normal position (although l it may have been positioned to actuate and then repositioned to normal). An l investigation was unable to determine if anyone was in the area to actuate the deluge system initiation switch. After isolating the deluge, inspections did not reveal any unusual or abnormal effects. After the turbine trip, the tur- '

bine bearing No. 3 vibration probe cable connection cover was removed, and significant water accumuation was found. This caused a grounding of the j connector, leading to a false high vibration signal. Evaluation of all parameters, including other vibration signals, and subsequent normal turbine operation did not indicate that an actual high vibration existed.

  • River Bend is a 936 MWe (net) MDC General Electric BWR located 24 miles I north-northwest of Baton Rouge, Louisiana, and is operated by Gulf States Utilities.

l l

(

14 o

The vibration probe cable connection covers were removed, and any water accu-nulation present was dried. The vibration probe (General Electric Model 3S7700VB100) was functio.nally tested and found to respond properly. The deluge activation switches werc modified to provide tamper-proof switch covers on the deluge activation switches, to reduce the probability of an inadvertent water curtain actuation. To provide access control and accountability for persons entering, card readers were installed on both doors to the fire protection room in the turbine building. This room contains fire protection isolation valves

> which can be manually opened to initiate the deluge system. In the interim,

, the deluge system manual isolation valve was closed and put under I administrative control, to preclude inadvertent actuation. (Ref.7.)

I l

}

l I

l 15 L _

1.6 References (1.1) 1. Duke Power, Docket 50-369, Licensee Event Report 86-13, August 28, 1986.

2. NRC Region I, Inspection Report 50-369 and 50-370/86-19, August 19, 1986 (1.2) 3. NRC Region I, Inspection Report 50-387; 50-388/86-09, June 10, 1986.
4. Pennsylvania Power and Light, Dockets 50-387 and 50-388, Licensee <

Event Report 86-21, June 20, 1986.

(1.3) 5. Tennessee Valley Authority, Docket 50-260, Licensee Event Report 86-06, May 23, 1986.

(1.4) 6. NRC Region II, Inspection Report 50-259; 50-260; 50-296/86-16, June 24, 1986.

(1.5) 7. Gulf States titilities, Docket 50-458, Licensee Event Report 86-39, June 18, 1986.

l i

I These referenced documents are available in the NRC Public Document Room at c 1717 H Street, Washington, DC 20555, for inspection and/or copying for a fee. (AEOD reports also may be obtained by contacting AE00 directly at 301-492-4484 or by letter to USNRC, AE00, EWS-263A, Washington, D.C. 20555.) ,

16

2.0 EXCERPTS OF SELECTED LICENSEE EVENT REPORTS On January 1, 1984, 10 CFR 50.73, " Licensee Event Report System" became effective. This new rule, which made significant changes to the requirements for licensee event reports (LERs), requires more detailed narrative descriptions of the reportable events. Many of these descriptions are well written, frank, y

and informative, and should be of interest to others involved with the feedback of operational experience, j

This section of Power Reactor Events includes direct excerpts from LERs. In t

general, the information describes conditions or events that are somewhat unusual l or complex, or that demonstrate a problem or condition that may not be obvious.

The plant name and docket number, the LER number, type of reactor, and nuclear steam supply system vendor are provided for each event. Further information may be obtained by contacting the Editor at 301-492-4493, or at U.S. Nuclear Regulatory Commission, EWS-263A, Washington, DC 20555.

Excerpt Page 2.1 Motor Control Center Automatic Transfer Design Inadequacy......... 17 2.2 Improper Raychem Installations Due to Procedure Inadeouacies...... 18 2.3 Degraded Auxiliary Feedwater Flow Due to Design Deficiency........ 19 2.4 Possible Loss of Standby Gas Treatment System Safety Function Due to Nonqualified Relay........................................, 21 2.5 Reactor Scram Due to Relay Malfunction /Line Separation 23 a t Reactor Wate r Level Transmi tte r. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2.6 Emergency Equipment Cooling Water System Cooling Capacity .

Found Inadequate for Some Postulated Accident Conditions.......... 20 2.7 Potential Loss of Shutdown Cooling Due to Improper Valve Lineup 27 During Maintenance on Shutdown Cooling Pump Shaft Seal............

l 2.1 Motor Control Center Automatic Transfer Design Inadequacy

)

Turkey Point 3; Docket 50-250; LER 86-23; Westinghouse PWR On.May 4,1986, while Unit 3 was at 100% power and Unit 4 was in cold shutdown, a design review of the proposed modifications for Unit 4 to the motor control

' center (MCC) D automatic transfer scheme, discovered a condition in which transfer of MCC D from a potentially operable bus to a potentially inoperable bus could occur. This scenario was determined to be applicable to the Unit 3 design as modified by recently installed plant change modification (PC/M)86-041.

This scenario involves a Unit 3 safety injection signal and a loss of offsite power on both units. Under these conditions, a relay race occurs which may i

17

( -

l l

l cause MCC D to spuriously transfer. Since a relay race results in an indeter-rrinate condition, it should be conservatively assumed that MCC D will transfer to its alternate, emergency supply diesel generator (EDG) A. Once MCC D has i transferred to EDG A, a subsegaent single failure of EDG A will leave MCC D in l a deenergized state since the automatic transfer device will not transfer the MCC back to EDG B. This condition could result in the loss of certain engineered safety features considered in the FSAR accident analyses, unless operator action is taken. Transfer of MCC D back to its normal supply (EDG B) can only be  !

accomplished by manually resetting the lockout relay on load center 4C.

An engineering evaluation per 10 CFR 50.59 and a justification for continued  ;

operation performed for this scenario determined that the postulated loss of MCC D will not result in unacceptable increases in containment temperature or pressure, or an increase in the site boundary doses beyond those postulated in the FSAR, provided the MCC is manually transferred to its normal power supply by operator action within 20 minutes.

Corrective actions include the following:

(1) Emergency operating procedures 3(4)-E0P-E-0 were revised to provide guidance to the operators to manually transfer the MCC D to its normal power supply within 20 minutes if it has transferred to a deenergized bus.

(2) The operators on-shift were briefed of the potential condition and the corrective actions taken.

(3) A change to PC/M 86-0421 is currently under review to correct this concern.

(4) The existing procedural requirements for independent design review and checking of proposed modifications will be reemphasized to the engineers involved with this design package.

(5) An independent review of PC/M 86-041 was performed to ensure that no other similar concerns existed. i 2.2 Improper Raychem Installation Due to Procedure Inadequacies Davis-Besse; 50-346; LER 86-21-01; Babcock & Wilcox PWR On May 9,1986, during a training session on proper Raychem terminution/ splicing it was discovered that past practices may have resulted in improper field installations. This condition resulted from the manufacturer's (Raychem's) installation and material requirements not being adequately incorporated into the station procedure and the design drawings.

A selected sample of 71 wire splices was inspected and it was determined that 67 of these splices did not conform with acceptance criteria provided by the ,

manufacturer (Raychem). The nonconformities included:

Improper Raychem heat shrink c'over size selection.

Improper finished diameters on Raychem heat shrink cover.

Inadequate overlap of heat shrink cover onto wire insulation.

Improper use of heat shrink cover directly on fabric cover on the wire.

18

Improper bending of heat shrink covers to accommodate installation in junction box.

Inspections for the locations of Raychem installations were conducted. A team of Raychem trained engineers and designers has been assigned to inspect each Raychem termination / splice installed during the current outage, and to provide specific instruction to the craft to assure all Raychem material and installation requirements are satisfied.

An inspection / repair program is being implemented to assure that past Raychem installations are either adequate or repaired. The splices in decay heat loop 2 and supporting systems have been inspected and replaced where necessary. All Raychem 1E and EQ splices will be replaced or evaluated as satisfactory by laboratory testing and engineering analysis prior to restart.

In addition, plant maintenance procedure MP 1410.24 has been revised to address 4 the manufacturer's material and installation instruction to ensure the oper-ability during accident conditions. Drawing E302A is in the process of being

( revised.

2.3 Degraded Auxiliary Feedwater Flow Due to Design Deficiency Catawba 1 and 2; Dockets 50-413 and 50-414; LER 86-17; Westinghouse PWRs On April 23, 1986, the unit initially entered hot standby. On April 29, 1986, an unacceptable flow rate to the upper surge tank (UST) was obtained from the auxiliary feedwater pump turbine (AFWPT) during a performance surveillance test.

Personnel issued a work request to investigate / repair valve 2CA-20. This valve was found to be fouled with pieces of iron magnetite and sand. Also, the valve's cascade piston would not respond due to rust. >

Prior to completion of the repair to the valve, personnel attempted to verify proper AFWPT flow to steam generator (SG) 28. Flow was unacceptable (less than 260 gpm) at normal SG pressure. A work request was then initiated to investigate / repair valve 2CA-181. The cage assembly of 2CA-181 was found to be fouled with iron magnetite and sand. Repairs of 2CA-20 and 2CA-181 were completed on May 2, 1986, and AFWPT flow to the UST was verified to be acceptable.

On May 15, 1986, motor driven AFW pump 2B failed to provide acceptable flow to the UST during an operations surveillance test. A work request was issued to investigate / repair valve 2CA-180. Personnel then attempted to verify proper motor driven AFW pump 28 flow to SGs 2C and 20. The flow rate obtained was

) unacceptable at normal SG pressure. The cage assembly of 2CA-180 was found to be fouled with iron magnetite and sand. Repair to 2CA-180 was completed May 16, 1986, and motor driven AFW pump 28 flow to the UST was verified to be acceptable.

(

19 '

l

On May 21, 1986, feedwater isolation occurreri due to a high-high SG level.

During the transient, personnel had difficulty maintaining AFW flows. Following the transient, three post-trip AFW flow graphs from reactor trips on May 17, 18, and 19, 1986, and the AFW preoperational test results were reviewed. AFW flow was found to be degraded to SGs 2C and 2D during the post-trip transients and the preoperational test. Personnel then attempted to verify AFW train B flow to the SGs with flow control valves full epen. The flow to SG 2C was found to be 185 gpm, and flow to SG 2D was found to be 210 gpm. Acceptable flow to each SG was greater than/ equal to 240 gpm. Personnel then issued work requests to investigate / repair 2CA-40 and 2CA-44 Repairs to both of these valves were completed on May 23, 1986, and motor driven AFW pump 28 flow to the SGs was verified to be acceptable (greater than 300 gpm to each SG).

This incident was due to a design, manufacturing, construction / installation deficiency. The AFW control valves' cage assemblies were not sized to assure that fouling of the cages would not occur. Although these materials have not )

been chemically analyzed, the material found appears to be iron magnetite, sand, and small diameter dirt or rock.

)

The sand should pass through the cage assembly. However, larger diameter iron magnetite has been found to be the major contributor to the fouling. The differential pressure across the strainer cage is expected to force the iron f magnetite through the cage so that fouling does not occur. This breakdown of the iron magnetite apparently has not consistently occurred.

The AFW flush procedure was reviewed. The methods used were consistent with accepted practice. All of the Fisher control valves utilizing cage assemblies were removed from the system for flush, and later were reinstalled. Approxi-mately 25 feet of 4 or 6-inch piping was not flushed, due to the inability to push flow through the AFWPT.

The preoperational test of the Unit 1 AFW system did not verify flow into the SGs. The Unit 2 CA system preoperational test verified that AFW flow was being supplied to the SGs. However, the specific values of the AFW flow were not included in the acceptance criteria for the test. The AFW system flow verifica-tion surveillance test has previously been performed in hot shutdown, with SG pressure not at normal pressure (greater than 1000 psig). P'erforming these surveillances at normal SG pressure would have indicated valve fouling problems earlier in the process. Three reactor trips occurred prior to discovery of the AFW valve fouling problems. The degraded AFW flow to SGs 2C and 2D was not recognized during the post-trip reviews of reactor trips on May 17,18, and 19, 1986. Currently, specific AFW flow is not reoCred to be reviewed during the post-trip review. Also, a standard group of L ansient monitor plots was speci-fied to be generated for the post-trip review. AFW flow to SGs C and D was not included in this group. Personnel plan to review the post-trip format to g include expected AFW flow values, as well as other improvements.

On May 22, 1986, the AFWPT failed to provide adequate flow to SG 28. Personnel adjusted the position of the AFWPT throttle control valves to provide additional flow to the SGs. It is believed that this is indicative of further valve fouling.

20  :

During the unit blackout test on May 27, 1986, AFW flows to SGs 2C and 2D were observed to be degraded (less than 300 gpm). On May 28, 1986, the unit entered hot shutdown. On June 9,1986, SG 2D was drained to approximately 65% wide range. Motor driver. AFW punp 2B was placed in service, and control valve fouling was verified. Valve 2CA-40 was cleaned and reassembled on June 10, 1986. Also on that day, SG 2C was drained to approximately 20% wide range. Motor driven AFW pump 2B was placed in service and control valve fouling was verified. On

, June 11, 1986, 2CA-44 was cleaned and reassembled.

On June 11, 1986, a Unit I reactor trip occurred. During the transient, AFW flow was observed to be slightly decreased from previous values, although still remaining acceptable. Personnel plan to monitor transient responses on both units for indications of AFW flow degradation.

Personnel may utilize motor driven AFW pumps to provide normal feedwater during unit startup. This could potentially contribute to AFW control valve fouling after the system is flow tested prior to entry into startup.

Corrective actions include the following:

(1) Cage assemblies were cleaned or replaced for valves 2CA-181, 2CA-180.

2CA-40, and 2CA-44; AFW control valve cage assemblies will be changed to variable size openings of 0.05 inches at minimum flow to 0.125 inches at full flow.

(2) SGs 2C and 20 were drained and AFW pump 2B flow was observed. Valves 2CA-40 and 2CA-44 were cleaned again; valve 2CA-20 was cleaned and returned to service.

(3) AFWPT flow control valves have been throttled open so that flow remains acceptable.

(4) Procedures PT/1&2/A/4250/068 were revisea so that AFW flow will be verified at SG pressure greater than 1000 psig, and results will be provided to the Unit Coordinator. The frequency of procedure PT/2/A/4250/068 has been

, increased so that proper AFW flow will be verified prior to each entry into startup. Af ter the fouling problems are corrected, the surveillance interval will be as required by technical specifications.

(5) The latest Unit I reactor trip transient was reviewed, and possible AFW flow control valve fouling was identified.

(6) The need to utilize main feedwater in non-emergency situations rather than the AFW system has been emphasized to the operators.

> 2.4 Possible loss of Standby Gas Treatment System Safety Function Due to Nonqualified Relay Grand Gulf 1; Docket 50-416; LER 86-20; General Electric BWR On June 3,1986, the Grand Gulf licensee was infonned by the Architect Engineer, I Bechtel Power Corporation, of a condition which could potentially cause a loss

> of the standby gas treatment system (SGTS) safety function during accident conditions. This situation is reportable pursuant to 10 CFR 50.73(a)(2)(v).

i

) 21 1

t

A 4-hour notification was also made in accordance with 10 CFR 50.72(b)(2)(ii).

On June 3,1986, the plant was operating at approximately 85 percent power.

Bechtel notified the licensee that postulated environmental effects on a non-qualified circuit located in the SGTS fire detection cabinet could cause the shutdown of both SGTS fans. The SGTS filter train fans were designed to automatically trip on a high temperature signal of 310 degrees F in the charcoal filter bed. This fan trip interlock should have been provided from a qualified class IE component. Bechtel, in conjunction with the licensee, also identified a violation of Regulatory Guide 1.75 separation requirements. Both l of the SGTS filter train fire detection cabinets were found to be powered from l a common non-class IE power supply while the class IE SGTS fan circuits are l

powered from their respective ESF divisional class IE power supplies.

I Utilization of the non-class IE power to these fire detection cabinets did not meet the Regulatory Guide 1.75 6-inch minimum separation requirement.

(

A material nonconformance report was initiated to document and track resolution of the condition. In order to allow operation of the plant to continue, the power supply breakers to the charcoal fire detection circuit were opened. This would prevent an inadvertent shutdown of the fans should the SGTS be actuated to mitigate an accident.

l The cause of the condition was due to design error. Bechtel had failed to identify this contact as safety-related when preparing the original NUREG-0588 equipment list and subsequent 10 CFR 50.49 list. This omission was identified 7

during the development of shutdown logic diagrams and safety function diagrams which show in a logic fashion the equipment required to be qualified in accordance with 10 CFR 50.49. The licensee has contracted an independent organization to develop these documents. An inquiry to Bechtel resulted in the described occurrence. An evaluation of other SGTS protective circuitry revealed that the equipment was environmentally qualified to 10 CFR 50.49. The control room fresh air unit filter train was reviewed and determined to be properly qualified. This condition was, therefore, determined to be an isolated case.

A design change is being made to remove the fan trip interlocks from the control circuitry. This will prevent failure of the nonqualified circuit from creating a trip of the fans. The existing high temperature (255 degrees F) and high-high temperature (310 degrees F) alarms will remain so that, if desired. -

the operator may shut down the fans manually. Additionally, the design change will provide class IE power to the fire detection cabinets from each respective engineered safety features division. This exclusive use of class 1E circuits within the fire detection cabinets will meet Regulatory Guide 1.75 separation requirements and retain protection of class IE power systems.

In its safety assessment, the licensee stated that a coninon mode failure of the charcoal bed temperature sensing circuitry during accident conditions could have caused a failure of both trains of the SGTS filter fans. The deletion of the fan shutdown interlock is justified on the following basis. The purpose of the temperature detection circuit was to sense a high temperature in the charcoal bed and provide an initial alarm at 255 degrees F and an alarm and i trip of the SGTS filter fans at 310 degrees F. Fire in the charcoal beds is (

22

(

1

(

1 likely because the heater elements in the filter train are located sufficiently away from the charcoal, there are no transient combustibles in this area, and the decay heat created by radiciodine loadings from a design basis accident are not sufficient to cause significant temperature increases. The deletion of the interlock would not preclude actuation of the alarms at 255 degrees F and 310 degrees F, which are well below the ignition temperature of the charcoal (626 degrees F). Also, upon actuation of the alams, manual shutdown of the i

fans could be initiated if desirable. The deletion of this interlock would not affect initiation of the fire suppression system for the charcoal bed as it is currently manually actuated.

2.5 Reactor Scram Due to Relay Malfunction /Line Separation at Reactor Water Level Transmitter Lacrosse; Docket 50-409; LER 86-18; Allis-Chalmers BWR On June 22, 1986, the reactor was operating at approximately 99% power. At 1010, the reactor scramed. The first out scram alam was " Main Steam Isolation Valve Not Full Open," though the reactor building main steam isolation valve (MSIV) was full open. The A train of the shutdown condenser initiated. Both forced circulation pumps tripped and the standby (1A) seal injection pump started.

Approximately 2 minutes after the scram, the operators observed that reactor water level safety channel 2 spiked downscale momentarily. Both high pressure core spray (HPCS) pumps, and both emergency diesel generators (EDGs) started, and the containment building isolated due to the momentary low water level sig-nal. Water level indicated approximately +18 inches (18 inches above nomal) on reactor water level safety channels 1 and 2. Shortly afterward, the MSIV actually closed on low main steam line pressure (</= 1000 psig). The wide range water level instrument and the narrow range water level recorder were indicating full scale.

The B train of the shutdown condenser initiated due to the actual closure of the MSIV. Tne shutdown condenser was removed from service. The EDGs and HPCS pumps were secured and returned to auto approximately 5 minutes after initiation.

Water level was approximately +18 inches on safety channels I and 2.

An increase in the activity indicated on the forced circulation pump cubicle monitor was noticed. This instrument monitors the containment building exhaust air prior to it being filtered.

When the auxiliary operator was in the airlock, preparing to enter the contain-I ment building, he heard an unusual noise and notified the control room. He j then investigated and observed steam and water at the east nuclear instrumenta-tion platform level in the area of the water level instrumentation. He exited the containment building and informed the control room at 1042. The Shift Super-visor, Auxiliary Operator, and a Health Physics Technician entered the contain-ment building dressed in protective clothing and full face masks. At 1100, the Shift Supervisor isolated reactor water level transmitter (LT) 50-42-301. The stainless steel tubing had separated at a 1/4-inch $wagelok coupling between 23

the valve and the transmitter on the reference leg side of the transmitter.

This transmitter supplies a signal to the reactor water level control system and the narrow range recorder in the control room.

When the leak was isolated, water level safety channel 1 momentarily spiked downscale. Both HPCS pumps and both EDGs started, and the containment building isolated. The HPCS pumps and EDGs were secured, and the containment building isolation valves normally open during plant shutdowns were reopened.

At 1150, the turbine building MSIV was closed, so that the reactor building MSIV could be opened for troubleshooting. (The two valves are in series.) 4 When the reactor building MSIV was opened, a loud buzzing noise started. This noise had also existed following the scram until the MSIV closed. The buzzing was traced to relay SK 19/1, which is in the MSIV not full open circuit. The MSIV was then closed and the noise stopped.

Following the occurrence of the leak, the water level channels readings did not coincide. As discussed above, the narrow range recorder and the wide range instrument pegged high. The safety channels indications also did not match.

Water level was controlled to maintain at least normal water level on the lowest indicating safety channel, which was channel 1. About 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> after the leak was isolated, it was noticed that the wide range water level indicator was read-ing approximately 68 inches, which corresponds to near the top of the core.

The lowest indicated reading on the chart was 63 inches, which corresponds to below the top of the core. Though the safety channels indicated adequate water level, water level was increased so that the wide range instrument indicated at least 70 inches to ensure adequate core coverage. A reaoing was taken from the reactor sightglass to obtain an approximate indication of actual water level.

The sightglass reading corresponded to approximately a 666-feet 8-inch elevation, which is slightly above normal water level. At that time, the wide range instru-ment indicated 68 inches, channel I was at +10 inches, channel 2 at +18 inches, and channel 3 at +23 inches.

A wide range water level transmitter was recalibrated. Its zero required con-siderable adjustment. Following calibration at 1510, the indicated level was approximately 30 inches higher. The other water level channels were also recali-brated. Safety channels 1 and 2 required slight adjustments (a couple of inches),

while channel 3 had experienced a shift of approximately 11 inches. The sepa-rated line was repaired and the recorder transmitter was recalibrated. Its zero had shif ted about 24 inches. While the wide range transmitter's output was too low, the other transmitters were found high ard their outputs were adjusted lower during the calibration. The condition of a high signal output corresponds to a higher than actual indication.

A reactor water level measurement system uses a comon standp'oe connected to l the reactor vessel as the variable leg for each transmitter used during reactor operations. There are three safety channels. Each of the transmitters i

(50-42-302, 303 and 306) for the safety channels has a separate reference leg.

The reference leg for transmitter 50-42-302 (safety channel 1), and the reference leg used for the narrow range recorder / reactor water level controller trans-mitter, 50-42-301, and the wide range transmitter, 50-42-305, both share the upper reactor vessel penetration with the vessel standpipe. The safety channels are considered safety-related; the other transmitters are not.

24 l

When the reference leg tubing for transmitter 50-42-301 separated, the indications for 50-42-301 and 50-42-305 both failed high, since they share a common reference leg. Since the standpipe shares the top reactor vessel tap with that reference leg, the standpipe experienced a pressure transient when the line separated and again when the break was isolated. This explains the j water level indication spikes which occurred at the time of the line separation and isolation. The reactor feed pump had already backed down due to the decrease in steam flow when transmitter 50-42-301 pegged high, so the tubing separation had no effect on the reactor water level control system, which controls feedwater flow to maintain water level during operation.

The reasons the wide range water level indication slowly decreased following isolation of the leak is that its reference leg had been at least partly emptied during the leak. Following isolation, hot reactor steam and water filled the wide range transmitter reference leg from the reactor vessel. As the reference leg cooled, the indication decreased.

When the reference leg sensing line for transmitter 50-42-301 separated, a hydraulic shock would have been transmitted through its diaphragm. The variabie leg for safety channel 3 is tied to transmitter's 50-42-301 variable leg sensing line. Therefore, the transmitter for channel 3 would also have experienced this hydraulic shock. The shock would have been muted by the narrow range variable leg header, so the other safety channel transmitters, whose variable legs are connected to the header, would not have experienced the same magnitude hydraulic shock. This could explain why the zero shift on channel 3's trans-mitter was greater than the shifts on the other safety channel transmitters.

The zero shift experienced by the safety channels 1 and 2 transmitters could have been due to either this hydraulic shock or the standpipe blowdown. The channel 1 transmitter would have seen the least pressure impact, since it expe-rienced a pressure drop on both its reference and variable sensing lines. After an equilibrium blowdown rate was reached, the differential pressure on all 3 safety channel transmitters would have been near nonnal.

The description of the sequence of events stated that both forced circulation pumps tripped at the time of the scram and the standby seal inject pump started.

It is believed that the forced circulation pumps tripped due to a temporary loss of seal inject flow during the transfer from station power to offsite power.

Both forced circulation pumps were restarted within 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> of the scram.

The safety channels transmitters are Foxboro, environmentally qualified, Model N-E130H transmitters. The wide range transmitter is Model N-E13DM and f the narrow range recorder transmitter is Foxboro Model T/613DN-HS2-0. Foxboro was contacted regarding the shift in the transmitter's zeroes. The Foxboro representative said the transmitters can experience zero shift during a fast differential pressure transient.

) The leak rate through the 1/4-inch transmitter line was quite small. There was no significant increase in the rate of retention tank water accumulation.

The water level lines were inspected and other fittings checked to be tight.

The SK 19/1 relay which had caused the scram was replaced. A diode in series with the relay coil had burnt out, causing the relay to chatter. The relay was an Allied Control, Model MHY0-12A, 115V AC relay, i

25 i

I Or June 23, the calibration of water level safety channel 3 was rechecked. A slight adjustment (less than 2 inches) was necessary. On June 24, it was noticed that water level safety channel I was indicating lower than the other channels. Channel I was recalibrated and adjusted back almost the amount it had been adjusted on June 22. Over the next week its indication was observed for signs of drifting; no drifting was noted. '

Tha safety significance of this incident was reviewed and was discussed with NRC personnel. Only one of the sefety-related water level cnannels experienced a significant zero shift, but all did experience some shift. The zero shift can happen to differential pressure transmitters if they experience a high differential pressure transient. Training will be conducted on this incident /

for the Operations Department. All crews have been briefed on the event. The major primary leak procedure already states that other specified indications should be used in conjunction with water level indication as indication of water inventory and primary system condition.

2.6 Emergency Equipment Cooling Water System Cooling Capacity Found Inadequate for Some Postulated Accident Conditions Fermi 2; Docket 50-341; LER 86-17; General Electric BWR .

The emergency equipment cooling water (EECW) system is des'igned to provide cool-ing water to maintain essential equipment within operating temperatures so that these devices are able to perform their reactor shutdown functions. The EECW system also provides cooling water to room coolers to maintain the ambient ,

(

temperature around essential equipment below maximum values and provide cooling to several nonessential devices. The nonessential service is based primarily on equipment location and not safe shutdown functions.

Calculations generated in response to an engineering evaluation request indicated that the EECW at Fennt may not be able to maintain adequate cooling to essential equipment during a postulated small break accident (SBA). The SBA would fill the drywell with high temperature steam, causing a high rate of heat transfer through the drywell coolers, recirculation pumps, and associated EECW piping.

The resulting higher temperature cooling water would limit the ability of the EECW to cool the essential devices downstream, thereby reducing the ability to perfonn their shutdown functions. This condition met criteria established in 10 CFR 50.73(a)(2)(ii), and was Jetermined reportable June 24, 1986.

To negate the impact of the SBA generated heat load on equipment essential for safe shutdown, an engineering design (EDP 5544) has been issued and is being implemented to change the operation of the EECW isolation valves to the drywell. {

These valves are normally open during EECW operation. Implementation of the

~

engineering change provides for automatic closure (auto-close) of the Division I .

and II (P44-F607A and P44-F606B) inlet isolation valves, should a high drywell )

pressure occur when EECW is operating. This will remove these two drywell heat (

loads from the remainder of the EECW system and allow the EECW system to provide adequate cooling to essential equipment external to the drywell. Calculations indicate that the modified EECW system cooling capacity can handle the essential j cooling loads with the drywell heat loads generated by the SBA removed.

26

--- )

l This unanalyzed EECW design condition was discovered while the plant was shut down for a scheduled maintenance and modification outage. The plant had never achieved power levels which would have jeopardized the EECW systems ability to perform it's shutdown functions.

The cause of this deviation was that Engineering personnel involved with the design of the EECW system overlooked the heat input associated with a SBA in (

' the design calculation of the EECW system. This heat input calculation was not i completed until design calculations associated with the response to the engi-reering evaluation request were generated. At the time of the discovery of this condition, the licensee was involved in an extensive design calculation update and verification program. Complete thermodynamic and hydraulic evalua-tions of the EECW and reactor building closed cooling water system were being undertaken. It is felt that these design calculations, which were being per- ',

fonned in parallel with the resolution of the potential oversight, would have identified the cmissions. The extensiveness of the reverification of plant design parameters preclude the likelihood of similar future occurrences of this nature.

This LER will be reviewed with the involved Engineering personnel. In addition,

=

a copy of this LER will be made reouired reading for appropriate Engineering personnel.

2.7 Potential loss of Shutdown Cooling Due to Improper Valve Lineup L During Maintenance on Shutdown Cooling Pump Shaft Seal Yankee-Rowe; Docket 50-029; LER 86-10; Westinghouse PWR On June 27, 1986, at 0137 hours0.00159 days <br />0.0381 hours <br />2.265212e-4 weeks <br />5.21285e-5 months <br />, during a maintenance outage with the plant in cold shutdown, main coolant was inadvertently drained to the low pressure surge tank (LPST). This could have resulted in a loss of shutdown cooling. Thi:

occurred while transferring from normal shutdown cooling to the alternate method of shutdown cooling using the LPST tooling pump and cooler per approved procedure OP-2162. Performance of this procedure was necessary because of the failure of the shutdown cooling pump's shaft seal. During this evolution, approximately 2000 gallons of water was drained from the pressurizer, the pressurizer level indication dropped quickly from 300 inches to an offscale low condition, and main coolant pressure dropped from 100 psig to 10 psig. The pressurizer did not empty. The Control Room Operator (CRO) immediately secured the LPST cooling pump, and the Primary Auxiliary Operator (PA0) isolated the flow path by shutting SC-V-614 and SC-V-661. The CR0 also started all three charging pumps and restored pressurizer level and pressure to 300 inches and 100 psig, respectively.

The root cause of this occurrence has been attributed to personnel error by the PA0. While conducting the alternate shutdown cooling valve lineup, CH-V-654 was not fully shut, which resulted in a flow path from the main coolant system

> to the LPST. The PA0 thought that the valve had completed its full travel when he had operated the manual valve during his valve lineup.

> Without any operator action, the valve lineup could have allowed the main coolant loops to be drained and a loss of shutdown cooling could have resulted. For this to occur, however, the event would have had to continue completely unnoticed. The LPST would have filled and subsequently vented water through a 27

________ ~ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

1" vent line to the gravity drain tank. High level alarms on the LPST and the gravity drain tank would have alerted the opertors to this event.

All members of the Operations Department involved in this event were instructed of its potential consequences and the need for strict procedural compliance in conjunction with the utilization of practical systems knowledge when performing plant evolutions. Plant personnel were made aware of this event and complete .

compliance with plant procedures and programs at all times was emphasized.

28

f 3.0 ABSTRACTS / LISTINGS OF OTHER NRC OPERATING EXPERIENCE DOCUMENTS 3.1 Abnormal Occurrence Reports (NUREG-0090) Issued in May-June 1986 An abnormal occurrence is defined in Section 208 of the Energy Reorganization Act of 1974 as an unscheduled incident or event which the NRC determines is significant from the standpoint of public health or safety. Under the provi-sions of Section 208, the Office for Analysis and Evaluation of Operational Data reports abnormal occurrences to the public by publishing notices.in the Federal Register, and issues quarterly reports of these occurrences to Congress in the NUREG-0090 series of documents. Also included in the quarterly reports are updates of some previously reported abnormal occurrences, and summaries of certain events that may be perceived by the public as significant but do not meet the Section 208 abnormal occurrence criteria.

Date Issued Report 5/86 REPORT TO CONGRESS Gli ABh0RMAL OCCURRENCES, OCTOBER-DECEMBER 1986 VOL. 8, N0. 4 There were five abnormal occurrences at NRC licensees during the period. Two occurred at licensed nuclear power plants, and three occurred at other licensees (industrial radiographers, medical institutions, industrial users, etc.).

The occurrences at the plants involved: (1) inoperable main steam isolation valves at Brunswick Unit 2; and (2) management deficiencies at Fermi Unit 2.

The occurrences at other licensees involved: (1) a diagnostic medical misadministration at Letterman Army Medical Center, San Francisco, California; (2) a therapeutic medical misadminis-tration at Queen's Medical Center, Honolulu, Hawaii; and (3) a diagnostic medical misadministration at Hospital Universitario, San Juan, Puerto Rico.

Also, the report updated information on: (1) the nuclear acci-dent at Three Mile Island (79-3), first reported in Vol. 2, No.

1, January-March 1979; (2) failure of the automatic reactor trip system at Cook Unit 2 (83-3), first reported in Vol. 6, No. 1, January-March 1983; (3) uncontrolled leakage of reactor coolant outside the primary containment at Dresden Unit 3 (83-6), first reported in Vol. 6, No. 3, July-September 1983, (4) a through wall crack in the vent header of the BWR containment torus at Hatch Unit 2 (84-2), first reported in Vol. 7, No.1, January-r March 1984; (5) loss of the main and auxiliary feedwater systems at Davis-Besse (85-7), first reported in Vol. 8, No. 2, April-June 1985; (6) the buildup of uranium in the ventilation system at Nuclear Fuel Services, Inc., Erwin, Tennessee (84-19), first 29 l L -_ _ - _ _ _ - - .

I Date Issued Report -

reported in Vol. 7, No. 4, October-December 1984; (7) a diagnos-tic medical misadministration at the Hospital of San Raphael, New Haven, Connecticut (85-8), first reported in Vol. 8, No. 2, April-June 1985; (8) a diagnostic medical misadministration at Mercy Hospital, Pittsburgh, Pennsylvania (85-9), first reported in Vol. 8, No. 2, April-June 1985; (9) a therapeutic medical 1 misadministration at the University Health Center of Pittsburgh's Joint Radiation Oncology Center in Pittsburgh, Pennsylvania (85-15), first reported in Vol, 8, No. 3, July-September 1985; '1 and (10) a therapeutic misadministration at the Hershey Medical Center, Hershey, Pennsylvania (85-16), first reported in Vol. 8, No. 3, July-September 1985.

In addition, items of interest that did not meet abnormal occurrence criteria but may be considered significant by the public involved: (1) a systems interaction event due to leakage of fire protection water at Hatch Unit 1; (2) suspension of the license for the Institute for Medical Research of Bennington, Vermont, due to possession of radioactive material in excess of those quantities authorized by the license; and (3) multiple check valve failures of the high pressure coolant injection turbine exhaust at Shoreham Unit 1.

30

3.2 Bulletins and Information Lutices Issued in May-June 1986 The Office of Inspection and Enforcement periodically issues bulletins and information notices to licensees and holders of construction permits. During the period, one bulletin, 24 information notices, and one information notice supplement were issued.

L Bulletins are used primarily to communicate with the industry on matters of generic importance or serious safety significance (i.e., if an event at one reactor raises the possibility of a serious generic problem, an NRC bulletin may be issued requesting licensees to take specific actions, and requiring them to submit a written report describing actions taken and other information NRC should have to assess the need for further actions). A prompt response by affected licensees is required and failure to respond appropriately may result in an enforcement action. When appropriate, prior to issuing a bulletin, the NRC may seek comments on the matter from the industry (Atomic Industrial Forum, Institute of Nuclear Power Operations, nuclear steam suppliers, vendors, etc.),

a technique which has proven effective in bringing faster and better responses from licensees. Bulletins generally require one-time action and reporting.

They are not intended as substitutes for revised license conditions or new requirements.

Information Notices are rapid transmittals of information which may not have been completely analyzed by the NRC, but which licensees should know. They require no acknowledgement or response, but recipients are advised to consider the applicability of the information to their facility.

Date Bulletin Issued Title 86-01 5/23/86 MINIMUM FLOW LOGIC PROBLEMS THAT COULD DISABLE RHR PUMPS (Issued to all General Electric BWR facilities holding an operating license or construction permit) .

Information Date Notice Issued Title 86-31 5/6/86 UNAUTHORIZED TRANSFER AND LOSS OF CONTROL 0F INDUSTRIAL NUCLEAR GAUGES (Issued to all power reactor facilities l holding an operating license or construction permit) 86-32 5/2/86 REQUEST FOR COLLECTION OF LICENSEE RADI0 ACTIVITY

> MEASUREMENTS ATTRIBUTED TO THE CHERN0BYL NUCLEAR PLANT ACCIDENT (Issued to all power reactor facilities holding an operating license or construction permit)

86-32 6/6/86 REQUEST FOR COLLECTION OF LICENSEE RADI0 ACTIVITY Sup. 1 MEASUREMENTS ATTRIBUTED TO THE CHERN0BYL NUCLEAR PLANT ACCIDENT (Issued to all power reactor facilities e holding an operating license or construction permit) 31 C -_ _ _ _ _ _ _ _ _ _ _

Information Date Notice Issued Title 86-33 5/6/86 INFORMATION FOR LICENSEES REGARDING THE CHERNOBYL NUCLEAR PLANT ACCIDENT (Issued to all fuel cycle and Priority 1 material licensees) 86-34 5/13/86 IMPROPER ASSEMBLY, MATERIAL SELECTION, AND TEST OF VALVES AND THEIR ACTUATORS (Issued to all power reactor facilities holding an operating license or construction permit) 86-35 5/15/86 FIRE IN COMPRESSIBLE MATERIAL AT DRESDEN UNIT 3 (Issued to all power reactor facilities holding an operating license or construction permit) 86-36 5/16/86 CHANGE IN NRC PRACTICE REGARDING ISSUANCE OF CONFIRMING LETTERS TO PRINCIPAL CONTRACTORS (Issued to all power  ;

reactor facilities holding an operating license or construction permit) j 86-37 5/16/86 DEGRADATION OF STATION BATTERIES (Issued to all power j reactor facilities holding an operating license or construction permit) 86-38 5/20/86 DEFICIENT OPERATOR ACTIONS F0LLOWING DUAL FUNCTION VALVE FAILURES (Issued to all power reactor facilities '

holding an operating license or construction permit) /

86-39 5/20/86 FAILURES OF RHR PUMP MOTOPS AND PUMP INTERNALS (Issued  ;

to all power reactor facilities holding an operating  ;

license or construction permit) 86-40 6/5/86 DEGRADED ABILITY TO ISOLATE THE REACTOR COOLANT SYSTEM FROM LOW PRESSURE COOLANT SYSTEMS IN BWRs (Issued to all power reactor facilities holding an operating license or construction permit) 86-41 6/9/86 EVALUATION 0F QUESTIONABLE EXPOSURE READINGS OF LICtNSEE PERSONNEL 00SIMETERS (Issued to all byproduct material licensees) 86-42 6/9/86 IMPROPER MAINTENANCE OF RADIATION PONITORING SYSTEMS (Issued to all power reactor facilities holding an operating license or construction permit) 86-43 6/10/86 PROBLEMS WITH SILVER ZE0 LITE SAMPLING OF AIRBORNE RADI0 IODINE (1ssued to all power reactor facilities holding an operating license or construction permit) '

32

Ir. formation Date Notice Issued Title 86-44 6/10/86 FAILURE TO F0LLOW PROCEDURES WHEN WORKING IN HIGH PADIATION AREAS (Issued to all power reactor facilities holding an operating license or construction permit, and to all research and test reactor licensees) 8E-45 6/10/86 P0TENTIAL FALSIFICATION OF TEST REPORTS ON FLANGES MANUFACTURED BY GOLDEN GATE FORGE AND FLANGE, INC.

f (Issued to all power reactor facilities holding an operating license or construction permit, and to all research and test facility licensees) 86-46 6/12/86 IMPROPER CLEANING AND DECONTAMINATION 0F RESPIRATORY PROTECTION EQUIPMENT (Issued to all power reactor facilities holding an operating license or construction permit, and to fuel fabrication facility licensees) 86-47 6/9/85 FEEDWATER TRANSIENT WITH PARTIAL FAILURE OF THE REACTOR SCRAM SYSTEM (Issued to all BWRs and PWR facilities holding an operating license or construction permit) 86-48 6/13/86 INADEQUATE TESTING 0F BORON SOLUTION CONCENTRATION IN THE STANDBY LIOUID CONTROL SYSTEM (Issued to all BWR facilities holding an operating license or construction permit) 86-49 6/16/86 AGE / ENVIRONMENT INDUCED ELECTRICAL CABLE FAILURES

\ (Issued to all power reactor facilities holding an operating license or construction permit) 86-50 6/18/86 INADEQUATE TESTING TO DETECT FAILURES OF SAFETY-RELATED PNEUMATIC COMPONENTS OR SYSTEMS (Issued to all power reactor facilities holding an operating license or f construction permit) 86-51 6/18/86 EXCESSIVE PNEUMATIC LEAKAGE IN THE AUTOMATIC DEPRES-SURIZATION SYSTEM (Issued to all BWR facilities holding an operating license or construction permit) 86-52 6/26/86 CONhUCTORINSULATIONDEGRADATIONONFOXBOR0MODELE CON'rROLLERS (Issued to all power reactor facilities holding an operating license or construction permit) 86-53 6/26/86 IMPRUPER USE OF HEAT SHRINKABLE TUBING (Issued to all power reactor facilities holding an operating license or construction permit) 86-54 6/27/86 CRIMINAL PROSECUTION OF A FORMER RADIATION SAFETY OFFICf'4 WHO WILLFULLY DIRECTED AN UNQUALIFIED INDIVI-DUAL D PERFORM RADIOGRAPHY (Issued to all holders of bypro!uct, source, or special nuclear material) 33 L

\

{

3.3 Case Studies and Engineerino Evaluations Issued in May-June 1986 I

The Office for Analysis and Evaluation of Operational Data (AE00) has as a pri-mary responsibility the task of reviewing the operational experience reported

, by NRC nuclear power plant licensees. As part of fulfilling this task, it se-l lects events of apparent safety interest for further review as either an en-gineering evaluation or a case study. An engineering evaluation is usually an imediate, general assessment to determine whether or not a more detailed pro- ]

tracted case study is needed. The results are generally short reports, and the effort involved usually is a few staffweeks of investigative time.

5 Case studies are in-depth investigations of apparently significant events or situations. They involve several staffmonths of engineering effort, and result in a formal report identifying the specific safety problems (actual or potential) illustrated by the event and recommending actions to improve safety and prevent recurrence of the event. Before issuance, this report is sent for peer review and comment to at least the applicable utility and appropriate NRC offices.

These AE0D reports are made available for information purposes and do not impose any requirements on licensees. The findings and recommendations contained in these reports are provided in support of other ongoing NRC activities concerning the operational events (s) discussed, and do not represent the position or re-I quirements of the responsible NRC program office.

Special Date Study Issued Subject

$602 5/86 AN OVERVIEW 0F NUCLEAR POWER PLANT OPERATING EXPERIENCE FEEDBACK PROGRAMS The primary purposes of this AE00 study were to: (1) deter-mine the characteristics of representative licensee operat-ing experience (0E) feedback programs; (2) determine how feedback documents from various NRC offices and industry are used in such programs; and (3) determine if there is a need for changes to the NRC's feedback program of NRC re-quirements governing licensee activities related to OE (

feedback. The principal findings and conclusions from the study are:

4 (1) Although, since TMI-2, many significant and worthwhile initiatives have been implemented to understand the lessons of experience, most plants are making moderate, not extensive, use of their in-house operating exper-ience, and in general are making less use of the large body of knowledge associated with events and concerns that originate elsewhere in the industry. To increase the effectiveness or OE reviews in order to correct past and potential operational problems, licensees and the industry or the NRC need to take further actions.

34

f I

f Special Date Study Issued Subject l

(2) The resources and attention being devoted by licensees to understanding and implementing the lessons from experiences at other plants are usually considerably less than those directed to in-house feedback. As a

> result, the yield from these activities in terms of corrective actions and improved operator knowledge and capability and the associated reduction in component p failures and plant events seems smaller than expected.

t (3) The NRC lacks information industry-wide on the range of effectiveness of OE review activities and the de-

! gree to which plants are meeting the established NRC requirements. At this time, however, the general NRC requirements do not provide a sufficiently definitive basis to permit a meaningful evaluation of the total process by the-NRC. Current industry efforts also do

, not provide a suitable basis for judging the effective-l ness of licensee activities in this area.

(4) The large volume of OE feedback is degrading the ef-l fectiveness of the feedback programs. Duplication and

! overlap are diverting resources available to effec-l tively use OE feedback. In addition, conflicting in-l formation provided by two or more sources poses major difficulties for licensees unless the conflicts are identified.

Based on this study, AE00 recommended that the effective-ness of OE activities should continue to be monitored.

Engineering Date Evaluation Issued Subject E514 5/27/86 CORE DAMAGE PRECURSOR EVENT AT TROJAN Rev.' 1 This study is an update of E514; issued October 8, 1985, which was summarized in Power Reactor Events, Vol. 7, No. 5, p. 33. This revision incorporates comments from NRC's Region V Office, and clarifies selected portions of the report.

i During 1984, five events occurred at the Trojan nuclear power plant which could have had serious consequences for equipment or personnel had the events occurred l under different circumstances. The potentially most serious event occurred on September 20, 1984, when multiple, independent undetected failures of safety-related components resulted in the partial loss of the emergency onsite power supply and the total loss of the safety-grade auxiliary feedwater system. The other four events are also discussed in detail in the report.

35 l

1

l l

Engineering Date Evaluation Issued Subject E514, Rev. 1 This study's findings indicate that the September 20 (continued) event was a severe accident precursor with a conditional core melt probability on the order of IE-02 = .01, depending upon the assumptions made. This is a very conservative probability as it does not take into ac- a count any surveillance tests detecting preexisting faults. It should be considered an upper bound.

Supplementing NRC Region V's recognition of the j significance of the event, this report calculates a conditional core melt probability based on the actual plant conditions at the time and thus empnasizes and quantifies the seriousness of the event.

This review determined that the reliability of the safety-grade auxiliary feedwater pumps was poor and resulted in the need to rely on the non-safety grade motor driven pump more than is desirable. The Region has taken steps to require improvement, and it appears that improvement has been made.

Collectively, the five events indicated a ' lack of at-tention to detail, a lack of good maintenance practices, and a lack of appreciation of the significance of operating experiences at other facilities. Positive actions to correct each of these deficiencies have been taken by Region V.

The Region V Administrator met with the licensee on October 12, 1984, to discuss the significance of the September 20 event. The Region emphasized that senior management must take more of an interest in the operation of the plant. They are also closely following the licensees's corrective actions, both short term and long term.

E606 5/27/86 LOSS OF SAFETY INJECTION CAPABILITY AT INDIAN POINT UNIT 2 On December 28, 1984, while the Indian Point Unit 2 reactor was critical, all three safety injection (SI) pumps were declared inoperable. Two pumps were inoper-able due to low suction flow and the third pump could not be manually started. The reactor was manually scramed. While it was subcritical and a cooldown was in progress, a spurious SI signal occurred. Since all SI pumps were inoperable and had been deenergized, no SI occurred. [

l 36

Engineering Date Evaluation Issued Subject E606 5/27/86 The malfunction of the SI pumps was caused by boric f (continued) acid crystallization and possible gas binding of the pumps. Two parallel leaky valves in the discharge line of the boron injection tank (BIT) enabled highly concentrated boric acid to flow through the low pres-sure discharge line and to precipitate in the pumps which were not heat traced. Degassing of the nitrogen

, cover gas which had been dissolved in the boric acid solution is believed to be one of the likely sources of the gas found in the pumps. The contribution of nitrogen from the isolation valve seal water system could not be verified.

The degradation of the SI pump performance due to boric

! acid crystallization is not a generic problem with

) Westinghouse (W) PWRs, and is site-specific. Most W PWRs have their BIT downstream of the SI pumps.

However, besides Indian Point Unit 2, a few cther W plants were designed with their BIT (or equivalent) l located upstream of the SI pumps. This study identi-fies some additional plants that have the same config-uration as Indian Point Unit 2, and notes that there f may be others.

Before the Indian Point Unit 2 event, many licensees of W plants had been reevaluating their main steam line break (MSLB) analyses to justify elimination of the BIT or operation with lower concentrations of boric acid in order to eliminate problems of boric acid crystallization. To date, at least 15 W plants I (including Indian Point Unit 2) have obtained approval from the NRC to remove their BIT and/or to operate with reduced boric acid concentrations.

In August 1985, Consolidated Edison Company (the li-censee) submitted a request to the NRC for amendment of the Indian Point Unit 2 license to remove the BIT.

The permission was granted in December 1985. The re-moval of the BIT eliminates the boric acid precipitation problem and also removes one source of nitrogen in the system, thus reducing nitrogen binding of the SI pumps.

This event was reported to Congress as an abnormal occurrence. The NRC has sent a generic letter to all licensees of W PWRs encouraging them to reassess their design basis steam line breaks to reevaluate the need for highly concentrated boric acid. The NRC also plans to initiate an assessment of potential safety concerns of boric acid leakage from the BIT (including 37

Engineering Date Evaluation Issued Subject E606 those located downstream of the SI pumps) for the W .

(continued) plants. Also, on a case by case basis, the conse- {

quences of BIT removal on plant safety following a design basis MSLB will be evaluated.

This study suggested that an information notice be issued to make the licensees of the potentially af-fected W PWRs more aware of this safety concern, and to emphasize the need to follow strict administrative controls to prevent similar occurrences until their BIT is removed.

b 38

f

)

f I

3.4 Generic Letters Issued in May-June 1986 Generic letters are issued by the Office of Nuclear Reactor hegulation, Division of Licensing. They are similar to IE Bulletins (see Section 3.2) in that they transmit information to, and obtain information from, reactor licensees, appli-cants, and/or equipment suppliers regarding matters of safety, safeguards, or environmental significance. During May and June 1986, three letters were issued.

Generic letters usually eitner (1) provide information thought to ae important in assuring continued safe operation of facilities, or (2) request information j on a specific schedule that would enable regulatory decisions to be made regard-1 ing the continued safe operation of facilities. They have been a significant

) means of communicating with licensees on a number of important issues, the reso-lutions of which have contributed to improved quality of design and operation.

Generic 0 ate Letter Issued Title 86-05 5/29/86 IMPLEMENTATION OF TMI ACTION ITEM II.K.3.5, " AUTOMATIC TRIP 0F REACTOR COOLANT PUMPS" (Issued to all licensees and applicants of B&W designed NSSSs) 86-06 5/29/86 IMPLEMENTATION OF TMI ACTION ITEM II.K.3.5, " AUTOMATIC 1 TRIP 0F REACTOR COOLANT PUMPS" (Issued to all licensees and applicants of CE designed NSSSs) 86-10* 4/28/86 IMPLEMENTATION OF FIRE PROTECTION REQUIREMENTS (Issued to all power reactor licensees and applicants) 86-11 6/25/86 DISTRIBUTION OF PRODUCTS IRRADIATED IN RESEARCH REACTORS (Issued to all non-power reactor licensees) i

)

l I

  • This generic letter had not been listed in the March-April 1986 issue of Power t

Reactor Events (Vol. 8, No. 2).

39 l - _ _ _ _ _ _

3 3.5 Operating Reactor Event Memoranda Issued in May-June 1986 The Director, Division of Licensing, Office of Nuclear Reactor Regulation I (NRR), disseminates information to the directors of the other divisions and program offices within NRR via the operating reactor event memorandum (0 REM) system. The OREM documents a statement of the problem, background information, the safety significance, and short and long term actions (taken and planned).

Copies of OREMs are also sent to the Offices for Analysis and Evaluation of Operational Data, and of Inspection and Enforcement for their information.

No OREMs were issued during May-June 1986.

I i

i 40 j

l 3.6 NRC Documentation Compilations The Office of Administration issues two publications that list documents made publicly available.

. The quarterly Regulatory and Technical Peports (NUREG-0304) compiles bibliographic data and abstracts for the formal regulatory and technical reports issued by the NRC Staff and its contractors.

. The monthly Title List of Documents Made Publicly Available (NUREG-0540) contains descriptions of information received and generated by the NRC.

This information includes (1) docketed material associated with civilian nuclear power plants and other uses of radioactive materials, and (2) non-docketed material received and generated by NRC pertinent to its role as a regulatory agency. This series of documents is indexed by Personal Author, Corporate Source, and Report Number.

The monthly Licensee Event Report (LER) Compilation (NUREG/CR-2000) might also be useful for those interested in operational experience. This document con-tains Licensee Event Report (LER) operational information that was processed into the LER data file of the Nuclear Safety Information Center at Oak Ridge during the monthly period identified on the cover of the document. The LEP summaries in this report are arranged alphabetically by facility name and then chronologically by event date for each facility. Component, system, keyword, and component vendor indexes follow the summaries.

Copies and subscriptions of these three documents are available from the Superintendent of Documents, U.S. Government Printing Office, P.O. Box 37082, Washington, DC 20013-7982.

41 l

UNITED STATES ,,c, ,, cu .,,

eoeraae 'es mo NUCl.EA2 REZULATORY COMMISSION WASHINGTON, D.C. 20665 gy]j.,

6FFICtAL SU$tNESS PENALTY FOR pnNATE USE,4300 US NRC ADM-DIV 0F PUB SVCS POLICY t PUB NGT BR-POR NUREG W-501 WASHINGTON OC 20555 i

I i

i l

i

~

t

/

l

_______ - _ -