ML20044C271
| ML20044C271 | |
| Person / Time | |
|---|---|
| Site: | Millstone  |
| Issue date: | 09/23/1991 |
| From: | Mroczka E NORTHEAST NUCLEAR ENERGY CO., NORTHEAST UTILITIES SERVICE CO. |
| To: | Hehl C NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| Shared Package | |
| ML16266A160 | List: |
| References | |
| FOIA-92-162 B13928, NUDOCS 9303190368 | |
| Download: ML20044C271 (15) | |
Text
.
' 6t= :
NORTHEAST UTILITIES senerai omee. . semen street. Bernn. Connecticut f)
J .m.. sse stm caccw. P O. BOX 270 mw *.n"w" wa' HARTFORD. CONNECTICUT 06141-0270 ,
L L J %%[17,%'.' . (203) ssw000 September 23, 1991 F Docket No. 50-336 !
B13928 RE: Employee Concerns Mr. Charles V. Hehl, Director Division of Reactor Projects U. S. Nuclear Regulatory Commission -
Region I 475 Allendale Road King of Prussia, Pennsylvania 19406
Dear Mr. Behl:
Millstone Nuclear Power Station, Unit No. 2 '
RI-91-A-0082 In earlier correspondence on this issue ve requested additional time to complete our evaluation of Issues 5 & 6. Our investigation is now complete and we provide the following response. As requested in .your. transmittal letter, our response does not contain any personal privacy, proprietary, or safeguards information. The material contained in this response may be released to the public and placed in the NRC Public Document. Room at your discretion. The NRC transmittal letter and our response have received controlled and limited distribution on a "need to know" basis during the preparation of this response.
ISSUE 5:
Procedure discrepancies exist between OP-2336E and SP-2617A for the restoration of the line-up of the radiation monitor (RE-245), and its associated sample pump. Operators routinely fail to ' perform OP-2336E, Section 5.1, Step 5.1.13 which is to immediately close A0V-244A/B and A0V-245 when securing from condensate polishing facility discharges. This failure to follow procedures results in the sample pump to radiation monitor (RM-245) continuing to operate when the tank discharge is secured.
9303190368 921217 1 I
HUBBARD92-162 'PDR
. Mr. Charlos U. H:hl, Director
. U. S. Nucle:r Regulatory Corsission B13928/Page 2 September 23, 1991
. Reques t:
Please discuss the validity of the above assertions. If any discrepancies are identified concerning procedure noncompliance, please discuss their significance on the operation of radiation monitor RH-245. Please discuss any corrective action taken or planned, to ensure operators are meeting procedural and technical specification requirements.
Response 5:
The assertion as stated is not valid. The two procedures mentioned in the assertion, OP 2336E - Condensate Polishing Liquid Vaste System and SP2617A
- Radioactive Liquid Vaste Discharges, are for two completely different purposes and would not be expected to have many steps in common. SP2617A is intended to verify surveillance requirements are met prior to and during '
potentially radioactive discharges whereas OP2336E is intended to provide a i procedure for the collection, processing and subsequent discharge of liquid i vaste from the Condensate Polishing Facility (CPF).
The surveillance procedure directs the operator to the operating procedure, and the operating procedure directs that the sump pump be secured and isolation and discharge valves be closed when the discharge is complete. ,
There is no procedural requirement that the operator is to "immediately l close" the valves at issue as stated in the assertion. The radvaste discharge processes for Millstone Unit No. 2 are designed to automatically secure from discharge without the need for an operator to be standing by.
The area operator then 'vraps up' from the discharge and records the required data for documentation of the discharge. OP2336E directs the operator to close A0V-244A, A0V-244B and A0V-245 "When Sump Discharge is complete." Ve find no elements of procedure noncompliance involved. No ,
corrective actions concerning procedures are planned at this time. l ISSUE 6: ;
The following discrepancies have been identified during an evaluation of Vork Order AV0-M2-91-04411. These discrepancies identify continued noncompliance with procedures and poor response of operations and management to recurring problems with radiation monitor RE-245.
- a. The sample pump continues to run when the tank discharge stops at 15%
tank level (TK-ll).
- b. The "Lov Flow" switch does not always see a lov flow condition when i TK-10 and TK-11 discharge pumps stop. The head of water in the pipe and tidal conditions affect the flow of water.
- c. Operations normally rely on the 15% tank level pump trip to stop flow causing a lov flow to trip shut RE-245 discharge valve, and A0V-245.
If A0V-244A/B are shut and no lov flow condition exists, RE-245 sample pump vill continue to run until A0V-245 is shut. ;
- d. Changes to OP-2336E vere identified in 1989 to prevent the problems' identified by AVO-M2-91-04411. However continued identified procedure ;
noncompliance by operations has caused repeated problems.
1
Mr. Charles U. Hthl, Director }
U. S. Nuclear Regulatory Commission B13928/Pagn 3 September 23, 1991 Request:
Please provide an assessment of the above discrepant conditions. If the {
assertions are valid, please discuss their safety significance and effect :
on ' operation of radiation monitor RE-245. Please discuss any corrective -
actions that are being used to correct the problems. ,
Response 6: I Assertion 'a' is valid as stated but is not a discrepant condition. The sample pump is not interlocked with tank level and is not designed to stop running as a result of tank level. There is no operational safety significance to the sample pump continuing to run after a discharge is .
completed. l Assertion 'b' is valid as stated. The tidal effect mentioned is a known contributor to sample pump impeller wear and failure resulting in corrective maintenance. There is no operational safety significance to the lack of a "lov flow" condition after a discharge is complete. The flov l switch is intended to detect inadequate sample flow during discharges and i serves no purpose after completion of the discharge. Because of a history l of excessive maintenance on the sample pump, the Engineering department is l processing a plant design change to replace the sample pump and make electrical circuit changes to ensure the sample pump trips when the i discharge pump trips. These changes vill correct the pump impeller vear resulting from tidal effects on CPF liquid vaste discharges.
i Assertion 'c' is valid as stated, but there is no operational safety ,
significance to the sample pump continuing to run. As stated in Response ;
5, the radvaste discharge processes for Millstone Unit No. 2 are designed ;
to automatically secure from discharge without the need for an operator to be standing by. !
The assertion of procedure noncompliance in 'd' above is not valid. The tidal effects on the lov flow switch and the radvaste discharge philosophy may have resulted in an undesirable maintenance history for the sample pump ,
at issue but there has been no procedure noncompliance. A plant design change intended to revise the interlock for the sample pump vill solve the ,
undesirable corrective maintenance history and any necessary procedure !
changes resulting from this design change vill be incorporated as a normal part of the design change process.
j i
1 e
}
i h
6 A . _ _ _ _ _
i !
e Mr. Charles V. H:hl, Director ;
, U. S. Nuclear Regulatory Commission B13928/Page 4 September 23, 1991 i Ve were made aware of these issues via an internal memo in May 1991, and we !
are pursuing corrective actions to effect changes to the CPP liquid radvaste system. After our review and evaluation of this issue, ve find that this issue did not present any indication of a compromise of personnel or nuclear safety. Ve appreciate the opportunity to respond and explain '
the basis of our actions. Please contact my staff if there are further questions on any of these matters.
Very truly yours, I NORTHEAST NUCLEAR ENERGY COMPANY ,
4
' cida (/
E.J.~MpicePresident Senior V I
cc: V. J. Raymond, Senior Resident Inspector, Millstone Unit Nos. 1, 2, and 3 E. C. Venzinger, Chief, Projects Branch No. 4, Division of Reactor Projects l E. M. Kelly, Chief, Reactor Projects Section 4A ;
J. T. Shedlosky, NRC, Millstone Nuclear Power Station t
i 6
[L n- s E'
- 8,, .
o 8
UNITED STATES NUCLEAR REGULATORY COMMISSION' f) ,
l
% REGloN I ' 'j o,, 475 ALLENDALE ROAD g e'[ KING OF PRUSSIA. PENNSYLVANIA 194061415 t
00T 111991; 1 j
Docket Number: 50-336 ---
Northeast Nuclear Energy Company l
A'ITN: 'Mr. John F. Opeka
'l Executive Vice President - Nuclear i Engineering and Operations Group P.O. Box 270 Hartford, Connecticut 06141-0270
Dear Mr. Opeka:
Thank you for informing us of the results of your reviews of the concerns listed in the enclosed l
table. We have performed verification inspections on selected issues, find'your responses'. ;!
generally acceptable, and plan no further actions on these issues at this time. This is not to say l that further independent reviews of special and/or contested issues will not take place. You will .I be kept informed of such verification inspections and independent reviews' by the normal j inspection report process.
l A copy of this letter as well as the referenced correspondence is being placed in the Public.-
Document Rooms and sent to the State of Connecticut. We appreciate your cooperation in these matters. ,
'?
Sincerely, j i
^
hp)fl 8 Edward enzinger, C .
React Projects Branch 4 i
Enclosure:
Table of NU's Responses i
cc wienet:
}
Public Document Room (PDR) i Local Public Document Room (LPDR)
State of Connecticut i bcc w/ encl: ;
Allegation File: As Stated in the Enclosure i W. npr.om i. o.no.c&.y E. Conner's Files j]
G. Kelly j py?% L'\ ' '2-ff- [
t 1 -
(l
/' i
i I
ENCLOSURE ,
{
f NRC NU RESPONSE SUBJECT OF ISSUE !
NUMBER RESPONSE DATE 91-037-1 & 4 A09556 08/09/91 TORQUE WRENCH USE & MOV f TORQUE SWITCH BALANCING ,
91-052-1 to 5 A09559 08/09/91 PROBLEMS %TTH PDCE M2-90-032 91-063 A09569 07/01/91 AUX BUILDING ACCESS CONTROL f i
91-064-2 to 5 B13905 08/09/91 QC FIT-UP FOR NEW SJAE MONITOR !
I 91-069-3 & 4 A09660 08/16/91 MODIFICATION TAGOUT & ;
REVIEW & RM CALIBRATION !
91-082-1 to 3 A09604 08/09/91 SG LOOP FOLDER DRAWINGS &
INSTRUMENTATION i 91-093 A09700 08/16/91 RAD MONITOR SURVEILIANCE ,91-109 A09632 07/27/91 RAD MONITOR TAGGING ERROR 4
91-118-1 to 4 A09657 08/16/91 INADEQUATE WORK CONTROL 1 I
91-137 A09658 08/16/91 PERSONNEL SAFETY TAGGING l B13926 09/16'91 ERROR 91-l M A09655 05 16'91 ~B* ACID FEED PUMP TAGOUT C 91-139 A09656 05/16791 *B* ELECTRO-HYDRAULIC PUMP ;
TAGOUT [
91-140 A09658 08/16/91 MOV LIMIT SWITCH TAGOUT [
91-142 A09658 08/16/91 CHILLER COMPRESSOR TAGOUT 91-143 A09716 08/16/91 RAD MONITOR AS-BUILT l CONFIGURATION 91-145 A09661 08/16/91 EEQ PROCEDURES91-147 A09658 08/16/91 4160 5%TTCHGEAR TAGOUT l 91-148 A09659 08/16/91 WORK CONTROL 91-149 A09658 08/16/91 MOV SAFETY TAGOUT i
t i
f
r 95 RECORD OF ALLEGATION PANEL DECISIONS SITE: A//,'/f; do4 -t PANEL ATTENDEES:
ALLEGATION NO.: k3 -4 f 8 3 Chairman - ~.T. T Us e in s f
DATE: B MAY9I (Mtg.@2345) Branch Chief - f. C. Weg.'ng ee PRIORITY: High @ tow Section Chief ( AOC) - E.61.1/elfy SAFETY SIGNIFICANCE: Yes No Unknown Others - C u). Whik e /2.C b ly CCNCURRENCE TO CLOSEOUT: DD @ SC 3. 6. CwllMs 3. S. S feAh CONFIDENTIALITY GRANTED: Yes
@ k b. ScIn k f. he* cts er (See Allegation Receipt Report)
IS THEIR A DOL FINDING: Yes No IS CHILLING EFFECT LETTER WARRANTED: Yes No HAS CHILLING EFFECT LETTER BEEN SENT: Yes No HAS LICENSEE RESPONDED TO CHILLING EFFECT LETTER: Yes No ACTION:
- 1) besp % h ;w u,v. i hih4 -80E47 is keAo ce vieto e LJ ces, L,. abrim J &aa0ctLaoo m sacLh9
- 2) Reu cc~s o<ceu ed ds 2 4 A -lls6e"f
- 3) ;
i 4) f NOTES:
b
4 4d
- < JUM . ! 1991 Docket No. 50-245 Mr. E. J. Mroczka Senior Vice President - Nuclear Engineering and Operations Northeast Nuclear Energy Company P.O. Box 270 Hanford, Connecticut 06141-0270
Dear Mr. Mroczka:
Subject:
Millstone Unit I Inspection 91-06 This refers to the routine safety inspection conducted by Mr. D. Dempsey of this office on April 9 through May 20, 1991, at Millstone Unit 1 in Waterford, Connecticut. The preliminary findings were discussed with Mr. S. Scace and others of your staff at the conclusion of the inspection.
Areas examined during the inspection are described in the enclosed repon. Within these areas, the inspection consisted of performance observations of ongoing activities, independent verification of safety system status and design configuration, interviews with personnel, and review of records.
Good performance by unit management and staff was noted in the overall conduct of the refueling outage and in response to unanticipated events. Focus was consistently maintained on the safety and quality of work.
Our review of the circumstances surrounding the dry tube event of May 1 found that the event was caused, in part, by your failure to recognize and address in procedures the potential for tube breakage during handling outside of the reactor vessel cavity. The event had low radiological significance due to its short duration, but suggests a need for more thorough preparation for non-routine evolutions. Potentially significant exposures were avoided by close attention to the work activity and prompt action by Operations and Health Physics personnel. We acknowledge your conservative and successful efforts to assure the safe conduct of the remainder of the dry tube replacements and subsequent local power range monitor work.
OFFICIAL RIEORD COPY I l 0, vm W Crmt wb b [
i consuhed to identify wno may have had access to the area. The ucensee was unable to identify who was responsible for the incident. Based on a review of exposure records, no l radiological overexposures were known to have occurred as a result of the incident.
The inspector verified that the door had bcen posted as an HRA door. No discrepancies were identined in the applicable HRA door key accountability logs er radiation work permits. l However, the inspector was concerned that the licensee's accountability system was unable to identify when and by whom the gate had been opened, and requested that the licensee address this issue.
Between 1987 and 1988, unlocked HRA doors were discovered on 16 occasions. Extensive corrective actions were taken at that time to resolve the incidents. These activities were i documented in NRC inspection repon 50-245/88-25. Since the present incident was the first identified since 1988, the inspector considered this to be an isolated case. Licensee immediate corrective actions were appropriate. In addition, the licensee is evaluating further options to enhance control of locked HRA doors during outage periods. The condition in which a locked HRA door was blocked open and unguarded for over 15 hours1.736111e-4 days <br />0.00417 hours <br />2.480159e-5 weeks <br />5.7075e-6 months <br /> is a violation of technical speci6 cation 6.12.2. However, based on the criteria of 10 CFR 2, Appendix C, section V.G.1 and the isolated nature of the incident, the violation is not being cited. The inspector had no further questions conceming this occurrence. ,
i 1.2 SRAUIR51 Dry Tube Event of May 1,1991 Event Summary and Inspection Scone During the removal of dry tube assemblies on May 1, with the reactor in cold shutdown for refueling, an instrument dry tube assembly broke while being removed from the vessel and became temporarily uncontrolled while workers positioned it for cutting. The highly radioactive end of the dry tube became unshielded as it came out of the water in the spent ,
fuel pool at 10:25 a.m., and momentarily caused a high exposure rate to personnel on the refueling floor (108 ft elevation of the reactor building). '
Pan of the tube rotated in such a way as to break the surface of the water in the refueling pool for several seconds. Workers handling the dry tube quickly retumed the assembly under water, evacuated the area, and controlled access to the refueling floor to assure personnel protection. Although workers received an unexpected radiation exposure, individual doses remained well within regulnory and administrative exposure limits. Plant safety systems automatically detected the increased radiation leveh and caused the automatic isolation of the reactor building ventilation and the initiation of the standby gas treatment system (SBGTS).
There was no release of radioactivity or contaminated material in the reactor building, and ;
there was no hazard to the public.
l l
l r
3 t 1
n
. NNECO stopped the work activity pending the development of an action plan and procedures to recover the broken dry tube. NNECO management also revised the removal and replacement procedures to assure more positive control of the dry tubes and assessed the event. The NRC resident inspector immediately responded to review the status of equipment and system performance, to review the immediate actions taken by NNECO, and to assess the i consequences of the event. Followup NRC inspection reviewed the NNECO actions to prepare for dry tube replacement; recover the broken assembly; develop modified replacement methods and procedures; assess the radiological significance of the event; and to assess the overall adequacy of NNECO controls to conduct the activity.
A chronology of dry tube replacement activities is provided in Attachment 1 of this report.
The following drawings are also attached to describe the activity: Figure 1 - shows the position of the dry tube and location of personnel when the end became unshielded; Figure 2 - shows the details of the dry tube plunger and upper end; Figure 3 - shows the dry tube dose rates measured by NNECO; and, Figure 4 shows the NNECO estimates for the personnel exposure rates while the dry tube was out of the water.
Drv Tube Replacement Process and Procedural Controls The incore dry tube assemblies house the moveable source range monitors (SRM) and intermediate range monitors (IRM). The dry tubes provide a path for the SRMs and IRMs to travel inside the reactor. The dry tubes are a pan of the reactor coolant system pressure boundary during normal reactor operations. Each dry tube is about 40 feet long and about 1 inch in diameter. The top end of a dry tube contains a spring loaded plunger assembly, which is captured by a detent in the core upper grid to hold the top end of the dry tube in place.
NNECO conducted inspections during the present outage of dry tube conditions as pan of the '
reactor vessel inservice inspection program. The inspections were done in accordance with recommendations by General Electric made in Service Information Letter 409, Revision 1, :
Incore Dry Tube Cracks, dated July 31,1986. As in previous inspections, NNECO inspections during the present outage did not identify dry tube cracks or degradation of the type identified elsewhere in the industry. However, NNECO did note that some upper ends of the dry tubes were out of the detents in the upper core grid. NNECO decided on April 24,1991, to replace all 12 dry tubes during the present outage.
Both dry tubes and local power range monitor (LPRM) assemblies are replaced with the reactor in a refueling condition. At Millstone 1, several LRPMs are changed out each refueling outage as a routine reactor maintenance activity. Dry tubes are typically changed out on a less frequent schedule, and at Millstone 1, the dry tubes were being changed out for the first time during the 1991 refueling outage. The process and tools for dry tube replacement is very similar to that used for LPRMs. ;
i 4
, Since the top 12 feet are in the core region, the top ends of the ory tubes become highly.
radioactive during reactor operation. Because the potential for high exposures exists during removal and handling, special controls are taken to assure worker protection. These precautions and controls include detailed procedures for the replacement process, including !
removing, handling and cutting the old tube, and insertion of the new tube; the use of quali6ed operators for the work and a licensed senior reactor operator to supervise the work; ,
the use of inplant and supplemental monitoring equipment and dedicated health physics i
personnel to monitor the source at all times; worker briefings to review the work activity and established controls; and, administrative and mechanical controls to assure that the dry tube stays shielded (under at least 5 feet of water) during removal.
NNECO developed special procedure SP 91-1-37 for dry tube removal to establish the above ;
controls for the work activity. The procedure was developed in consultation with General Electric. NNECO visual inspections earlier in the outage had failed to detect the cracking ,
seen at other facilities. Since industry experience had shown dry tube embrittlement and cracking could occur, NNECO also incorporated special controls in the procedure (prior to the May I event) in case the dry tubes broke during removal. NNECO evaluations concluded that the most likely time that the dry tube might fail was during the application of the
- break away" force by the handling tool attached to the plunger assembly when first trying to lift the dry tube out of the vessel. The added procedure controls focused on how to contain a loose part should the dry tube break while lifting the assembly offits seat. NNECO also modified ,
the normal method of attaching the tools to the cold end of the assembly while pulling the l tube from the reactor to the spent fuel pool (SFP), so as to minimize stress on the plunger assembly during the transfer. t i
in spite of these precautions, the procedure controls were demonstrated by the May I event to be inadequate, in that the NNECO evaluation failed to anticipate the stresses applied to the plunger assembly during the cutting process in the SPF.
Details of Mav 1 Event j
A pre-job briefing was conducted with all personnel involved with the activity prior to the start of work on April 30. ne briefing included a review of the radiation work permit controls and a review of an incident at another facility while handling dry tubes. The e briefing also covered the replacement process per special procedure SP 91-1-37, SRM/IRM t Dry Tube Replacement.
Two dry tubes had been removed from the reactor and cut on April 30; the dry tube in core
- location 36-33 bound in the guide tube and was left for removal at a later date. Health Physics supervision witnessed the pre-job briefing and the removal of the first dry tube. On May 1, activities were in progress on the refueling floor to remove a dry tube assembly (#4) from its position in core location 28-33. There were seven workers on the refueling floor on May 1. including a senior reactor operator (SRO) in charge of activities. Other workers l 5
1 k
~. i included two plant equipment operators, two health physics techmcians and a worker assigned ,
to decontaminate equipment removed from the SFP. A worker assigned to fire watch duties was on the floor, but was not involved in dry tube replacement.
Dry tube #4 was removed from core location 28-33 and was moved horizontally from the i reactor cavity through the transfer chute with the upper (plunger) end in the nonheast corner of the SFP. Iri accordance with SP 91-1-37, the instrument handling tool was attached to the i plunger end of the dry tube, and the tool was in turn suspended from the monorail hoist on the refueling bridge. A second line with a "C" clamp was also attached to the cold end of the dry tube near the core plate boss. While being moved into position in the SFP to cut off -
the upper radioactive section, the dry tube unexpectedly broke at 10:25 a.m. just below the point where the handling tool was attached to the plunger assembly. The hot end rose to the ,
surface of the pool since the tool attached at the core plate boss acted as a fulcrum when the ;
heavier (cold) end sank into the reactor cavity. ,
The highly radioactive end of the tube quickly rose up and came out of the water, exposing about a three foot section of the tube end. The assembly was out of the water for about 2 seconds as the senior reactor operator released the handling tool on the cold end of the tube, allowing the assembly to sink back into the SFP. The unshielded dry tube end caused a momentary high radiation field that exposed workers working around the SFP, and caused area radiation monitors to alarm. ,
1 The high radiation levels in the spent fuel pool area were detected by an area radiation monitor (ARM) permanently installed on the west end of the refueling floor, hand held radiation monitors (teletectors) used by health physics workers covering the job, and two ponable area radiation monitors installed on the side of the refueling bridge and on the side ;
of the SFP. The elevated radiation levels caused the west area radiation monitor to alarm tsetpoint at 90 mrem /hr) and to initiate an autamatic start of the standby gas treatment system (SGTS).
I immediate actions were taken to assure personnel safety by the SRO in charge of the activity when the west ARM sounded. The actions included the following: to shield the detector, the crane operator was directed to lower the monorail hoist while the SRO lowered his lift pole attached at the cold end of the assembly; the lifting tool was lowered to the cask lay down !
area in the SFP and the refueling bridge power supply was tagged off under shift supervisor control to preclude inadvertent lifting of the tool pending the development of a recovery plan.
After verifying the dry tube was secure in the spent fuel pool, the refuel floor was evacuated and posted as a technical specification locked high radiation area. Pocket ionization chambers (PICS) for all personnel were checked immediately for exposure. All PICS were on scale and the highest reading was 100 mrem.
Plant operators restored the reactor building ventilation system to a normal configuration at 10:38 a.m. after verifying conditions on the refueling floor were acceptable. The SBGTS ,
operation was reported to the NRC at 11:05 a.m. on May I as an engineered safeguard 6
, system actuation in accordance with 10 CFR 50.72(b)(2)(ii). Plant information report 1 60 was initiated to document the event and track the NNECO followup actions. NNECO subsequently submitted LER 91-015 to summarize the event, its causes and the corrective ;
action to prevent recurrence.
Inspector review of the replacement activities for dry tube #4 determined that procedure and administrative controls were established as required for dry tube removal (a minor exception is discussed below in the section on work permits). Plant systems were aligned as required to ,
support the activity. NNECO personnel followed the procedure requirements of SP 91-1-37.
However, the original procedure method was proven to be inadequate to control a dry tube that broke while in transit. Operations personnel responded well to recognize the broken dry tube and immediately return it to a stable, shielded position.
l Dose Assessment f
The assessment of the radiological consequences of this event was described in NNECO memorandum MP-HPO-91055. The inspector reviewed the assumptions and methodologies :
used by NNECO to estimate the dose rates from the exposed dry tube, and the resulting personnel exposures. NNECO interviewed each worker involved in the job and reconstructed the event to determine the positions and actions of all personnel when the dry tube was out of !
the water. NNECO used the exposures recorded from the worker PICS and thermolumine-scent dosimeters, adjusted for a dose gradient, to estimate personnel exposures. The inspector concluded that the above approach resulted in a conservative estimate of worker [
exposure.
NNECO surveys determined that the exposed dry tube end had a contact dose rate of 24,000 R/hr. This source created a dose rate field of 370 R/hr for the worker closest to the unshielded tube when it was out of the water. Actual personnel exposures from the event raaged from zero mrem (for the fire watch) to a high value of 154 mrem. Cumulative ;
personnel exposure was 0.52 Rem. These exposures are below the "available exposure"
. l limits established for the workers by the radiation work permit, and well below the NNECO ,
administrative limits as well as the 10 CFR Part 20 regulatory limits.
Although the event had low actual radiological significance due to the short duration of the ,
exposure to the unshielded tube, the event had the potential for significant overexposure had the tube remained out of the water for any reason. NNECO estimated that, for a worst case scenario where the dry tube stayed unshielded, NRC exposure limits (3000 mrem / quarter) would have been exceeded in about 0.5 seconds, and a postulated dose of 35 rem would occur in the 20 seconds it would take for personnel to evacuate the area. An unshielded tube could have resulted in a projected site boundary dose rate of 17 mrem /hr at the closest water boundary and 0.0001 mrem /hr at the nearest land boundary.
7
1 NRC inspection determined that NNECO had established good health physics controls to monitor the dry tube replacement activities. Health physics action at the time of the event was proper to reduce the potential for further exposures. The attention to the work activity, and the quick action by operations and health physics personnel were responsible for mitigating the event and preventing a significant worker overexposure. NNECO estimates for the resulting dose rates and exposures were conservative. .
Revised Handline Procedures NNECO developed SP 91-1-38, Dry Tube Recovery, to safely recover dry tube #4 from the '
SFP, The procedure incorporated enhanced health physics controls, inspection of the assembly underwater to ascertain its exact condition, and the use of additional tools (rigid poles) to ensure stronger positive control over the tube during recovery. The procedure was used successfully on May 3 to retrieve dry tube #4, cut the irradiated end and store it underwater, and to remove the cold end from the SFP.
NNECO changed the dry tube removal procedure (SP 91-1-37) to modify the methods to attach tools to the dry tube to better control the tube when removing from the vessel, while transferrir:g the tube from the vessel to the SFP and while cutting the tube in the SFP. Two methods were developed and used. General Electric representatives were also consulted to develop the revised methods and tools. r The first method revised the removal process by employing a new portable cutter that would allow cutting the upper 12 foot section of the dry tube as it hung vertically over the reactor vessel. This method also used revised attachment points to the dry tube to minimize stress to the plunger assembly. This method eliminated the need to bend the tube, thereby eliminating a mechanism for the tube to breach the water surface during cutting. This method was used successfully for continued dry tube removal from May 7 until May 15, after which it was abandoned due to failures of the cutting tool. The second method used, after failure of the ;
portable cutting tool, improved handling by providing for attachment to the dry tube at three ,
locations: at the plunger assembly; at the core plate boss and below the plunger assembly with a rigid poll. This method was successfully used for the remainder of the dry tube removal activities. 6 On May 8, operators had panially removed another 40-foot-long dry tube from the reactor vessel in preparation for cutting the tube in place in the vessel. The tube snapped off at the same plunger location as had occurred on May 1. The inspector determined that even if the new procedure controls had not been in place at this time, no part of the tube could have ;
breached the surface of the reactor cavity. The inspector also concluded that use of the new ;
tube removal method did not cause the break.
i 8
The inspector attended t... PORC meeting for the development o, dry nibe handling methods and interviewed engineering personnel during the development of the revised procedures.
NNECO briefed the inspector throughout the development of the new procedures. NRC inspection determined that the revised methods would assure better positive control over the -
hot end of the dry tube at all phases in the removal process. ,
Work Order Controls i NNECO suspended activities on May 1 after dry tube #4 came out of the water. During a post-event review of the procedures package, engineering personnel noted that an automated work order (AWO) had not been prepared as intended prior to the start of the job on April 30. An engineering supervisor requested that a staff member prepare an AWO for the dry tube activity, in a manner similar to other refueling related work. The AWO was not written due to an oversight by the engineering staff. The primary purpose of using the AWO is for traceability of material issued from the warehouse used for the activity and to apply QC surveillance.
- Millstone 1 Engir.eering personnel initiated AWO 91-05894 in the afternoon and dated the
- form as initiated on 5/1/91. The work schedule section of the form listed the " scheduled start date" as 4/30/91, which was the original date for the removal of all 12 dry tubes. Three of 3
the 12 dry tubes had been removed as of May 1. Inspector interviews with engineering personnel indicated that the preparer believed the proper entry for the section was the date on which the activity began. Engineering perso... el presented the AWO to Millstone QC for -
review and approval as required by procedure. OC deferred approval of the AWO pending a ,
review of the circumstances as to why the AWG was generated late, and how the activity proceeded without the AWO in place. These topics were discussed with Operations personnel ,
on May 1.'
1 During the OPS /QC discussions, a question arose as to the proper implementation of the !
requirements of ACP-QA-2.20C, Work Orders, which states an AWO need not be generated f
- for "... routine plant evolutions performed by Operations Department personnel which
- Is denned in operating procedures; or, is considered routine by the SS/SCO; and, is not considered to be maintenance." Operations personnel believed that an AWO would not be needed since dry tube removal is covered by an operations procedure and is similar to LPRM removal, which is a " routine" operations activity during refueling. QC personnel believed q that the ACP would require an AWO to be issued since the change out of vessel components could be considered a maintenance activity, i This matter was reviewed by QC management and discussed again with the Operations Manager in a meeting on May 3. NNECO determined that an AWO was required for dry tube replacement to meet the ACP requirements. The Operations Manager discussed ACP 2.02C with operations personnel to assure the requirements were clearly understood, and to enfy that AWO controls were being applied to other activities that the operators might consider to be " routine." NNECO plans to change ACP 2.02C to clarify its requirements. !
9 I
i 4
5
" The inspector reviewed tne controls intended to be applied by the AWO and noted that the !
controls established for the activity by SP 91-1-37 met or exceeded those required by the AWO. The AWO did require a QC hold point for the work, which was not performed for >
the three dry tubes removed between April 30 and May 1. The hold point required QC to j verify that torque values were properly applied at 50 ft-Ibs as required by Step 6.2.19.6 for t the installation of new dry tubes. As of May 1, no new dry tubes had been installed. A second QC check was a verification that OPS Form 328 E-1 was completed as a prerequisite to the replacement activity. The inspector noted that the checklist was performed for dry .
tube replacement.
In summary, engineering personnel identified and corrected the failure to apply intended work order controls for the dry tube replacement activity. The inspector determined that the lack of an AWO on May I had no bearing on the dry tube event. The Station QC Group was effective in assuring that the programmatic requirements for work orders were clearly .
understood and implemented. The inspector concluded that licensee corrective actions were I approp6 ate.
- Findings and Assessment f
The cause for the unplanned personnel exposure on May I was the unexpected failure of the l dry tube after removal from the reactor vessel. The probable reason for the failure was the t
cracking mechanism identified in SIL 409 (embrittlement and cracking due to dry tube age),
and was not due to the replacement methods. NNECO is reviewing its schedule for dry tube replacement and plans to change the assemblies on a more frequent schedule during future ;
- outages.
NNECO appreciated the need for special controls to be applied during dry tube replacement, e since it was the first time the dry tubes were replaced at Millstone 1. Special controls were applied in recognition of problems experienced by the industry. These controls focused on personnel safety in a manner commensurate with those applied to LPRM replacement, a routine refueling activity. NNECO also recognized the potential for loose pans generation j during the dry tube removal as a reactor safety issue, and controls were incorporated in the i procedures to address this issue. :
i NNECO also recognized the potential for the dry tube assemblies to break during the removal process. In spite of this effon devoted to the process prior to removal of the dry tubes, ;
NNECO failed to anticipate that the dry tube could break after the tube was removed from '
the vessel, and could result in a handling problem while transporting the tube into the spent l fuel pool for cutting and storage. The failure to recognize this possibility and to address it '
during the review process is a contnbuting cause for the event. l 10 i
L I
i 5
After the May I event, ., DECO followup was conservative and sgressive to assure positive
"~ controls would be maintained over the dry tubes and LPRMs for the remainder of the replacement activities. Health physics controls were strengthened. Health physics, i operatic >ns, and unit management participated extensively in post-event critiques, recovery planning. and the development of modified procedures. Pre-job briefm' gs were ,
comprehensive and detailed, with workers asking good "what if" questions. PORC and engineering reviews were very thorough to develop recovery and modified removal procedtres. Overall, NNECO took a sound, conservative, non-hurried approach to resolve the issue.
NNECO has recognized the need to incorporate lessons teamed from the May I experience ,
into procedures and will be discussed during the planned post-outage critique.
4.3 (Closed) Unresolved Item 50-245/88-25-01: Reportability of Violations of Technical Specification Administrative Requirements t This item involved the licensee position that violation of administrative technical specifications concerning high radiation area doors are not reportable to the NRC per 10 CFR 50.73. Arguably, a condition which would permit uncontrolled access to a high radiation area and potentially result in personnel overexposure could be constmed as a plant operation or conc!ition prohibited by technical specifications. However, the inspector noted that the ,
intent of the LER mie is to aid the NRC in identifying operations or conditions that might '
lead to serious accidents or threats to public safety and to provide the information for r
engineering studies of operational anomalies, trends, and pattems. The inspector concluded that the licensee's position was acceptable. This item is closed.
5.0 31A INTENANCE/ SURVEILLANCE (IP 61726, 62703, 61701, 60710, 57050, ,
61715) !
5.1 Observation of Maintenance Activities The inspector observed and reviewed selected portions of preventive and corrective maintenance to verify compliance with regulations, use of administrative and maintenance procedures, compliance with codes and standards, proper QA/QC involvement, use of bypass i
jumpers and safety tags, personnel protection, and equipment alignment and retest. The followmg automated work orders were included: .
M 1-90-10567, Inspect "A" Low Pressure Turbine l I
M1-91-06246, Evaluate and repair cracks in "A" Turbine L-3 wheel M l-90-02795, Repair valve 1-SG-6 M 1-91-00940, Repair valve 1-SG-7B
-- M 1-87-07810, Perform service test of *B" main station battery i
M 1-91-02499, Replace valve 1-SW-7A M1-90-02717 Perform VOTES test of valve 1-IC-1 11 ,
__ _ . _ _ .__ _