IR 05000331/1997006
| ML20216B117 | |
| Person / Time | |
|---|---|
| Site: | Duane Arnold  |
| Issue date: | 09/02/1997 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML20216B075 | List: |
| References | |
| 50-331-97-06, 50-331-97-6, NUDOCS 9709050233 | |
| Download: ML20216B117 (41) | |
Text
-,,
-c
,
U. S. NUCLEAR REGULATORY COMMISSION REGION lll Docket No:
50 331 License No:
DPR-49 Report No:
50 331/97006(DRS)
Licensee:
IES Utilities incorporated 200 First Street S.E.
P.O. Box 351 Cedar Rapids, lA ' 52406-0351 Facility:
Duane Arnold Energy Center Dated:
March 24,1997 - July 1,1997 Inspectors:
R. Westberg, Team Leader
- E. Plettner, Reactor Engineer M. Holmberg, Reactor Engineer R. Winter, Reactor Engineer J. Millanda, NRC Contractor M. Bagele, NRC Contractor Approved by:
M. Ring, Chief, Lead Engineers Branch
,
Division of Reactor Safety
'
s I
DO P.DR
,..
i
-.,
_
__m_.__._
_
_ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ -___-__
.
-,,--
4-
,_
.,
EXECUTIVE SUMMARY Duane Arnold Energy Center NRC Inspection Report _50-331/97006(DRS)
The inspection was a system operational performance team inspection and included aspects of licensee operations, engineering, and maintenance. Additionally, the inspection
-
> included assessment of the effectiveness of licensee controls in identif'/ ng, resolving and i
preventing problems; Ooerations :
The team considered the general conduct of operations to be professional and
safety-conscious (Section 01.1).
Use of Technical Specification Interpretations associated with the residual heat e
- removal (RHR) system was acceptable and the team considered their removal from Technical Specifications an appropriate improvement (Section 03.1). -
Auxiliary operators appeared knowledgeable about the RHR, essential service water,
and reactor recirculation systems and procedures (Section 03.2),
A violation was identified for failure to assure that corrective actions taken in e:
response to a previously cited violation were adequate to preclude repetition (Section 03.2).
!
Use of procedures in the simulator that were not the latest revision was considered
>
e-a program weakness (Section 03.21.
The problem identification and corrective action system, the action request (AR)
e-system, appeared to be well established and formalized. The AR system maintained a good capability for problem identification and appeared to be well utilized by plant personnel to document problems in a variety of areas with a sufficiently low initiation threshold level to allow identification of re:atively minor deficiencies (Section 07.1).'
f The PH excursion fact finding meeting was thorough and management was involved e'
in the process. The important issues were effectively identified (Section 07.1.2).
.
. Recent efforts had produced more rigor in root cause analyses and there appeared _to
.*
be sufficient numbers of in-depth analyses (Seccion 07.1.3).
- '
Programs that implement operational experience feedback contained sufficient responsiveness and appeared to have mechanisms to notify and train appropriate
_ ersonnel on the issues (Section 07.1.4).
p Appropriate mechanisms appeared to be in place for self-assessment and quality
assurance activities and did identify and correct a number of plant and organization
- problems; however, not having some mechanism to independently check design calculations was a weakness in light of the numerous mathematical errors,
.
._
_ _ _ _ _ _ _ _ _.
_ _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ _.
-_
.3 =
....
g.
--4-inconsistent source data and the undocumented assumptions found by the team (Section 07.1.5).
The operations committee and the safety committee each appeared to ask sufficient e
- critical questions and to seek a thorough understanding of issues presented to them.
The committees gave appropriate attention to issues and generally appeared to focus on the right level of problems (Section 07.1.6).
Maintenance
'
Inaccuracies in the methodology used to establish pump reference data for the e
replacement pump could potentially impact subsequent evaluations of pump performance (Section M1.1).
The general material condition of the RHR system was good. No liquid leaks were e
. observed during the walkdowns; however, the RHR system was not in operation during the inspection (Section M2.1.1).
+
Relatively_ narrow interpretation of the American Society of Mechanical Engineers e
Code appeared to contribute to the lack of timely corrective actions for an identified flaw. A violation was identified for failure to promptly correct the flawed and potentially nonconforming valve body prior to returning the valve to service (Section M2.1.2).
The station equipment monitoring program had the potential to improve component e
and system reliability (Section M2.2),
A violation was identified for failure to ensure design acceptance criterion as e-described in the Updated Final Safety Analysis Report (UFSAR) were incorporated
. into the surveillance acceptance testing criterion for valves MO 2003i MO 1905
--
MO 2117, and MO-2137 (Soction M3.1).
Enaineerina A violation was identified for failure to perform a documented independent review e
and assign a calculation control numbe' for the calculation documented in AR 95-0464 (Section E1.1.1).
- .
A violation was identified for failure to follow the requirements of the design verification procedure (Section E1.1,2).
The team i&ntified numerous inconsistencies in design basis calculation inputs for
- -
RHR and relatcd systems..The team considered these calculation input errors and/or
- inconsistencies a weakness in the licensee's calculation control process (Section E1.1.3),
A violation was identified for failure to perform an operability determination in e
accordance with the action request procedure (Section E2.4).
.__ _ _ _ _ _ _ _ _ _ _ _ _ -
.:,
..
,
.e e-
. The team concluded that the use of a 184 degree F bonnet temperature in CAL 97 002, which evaluated pressure locked valve V19148 was a non verified and non bounding input assumption and a violation of 10 CFR 50, Appendix B, Criterion Ill, " Design Control" (Section E2.4),
For the RHR heat exchanger (HX) performance monitoring procedure, the team o
-
identified weaknesses in confirming the design basis performance and minimum i
performance criterion used in the current HX monitoring procedure. Additionally,_
the licensee had concluded that due to the inaccuracy of the installed temperature-instruments the current HX performance monitoring data was not reliable enough to use as a predictive toe. 1ection E3.1).
Although the design basis documents and UFSAR generally appeared to adequately
define and document the design basis, the team identified numerous inconsistencies
- in the UFSAR, Design Basis Documents and supporting calculations associated with containment pressures and temperatures assumed in evaluating adequate net -
positive suction head for the RHR pumps under design basis accident scenarios (Section E3.4).
I
-
.
l
-
- _ _ _ _ _ _ _ _ _ _ _. _ _ _ _ - _ _
.,
..
,
BAD 9111te_titila LOoeratior a
,
Conduct of Operations 01.1 Control Room Observations a.
{nsoection Scone (93801)
The team performed frequent observations of control room activities throughout the inspection, b.
Observations and Findings Thorough shift turnovers were observed during the inspection. Operations personnel were knowledgeable of plant conditions and responded promptly and appropriately to alarms. The team considered the staffing levels appropriate to support control room activities, c.
Conclusion The team considered the general conduct of operations to be professional and safety-conscious.
Operational Procedures and Documentation O3.1 Review of Duane Arnold Technical Soecification Interoretations (TSis)
a.
Insoection Scoce (9900)
The team reviewed Duane Arnold Energy Center's (DAEC's) TSis to determine which TSis affected the operation of the residual heat removal (RHR) system and to determine their acceptability, b.
Observations and Findinas DAEC had eight TSis in the approved TSI Manual; four of which were for power operation that directly involved the RHR system (these were reviewed in detail by the team). No TSis associated with the RHR system were identified that violated the Technical Specifications (TSs), allowed a less conservative operation than the TS, or changed the intent of the TSs. The licensee was in the process of converting to standard TSs and expected to fully implement the program by the end of 1997.
Subsequent to the on-site portion of the inspection, the team was informed that effective July 25,1997, the TSIs would be removed from the TSs.
_.
___
.,
..
.
c.
Conclusion The team concluded the use of TSis associated with the RHR system had been acceptable and considered their removal from TSs an appropriate improvement.
03.2 Review of Ooeratina Procedures a.
Insoection Scone (93801)
The team reviewed operating procedures and alarm response procedures for the RHR system in various modes of operation and expanded the review to include the Essential Service Water (ESW) and Reactor Recirculation (RR) systems. Most of the procedures associated with this review, were walked through usirig the site specific simulator and in select instances, the team observed actual performance of system surveillanco procedures. The following procedures were reviewed in detail:
Operr+ing instruction (01) 149, " Residual Heat Removal System,"
e Revision 51, dated January 10,1997 01454, " Emergency Service Water System," Revision 24, dated
November 2,1996 Annunciator Response Procedure, "'A' RECIRC MG DRIVE MOTOR TRIP OR e
OVERLOAD," Revision 6, dated October 22,1996 STP 46E001-SLO, "Dai!y Jet Pump Operability Single Loop Operation,"
e Revision 8, dated December 20,1996 STP 47A016-Q, " Containment Purge and Vent Valve Quarterly Leakage
Integrity Test," Revision 2, dated February 12,1997 STP 45A002 O, "LPCI Systom Cuarterly Operability Tests," Revision 19,
dated August 29,1996 STP 43F004, * Single Loop Operation Procedure," Revision 9, dated
February 14,1997 Abnormal Operating Procedure (AOP) 149, " Loss of Decay Heat Removal,"
o Revision 8, dated December 13,1996 AOP 255.2, " Power / Reactivity Abnormal Change," Revision 12, dated
November 6,1996 b.
Observations and Findinas Previous NRC inspections had identified problems with technical inaccuracies in operational procedures and the following corrective actions were taken as described below:
i
1
.
_ - __ __ __
- ,
.
.
.
.
.
The response to violations in NRC Inspection Report 96005, dated September 27, 1996, (NG-96-1968) stated in Section 3 for response to Violation 2, that the Quality Assurance (QA) department would perform a review of plant procedure deficiencies associated with the procedure review process. Part of the corrective actions taken by the licensee included a letter to Operations from the QA department dated November 27,1996, to refresh and heighten the procedure owner's responsibility for the following requirements: ACP 106.3, " Procedure Revision Process," stated that the procedure owner shall review the procedure and Document Change Form (DCF) for the following: 1) technical accuracy (this includes labels and plant configuration); 2) compliance with requirements:
3) comp!eteness; 4) adequacy of the safety evaluation applicability review (SEAR)
per ACP 103.2; 5) impact on other procedures; 6) impact on worker qualifications; and 7) insertion of quality control inspection points.
The response to violations in NRC Inspection Report 96011, dated February 26, 1997, (NG 97-0343) stated in Section 2 for response to Violation 1, that shift orders were issued on November 5,1996, to the operations department emphasizing that the Operations Shift Supervisor should verify that DCFs are incorporated prior to authorizing any Surveillance Test Procedure (STF). As a result of a previously identified instance of failure to incorporate DCFs that occurred in August and September of 1996, two additional root cause analyses were performed. Both root cause analyses concluded that DCFs were not always being properly incorporated into their associated procedures. The process of incorporating DCFs was also reviewed as part of the operations department self-assessment and the foilowing additional actions have been taken to prevent recurrence of similar personnel errors:
ACP 106.3, " Procedure Revision Process," was revised to require initiators of
temporary DCFs to include a markup of the affected procedure and require pages affected by the DCF to be conspicuously marked.
The following procedures have been created or revised to implement the
requirements of ACP 106.3:
Maintenance Directive 46, " Maintenance Procedure Revisions"
Plant Chemistry Procedure 3201.15, " Marking Procedures for
Temporary DCFs" Operations Department Instruction 5, " Marking Procedures for
Temporary DCFs" Procedure Department Instruction 6, " Temporary DCF Processing" e
Radwaste Department Instruction 2, " Marking Procedures for c
Temporary DCFs" In addition, Updated Safety Analysis Report (UFSAR) 17.2, Appendix A, Section 6.11 was changed to replace the previously required biannual review to a
-
-
.
.
...
.
_...
._
_
-_ -
..
.
single review process that will now serve two functions, one to address the immediate change and second to ensure the procedure content is accurate.
The team obser/ed that modifications, reviewed by the engineering team members, w are incorporat ' into the operations procedures. However, technical inaccuracles
. vere identified d.
ig the review. All examples of inaccuracies were provided to e appropriate departments during the inspection and corrective action documents w *e generated as noted in the following representative examples:
1.
01149, Section 11.0, Step 5.b, identified the torus level recorder on Panel 1CO3 as LR-4325; the correct designation was UR-4325. The torus level recorder on Panel 1C29 was identified as LR-4385; the correct designation was LR-4385B.
2.
01454, Step 3.14, stated, " Verify that system operability has been established by successful performance of STP 48 COO 1 within required surveillance frequency " The correct statement was " Verify that system operability has been established by successful performance of STP 48EOO1-0 within required surveillance frequency."
Procedure work request OPS 97-04-34 was initiated to correct the inaccuracies.
3.
Annunciator Response Procedure "A" Recirc MG Drive Motor Trip or Overload, Step 3.1, stated, "If A MG SET MOTOR AMPS are greater than 565 amps and does not trip,immediately reduce the speed of the A Recirc MG until this alarm clears. Comply with Tech Specs 3.6F for Recirc Pump Speed Mismatch Limitations"; the correct statement was, "If A MG SET MOTOR AMPS are greater that 565 amps, trip the A Recirc MG."
The procedure also contained two steps marked as 3.3.
4.
STP 46EOO1 SLO, Step 7.1.11.c, stated, "lF tho answer to 7.1.11.a is NO, notify the OSS immediately and follow the instructions given in General Instruction 4.3 N/A Step 7.1.11.b." The correct statement was "lF the answer to 7.1.11.a is NO, notify the OSS immediately and follow the instructions given in General Instruction 4.4. N/A Step 7.1.11.b " The same error as above existed in Ster 7.2.11.c of the procedt.re.
A procedure work request, STP 97-04-58, was initiated to correct the inaccuracy.
5.
STP 47A016-O, Step 7.1.12, failed to contain the pressure requirements for the test. It was assumed to be the same pressure stated in Step 7.1.9 which was 43 pounds per square inch gauge (psig) (+ 2,-O psig). However, because of the time necessary to perform Steps 7.1.10 and 7.1.11, the pressure could have decayed to below 43 psig and thus made the test invalid.
AR 971152971152was initiated to evaluate and determine a corrective action.
,
.
..
..
.
.
..
...
.
.
_ __-_ - _ _____ -_-____ - _ _
_ _ _ _ _
_-
i,
-.
<
,
,
6.
STP 47A016 0, Step 7.1.15, stated, " Confirm " Pressure Drop" obtained in Step 7.1.13 is less than (<) 10 psi over the ten minute period." The correct statement was " Confirm " Pressure Drop" obtained in Step 7.1.14 is less than (<) 10 psi over the ten minute period." The team noted that the above error had existed since the procedure was changed from STP 47A016 Revision 8 to STP 47A016-0 in 1995. The procedure had been performed.-
once a quarter for a total of six times since the revision and the error had not been identified.
Procedure work request, STP 97-04 58, was initiated to correct the
- inaccuracy in Step 7.1.15, 7.
STP 45A002 0, Step 7.4.55, failed to have American Society of Mechanical Engineers (ASME) indicated aftor the closed stroke time acceptance criteria.
Licensee personnel stated that they were taking no actions to correct the above technicalinaccuracy since the valve is timed with the ASME criteria in Step 7.2.55. However, notes in the procedure allowed an option to not complete Step 7.2.55. Therefore, the team considered the procedure to be in error.
During the procedure walkthroughs, the team noted that auxiliary and control room operators appeared knowledgeable about the RHR, ESW, and RR systems and procedures. The team also noted that four procedures used in the simulator were not the latest revision of the procedure The procedures were as follows:
Annunciator Procedure 1C04A A-4, ""A" Recirc MG Drive Motor Trip or e
Overload," Revision 6, dated April 15,1996. The correct revision was 7, dated October 22,1996, Annunciator Procedure 1C04A A-1, ""B" Recirc MG Drive Motor Trip or e
- Overload," Revision 5, dated April 29,1996. The correct revision was 6, dated October 22,1996.
STP 43F004, " Single Loop Operation Procedure, Revision 8, dated
November 6,.1996. The correct revision was 9, dated February 14, 1997.-
STP 48E0001-0, "rmergency Service Water Quarterly Operability Test."
-*
Revision 3, dated May 24,1995. The correct revision was 4, dated April 10.
1997.
c.
Conclusion
.
The corrective actions implemented by November 27,1996, for previously identified violations in Inspection Report 96005, dated September 27,1996, were inadequate because seven of the procedures reviewed during the inspection were found to contain similar technicalinaccuracies, e-
. _ - _ _ - _ - _
_ - _ _ _ - _
___-_ _ _ _ _ _ _ _ _ _ _ _
i
+
.
,
_ *
<
,
Failure to assure that corrective actions taken in response to a previously cited violation were adequate to preclude repetition was a violation of 10 CFR 50, Appendix B, Criterion XVI (50-331/97006 01a(DRS)).
The team considered the use of procedures in the simulator that were not the latest revislSn, a program weakness.
Operations Staff Training and Qualification O5.1 Ooerator Trainina on RHR The team reviewed several training procedures and interviewed a training and simulator instructor. Based on statements made by the instructors, training on the RHR system, and all its modes, was taught frequently. Training on specific portions, such as engaging shutdown cooling and implementation of EOPs, was taught more frequently in both class room and simulator. No discrepancies were identified by the team.
Quality Verification of Operations 07.1 Licensee Controls in Identifvina. Resolvina and Preventina Problems 07.1.1 Corrective Action Proaram a.
Insoection Scoce (40500)
The team reviewed deficiencies tracked in the AR system, the overall site wide corrective action tracking program. The review included actions taken and interim or deferred resolutions.
J
b.
Observations and Findinas The team noted that the AR system had established a single process for identifying and correcting plant problems. This process had been in place for several years after replacing many independent deficiency tracking systems and was, generally, well utilized by plant personnel. The problem identification / corrective action process
'
included initiation by any plant personnel and discussion of new ARs at a small but diversely attended meeting immediately after the morning meeting. This meeting continued on into a screening for more complex ARs to discuss alternative solutions or due date extens!ons or to forward large cost ARs to the Safety Committee evaluation panel for their review.
The AR system incorporated procedural steps to address reportability and operability determinations. However, the team identified that presently there was no method to flag or sort ARs which involved operability or reportability issues. Because the AR system already had codes for sorting ARs into some categories for trending, the licensee was considering adding a special code for operability and reportability issue ARs.
1
_ _ _ _ _ _ _.
____-
__
i,
.
o
.
I Although the programmatic controls in this area appeared to be sufficiently comprehensive, at times the AR program was not rigorously implemented. The team noted some instances where there appeared to be insufficient questioning by plant personnel about potential problems. For example:
An'AR was initiated for a flaw in the RHR pump D check valve after the team
identified weak problem documentation and analysis in the original AR. The licensee performed further corrective action, An AR was initiated when the team questioned the possibility of o
preconditioning a pressure detector, since a note for non normal conditions allowed cycling the detector before the readings were taken in a surveillance.
The team noted that there was inconsistency and lack of clarity in supporting writeups for the conclusions reached by some AR investigations; however, the licensee appeared committed to further improvements and refinements to the AR process, c.
.ConclusioD The licensee's problem identification and corrective action system and the AR syctem appeared to be well established and formalized, it maintained a good capability for problem identification and appeared to be well utilized by plant personnel to document problems in a variety of areas, in addition, it had a sufficiently low initiation threshold level to allow identification of relatively minor deficiencies.
07.1.2 Problem Identification and Correction a.
Insoection Scoce (40500)
The team reviewed a sample of operational events, testing, modification and maintenance activities, including some with safety evaluations or operability determinations, to assess the licensee's ability tc identify and correct problems.
b.
Observations and Findina_ s The team noted that proper initialidentification and correct characterization of problems was generally made. For example, the team attended a fact finding meeting concerning circumstances where the circulating water chemistry control system allowed the PH level of discharged water to possibly be less than 6.0. At that PH level, water would be discharged outside the boundaries of the state environmental permit. The licensee reported the occurrence to the State of Iowa and because this outside agency was notified the PH excursion became a reportable event to the NRC. The fact finding meeting focused on understanding the reasons behind the occurrence and reconvened the next morning to hear additional personnel
'
who were involved in the chain of events. Troubleshooting of the PH sensor called for changing the plant valve alignment and isolating the PH sensor into a stagnant line configuration which in turn caused the "non-actual conditions" indications feeding into the control circuit to continuously run the acid pump creating a PH
_ _ _ _ _ _ -.
-_
_ - - _ _ - _ _ _
_
,
.
.,
,
excursion. The basic flaw was the assumption by maintenance and operations departments that the acid pump could easily be run in manual, when in fact, it had never been run in manual, in addition, the Chemistry Department had reached an internal opinion that it should never be run in mamal. The licensee planned further action with a root cause investigation to deterniN appropriate corrective actions.
The team determined that safety evaluation screenings reviewed with maintenance work packages were properly documented, but were typically minimalin scope, in most cases there was good followup on material condition issues and aggressive resolution for most emergent safety issues. Engineering performed Probabilistic Risk Assessment for maintenance and unusual operating conditions allowing the plant to understand the level of risk.
c.
Conclusion The team concluded that the fact finding meeting was thorough and that management was involved in the process. The licensee was effective in identifying the important issues.
07.1.3 Root Cause Analvsis Proaram a.
Insoection Scone (40500)
"ine team reviewed root cause and a similar program, the Human Performance
/
Effectiveness System (HPES), to identify any strengths and any weaknesses,
\\.
b.
Observations and Findings I'
The action request system procedure, ACP 114.5, required a root cause investigatic.1 for level 3 or higher ARs. Some level 4 ARs which involved more complex situations received root cause investigations by use of a solution team.
The solution team was generally composed of personnel from the responsibie department that would follow the root cause analysis procedure, ACP 114.3, and make determinations. Their approach could be more solution oriented using engineering judgment than a rigorous root cause methodology performed by an independent party.
The team reviewed an instance where the initial analysis was less than rigorous, in 1995, CV4302 failed to cycle and a temporary modification moved wires from the first two relay contacts to two unused contacts. At that time, the cause was believed to be a failure in the relay base and that a part would have to be ordered.
Subsequently, the relay base was replaced and the temporary modification wiring was restored to the original configuration. Several months later, the problem recurred, but troubleshooting was inconclusive on the cause of the intermittent problem. During the inspection, the problem occurred again and the licensee concluded that the probable cause was relay A718-K4302. The relay was replaced.
The removed relay was inspected at the instrument shop and one set of contacts appeared to have less closing tension creating a higher contact resistance. To verify this hypothesis a test circuit was set up to periodically test the contact resistance
l
_ _ _ - _ - _
_ - - - - _ _ _ - _ _ _ _ _ _ _ _ -
<
.
.
over a period of several weeks to cor.fum the intermittent high contact resistance.
The team noted that tha current roct cause analysis appeared to be thorough.
In another example, the root cause analysis appeared to be fairly rigorous but a final solution was still not obtained for the spurious runback to approximateiy 45 percent speed of the B MG set in January,1996. The licensee theorized that electrical noise was picked up on the wiring to the 45 percent input to the controller. The cabling has no shielding and the wires are not a twisted pair making the circuit vulnerable to noise. Additional troubleshooting had occurred and several alternative solutions have been investigated, but the licensee was committed to finding a final resolution.
In another example, the root cause analysis reviewed nine failures to the 480 VAC load center breakers. Two categories were estehlished;less than adequate maintenance practices, namely, sliding the breakers across the floor could jostle the close latch release rod and damage the nylon coupling bushing, and at every preventative maintenance the nylon coupling bushing was overtightened. The team noted that this root cause appeared to be sufficiently self-critical.
Last year there were 41 investigations of either root cause or HPES, a similar in-depth investigation. Human performance effectiveness was generally concerned with breakdowns such as self-checking errors and independent verification issues.
Human performance errors were tracked by the QA department.
c.
Conclusion l
The team concluded that recent efforts had produced more rigor in the root cause analyses work. In addition there appeared to be a sufficient number of in-depth analyses that were performed by the licensee, 07,1.4 Goerational Exoerience Feedback Proaram a.
Jnsprction Scoce (40500)
The team evaluated the adequacy of the licensee's programs that implement operational experience feedback.
b.
Observations and Findinas The licensee maintained a process to review operating experience reports, such as significant event reports, significant operating event reports, and significant event notifications generated by the Institute of Nuclear Power Operations, NRC notifications, vendor reports, and reports from similar facilities.
The AR system was utilized to document industry and operational event issues.
Although some historical operational feedback issues were screened with a narrow viewpoint and had weak resolution, recent issues had good initial screenings and sufficient resolution and tracking.
l
_ _ _ _ _ _ _
- _ _ _ _ - _ - _ _ _ _ _ _ _.
_ __
i,
..
.
Althou0h the team identified no programmatic concerns, soms implementation weaknesses were noted. For example, even though vendor supplied information -
was generally well screened, Section E2.2 of this report provides an example where a service information letter was implemented with subjective enginearing judgment.
c.
Conclusion The team concluded that programs which implement operational experience feedback contained sufficient responsiveness and appeared to have mechanisms to notify and train appropriate personnel on the issues.
07.1.5 Self Assessment Proaram a.
Insoection Scone (40500)
The team evaluated the effectiveness of the licensee's self assessment capability by reviewing Engineering self assessment reports and Quality Assurance (QA) quarterly assessment reports, b.
Observations and Findinas The team observed that Engineering self assessments had numerous recommendations. However, the self assessments and recommendations appeared to focus on programmatic aspects, such as, improve communications and align expectations between Engineering, Operations and Maintenance.
The team noted that the QA program emphasized closer and more frequent interaction with the line departments. The quarterly assessment report brought relevant data together in one place which demonstrated that a substantial amount of effort was expended in improving the QA programs. QA audits were detailed with comprehensive audits targeting most of the line organizations onsite. ' Trending information was available in the Quality Assurance Trend report.
The team identified that QA did not review calculations except when a calculation was performed during the time that QA was reviewing modifications, c.
Conclusion The team concluded that from a programmatic standpoint, the licensee appeared to have appropriate mechanisms in place for self-assessment and quality assurance activities. QA had identified a number of plant and organizational problems. The
.
team concluded that not having some mechanism to independently check design calculations was a weakness due to the numerous mathematical errors, the inconsistent source data and the undocumented assumptions found by the team.
.
.- _ _ _ _ _ _ _
)
.
'
,
o
.
07.1.6 On-site Review Committee a.
insoection Scone (40500)
The team evaluated the effectiveness of the operations committee and the safety committee by reviewing committee minutes and attending meetings, b.
Qbservations and Findinas The team noted that the agenda included action issues on the immediate horizon and issues that were more than a year away from action. Usually several options were discussed and considered. The format of a written agenda was valuable although the team noted that not all participants had a chance to study the topics to be discussed before attending the meeting. Management commitmeat was shown by the Nuclear Vice President being actively involved in the one meeting that the team attended.
c.
Conclusion The team concluded that the operations committee and the safety committee (equivalent to onsite and offsite review committees) each appeared to ask sufficient critical questions and to seek a thorough understanding of issues presented to them.
The team also concluded that the committees gave appropriate attention to issues, raised questions freely, and generally appeared to focus on the problems.
II. Maintenance M1 Conduct of Maintenance M 1.1 Ecolacement of the "A" Residual Heat Rq oval Service Water (RHRSW) Pumo m
a.
tnsoection Scoce (93801)
The team observed portions of the A RHRSW pump,1P022A, replacement and reviewed the completed preventative maintenance action request 1100316.
b.
Observations and Findmas On April 14,1997, AR 97-1143 documented that the A RHRSW pump had recorded post-maintenance vibration readings above the inservice testing alert range. it was determined that the increased vibration had been caused by a rub in the pump column section in a rubber bushing and a decision was made to replace this pump with a rebuilt spare pump. The replacement pump had been rebuilt (included new stainless steel impellers) by maintenance personnel and therefore had no applicat:'e vendor pump performance curve. Post-maintenance testing was performed to record six points which were plotted on a vendor pump curve for a similar vendor rebuilt pump. The team identified that establishing a pump reference curve in this manner did not account for installed instrumentation inaccuracies with respect to instrument accuracies used to establish certified vendor pump curves, nor were fluid density temperature corrections applied for the actual temperatures of the baseline
_ _ _________
_ _ _ _ _ _ _ _ _ _ _ _ - __
-i,
-
e
-,
testing, vice that used to establish the vendur pump cu es. These inaccuracles could potentially affect the analysis of the subsequent inservice testing data used to compare the measured pump performance and monitor for potential pump degradation.
The team observed that no difficuities were encountered during botting of the replacement pump, which was accomplished in accordance with procedure GMP-MECH-01, " General Botting Requirements," Revision 8.
c.-
Conclusion
The replacement of the A RHRSW pump was performed without difficulty.
However, the team identified inaccuracles in the methodology used to establish pump reference data for the replacement pump, which could potentially impact subsequent evaluations of pump performance.
M2 Maintenance of Facilities and Equipment M 2.1 Material Condition of the RHR Svstem M2.1,1 Svstem Walkdown a,
insoection Scoce (93801)
The team walked down selected portions of the RHR system, b.
Observations and Findinos The following discrepancies were identified by the team during the system walkdown:
MO 1921 and MO 1913 had 3/4 inch bypass lines installed to prevent e-pressure locking during work performed in December of 1996, under an Engineering Maintenance Action (EMA). These bypass lines were not shown on the system drawing, M-119, Revision 63. The closure process for EMAs required updating of the system drawings. The EMA which had installed this modification had not yet been submitted to engineering for closure.
ACP 109.1, " Engineered Maintenance Action," Revision 2, required that the
-
EMA should be closed "within a reasonable time period, using 120 days as a guideline, froiii the date the Engineered Maintenance Action (EMA)is transmitted to Engineering for closure...."; however, the team noted that this procedure did not specify an overall time requirement for the EMA closure pror:ess.
MO 2004, MO-1939, MO 1940, MO-1941, MO-1949A, MO-19498 and
- -
many of the manual RHR system valves had labels locally at the valves with abbreviations and/or descriptions (such as for the associated pump or heat exchanger) that did not match the valve lineup wording exactly. The licensee reviewed the AR database and determined that there had been no history of problems on the RHR system attributed to valve labeling.
_-
_ - - _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
_ _ _ _
_
.
,
.
i MO-1921 and V19190 lacked identification tags locally t.t the valve. The e
licensee initiated AR 97-1167 and 971167.01 to label these valves, V19-190 (third isolation valve on the B RHR HX shell side drain) was installed o
on the B train of RHR and no counterpart valve existed on the A train of RHR.
This valve was not identified on the valve lineup, however it was shown on the system drawing. This valve had been added to the system drawing after performance of an as-built walkdown in 1992; however, the reason for the installation of this valve was unknown. The licensee initiated AR 1167 to address this issue and reportedly planned to add this valve to the system lineup, The team questioned the location of pipe reducers near MO-1939 as shown o
on the system diagram. AR 97-1144 was issued to correct drawing M-119, which did not show the pipe reducer upstream of MO 1939.
MO 1966 and MO-1967 were abandonnd in place but remained as part of
the system and were not on the system lineup. The licensee reported that the tagout procedure, ACP 1410.5, would ensure that these valves were in the correct position after maintenance which would be the only evolution that would manipulate these valves, internal elements of sightglass flow indicators FG 2063 and FG 2064 in the e
ESW lines providing cooling flow for RHR pumps A and D seal water coolers had corroded extensively. The team could not determine if the internal check valve element would be free to operate and thus could impact cooling of the pump seal water cooler. However, no flow problems were recorded during previous ESW cooling flow testing, which had been performed using an ultrasonic flowmeter and thus did not rely on the sightglass flow elements.
c.
Conclusion
The team concluded that the general material condition of the RHR system was good. No liquid leaks were observed during the walkdowns; however, the RHR system was not in operation during the inspection.
M2.1.2 Action Recuest Reviews a.
Insoection Scooe (93801)
The team reviewed all RHR-related ARs issued in the past three years and associated corrective actions taken for system or component deficiencies in the RHR system, to assess system material condition history.
b.
Obserrations and Findings Flaw Found in the Body of Valve V19-OO1 On March 23,1995, a 3/4 inch long surface breaking linear indication (flaw) was discovered during an ASME Section XI Code required magnetic particle examination
_ _ _ _ _ - _
_ - _ -_ - _ _ _- __ _ _
'l
-
.,
-
,
of valve to pipe weld, RH1 CF013, which was documented in AR 95-0464. The flaw was in the base metal of the check valve body of the RHR pump D discharge cFeck valve,- V19-001,0.06 inches outside the Code mandated inspection boundary
,
and was oriented circumferentially along the measured length. The valve was accepted for continued service based on an engineering evaluation using a fracture mechanics analysis which assumed an initial flaw depth of % the valve body wall thickness as an input to this calculation. This flaw depth was an assumed value without a documented nondestructive examination to validate or bound the assumption.
The team questioned the basis for not performing a documented volumetric nondestructive examination to determine or bound the t,ubsurface extent of this fiaw. A written response was provided which stated that the option of performing an Ultrasonic Examination (UT) of the flaw area was not feasible because performing UT on cast materials doc: not provide reliable results. However, the team's inquiries prompted the licensee to perform an ultrasonic examination of this area. The team observed this examination end concluded that it was feasible to examine the valve body casting using UT to bound the depth of the flaw within limits assumed in the fracture mechanics analysis. On April 10,1997, the flaw was removed by surface grinding.
During a phone call held with the Office of Nuclear Reactor Regulation (NRR) staff on April 7,1997, the licensee was informed of the NRC staff position that ASME Section XI Code rules should have been applied to this flaw and that the fracture mechanics analysis was required to be submitted to the NRC for review in-
'accordance with the ASME Code,Section XI,1980 Edition Winter 1981 Addenda, Article IWB-3125(b). A written response was provided for this issue, which stated
"We (DAEC) feel that the flaw was adequately characterized using engineering judgment and informational (undocumented) ultrasonic examination to supplement the magnetic particle examination." Additionally, "The evaluation was performed using ASME Section XI, Appendix A only as a guide realizing that the flaw did not fall under the requiroments of ASME Section XI " Subsequently the fracture mechanics analysis ior this flaw was submitted to the NRC for review in a letter dated April 18,1997.
Had ASME Code Section XI requirements been applied to this Class 2 component, the flaw depth would have been determined as part of the flaw characterization to support the fracture mechanics analysis. Additionally, the Code required scheduling followup inspections of this component to measure flaw growth after initial discovery and inspections of similar components / areas. These Code corrective actions were not followed in dispositioning this flaw nor were comparable measures taken to confirm the integrity of this valve.
c.
Conclusion -
The relatively narrow interpretation of the ASME Code appeared to contribute to the-lack of timely corrective actions for this flaw. Failure to promptly co. rect the flawed and potentially nonconforming valve body prior to returning the valve to service was a violation of 10 CFR 50, Appendix B, Criterion XVI (50-331/97006-01b(DRS)).
....
_ _ - _ _ -
_ _ - _
e
.
,
,
M2.2 RHR System and Comoonent Monitorina Proarams a.
Insoection Scooe (93801)
The team evaluated the equipment performance monitoring programs in place for the RHR system. The team reviewed the " Component Failure Trending Program for Electrical and Meenanical Components Equipment Monitoring Manual," Revision 2, which established the guidance and instructions to identify, analyze and trend repeat and common mode failures of components. The scope of the program covered by this manual included 52 risk significant systems with 23,000 total components, b.
Observations and Findings
)
The RHR system engineer used a RHR system specific equipment monitoring progrum established to trend system unavailability for the low pressure coolant injection mode, containment spray mode and functional failures for the shutdown cooling mode and torus cooling mode. The system engineer maintained this program database, which included RHR system technical specification (TS) required
'imiting Condition for Operation (LCO) entries, the reasons for the system LCO
.
(ntries and total time spent in an LCO. Additionally, the system engineer demonstrated the ability to monitor and trend RHR pump bearing temperatures from on-line data retrieved in this monitoring system.
For the station wide equipment monitoring program, a relational database of system components had been established by the program engineering group. This progrem had been available for about six months and was accessible for use by engineering and maintenance department personnel on the station wide computer system. This system was capable of trending similar component failures across multiple systems.
The RHR system engineer had received training within one month of the inspection on the use of this equipment monitoring program. The information recorded in the equipment monitoring program database had a limited time history, typically dating back to 1995 or 1994 and was categorized into the following program categories:
Thermography, inservice Testing (IST), Vibration, motor operated valve (MOV) and Lube Oil. The RHR system engineer demonstrated the ability to use these portions of the program to display applicable parameters for RHR system valves and pumps.
c.
Conclusion The team concluded thc. the station equipment monitoring program had the potential to improve component and system reliability.
M3 Maintenance Procedures and Documentation M3.1 RHR Svstem Surveillance Test Procedurgs a,
lasoection Scoce (93801)
,
The team reviewed the most recently completed RHR system surveillance test procedures for compliance with the IST program, UFSAR, and TS requirements.
l
_ _ - - _ - _ _ - _ _ _ _ _ _ _ - -
- .
.,
,
b.
Oluanfations and Findinos UFSAR table 6.3.2 listed a maximum Low Pressure Coolant injection (LPCI) injection valve opening time of 18 seconds. STP 45A002-Q, " Low Pressure Coolant injection Operability," Revision 19, listed a maximum valve opening time of 18.4 seconds for MO 2003 (inboard injection isolation valve, train B). UFSAR change request 96104 (approved December 12,1996), had been initiated, which changed the maximum opening time for these valves to 28 seconds per the current Loss Of Coolant Accident (LOCA) analysis. NEDC 31310P, Supplement 1, dated August 1993
"DAEC SAFER /GESTR LOCA Loss-of Coolant Accident Analysis," had been submitted (which supported UFSAR change request 96104) to the NRC as part of the Core Operating Limits Report (COLR) for Cycle 13. The team could not detarmine if using a LOCA analysis submitted as part of the COLR satisfied 10 CFR 50.59 requirements for NHC review and approval, since a written NRC approval (e.g., safety evaluation) was not issued for this COLR submittal. However, no recorded opening times for the LPCI injection valves in excess of 18 seconds were found, based on review of records dating back to 1987.
On April 24,1997, the team identified that STP 45A002-Q, Revision 19, listed the
_
maximum acceptable closin0 time for MO-2003 as less than or equal to 19.5 seconds and less than or equal to 18.4 seconds for MO 1905. These times exceeded the design closure time of 18 seconds for the RHR system shutdown cooling discharge isolation valves described in UFSAR Section 7.3.1.1.1.7. The licensee reported that no closure times in excess of the UFSAR values had been recorded for these valves. DAEC personnel considered the UFSAR values for the closing times of these valves to be " nominal design values " This explanation was not consistent with the basis for these times as described in the UFSAR which stated, in part, "To meet the requirement that automatic Type A valves be fully closed in time to prevent the reactor vessel water level from falling below the top of the active fuel as a result of a break of the line that the valve isolates,...." and the fact that no tolerances were specified for the UFSAR " Design Closure Times.'
The team identified stroke times of 22 seconds for the RHR LPCIinjection valves in the DAEC ASME Valve Stroke Time Data Book. These same times were reflected in the BECH-E200 MOV Data Sheets and VOTES test procedure acceptance criteria for these valves. The opening and closing times for these valves were specified in the FSAR as 18 seconds. The RHR system engineer and lead MOV engineer, stated that surveillan :e and VOTES test personnel relied on the ASME times as the proceduralinput to their test programs. Subsequently an AR was initiated to investigate this issue. At the conclusion of the inspection the licensee identified that the Core Spray System injection isolation valves MO 2117 and MO 2137 had recorded closure times in excess of the UFSAR values. On February 10,1997 the licensee had recorded valve closure times for MO-2117 and MO-2137 of 8.13 seconds and 8.16 seconds respectively and had specified an acceptance criterion of 10 seconds maximum in STP 45A001-Q, "Cors Spray System Quarterly Operability," Revision 23. These valve closure times and maximum allowable seceptance criteria were contrary to UFSAR Section 7.3.1.1.1.7, which specified a design closure time of 8 seconds for these valves.
___
_ - - _ _
. _ _ - _ _ _
-
.
.,
.;
,
--
.
The licensee also identified the following additional valves with ASME Code Section XI acceptance criteilon different from that established in the UFSAR:
Recirculation System valves MO-4627, MO-4628, MO 4629, MO 4630; Core Spray -
System valves MO-2115, MO 2135; and RHR System valves MO 2004 and MO-1904. However, the actual recorded stroke time values for these valves appeared to be conservative with respect to the UFSAR values, c.
Conclusion s
The licensee revised LOCA analysis was used to change UFSAR design opening times for the LPCIinjection valves. Since this revised LOCA analysis had not received specific NRC review and approval, the team could not determine if 10 CFR 50.59 requirements had been met. Pendin9 review of this practice by NRR, this issue was an Unresolved item (50-331/97006-02(DRS)).
RHR system surveillance testing procedures reviewed correctly implemented the IST program and TS requirements; however, more restrictive UFSAR desig, values were not incorporated into surveillance testing procedures. Failure to ensure design acceptance criterion as described in the UFSAR were incorporated into the surveillance testing acceptance criterion for MO 2003, MO 1905, MO-2117 and MO 2137 was a violation of 10 CFR Part 50, Appendix B, Criterion XI (50 -
-331/91006-03(DRS)).
M3.3 Maintenance Work Packsoes a.
Insoection Scoce (9380.11 The team reviewed the following maintenance work packages (both preventative and corrective) for work on the RHR and RHRSW systems to evaluate the potential impact on system operability and 10 CFR 50.59 safety evaluation applicability.
-
,
RHR - A15819, A26702, A26703, A27549, A27550, A28004, A36004, A11808, I
A15819, A19682, A20260, A20917,' A21155, A27052, A33319, A33320, A23969, A33321, A33322, A27051, A23142, A26422, A28004, A35498, A36004,Ai2651,A34583,A24399,A18595,A35399,1081568,1081569 RHRSW - A29051, A19397, A34248, A36816, A21401, A22100A, A26895, 1087189,1094478,1079328,1077566,1079446, A20919, A21007, A22902, A27856,A19366 b.
Observations and Findina_ s A safety evaluation screening had been properly documented where applicable for tha maintenance work packages reviewed. However, these safety evaluation screenings were typically minimalin scope, consisting of a documented safety evaluation screening checklist for the principle maintenance work being performed.
These screenings were limited in their technical basis, in that documented reasons (when provided) for checking the safety evaluation screening blocks, relied on engineering judgment, vice calculations and analysis.
1
- _ - ____ -__ - _
?
'l
,
Documented entries into applicable TS LCOs had been made (where required) for equipment that had become inoperable or potentially inoperable. The maintenance activities reviewed had been completed such that an operability concern did not
-
exist after the maintenance had been completed.
c.
Conclusion The team concluded that the safety evaluation screenings associated with the maintenance activities reviewed were minimalin depth and scope. Evaluation of the potential impact on component and system operability for maintenance activities appeared to be adequate.
M4 Maintenance Staff Knowledge and Performance M4,1 Post-Maintenance Test Proaram a,
insoection Scoce (93801)
The inspection team reviewed Procedure MD 024, Revision 12, "DAEC Maintenance -
Department Post Maintenance Testing (PMT) Program." The team examined the roles and responsibilities section of the procedure to ascertain whether proper interfaces were delineated between the IST, Equipment Qualification, MOV, Plant -
Engineering, and Operations personnel. The team interviewed a number of individuals from these organizations relative to their scope of work asaignments and -
responsibilities for PMT activities, b.
Observations and Findinas Attechments 1 through 4 of the PMT procedure provided guidance in selecting appropriate PMTs related to corrective maintenance activities related 'a pumps,
- valves, relays, and other critical equipment. The team identified no discrepancies or inconsistencies and all personnelinterviewed were cognizant of tneir technical and organizational roles and responsibilities associated with PMTs as described in M D-024.
c.
Conclusion The team concluded that personnelinvolved in post maintenance testing were knowledgeable of their roles and responsibilities and that DAEC Procedure MD 024 contained sufficient detail for identifying appropriate PM7s for safety related
. equipment, s
_ - - _ _ _ _ _ _ _ _
. - - - _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _.
!
-
.
,
,
'
e
..
lit. Engineering E1 Conduct of Engineering E1,1 Engineering Calculations and Evaluations
[
E1.1.1 Fracture Mechanics Evaluation of a flaw in the body of valve V19-001 a.
Insoection Scone (9380U Team rev:ewed the fracture mechanics analysis performed on March 29,1995, used to accept a flaw in the valve body of RHR pump D dise;iarge check valve, V19-001, as documented in AR 95 0464. (Section M2.1.2 of this report also describes this y
issue.)
b.
Observations and Findinas On March 23.1995, a 3/4 inch long surface breaking linear indication was discovered in the base metal of the check valve body of the RHR pump D discharge check valve (see Section M2.1), This valve was accepted for continued service based on a fracture mechanics analysis performed on March 29,1995,and documented in AR 95-0464.
On April 7,1997, the team identified that a fracture mechanics calculation performed on March 29,1995, to accept the flaw in check valva V19 001, lacked a documented independrnt review of the calculation and a calculation control number.
c.
Conclusion The team identified several examples in addition to AR 95-0464 where the design control procedural requirements were not followed for calculations associated with safety-related comnonents (see Section E1.1.2 below alto),
Engineering Department procedure 1203.21, " Engineering Calculations," Revision 3 Gection 3.2, required that engineering calculations be verified independently by another individual technically qualified in the same subject and who did not participate in the original calculation. The procedure also required engineering calculation numbers for allies Utilities Inc. ard all supplier generated calculations.
Failure to perform a documented independent review and arsign a calculation control number for the calculation documented in AR 95-04 34 as required by procedure 1203.21 was a violation of 10 CFR 50, Appendix B, Criterion V (50-331/97006 04a(DRS)).
E1.1.2 B RHR HX Aelief Protection Evaluation a.
insoection Scoce (93801)
The team reviewed EMA A26702G, Revision 2, which replaced the existing B RHR HX shell side relief valve PSV 1953 and gagged and abandoned in place PSV 1952.
i
_ _ _ _
_ _ - _ _ _ _ _ - _ _ _ _ _ _ _ _ - _
/,.',*
'
b.
Observations and Findings EMA A26702G, Revision 2, stated the following concern:
"Th>> Engineering Evaluation in Rev 1 discusses the bellows seal on the 4 by 6 steam service valvo and takes credit for the ability of this seal to isolate the exhaust line to the torus from the top works of this valve. An additional bature of the bellows sealis that it also isolates the effect of back-pressure in the exhaust line from the set pressure of the valve. The torus has a potential to be pressurized during and after an accident which would have the affect of raising the set point of a valve without a bellows seal by en amount equal to the back pressure. The installed 4 by 6 relief valve will be gagged and neither the 3/4 by 1 valve which is installed nor the Crosby 900 series to be installed have a bellows. Describe why it is acceptable to defeat this feature."
In response to this conceri. the engineering evaluation stated the following:
"The stresses caused by this short-lived over pressurization on the vessel are less than ASME Code allowable."
in support of this statement an engineering calculation was documented in the design verification summary report section of this EMA and dated September 19, 1996. This calculation evaluated the potential stress affects on the B RHR HX shell that would be created if the HX were to be over pressurized and a 25 psig back-pressure from the torus was applied to the HX shell relief valve (A LOCA would create this back-pressure).
On April 24,1997, the team identified the following:
The incorrect design Code was used for stress level acceptance criterion in this calculation. Calculated stresses in the HX shell were compared against allowab.e stresses from the ASME Code Section Ill, which were less conservotive that the original design Code, which was Section Vill, Division 1 of the 1968 Edition, Winter Addenda, e
An incorrect /nonconservative input value was used in this calculation.
PO 205-AA662, "RHR Heat Exchangers General Description," Revision 2 stated that a corrosion allowance of 0.1 inch was included on all carbon steel surf aces. The full waII RHR shell thickness of 7/8 inch was used in the calculation vice reducing the wall thickness to n. appropriate original design value by subtracting the corrosion allowance from the wall thickness, which introduced a nonconservative error.
Design assumptions, applicable Code Edition and Addenda were not
documented in this calculation.
Licensee personnel stated that this calculation was a " scoping calculation" and thus net subject to the design control requirements of procedure 1203.31. However, the team identified that this calculation was checked under both the design review and
!
l
_ _ _ _ _ _ _ _ _
. ','
.
,
alternate calculation blocks of the EMA Design Verification Summary Report. Both design reviews and alternate calculations fall within the scope of procedure 1203.31.
,
The safety rotated function of the RHR HX is to remove heat from the containment fol'owing a LOCA and thus it would be in service when the LOCA Induced back-pressure would affect the RHR HX relief valve. Since the shutoff head of the RHR pumps was below the relief valve setpoint with the HX in service, the tearn identified no credible sequence of events that would challenge the integrity of the RHR HX The licensee concluded on September 13,1996, in memorandum "RHR HX Thermal Relief Backpressure Considerations," that the effects of containment
<
onck pressure were not considered in the original design of the relief valves for the RHR HXs and were not required to be considered per the original construction Code, since back pressure induced during a LOCA event would not meet the Code definition of a constant back pressure, c.
Conclusion Engineering Department procedure 1200.31, "Desigr, Vcrifi::ation Procedure,"
Revision 6, Section 3.4, required " Design review of design document (s) or modifications will be sufficient to verify the appropriateness ni the design input; including assumptions, design basis, and applicable regulations; codes and standards; and that the design is adequate foi 'he Imanded application..."
Failure to follow the requirements of 1206.31 for EMA A?,6702G, Hevision 2,is a vlotation of 10 CFR 50, Appendix B, Criterion V (50 331/97006 04b(DRS)).
- E1.1.3 Calculations of Not Positive Suction Head (NPSM) Available for the RHR/ Core Sorav Pumps a.
lagoection Scone (93801).
The team reviewed BECHTEL calculations fAC-40B, January 9,1970, and 423 N 005, November 7,1986. The team also reviewed NUTECH calculation 25.2644.0402 and CAL M97 007, " Care Spray /RHR Pump NPSH calculation."
j b.
Observations and Findinas Calculation Nos. MC 40B and 423 N-005 contained the following errors; however,
,
all had been previously identified by licensee personnel:
e-Incorrect torus water level _(neither MC-40B nor 423 N 005 take post accident water level changes into account).
(
Erroneous friction losses (MC 40B and 423-N 005 both calculate friction
'
using erroneous piping configurations).
- -
Incorrect assumption for torus water inhibitor (MC-40B assumes the torus water has a chromate-type inhibitor additive,'which is incorrect and affects friction loss calculations).
.:.
__.:
_
=
= = _ _
-
__
_ _ _ _ _.
- _ _ _ - - _ _ _ _ - _ _ _ _ _ _ _ _
',.. ', '
e incorrect torus water temperaturo (MC 408 has a transposition error from one part of the calculation to the summary matrix page),
e Incorrect and non-conservativo suction strainer head loss value of one foot head loss maximum wore used (MC 40B and 423 N-005 both assumo this vWue).
On April 7,1997, the licenseo issued CAL M97 007, *NPSH for Coro Spray and RHR Pumps." The team reviewed this calculation and found it to be an improvement over previous calculations evaluating the available NPSH to the RHR pumps. The licensee issued AR 97 0819 to address the errors in existing RHR and Core Spray pump NPSH calculations.
NUTECH calculation 25.2644.0402 identified the head losses across the Emergency Core Cooling System pump suction stroinors of 1.66 feet which exceeded the 1.0 foot maximum value specified in the UFSAR. This NUTECH head loss number was
-
subsequently used as an input to the DAEC Coro Spray /RHR Pump NPSH calculation, CAL M97 007. In this calculation the licensoo concluded that a negative NPSH would exist for the Core Spray pumps in one mode of operation during maximum flow conditions, which was characterized as being an insignificant amount. Licensee engineering staff reported that the negative NPSH for the Core Spray pumps would only occur after fifteen minutes of operation in one mode of operation at run out flow conditions. The licensee reported that operators would throttle back the Core Spray pumps to rated flows within fifteen minutes of the accident and that this was consistent with the UFSAR.
j Pending further review of supporting documentation by the NRC to confirm that I
reliance on operator actions to preclude a negative NPSH was concistent with the sys'em design basis, this issue was considered an unresolved item (50 331/97006-OD(DRS)).
c.
Conclusion i
Between April 7 and April 25,1997, the team reviewed multiple erroneous input assumptions in calculation MC 408 and 423 N 005 associated with the available NPSH for the Core Spray and RHR pumps that had been previously identified by licensee engineers.
in addition, the team identified numerous inconsistencies in design basis calculation inputs for RHR and related systems. The team considered these calculation input errors end/or inconsistenclos to indicate a weakness in the licensee's calculation control process.
E1.1.4 Electrical Distribution Svstem Calculations a.
insoection Scone 193801)
The team reviewed direct current (DC) Input Design Log X10021/IELOCO, Filo IELO50,0150.02, " Miscellaneous Unidentified Design input Sheets," calculation X10021.0200.06, Calculation X10021.0200.05, "125V DC Elec'rical Distribution
--
}
/
_ _ _ _. _ _ _ _ _ - -. _
. - - - - - _ - _ _. _ _ _ _ _ _ _ _. _.
.
'
.
. << s
<
System Load Flow and Voltage Drop Calculations," Revision 0, "250V DC Electrical Distribution System Load Flow and Voltage Drop Calculations," Revision 0, and calculation X10021.0200.08, "125V DC Electrical Distribution Short Circuit Calculation," Revision O.
b.
Ohicrvations and FindjD92 The team identified design input inconsistencies with the following calculations:
DC Design innut too Xl0021/IELO50. File IELO50alfD42:
The cable length for the cable from 104 to 1D40 (39 feet) was not in
accordance with the referenced source.
The cable type for the cable routed from 1D42 to MO 1909 was shown as
- 12 AWG, The reference showed this cable as a #10 AWG and #12 AWG, No explanation was given for using #12 AWG only.
The locked rotor and running current values for MO 1909 did not agree with
Calculation CAL E88 005, Revision 2, June 12,1995.
The load data for inverter 1D45 '.48.5 amperes) did not agree with
Calculation CAL E88-005 (40 amperes).
The starting current for MO2401 was shown as 32 omperes and the running
current was shown as 65 amperes. Normally the starting current is higher than the running current, A communication record with the battery charger manufacturer stat 3d that o
the worst case short circuit contribution from the 125V and 250V battery chargers is 10 times the battery charger rated current. However, the battery system short circuit calculations utilized 150 percent of rated current as the short circuit contribution of the battery charger.
CalculadonjGQQ2LO200.05. "125V DC Electrical Distribution System Load Flow and VoitagtDr.co Gimilalkt":
inverter 1015 wa4 moceled as a constant load of 272,4 amperes which did e
not agree with calculation CAL E88-005, Revision 2 June 12,1995, which showed this load fis 82 amperes.
Utilidng half cf tho,tsrting current for pumps 1P227 and 1P228 that start
simultaneously was not explained and may be non conservative, The current for stuting MO-2512 was shown as 65 amperes in the load o
schedule. One half of shat current is 33.5 amperes vice 15.85 amperes used in the calculation.
_ _ _ _ _
,
- s a
Calculation XIOO21.0200.06. *250V DC Electrical Distribution Svalem Load Flow and Voltage Droo Calculations":
Inverter 1D45 was modeled as a constant load of 130.8 amperes which did
not agree with calculation CAL E88 005, Revision 2, which showed this load as 40 arnperos, Two motor-operated valves had voltages less than the minimum acceptable o
voltages during starting. The calculation stated that the system was still functional since the model was ultra conservative and starting currents only occur for one instant in time. This statement could not be verified, The cable lengths and cable resistances d;d not agree with CAL E88 00b, o
Revision 2, or with the Design input Sheets.
Calculallon X10021.0200.08. *125V DC Electrical Distribution Short Circuit Calculation":
The calculation did not include the contribution of the inverter to the short
circuit and the methodology section did not explain why this load was not included.
The assumption that the battery was being equalized at a maximum of
137.46 volts did not account for voltmeter inaccuracles, it was assumed that the short circuit contributed by the battery charger was p
e b
150 percent. This value did not agree with the communication record from the battery charger manuf acturer in the Design input Sheets, The cable resistance for the cable from the DC MCC to 1P227 and 1P228 o
was adjusted for room temperatures of 32 degrees C and 149 degrees C.
The reference for Environmental and Seismic Service Conditions, Qual-SC101, Revision 1, did not confirm these temperatures, c.
Conclus]on Pending NRC review of the licensee's resolution of the inconsistencies, this was considered an Unresolved item (50 331/97006 06(DRS)).
E1.3 Mechanical Modifications a.
Insoection Scoce (93801)
The team reviewed mechanical modification package C/ MAR A23967, originated October 19,1994, which drilled a hole in the disc of MO 2003, and C/M AR A33320, originated Jure 24,1996, which installed a pressure relief line on MO 1921.
_-_
- _ _ - _ _ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _ _
',',. ',
..
b.
Qbservations and Findingg Modification A23967, provided a pressure relieving mechanism for the RHR Loop A LPCI inboard injection valve MO 2003 to alleviate potential pressure locking. No discrepancies were identified and the package was sufficiently comprehensive for this work activity.
The torus suction valves were previously determined to be susceptible to a thermally induced pressure locking scenario (ref. Pressure Locking and Thermal Binding (PL/TB) engineering evaluation, dated February 23,1990). This evaluation concluded that, although the torus suction valves were potentially susceptible to this condition, no field modifications were required due to the assumed low temperature difference the valves would experience. The originator of the evaluation concluded mixing was not a consideration in this case, while the reviewer of the evaluation did not agree. At issue was the question of whether or not any fluid rnixing would occur in this particular plant configuration. This disagreement on the validity of this low temperature assumption prompted the licensee to acquire in-plant temperature data on the subject valve in order to verify the actual valve bonnet temperature (AR Addendum No.: 901001.01). The measured temperature of the valve bonnet was 204 degrees F, which invalidated the 130 degree F assumption made in the PL/TB evaluation, and the pressure relief line modification A33320 was subsequently performed. The team identified no discrepancies in the maintenance work instructions associated with modification A33320, associated EMA and safety evaluation applicability review, pipe stress evaluation, or PMT documentation, c.
Conclusion Licensee measurements of RHR suction valve bonnet temperatures confirmed that initial engineering assumptions of bonnet temperature used to evaluate the RHR torus suction valves for potential pressure locking susceptibility were non-conservative, Licenseo modifications were performed that adequately addressed the potential pressure locking concerns; however, the reasons for the initial non-conservative engineering input assumptions used in the evaluation were not addressed.
E2 Engineering Support of Facilities and Equipment E 2.1 Evaluation of Shutdown Cooling Qnerations a.
Insnection Scone i93801)
The team reviewed AR 95 2371, December 12.1995, which requested an engineering evaluation of the General Electric (GE) Service Information Letter (SIL)
69 requirements associated with the operation of shutdown cooling using RHR pumps and recirculation pumps in parallel configurations.
_ - _ - _
---
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
','4
'
.
b.
Observations and Findinas The engineering evaluation of GE Sll 09 was completed on January 31,1990, and was used as the basis to incorporate a non vendor recommended RHR and recirculation system configuration into RHR 01149, " Residual Heat Rernoval System," Revision 51. The engineering evaluation of GE SIL 09 recommended revising the operating procedure, RHR 01149, to include a configuration with the B
_
recirculation loop pump in service at minimum speed with the RHR pumps injecting into the A recirculation loop. For this configuration GE SIL 69 stated, " Note that Case IVB presents a potential suction head problem for both the RHR and recirculation pumps," and, "Cavitatit,n on RP more likely than in IVA due to NPSH reduction."
The team identified that this evaluation lacked calculations to demonstrate that adequate NPSH would be available in this configuration for the RHR pumps and recirculation pump B. This evaluation relied entirely on engineering judgement to conclude that adequate NPSH existed for the RHR and recirculation system pumps in this configuration.
c.
Conclusion The team questioned the validity of relying entirely on engineering judgement, which prompted the licensee to perform calculations of the NPSH available to the RHR and recirculation system pumps. Preliminary calculations indicated that adequate margins for NPSH were available to ensure cavitation would not occur in these pumps.
Pending NRC review of the completed NPSH calculation, this issue was considered an Inspection Followup item (50 331/97006-07(DRS)).
E2.3 finaineerina Evaluation Proaram for Pressure Lockina/ Thermal Bindino a.
Insoection Scone (93801)
The team reviewed the "DAEC/ Program Engineering Evaluation for Pressure Locking and Thermal Binding," dated February 13,1996. This document was generated in response to NRC Generic Letter (GL) 95 07.
b.
Observations and Findinas Specific detailed evaluations were performed for valves which were screened out as
'
susceptible to PL/TB. These evaluations considered the valve design and operating basis as described in the UFSAR, and applicable RHR operating instructions to ensure worst case conditions were considered, The team considered the general and gate valve specific screening criteria for PL/TB scenarios consistent with concerns raised by generic letter 95-07. However, the evaluation had concluded that four pump torus suction valves, MO 1913, MO 1921, MO 2012, and MO-2015, were potentially susceptible to pressure locking when re-aligning the system to the LPCI mode of operation. These valves were subsequently dispositioned as acceptable (e.g., requiring no corrective actions). This was based 30-I
. _ _ _ _ _ _
.
4...
.
on the assumption that the heatup would be less than 10 degrees F. The temperature rise was described as less than normal ambient room temperature cha ps and was therefore considered insufficient to cause pressure locking.
c.
Conclusica The team concluded that this assumption was unverified and potentially non-conservative. Ponding licensee verification of this assumption and subsequent NRC review, this was considered an Unresolved item (50 331/97006 08(DRS)).
F.2.4 Pressure Locked shutdown Coolina Manual isolation Valve a.
In50ection Scone (93801)
On January 15,1997, the 18 inch shutdown cooling manual isolation valve V 19-148, could not be opened due to heatup and pressurization of trapped coolant in the valve bonnet (pressure locked). The team reviewed AR 97 0094, which was issued on January 10, to document this event and the associated engineering calculations and evaluations.
b.
Observations and Findings initially only an externalinspection of this valve was performed for evidence of leakage. An engineering evaluation and calculation, CAL M97 002, was completed on February 3,1997, and it concluded that no gross yielding of the bonnet had occurred based on a simplified limit load analysis. However, this calculation demonstrated that pressures could have exceeded 4000 psig in the valve bonnet, which was beyond the vendor proof testing for this valve. The team identified that no operability determination had been documented related to this event, which potentially challenged the integrity of the valve bonnet, which formed a non isolable portion of the reactor coolant pressure boundary.
CAL M97 002, which evaluated the pressure locked valve V19148, used an input of 184 degrees F for the final temperature of the valve bonnet based on a straight average between the drywell temperature and measured reactor coolant system temperatures, without a supporting thermal analysis to confirm or bound this temperature input. The team noted that a smallincrease in the valve bonnet temperature input could have reeulted in the calculated stresses exceeding gross yielding for the valye bonnet assembly.
Subsequent to the on site portion of the inspection, on May 22, an outside contractor specialist evaluated the potential pressure developed in the velve bonnet from thls event. The outsiou contractor evaluation used an estimated bonnet temperature value of 250 degrees F. Additionally, the licensee performed CAL M97 013 (draft), which included a thermal analysis of the valve bonnet temperature and concluded that it would have reached 240 degrees F. The licensee concluded that it was highly unlikely that the pressure locking incident would prevent this valve from performing its safety function as a leak tight pressure boundary, however, the contractor recommended an inspection of the valve to ensure damage had not occurreo. The valve vendor performed an external
___ __
,
,..
,
inspection on May 24, and found no evidence of gross yielding of the retainer ring, as evidenced by the bonnet retainer cap screws tight fit against the bonnet retainer.
.
During review of CAL M97 002, 'V19148 Bonnet Pressure," the team noted that numbers in an original calculation had been erased and new numbers substituted in their place without revising the calculation and obtaining new review and approvel signatures. On April 24,1997, the licensee issued AR 97 0856 to address this issue. Pending NRC review of the evaluation of AR 97 0856, this was considered an Unresolved item (50-331/97006 09(DRS)).
s c.
Conclusion ACP 114.5 Revision 9, paragraph 3.1.(2),(a) required review and documentation of operability evaluations on the action request form.
Failure to perform an operability determination in accordance with ACP 114.5 for the pressure locked condition of valve V19148 was a vlotation of 10 CFR 50, Appendix B Criterion V (50 331/97006-04c(DRS)).
The team concluded that the use of a 184 degree F bonnet temperature in CAL M97 002, which evaluated pressure locked valve V19148 was a non verified and non bounding input assumption.
Failure to verify an input assumption for bonnet temperature in calculation CAL M97 002 was a violation of 10 CFR 50, Appendix B, Criterlon ll1 (50-i 331/9700610(DRS).
E3 Engineering Procedures and Documentation E3.1 HHR Heat Exchanaer Performance Monitorina Te.ng a.
insoection Scooe (93801)
The team intt.rviewed cognizant engineering department personnel and reviewed the following documents related to RHR heat exchanger performance monitoring:
NG 93 5151, " Rationale for Testing Plant for 1E201 A & B," dated
December 3,1993 UFSAR DAEC 1, Figure 5.412, " Residual Heat Removal Systems"
NG 97 5151, " Rationale for Testing Plant for 1E201 A & B," dated
December 3,1993 EMP 1E201 HT, " Equipment Monitoring Procedure * Revision 0,
Attachment 1, data sheets for surveillance testing of the RHR HX in 1994, 1996 and 1997 NG 97-0550, "RHR Heat Exchanger Testing Results for 1996 and 1997,"
dated March 17,1997
_
- _ - - _ -
- _ - _ _ _ _ - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
_ _____
,
t,
.
<
,..
,
b.
DhacLVAllens and Findings EntcallallyJnanalyzed RHR HX Inlet temneratures NG 97 550 concluded that assun ing the service water inlet temperature was equal to river vvator temperature at the inlet to the RHR HX was not valid. Based on surveillance data from March of 1997, a correction of 4.20 degrees F for the A RHR HX and 3.5 degrees F for the B RHR HX n'ust be added to the measured river water temperatures to accurately predict the service water supply temperatures at the inlet to the RHR HXs. Therefore,if the river temperature reached the design basis temperature of 95 degrees F assumed in the UFSAR, this could result in a potential intet temperature of up to 99.4 degrees F at the intet of the A RHR HX. This temperature is above the 95 degrees F assumed for the service water intet temperature in the post l.OCA RHR modes C 1 and C 2 of UFSAR Figure 5.412, as well as other accident mode scenarios. The licensee issued AR 97 967 to evaluate this concern identified by the team. Pending NRC review of the evaluation of AR 97 967, this was considered an unresolved item (50-331/9700011(DRS)).
Weaknesses in the RHR HX Performance Monitorino Procedure NG 93 5151, * Rationale for Testing Plant for 1E201 A & B," dated December 3, 1993, established the basis for procedure EMP 1E201 HT The procedural basis included collecting data for monitoring HX performance and trending. The team identified the following weaknesses with procedure EMP 1E201 HT:
The design cooling performance data for the RHR HX is expressed as a
coefficient (or K factor) of 502,000 British Thermal Unit per hour per degree Fahrenheit (BTU /HR"F) as used in figure 5.413 of the UFSAR. For compa9 son of measured RHR HX performance to design, EMP 1E201 HT used this 502,000 BTU /HR"F for the design value which differed from the comparable vendor design and UFSAR LOCA values. From the vendor (Perfex Corporation) specification sheets the team calculated a K factor of 502,500 BTU /HR"F for the same flow conditions established in the monitoring procedure. From the UFSAR Figure 5.412, for post LOCA mode C 2 which has the same flow rates as established in EMP 1E201-HT, the team calculated a K factor of 502,941 BTU /HR"F. Thus, the team considered the reference design value used in EMP 1E201-HT slightly less conservative than the performance required (K factc,r) under LOCA Mode C 2 and comparable vendor design data.
Procedure EMP-1E201 HT, lacked a well defined criterion for rejecting HX
performance measured during this procedure. The procedure required notification of the System Engineer for further evaluation if the measured HX performance was less than 90 percent of design. This criterion appeared to be subjective in that a definitive limit at which the HX performance would be considered inadequate and/or inoperable had not been established. The licensee initW 9 an addendum to AR 97081697081601 to evaluate and prepare a more fermalized basis for the HX performance acceptance criterion.
. _ _ - - _ _
- ____ _ ___ _ ___ - -_______
_
i,
- .
...
,
The 90 percent of design HX performance evaluation point established in
EMP 1E201 dT was based on the available margin to reach the torus temperature limit of 200 degree F for safety relief valve operation (documented in NG 93 5151). The calculations to demonstrate the acceptability of the 90 percent HX performance did not include additional margins to account for temperature or flow instrument inaccuracles. Thus, the tearn considered that the 90 percent evaluation point may not conservatively bound the required HX performance.
Measured RHR HX performance data from EMP 1E201 HT had been plotted for trending purposes. In EMP 1E201 HT, performed on RHR HX 9 on March 28,1996, the measured HX performance was 90 4 percent of the design value and the HX was returned to service until cleaning was performed in October of 1990. Based on review of the data recorded up to March of 1996, the team identified that trending data could have predicted a drop in th6 B RHR HX performance below the 90 percent evaluation point prior to the cleaning. Licensee personnel reported that they would not use this data to predict when the RHR HX would fall below 90 percent of the design value, since the data was not accurate enough to support this type of performance prediction. The team notud a wide 3 ariation in recorded HX performance with recorded values ranging from 90.4 percent to 163 percent of the design reference value, which supported the licensee's position. NG 97 0550 conc!uded that the principle reason for the wide variation in measured HX performance was that more accurate temperature monitoring equipment was needed to obtain accurate data to support trending HX performance. Additionally, licensee personnel stated that the HX monitoring program met or exceeded GL 8913 commitments which required only inspections and cleanings of the HXs, The RHR HXs had been inspected and cleaned each refueling outage until 1988. In 1991, the cleaning frequency was extended to every fourth refueling outage, based on the previous visual examination history, which reportedly had recorded only silting / sediment deposits with no tubo blockage.
c.
Conclusion The team identified a concern with respect to a postulated condition created by the RHR inlet temperature exceeding the river water temperature at the design basis maximum river temperature of 95 degrees F, which would be beyond temperatures assumed in the accident analysis.
For the RHR system HX performance monitoring procedure. the team idenblied weaknesses in confirming the design basis performance and minimum performance criterion used in the current HX monitoring procedure.
E3.2 Docurnentation of Deslon Basis E3.2.1 Uodated Final Safety Analvsis Review a.
1030ection Scopa A recent discovery of a licensee operating their facility in a manner contrary to the UFSAR description highlighted the need for a special focused review that compares
__
__
_ _ _ _ _ - _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _
i,,.*,,
,
plant practices, procedures and/or part. meters to the UFSAR description. While performing the inspections discussed in this report, the team reviewed the
_
applicable portions of the UFSAR associated with the RHR and supporting systems, b.
Obsafyations and Findinas The following inconsistencies were noted between the wording of the UFSAR and the plant practices, procedures and/or parameters observed by the team, Table 5.4-4, " Design Data of the RHR System Equipment," listed 20.1 e ;0'
o BTU /HR Heat exchanger duty at 125 degrees F inlet temperature, which is below the limiting design basis duty for the heat exchanger per figure 5.412 of the Updated Safety Analysis Report for accident (36.6 e 10' BTU /HR, or 51.3 e 10' BTU /HR) and shutdown cooling modes (98.5 e 10' BTU /HR).
Figure 5.413 (graph of the RHR heat exchanger duty vice suppression pool
temperature) represents RHR HX duties less than those required for HX performance in post LOCA modes C 1 and C 2 of Figure 5.412 of the UFSAR.
Page 6.2 71 of UFSAR Section 6.6.6.3.1 states, "For the complete spectrum e
of steam leaks, the time for the containment wall to reach the design temperature of 281 degrees F is greater than 2000 sec. It was evident from these results that sufficient time was available for the operator to limit the drywell wall temperature to less than the design temperature." The licensee
{
provided figure 14.0.36, " Loss of Coolant Accident, Drywell Temperature Response," from lowa Electric Light and Power Duane Arnold Energy Center
- Preliminary Safety Analysis Report, which did not support this UFSAR statement. This figure showed that drywell temperatures would approach containment design temperatures in approximately ten seconds, Section 6.3, pages 6.319 and 20 stated that the RHR motor design allows e
the motor to run at 120 percent of fullload current for 25 minutes before operator corrective action is necessary. A review of the calculation utilized to support this statement revealed that the intersection of 120 percent of the motor full load current and the motor safe heating cWye la c'ocer to 1000 seconds or approximately 17 minutes. 'Ihe team noted that the curve that was used for this analysis was difficult to read, however, due to the importance of being conservative when determining possible motor damage, a conservative time should have been chosen.
Section 8.2.2.2.4 stated that the buses supplied safety loads from 183,184,
189, and 1820 above the required 70 percent (322V AC) or 80 percent (368V AC), which does not agree with the voltages listed in Table 8.2-1.
Table 8.21 did not include safety related MCCs 1821,1891,1846,1834A, e
and 1B44A.
Section 8.3, page 8.3-4 states that the 480V MCCs are the indoor metal e.
clad type and supply power to 250 horsepower motors and below. Licensee
_
~
_
__ -
_
-
. =.
=
_
_ - _ _ _ - - _ _ _ - _
- _ _.
. _ _ _ _ _ __
___ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _
a
.,
.
,e
,
,
engineering staff confirmed that the MCCs are not metal clad and oo not supply motors larger than 100 horsepower.
Pending NRC toview of the licensee's resolution of these UFSAR and design basis related inconsistencies, this was considered an Unresolved item (50 331/97006-12a(DRS)).
E3.2.2 Supoort System Deslan Basis Document (DBDj a.
insoection Scone (938011 The team reviewed the following support system DBDs:
DBD R22 002, Auxillary AC Power System DBD, Revision 2,8/1/95 e
DBD R42 002,250V DC System DDD, Revision 3,8/30/95
DBD R42 001,125V DC System DBD, Revision 3,8/22/95 b.
Observations and Findinas The DBDs appeared to adequately define the design basis with good documentation of the references required to support the system design. The team noted a list of outstanding open items (some were opened as early as 1992) for which the licensee had initiated ARs. The licensee staff stated that these issued would be resolved by September 26,1997.
E3.3.3 Desian Basis for Available NPSH for RHR Pumos inconsistent a.
insoection Scone (93801)
The team reviewed UFSAR, DBD and original design basis specifications, and calculations associated with the available NPSH to the RHR pumps under design basis accident modes of operation, b.
Observations and Findings The team identified a number of inconsistencies between the UFSAR, DBDs, and design calculations relative to containment pressure, torus water temperature, and the relationship of these parameters to RHR pump NPSH during the LPCI mode of operation. Containment pressure inputs ra*1ged from 0 psig to pressurized conditions, torus water temperature essumptions were 130 degrees F or 150 degrees F depending on the particular document. Additiona!ly, the time related dynamic relationships of these vu eblos in terms of the impact on NPSH available to the RHR pumps was unclear (e.g., between short term pump operation at run out flows versus long term rated flow conditions), ha team identified the following documents with inconsistencies as described above:
UFSAR Section 1.8.1 (pressurized containment)
(
l
..
_
<;
- .
,,..
UFSAR Section 6.3.2.2 (0 psig and maximum fluid temperature)
e UFSAR Figure 5.4.15 (14.7 16.5 pounds per square inch absolute (psla))
UFSAR Figure 6.2 47 (150 degrees F)
e DBD E11001 Section 4.10.4 (14.7 pela and 130 degrees F)
e
DBD E11001 Section 6.10 (preesurized contalnment)
.
e DBD E11001 issue (pressurized containment)
i DBD A61001 Section 5.2.10.2(15)(saturated torus water at atmospheric e
pressure)
f
-*
GE Spec. 22A1341 Section 4.2.1.6 (14.7 pela and 130 degrees F)
NEDO 10139 Section 3.3.2.1 (operating at all pressures / temperatures
resulting from a LOCA )
NPSH Calculation MC-400 (19.7 psla and 197 degrees F for Mode C 2)
- -
NPSH Calculation MC 41D ( 14.7 psia and 150 degrees F)
e
,
Ponding NRC review of the licensee's resolution to these UFSAR and design basis related inconsistencies, thle was considered an Unresolved item (50 331/97000-12b(DRS)).
E3.4 Conclusions on Documenistinn of the Deslan Basis Although the DBDs and UFSAR generally appeared to adequately define and document the design basis, the team identified numerous inconsistencies in the UFSAR. DBDs and supporting ca:culations associated with containment pressure and temperatures assumed in evaluating adequate NPSH for the RHR pumps under design basis accident mode scenarios.
E8 Miscellaneous Engineering issues E8.1 (Closed) IFl 50 331/96013 04: Pressure Locking of RHR Shutdown Cooling Manual Isolation Valve review licensee's operability evaluation; On January 15,1997, valve V19140 could not be opened due to pressure locking.
The team reviewed the licensee corrective actions for this occurrence and identified two violations and an unresolved item (see Section E2.3.2). This item is closed.
E8.2 (Closed) VIO 50-331/9?O16 01: a. Multiple cases of PCNs used to chango meaning, content, and application of procedures; b. Multiple cases of routine
_
informal delegation and missing signatures; and c. Multiple cases of procedures approved by other than the plant superintendent.
_
e:
.
.i,
.
The team's review of procedures during the inspection found no instances of proceduro change notices being used to change the intent of procedures. The team also noted that all procedures reviewed had boon properly signed and approved.
This violation is closed.
V. Managtment Meetingt X1 Exit Meeting Summary On April 25,1997, the team presented the results of the on site portion of the inspection to membots of licensee management. The licensee acknowledged the findings presented.
The team asked the licensee whether any materials examined during the inspection should be considered propriotary. All proprietary information was returned.
Subsequent to the April 25th meeting, additionalinformation was transmitted to the team.
This information was reviewed by the team from June 9 30,1997, and resulted in a final phone exit on July 1,1997.
_.
_
__
_
_
_ _ _ _ _
.
.
.
.
.
. _ -
.
_ _ _ _ _ - - - - -
-.
,
'
.
,. *.
PARTIAL LIST OF PERSONS CONTACTED Licensee P. Bessette, Manager, Engineering J. Franz, Vice President, Nuclear G. Middlesworth, Plant Manager R. Murrell, Regulatory K. Peveler, Manager, Regulatory HaC C. Upa, Resident inspector M. Jordan, Chief, Projects Branch K. Riemer, Senior Resident inspector M. Ring, Chief, Lead Engineers Branch INSPECTION PROCEDURES USED IP 40500:
Effectiveness of Licensee Controls in Identifying, Resolving, and Preventing Problems IP 93801:
Safety System FunctionalInspection IP 9900:
Licensee Technical Specifications Interpretations ITEMS OPENE0, CLOSED, AND DISCUSSED DJtened 50 331/97006-01a VIO Corrective actions were not adequately taken in response to a previous violation on inadequate procedure reviews 50 331/97006 01b VIO Failure to promptly correct the flawed and potentially nonconforming valve body prior to returning the valve to service 50 331/97006 02 URI More restrictive UFSAR design values were not incorporeted into surveillance testing procedures 50 331/97006 03 VIO Design acceptance criterion as described in the UFSAR were not incorporated into the surveillance acceptance testing criterion for MOVs 50 331/97006 04a VIO Failure to perform a documented independent review and assign a calculation control number 50 331/97006-04b VIO - Failure to follow the requirements of the design verification procedure 50 331/97006-05 URI further review of supporting documentation by the NRC to confirm that reliance on operator actions to preclude a negative NPSH was consistent with the system design basis 50-331/97006-06 URI Pending review of the licensees resolution of the electrical calculation inconsistencies 50 331/97006-07 IFl Pending NRC review of revised NPSH calculation
_ - _ -
_
. _ _ _
. ',
.,
,.i i ITEMS DPENED, CLOSED AND DISCUSSED (cont'd)
60 331/97006 08 URI Pending licensee verification of this assumption and subsequent NRC review 50 331/97006-09 URI Review of calculation modified without proper design control measures 50 331/97006 10 VIO Faisare to verify input the assumption for bonnet temperature in calculation CAL M97 002 60 331/97006 11 URI Pending NRC review of the evaluation by AR 97 967 50 331/9700612a URI Pending NRC review of the resolution of UFSAR and design 50 331/9700612b bosis related inconalstencies C1010.d 50 331/96013 04 IFl Pressure Locking of RHR Shutdown Cooling Manualisolation Valve review licenset's operability evaluation.
50 331/93016 01 VIO a. Multiple cases of PCNs used to change meaning, content, and application of procedures; b. Multiple cases of routine informal delegation and missing signatures; and c. Multiple cases of procedures approved by other than the plant superintendent.
-
-
. _. =...
..
..
_ _ _ _ _ _ _ _
_ _.
_ _ _ _ _ _ _ _ _ _ _
.$
gie s
\\
LIST OF ACRONYMS USED ACP Administrative Control Procedure AOP Abnormal Operating Procedure AR Action Request ASME American Society of idechanical Engineers
.
BTU /HR"F British Thermal Unit per hour per degroo Fahrenholt COLR Core Operating Limits Report DAEC Duane Arnold Energy Center DBD Design Basis Document DC Direct Current DCF Document Change Form EMA Engineering Maintenance Action EOP Emergency Operating Procedure ESW Essential Service Water GE General Electric GL Generic Letter HX Heat Exchanger HPES Human Performance Effectiveness System IFl Inspection Followup Item IST Inservice Testing LCO Limiting Condition for Operation LOCA Loss of Coolant Accident LPCI Low Pressure Coolant Injection MOV Motor Operated Valve NPSH Net Positive Suction Head NRC Nuclear Regulatory Commission NRR Office of Nuclear Reactor Regulation
Operating Instruction PL/TB Pressure Locking and Thermal Binding PMT Post Maintenance Testing PSIA Pounds Per Square Inch Absolute PSIG Pounds Per Square Inch Gauge QA Quality Assurance RHR Residual Heat Removal RHRSW Residual Heat Removal Service Water RR Reactor Recirculation Sll Service Information Letter STP Surveil'ance Test Procedure
.
TS Technical Specification TSI Technical Specification interpretation URI Unresolved item UT Ultrasonic Exarnination VIO Violation
. _ _ _ _