W3P89-3066, Forwards Summary of Info Presented at 890601 Enforcement Conference Re Qualification for Submergence of Cables & Splices
ML20245E805 | |
Person / Time | |
---|---|
Site: | Waterford ![]() |
Issue date: | 06/21/1989 |
From: | Burski R LOUISIANA POWER & LIGHT CO. |
To: | Callan L NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
References | |
W3P89-3066, NUDOCS 8906270457 | |
Download: ML20245E805 (130) | |
Text
{{#Wiki_filter:_. _ _ _
- s. $
LO POWEUISI R &A NA / 317 LIGHT NEWBARONNE STREET ORLEANS, LOUIStANA P. O. BOX 60340 70160 * (504)595-3100 UTIUTIES SYSTEM June 21, 1989 W3P89-3066 ! A4.05 ) QA i l 1 U. S. Nuclear Regulatory Commission I ATTN: Document Control Desk l Washington, D.C 20555 ATTENTION: L. J. Callan ! 1
Subject:
Waterford 3 SES Docket No. 50-382 , License No. NPF-38 I LP&L Report on Submergence Issues and Sumary of Information Presented At The Enforcement Conference on June 1, 1989 The subject report is hereby submitted as a follow-up to the enforcement conference held on June 1, 1989. The report addresses the qualification for submergence of cables and splices. LP&L has concluded the cables are fully qualified based on the test reports in the EQ files. LP&L's analysis also demonstrates splices are qualified for submerged applications based on the available test documentation. As discussed at the June 1,1989 Enforcement Conference, the LP&L basis for the submergence qualification of cable and splices was audited by the NRC in 1983, and the NRC formally accepted the LP&L basis. LP&L believes that.in accordance with regulatory policy, this issue should therefore not be considered for enforcement action. In regards to LP&L's presentation at the enforcement conference, three questions were raised by the NRC which required additional research. LP&L has concluded that the related issues do not adversely impact submergence qualification of cables and splices. The questions and issues are addressed. l in the report in detail where applicabie. Briefly, the questions and LP&L's
- conclusions are the following
1
'W g PNU .
j "AN EQUAL OPPORTUNITY EMPLOYER"
1 W3P89-3066 Page 2 June 21, 1989 i i Testing On Cables The NRC raised the point that extensive testing had been performed , on cables with the same insulation, and the IR values varied ; significantly. LP&L contacted the cognizant utility and reviewed i the details of the testing. LP&L determined the tests were performed to assess installation damage and were not conducted per ; IEEE-383, and thus the results would not be applicable to j demonstrating submergence qualification after exposure to a LOCA, 1 4 l Testing Sequence For Cables and Splices The NRC raised the question whether testing a given cable or splice ; sample for LOCA and then a different, but identical, cable or l splice sample for submergence was a valid methodology. Test results ! for all cables and splices indicate the submergence IR values were i comparable or superior to LOCA IR values. Additionally, Tefzel and j irradiated Rockbestos XLPE cable samples were subjected to LOCA and ] submergence testing in series. The IR values during the submergence J phase were better than the IR values during the LOCA phase, demonstrating the effects of LOCA testing are not enduring and thus are not significantly additive to the effects of the submergence testing. This issue is addressed in Appendix D, Section VII. LP&L CI 261491, IR Values for No. 35 Tape Splices The NRC pointed out the IR values for splice configurations with No. 35 tape only in LP&L CI 261491 appeared to be low values. The IR values are expressed in Meguhms, and the IR values are 10,000,000,000 ohms and 3,000,000,000 ohms, relatively high values for resistance. Nonetheless, splices constructed with No. 35 tape only are not representative of the T95/35 splices installed at Waterford 3, nor has LP&L relied on the test results of the No. 35 tape only splices to confirm submergence qualification. None of the IR values within Table 1 were sufficiently low enough to be , classified as a failed splice. These values were reviewed and found l to be accurate and consistent with the data presented in Appendix E. l
.. . I 1
W3P89-3066 4 Page 3 June 21, 1989 I LP&L believes the report is a thorough and complete analysis which confirms the qualification of cables and splices based on documentation existing in EQ files. We also believe a detailed evaluation of our actions will demonstrate ; LP&L actions on this matter have been proactive, timely, and thorough. l Finally, we perceive our actions have been consistent with, and meet, regulatory requirements and intent. l The documentation deficiencies identified are not safety significant, although i we will pursue vigorously, correction of such deficiencies in an effort to reflect current regulatory guidance on this matter and to enhance our EQ ! program. Please contact me or J. E. Howard should there be any questions regarding this matter. 0)/LJ R. F. Burski, Manager Nuclear Safety & Regulatory Affairs RFB/RJM/pz 1 Enclosure l I cc: E. L. Blake, W. M. Stevenson, F. J. Hebdon, D. L. Wigginton, { R. D. Martin, W-3 NRC Resident Inspector's Office ; 1 i l l , i l
i l LOUISIANA POWER AND LIGHT COMPANY ., i WATERFORD STEAM ELECTRIC STATION, UNIT N0. 3 j Docket 50-382
)
i
SUMMARY
OF INFORMATION PRESENTED AT THE ENFORCEMENT CONFERENCE REGARDING SUBMERGENCE ISSUES I June 1, 1989 l l i 1 i i l 1 l
)
1 l 1 Louisiana ' Power &-Light Company l i Waterford Steam Electric Station, Unit 3 Docket 50-382 Unresolved Item 382/8831-01: -) . Submergence Issues 3 l June 1,.1989 - 1 l l I. INTRODUCTION j Louisiana Power & Light Company (LP&L) received a full power operating license for Waterford Steam Electric Station, Unit No. 3 (Waterford 3) 1 from the NRC on March 16, 1985. At that time, LP&L had implemented a -i comprehensive program to address the environmental qualification (EQ) of electrical equipment. Both LP&L and the NRC (Safety Evaluation Report j. (SER), Supplement 8), had determined that the plant was in compliance j with the requirements. of 10 CFR 50.49. ; 1 An NRC inspection team conducted an inspection at the plant on December J 8-12, 1986 to review again LP&L's EQ program. This inspection' led to- : issuance of Inspection Report 50-382/86-32 on April-16, 1987, in which: ] the NRC identified.several potential violations and open items. - Several-of the issues resulting from the first round.EQ. inspection were the subject of an Enforcement Conference held on October 4, 1988. One issue i involved the potential submergence of ITT Barton . level transmitters. However, this submergence issue was ultimately closed by the NRC without a violation (see Inspection Report .50-382/89-05, dated March 17,1989). The present issue arose subsequent to the first round EQ inspection -i and was identified by LP&L in Condition Identification (CI) Report No. ; 260306, dated December 15, 1988. Specifically, following the Barton transmitter issue, LP&L reviewed other similar equipment installations. As a result of this proactive effort, LP&L found Okonite T95/35 taped splice assemblies on a safety-related instrument (SI-ILT-7145-A) l potentially located below the postulated flooding level inside containment. The CI report was generated because the EQ file did not specifically address qualification for submergence of splices. However, adequate documentation to support submergence qualification was contained , in the file. The matter was subsequently documented as an Unresolved . Item in Inspection Report 50-382/88-31, dated February 9,1989, wherein -' the' NRC noted that LP&L was in the process of responding to.the question of submergence qualification. Following further inspections. conducted from February 13 - 17, 1989, the NRC classified the matter as an Unresolved Item and an " apparent violation" in Inspection Report 50-382/89-05 dated March 17, 1989. t
At the Enforcement Conference of June 1,1989, LP&L provided information to
. resolve this issue. LP&L also separately addressed the issue of qualification -
for submergence of cables. In summary, LP&L has concluded that cables are fully qualified based on the test reports in the EQ files. With respect to the splice issue, LP&L's analyses show that the' subject splices are'also. qualified for submerged applications bcsed on the existing test documentation. Ihis issue involves no more than a documentation deficiency, in that the files did not include an analysis'to show how the test data addressed submergence. LP&L will update the file and stands committed to resolving the splice issue. LP&L further maintains that escalated enforcement action is not warranted for this issue. because. (1) the Staff had previously reviewed and concurred. with LP&L's methodology for submergence qualification of cables and splices, (2)- the conditions lack safety significance in that LP&L's confirmatory analyses show that the cables / splices are qualified for submergence, (3) LP&L 'could not reasonably have done more to address this issue earlier given the evolutionary nature of the submergence concern ~ and (4) LP&L identified the issue and took prompt steps to address and resolve the NRC concern. II. BACKGROUND A. Chronology of Events As part of a self-initiated comprehensive review of EQ equipment status, LP&L identified a cable splice located below the' postulated containment flood elevation. This cable ' serves an environmentally qualified Safety Injection , Sump Level Transmitter, SI-ILT-7145A. LP&L subsequently documented the ! condition in Condition Identification (CI) Report No. 260306 on December 15, ) 1988. The CI report states that the EQ files did not explicitly address submergence of cable splices. On December 15, 1988, LP&L performed a preliminary enviropental qualification engineering evaluation and cc:.:1uded that the affected equipment was operable and qualified for. its intended use. This engineering evaluation was based on successful qualification testing of Ukonite taped splices conducted in a LOCA environment. The LOCA environment was judged'to be more moisture intrusive i than water submergence, given submergence at a depth expected during postulated accident conditions. The evaluation noted the.need for a more -; detailed confirmatory review. In addition, subsequent to this operability ' determination, LP&L undertook a comprehensive effort to identify other Okonite taped splices potentially subject to submergence. 1 As a result of subsequent NRC concerns regarding submergence qualifications. of - j certain Okonite splices and potentially submerged cable, on February 17,1989, ' LP&L entered its N0P-019 process to address the issue (CI Report No. 261491). On February 23, 1989, LP&L prepared and informally provided the NRC with a detailed engineering evaluation supporting the continued operation of the . plant. The evaluation also discussed submergence qualification for EQ Okonite taped splices and electrical field-run cable. LP&L later supplemented that i evaluation.with a more detailed evaluation of cables potentially subject to i submergence.
]
m s In Inspection. Report 50-382/89-05, the Staff formally identified Unresolved Item 382/8831-01 regarding- the submergence qualification of Okonite T95/35 , spliced tape assemblies as an apparent violation of 10 CFR 50.49. This j item follows from the LP&L-identified non-conformance CI No. 260306. In the i Inspection Report, the inspector observed that the test reports in the EQ file for the splices involved LOCA testing on 600 volt Rockbestos Firewall III XLPE and 1000 volt Rockbestos Pyrotrol III instrument and control cable. The ; inspector concluded that LP&L uses other applications of T95/35 tape splice assemblies at Waterford 3 below flood levels on cables manufactured by Samuel i Moore, Anaconda, and ITT Barton (Tefzel). The inspector observed that LP&L i had not analyzed the material compatibilities of these splice configurations to the jacket and insulation material of these other manufacturers. l 1 In response to issues discussed in NRC Staff's March 17, 1989 Inspection Report, several telephone conference calls were held between LP&L and the NRC ) regarding the adequacy of the Waterford 3 submergence qualification analysis i for splices. During the April 17, 1989 conference call, the Staff stated I that LP&L's EQ analysis far cable submergence did not address (1) the chemical composition of the long term immersion test fluid was similar to the chemical I composition of fluid expected to collect in the plant sump subsequent to a l LOCA and (2) whether the duration of the submergence tests were long enough to ; cover the operability times for the associated equipment.
]
Accordingly, LP&L provides below a separate discussion of submergence ; qualification of cable and cable splices. With respect to cable splices, LP&L ' also provides a discussion addressing the inspector's material compatibility concerns. The following discussions supplement analyses already provided to ; the NRC or contained in the EQ file for the subject cables. ' A detailed chronology is provided in Appendix A. B. General Considerations In this document LP&L provides a summary of its comprehensive evaluation of qualification methodologies for the Okonite splices and the cables that could be submerged subsequent to a design basis event. : LP&L's detailed engineering evaluations for cables and splices subject to submergence confirms the cable types and cable splices will perform their design functions and are qualified. LP&L concludes that qualification of cables for submergence is supported based on test information already included in the EQ files. The same type of test information also provides the basis for qualification of cable splices. However, the qualification files for the Okonite taped splices will be supplemented to more explicitly incorporate the analyses of submergence presented in this report.
In addition to this technical conclusion LP&L also believes that enforcement action should not be taken on this issue based on the circumstances in which it has arisen. We believe that alternative methods for submergence j qualification is an issue potentially aftecting a number of operating plants, ! and thus is not unique to Waterford. LP&L has taken a proactive, responsible j approach to be one of the first licensees to address and resolve the issue. l 1herefore, in recognition of the generic nature of the issue, and the fact I that it is an evolving technical matter, the issue should not be placed in the i context of enforcement. Stated another way, the issue of submergence l qualification by test and analyses had evolved subsequent to the November 30, ' 1985 deadline and does not warrant enforcement action under either the modified EQ Enforcement Policy of Generic Letter 88-07 or the general ! Enforcement Policy of 10 C.F.R. Part 2, Appendix C. LP&L believes the NRC should not consider enforcement on this matter because the NRC previously reviewed ard accepted the LP&L basis for submergence qualification. During January 4-6, 1983, the NRC performed an equipment qualification audit for Waterford 3. The NRC specifically reviewed the EQ files for Okonite T95/35 taped splices and 600V power and control cable. The NRC also evaluated the test results and methodologies for submergence qualification of these splices and cables. The questions raised by the NRC, specifically items 17 and 23 of LP&L letter W3P83-0320, are the same issues in substance currently undEr consideration. LP&L via letter W3P83-0320 not only i provided information to demonstrate submergence qualification of the splices j and cables, but also described in detail the methodology for submergence 1 qualification of the splices and cable. A key point is that the NRC reviewed and evaluated the same equipment and j technical issues currently under consideration. The test conclusions and the 1 methodology description were provided to the NRC via LP&L letter W3P83-0320, l dated January 27, 1983. The NRC in the Waterford 3 SSER Supplement 5, section ) 3.11.4.2, discussed the 1983 EQ audit and accepted the LP&L response on the i submergence issue. This letter was also referenced in that SSER Supplement. ) The NRC by virtue of the evaluation of identical equipment and issues and the ] NRC formal acceptance of the LP&L response established a "previously accepted i staff position." LP&L believes sufficient detail and information is available 1 in the LP&L response to reasonably conclude the NRC established a "previously l accepted staff position." LP&L maintains that even if the NRC judged a deficiency existed, pursuant to regulatory policy, a regulation non-compliance did not exist, and thus, regulatory enforcement should not be considered. Under Generic Letter 88-07 terminology, this is not a " deficiency" of which LP&L " clearly knew or should have known. In the past, various details ano questions have been raised amongst LP&L, industry, and the NRC, which demonstrate the requirements for the methodology and documentation for submergence qualification are not entirely definitive, but subject to a degree of interpretation and evolution. LP&L has maintained, and continues to maintain, that qualification was established by full LOCA testing, a high voltage withstand test (hi-pot test), and a long term water immersion test. This position is reasonable and supportable. The NRC has never previously articulated precise guidelines or requirements on either acceptable methodology in this area or the documentation necessary to report submergence qualification. l l
The Seabrook record provides only one example of this lack of clear methodology to address the issue. We acknowledge that the Staff does not presently agree that its Proposed Findings of Fact filed in the Seabrook licensing proceedings are applicable to our current issue. However, as discussed in our January 23, 1989 submittal,theNRCrecommended(SLB concurrence with a similar submergence qualification methodology. Even though a full discussion of the Seabrook record is provided in Appendix B, Seabrook is not relied upon as the basis for LP&L's position. It merely reflects the evolving nature of this issue and is one more indication why LP&L should not clearly have known of any deficiency in this area. Similarly, under the terminology of the general Enforcement Policy, this was a matter outside LP&L's control, at least in the sense it was not avoidable despite LP&L's reasonable efforts to meet EQ requirements as they were. previously understood in the industry. See Enforcement Policy, Section V. A. In addition, the discovery of additional cables and splices that could be : submerged was the result of a comprehensive, proactive effort by LP&L (to ' investigate submergence issues following the Barton transmitter concern) and, i as such, would warrant an exercise of enforcement discretion pursuant to Sections V.G.3 and V.G.4 of the Enforcement Policy. In summary, LP&L stands ready to resolve this matter prospectively as a technical or engineering matter. However, in recognition of LP&L's reasonable efforts to achieve compliance, this matter should not be the subject of escalated enforcement. III. LP&L'S POSITION ON THE ISSUE A. Submergence Qualification of Cables i
- 1. LP&L Position Regarding the Existence of a Violation LP&L maintains that the subject cables have been and are fully qualified i for use in submerged applications. The cables subject to potential .
submergence are identified in Appendix C. As support for our position, ! LP&L provides the following discussion. l The subject cables are qualified based on the test reports that were in I the EQ files. There are: Full LOCA Test (NUREG 0588) High Pot Test (IEEE 383) long Term Water Immersion Test (IPCEA) LP&L's analyses of this test data shows that the cables can survive a LOCA and separate submergence in that critical cable characteristics are not adversely impacted by the relevant test parameters. In addition, the test parameters adequately bound expected submergence conditions at Waterford 3. 1/ Public Service Company of New Hampshire, et. al. (Seabrook Station, Units T and 2), Docket Nos. 50-443 OL-1 and 50-444 OL-1, "NRC Staff Proposed Findings of Fact and Conclusions of Law," filed by Robert G. Perlis (Counsel to the NRC Staff), dated November 26, 1986. 1
- 2. Further Analysis 'of Cable Qualification LP&L performed a' functional analysis of the failure modes' for-currently installed cable subject to submergence. The analysis -
evaluates the critical operating characteristics for cable and ] establishes why LOCA and long term water immersion testing is equivalent. j to submergence testing by showing that those characteristics would not be adversely. impacted during actual submergence. l Appendix 0 provides LP&L's detailed functional analysis. The analysis s concludes that insulation resistance.(IR) is the critical functional 'i factor that must be maintained above a threshold value to ensure that ] equipment will operate properly. The primary environmental. conditions - that could affect-IR are temperature, pressure, and the' chemistry of the
. submerging fluid. LP&L also evaluated effects of radiation.
As discussed further in Appendix D, the subject cables have been successfully qualified by testing or_ similarity analysis:for.LOCA. conditions. The Staff specifically concurred with .the LOCA qualification , of certain cables and generally concurred with LOCA qualification of . ] other cables in NRC Inspection Report 50-382/86-32, dated April 16, 1987. I Therefore, LP&L begins its discussion with the premise that LOCA qualification of cables has been satisfactorily established.- ] In summary, the Appendix D analysis and discussions below demonstrate that 1) IR will likely be higher under the submerged-condition in that sump temperatures are-lower than the LOCA environment, 2) the increase in pressure due to submergence is minimal, and.3) sump. chemistry is similar to LOCA environment chemistry regarding. chemical composition and the , effect on materials,
'd i .It is also noteworthy that the subject cables 'are contained in conduit, j thereby limiting the cable exposure to pressure and chemicals and ;
maximizing the amount of heat transfer that takes place prior to reaching i the cable jacket. 1 The types of cables involved are identified in Table 1. 'The specific 1 cable applications are specified in Appendix D. Each of the cables have the same failure / functional criteria. .Therefore, LP&L's analysis is - generic with regard to IR being the essential functional element. A. specific discussion of the effect of each environmental parameter on IR j is provided below.
- a. Temperature H LOCA tests illustrate that IR varies inversely with l temperature. The peak LOCA qualification temperature at Waterford 3 is 340 F. The peak sump temperature will not exceed this value. 4
-l ]
I l l 1 __.__________________b
u FSAR Figure 6.2-4 illustrates that.for a worst case design i basis event, maximum, containment temperature exceeds. ; anticipated worst case . sump. temperature. During a LOCA,. cables located below the' submergence level will initially. experience. a containment vapor space temperature of approximately 269 F. After the sump fills, however, the now submerged cable jacket. will cool to. an equilibrium temperature consistent with the sump. Eventually,: sump _ temperatures rise abcVe the containment atmosphere temperature and are bounded by LOCA test values'and , analysis temperature; however, temperatures do not return to ' ]; LOCA levels. Therefore, the cooling effect due to the cable being submerged will likely cause an increase in IR, thereby providing increased IR margin. LOCA test results on potentially submerged cables are provided inthecable' submergence' evaluation (AppendixD).whereeach.. cable is addressed separately.- Tables provide IR values'at' different intervals'in the test sequence i.e., prior to aging and pre-LOCA and post-LOCA. The data indicates that the test cable IR values essentially remained unchanged or improved between~the pre-LOCA and post-LOCA values. The most significant. degradation occurred between the pre-aging and post-aging values, but remained within _ acceptable limits. .IR. values were found to vary with temperature during the temperature dwells of. the LOCA test as opposed to any other-f acto rs. ! Quantification of the amount of IR change is speculative; . ! however, we can conclude'that the IR margin of the insulation - will likely increase. As discussed further below, for other - ! environmental properties that could negatively affect the IR margin,- the decrease in IR margin ~would be minimal. , l In summary, LOCA testing 1-. sufficient to demonstrate that the cables are qualified for temperature effects due to ] ' submergence. Separate subanergence tests (at constant temperature) indicate minimal IR change. b) Pressure 3 Prior to submergence, the cable will be exposed to a maximum , LOCA-related containment atmosphere pressure of.44 psig. .The 1 cable LOCA tests adequately bound Waterford 3 design basis event _ pressure conditions. Upon being submerged, the cables q will experience hydrostatic and atmospheric pressure. ! However, the hydrostatic pressure resulting from submergence i will have relatively minimal effect on water absorption by.the cable jacket and insulation. Therefore, qualification for. pressure effects due to LOCA will adequately ' demonstrate submergence qualification. i
)
L _ -_- _ -- - - - - _ - - - - _ - _ - _ _ - - - - - - - - - - - . _ - _
The qualification test reports and-the plotted IR' values versus time of the LOCA tests suggest that pressure l induced moisture absorption has a negligible influence-on IR values. Pre-LOCA and post-LOCA IR values remain ~ essentially, unchanged indicating that no significant moistureiabsorption occurred even.at the
.high gauge pressures obtained ,during the tests. IR values were noted to change at major temperature dwells but increased as temperature and pressure ' decreased.' . If water absorption had occu'rred during these LOCA tests (because of pressure) it is expected that IR values would not.have been able to recover to .the extent that they had during'the test and as indicated by.
the' pre-LOCA and post-LOCA IR values. Similarly, long-term immersion tests demonstrate changes in.IR (before and after'the. test) that were not significant enough to adversely affect. component operation.- The pressure during these LOCA tests-ranged from 66'to 117 psi. This pressure combined with chemical spray was -sufficient to promote water absorption at- a level that exceeds expected sump conditions. c) Chemistry The LOCA tests utilize a conservative combination' of chemicals during 'the chemical spray sequence in that chemical i combinations utilized by these tests bound actual LOCA i conditions at Waterford 3. - LOCA atmospheric contaminants would eventually condense and drain to the containment sump containing the subject cables. The sump chemistry is determined by the chemistry of: the -{ Reactor Coolant and Refueling Water Storage Tank. The ! resulting composition -for Waterford 3 would be approximately 2000 ppm Boron, as boric acid, buffered to a pH. of approximately 7 by trisodium phosphate docecahydrate. There is-negligible effect on sump chemistry as a result of' discharge of the Safety Injection Tanks because of the ;1imited volumes of these units. The LOCA test spray chemicals will therefore 1 generally bound the contaminants in the . sump '(the submerged 4 chemical environment). Moreover, the potentially submerged 1 cables are contained in conduit, which further decreases the ! effect of chemical contaminants in the sump fluid. I LP&L also evaluated the impact of the chemical constituents of the sump not present in LOCA tests. The constituents of the sump are TlTi ppm quantities 1of solid fission products and i iodine, and (2) materials washed from containment surfaces as a result of the containment spray. The exact concentrations'of J materials washed from containment surfaces as a result of the H spray' system could only be determined from an actual' accident. ] The expected' primary constituents would be calcium and silicon j from uncoated surfaces, dust and dirt from pipe racks and cable < trays, small quantities of greases and oils from the surfaces j of rotating equipment, and small quant; ties of metals as a result of corrosion of exposed metal surfaces. j 1 q
LP&L has performed an evaluation assuming 1% of fission products contained in the sump. The radiation effects of these products are bounded by LOCA test irradiation. No chemical interaction with the cable jacket or insulation are expected. With respect to these other contaminants, LP&L has concluded that no adverse effects on the jacket or insulation material are likely. This is based on LP&L's review of manufacturers' data which indicates the cable insulation jacket material to be generally chemically resistant to these contaminants. See Appendix D section VIII for more detailed discussion cn sump chemistry.
- 3. LP&L's Position Regarding Whether Enforcement Action Should Be Taken Under the EQ Enforcement Policy or the Standard Enforcement Policy.
If, contrary to LP&L's position above, the Staff concludes that the finding constitutes a violation, LP&L maintains that for the reasons noted in Section II above, the finding does not warrant enforcement action under Generic Letter 88-07. Moreover, the finding does not warrant enforcement under the standard Enforcement Policy (10 C.F.R. Part 2, Appendix C.)
- 4. LP&L's Position Regarding Safety Significance of the Findings Regarding Cable Submergence
\
LP&L does not believe that the Staff's findings are safety significant in that it has been demonstrated that the cables ] are qualified for submergence use. In summary, considering the j fact that the components have significant IR margin, it is highly unlikely that the cable or the associated equipment will fail during, or subsequent to, a LOCA event. l
- 5. Other Considerations j The plant has been operating with the subject cables below I postulated flood level for several years. The cables are routed through the containment. floor and relocation or replacement would require significant modifications and shutdown time. In addition, it is our understanding that several other plants have qualified potentially submerged cables based on the same type of information as presented by LP&L.
In addition, LPSL acquired the services of two independent cable experts, William A. Thue and Nissen Burnstein, to' review. i its approach to submergence qualification. Both confirmed that l the approach is sound and the methodology valid. In addition, they expressed confidence based on their experience and LP&L's analysis that the cables will perform their intended function. l ,
- - _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - - - - - - - _ _ _ . - - d
TABLE 1 i l CABLES / TYPES SUBJECT TO SUBMERGENCE MANUFACTURER INSULATION JACKET ANACONDA EPR CSPE OR CPE OKONITE EPR CSPE SAMUEL MOORE El CSPE l ROCKBESTOS XLPE CSPE CERR0 XLPE CSPE ROCKBESTOS(TRIAX) XLPE CSPE ,
'I COMBUSTION ENGINEERING QUARTZ SEMI-RIGID !
i (SUPPLIER) STAINLESS STEEL
A q l B. Submergence Qualification of Splices / Material Compatibility of Splices and Cables
- 1. LP&L Position Regarding the Existence of A Violation l l
LP&L maintains that (1)' the subject s ! use in submerged applications and (2)plices are splices the Okonite fully qualified are for , materially compatible with cable applications used at Waterford 3.- l As support for this position, LP&L-provides the following j discussions. Qualification is based on the test data discussed above that was included'in the EQ file. LP&L acknowledges, however, " that documentation in the EQ file on November 30, 1985, did not .) fully discuss the potential for submerged use of the splices. That j analysis will be added to the file. i J
- 2. Further Anaiysis of Splice Submergence Qualification j
- a. Submergence l
LP&L has performed a functional analysis of failure modes for-currently installed splices subject to submergence. The analysis '{ evaluates the critical operating characteristics for splices and -! establishes why LOCA, high-pot, and long-term submergence testing-qualifies the splices, Appendix E provides LP&L's detailed functional analysis for cable ) splices. As you will note, many of the assumptions and methodologies are the same as those for submerged cables discussed above. The use of the same methodology is justified because'the failure mechanisms have been determined to be identical. Specifically, the critical characteristic for a submerged (or non-submerged) cable is IR because the integrity of the i signal / voltage must'be maintained to ensure that the component functions in accordance with its design. The critical characteristic for a cable splice (submerged or non-submerged) is also IR in that a splice offers another pathway for leakage currents , and hence. IR degradation. Given that IR is the critical < characteristic, the LOCA test, high-pot, and long term immersion test is sufficient to qualify a splice for use in potentially ; submerged locations just as a LOCA test, high-pot, and long term j immersion test is satisfactory to qualify cable for use in ; potentially submerged locations. LP&L maintains that in comparing a splice to a cable for analysis > purposes, one can justifiably assume certain conditions. First, the : analysis assumes that the workmanship regarding the splice meets l procedural standards. LP&L believes it is reasonable to assume that workmanship satisfies this standard. Second, the analysis assumes l that whether by mechanical or chemical methods, the splice creates a l seal with the cable, thereby becoming an extension'of the cable jacket. The Staff has not questioned LP&L's procedures for forming Okonite splices and therefore, it is reasonable to base this-analysis on the existence of properly constructed splices. The second assumption relates to the issue of material compatibility, ! which is addressed below. i _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _l
B. Submergence Qualification of Splices / . Material Compatibility of Splices and Cables
- 1. LP&L Position Regarding the Existence of A Violation LP&L maintains that (1) the subject splices are fully qualified for use in' submerged applications and (2) the Okonite splices are materially compatible with cable applications used at Waterford 3.
As support for this position, LP&L provides the .following discussions. Qualification is based on the test data discussed above that was included in the EQ file. LP&L acknowledges, however, that documentation in the EQ file on November 30, 1985, did not fully discuss the potential for submerged use of the splices. That analysis will be added to the file.
- 2. Further Analysis of Splice Submergence Qualification
- a. Submergence LP&L has performed a functional analysis of failure modes for currently installed splices subject to submergence. The analysis evaluates the critic 01 operating ' characteristics for splices and=
establishes' why LOCA, high-pot, and long-term submergence testing qualifies the splices. Appendix E provides LP&L's detailed functional analysis for cable splices. As you will note, many of the assumptions and methodologies are the same as those for submerged cables discussed above. ~The use of the same methodology is justified because the failure mechanisms.have been determined to be identical. Specifically, the critical characteristic for a submerged (or i non-submerged) cable.is IR because the integrity.of the j signal / voltage must be maintained to ensure that the component i functions in accordance with its design. The critical l characteristic for a cable splice (submerged or non-submerged) is ! also IR in that a splice offers another pathway for leakage currents and hence, IR degradation. Given that IR is. the critical , characteristic, the LOCA test, high-pot, and long term immersion ' test is sufficient to qualify a splice for use in potentially submerged locations just as a LOCA test, high-pot, and long term immersion test is satisfactory to qualify cable for use in potentially submerged locations. LP&L maintains that in comparing a splice to a cable for analysis purposes, one can justifiably assume certain conditions. First, the analysis assumes that the workmanship regarding the splice meets procedural standards. LP&L believes it is reasonable to assume that ' workmanship satisfies this standard. Second, the analysis assumes q that whether by mechanical or chemical methods, the splice creates a l seal with the cable, thereby becoming an extension of the cable j jacket. This assumption relates to the issue' of material ' compatibility, which is addressed below. 4
) \
- b. Materials Compatibility Inspection Report No. 89-05 raises the issue of compatibility of the Okonite tape splices to the cable materials utilized at Waterford 3. The splice configuration' at 1 Waterford 3 consist of cables manufactured by Rockbestos, i Samuel Moore, Anaconda, Okonite and tefzel insulated conductors I from Barton transmitters.
LP&L has concluded that material compatibility does not impact the integrity of the splice with respect to moisture intrusion. 4 The tape and cable insulation mold to form a mechanical seal that is not chemically dependent. This mechanical seal is formed by the elastic and self adhesion properties of the insulating tape which produces a moisture-proof barrier. LP&L has also analyzed the present Okonite splice / cable 1 configurations to address any other issue related to the compatibility of the tape to the associated cables. In 3 sunmary, LP&L is unaware of any chemical interactions j between the splice material and the cable jacket / insulation 4 that would adversely affect the cable-to-splice seal. Tefzel ! in particular has been determined to be chemically inert except ) with nitric / sulfuric acids and organic bases at high i concentrations near their boiling point. These conditions J will not exist in the postulated submerged environment. I Appendix E provides further discussion of the potential for j chemical interactions and additional bases for why the l as-installed splice configurations are qualified.
- 3. LP&L's Position Regarding Whether Enforcement Action Should Be Taken Under The EQ Enforcement Policy or the Standard Enforcement Policy If, contrary tc LP&L's position above, the Staff concludes that the finding constitutes a violation, LP&L maintains that for the reasons noted in Section II above, the finding does not warrant enforcement action under Generic Letter 88-07.
Moreover, as further discussed herein, the finding does not i warrant enforcement under the standard Enforcement Policy (10 C.F.R. Part 2, Appendix C.)
- 4. LP&L's Position Regarding Safety Significance Of The Findings Regarding Splice Submergence And Material Compatibility LP&L does not believe that the Staff's findings are safety significant in that the subject splices are qualified for their intended use based on available test documentation. At most, the Unresolved Item with respect to cable splices represents a documentation deficiency with no safety significance.
I _ _ _ _ _ _ _ _ _ _ _ _ I
In addition, LP&L has performed a functional analysis 'and determined that of the sixteen (16) devices involved, eight (8)- are not necessary for long term post-accident conditions. ' With respect:to the remaining eight (8),' LP&L has determined that in each case, failure of' the instrument would not prevent the operator from achieving and maintaining safe shutdown conditions. IV. MITIGATING CIRCUMSTANCES If the NRC disagrees with LP&L and believes that enforcement action is necessary with respect to submergence qualification, LP&L believes that' no civil penalty .is appropriate. As discussed above, this issue lacks safety significance given the analyses that confirm qualification of the cables and cable splices for submerged conditions. In addition, submergence qualification of cables and cable splices is an evolving issue which has been addressed responsibly and comprehensively by LP&L.. It is, therefore, not a matter in which LP&L " clearly should have known" of EQ deficiencies. Moreover, the Staff concurred with LP&L's cable and splice methodology in 1983. (see.the discussions.in Section II B above). In addition, if a civil penalty is contemplated for this issue, full mitigation should be allowed because of the following considerations:
- 1) LP&L identified the issue of potentially submerged cable'and splices related to the Safety Injection Sump Level Transmitter and promptly initiated CI Report No. 260306 in December 15, 1988. LP&L also promptly determined operability and initiated a confirmatory engineering evaluation.
- 2) Upon further review, LP&L identified additional cables and splices potentially subject to submergence. This equipment was appropriately incorporated within the scope of LP&L's operability determinations and detailed confirmatory engineering evaluation.
- 3) Upon further articulation of NRC concerns regarding submergence qualification, LP&L generated CI Report No. 261491.to initiate it's N0P-019 process. LP&L has acted responsibly and conservatively on this matter and kept the NRC fully informed.
- 4) LP&L had clearly demonstrated best efforts to complete all EQ requirements before November 30, 1985.
- 5) The issue required no corrective action under the EQ program, other than the addition of documentation to the EQ file for cable splices. ,
I
- 6) The documentation deficiency is not safety significant. i
- 7) LP&L is committed to resolving this issue.
i a j
APPENDIX A CABLE / SPLICE SUBMERGENCE ISSUE ,
-a
Background
Inspection Report 86-32 Identified potential enforcement / 4/16/87. unresolved item due to lack of- ! qualification documentation for.1'IT I Barton transmitters.(submergence) Inspection Report 88-27 Proposed a Severity, Level IV violation-11-23-88 8827-01 for failure to demonstrate qualification for -submergence' for ITT Barton' level ~ transmhters Non-conformance LP&L identified cable splices for CI #260306- Safety Injection Sump Level Transmitter 12-15-88 as being.potentially submerged; EQ file for Okonite splices did not address qualificatica documentation for submergence
'l LP&L Letter W3B88-2065 Denied violation 8827 01 concerning dated 1-23-89 submergence qualification of ITT Barton level transmitters and addressed submergence qualification 'for Okonite taped splices-NRC Letter dated Withdrew proposed violation on sub-2-6-89 mergence issue for ITT Barton level . 1 transmitters ;
Inspection Report 88-31 NRC inspector questioned the basis for
~
2-9-89 qualification of tne submerged splices. J Since LP&L was in the process of responding to the question of submergence qualification, this was considered an urresolved item (8831-01) Inspection 89-05 Inspector reviewed qualification ! 2-13 thru 2-17-89 documentation for submerged splices (' 0CA test followed by "Hi-Pot" Immersion test and analysis) and noted a l- possible violation based 6n NRC position that the above test reports are not i sufficient to demonstrate qualification ' for submergence Non-conformance Generated to initiate NOP-019 process to CI #261491 address NRC Staff concern for sub-2-17-89 mergence qualification. Initial Engineering Evaluation noted a disagreement in technical judgement and 1 made operability recommendation based on existing qualification documentation ; l _ _ - - _ - .s
Detailed Engineering Evaluation Developed for . equipment identified. 23-89 as'having taped splices potentially subject to submergence, supports-operability.and continued operation.. Conference Call with NRC' .The NRC Staff responded to LP&L Region IV and NRR 3-7-89 request for clarification on their position on qualification testing for submerged. applications. Inspection Report 89-05 Closed the issue on submergence 3-17-89 qualification for Barton transmitters (8632-02).- Identified an " apparent violation" for failure to fully establish qualification for Okonite. taped splices for submerged environment. Detailed Engineering Evaluation LP&L supplemented the prior evaluation (for EQ cables subject to for Okonite' splices subject to submergence) submergence,' addressing in more detail 4-5-89 cables identified as being potentially subject to' submergence. Conference Call with NRC NRC related concerns or questions on Region IV cable evaluation in area of 4-17-89 submergence test' durations and-similarity between tested fluid and expected sump contents-post LOCA. l l l l .;
)
u
APPENDIX B j DISCUSSION OF SUBMERGENCE QUALIFICATION IN SEABROOK PROCEEDING l LP&L has continued to maintain that submergence qualification was established by full LOCA testing followed by a hi;;h voltage withstand test (hi-pot test) { and a long term water imersion test. As described in our January 23, 1989 j i esponse to Inspection Report 50-382/88-27, the hi-pot test consists of ! mechanically stressing the cable, imersion it in water, and subjecting it to j many times its rated voltage while imersed. Cables were also subjected to a j separate long-term immersion test, during which they were imersed in water at 90 C for 26 - 52 weeks. While immersed, the cables were energized at rated voltage and subjected to periodic parametric tests to identify structural degradation. In the Seabrook licensing proceeding (Docket Nos. 50-443 OL-1 and 50-444 OL-1), the New England Coalition Against Nuclear Power (NECNP) attempted to demonstrate that certain cable at that plant had not been specifically tested for submergence for 30 days, and therefore, the cable was not submergence qualified. The Staff stated, however, in its proposed findings of fact:
"24. There is no question that the cable was not tested under submerged conditions for 30 days. Rather, it was thermally aged and irradiated, and then subjected to pressurized steam, high temperature, chemical spray, and humid environment, after which it was submerged briefly [hi-pot test]. Woodward TR.
395-402, 405-06; NECNP EX. 5, Reference 14 at 1. ! The test conditions sumarized a worst case ] environment, combining the anticipated environmental conditions for a loss-of-coolant (LOCA) accident with those for a steam line break. Woodward, Tr. 396, 300-02. l
- 25. No evidence has been aresented before the Board which would indicate that tie Applicant's conclusions are ;
invalid. On the contrary, Staff witness Walker confirmed that environmental qualification, for submergence may be shown either by submercience testing or by a demonstration that the equipment in i question can be operated in a submerged condition. l ....The Board accepts Applicant's determination that the cable has been demonstrated to be qualified for submergence for 30 days." NRC Proposed Findings of Fact and Conclusions of Law, Docket Nos. 50-443 OL-1 and 50-444 OL-1, dated November 26, 1986, at p.12. l l
i Note that the Seabrook analysis involved an immersion test that apparently did not consider chemical differences between the immersion fluid and expected post-LOCA sump fluid. The Staff concurred with that methodology by j publication of its Proposed Findings of Fact. ) Notwithstanding the above Staff position, LP&L has agreed to go beyond the scope of previous requirements and again establish that a LOCA test with a-subsequent hi-pot imersion and a long term immersion test (not required by i the Staff in the Seabrook proceedings) conservatively qualify cables for use < in submerged applications. As discussed in LP&L's engineering analyses and ; presentation to the NRC, we have compared LOCA parameters (e.g., temperature, i humidity, pressure, chemical composition and radiation of the environment) to anticipated submergence parameters and concluded that the critical functional 4 parameter, insulation resistance (IR), would be maintained above required i levels and therefore the cables would be able to perform intended design ! functions. ]
'l I )
l 2 i
I APPENDIX C j i EQ CABLE DESCRIPTION , Cable B/M No. Cable Manufacturer Cable Description 025-09 Three conductor (600V) Power Cable - #8 AWG l ANACONDA Ins. - 45 mils Ethylene Propylene Rubber i (EPR) with 15 mil l Chlorosulphonated ', Polyethylene (CSPE) J jacket i Jacket - 60 mils CSPE overall i OK0 NITE Ins. - 60 mils EPR with 25 . mil Hypalon (CSPE) 'q jacket Jacket - 80 mils Hypalon (CSPE)overall , l ROCKBESTOS Ins. - 45 mils Cross-linked I Polyethylene (XLPE) Polyolefin with 15 mil CSPE jacket Jacket - 60 mils (CSPE) overall
) -_ua__._________--
s
-j il APPENDIX C (Cont'd) ')
i EQ' CABLE DESCRIPTION
\
~ Cable B/M No. Cable Manufacturer Cable Description - l i 4 050-03 Two conductor (600V) Non-shielded contral cable >
- #14 AWG ANACONDA Ins. - 30 mils EPR (flame ;
retardant) . Jacket - 45 mils chlorinated { polyethylene (CPE) -1 (f1ame retardant) overall OKONITE Ins. - 30 mils EPR with 15 l mil Neoprene jacket-Jacket - 45 mils Hypalon 4 (CSPE) overall j j ROCKBESTOS Ins. - 30 mils XLPE (Polyolefin) (flame retardant) Jacket - 45 mils Hypalon j (CSPE) overall l I SAMUEL MOORE Ins. - 30 mils EPR (flame J retardant) Jacket - 45 mils Hypalon .. (CSPE) overall l i I _ _ _ _ _ - _ - _ _ _ _ _ _ _ - . _ _ . i
APPENDIX C (Cont'd) EQ CABLE DESCRIPTION 1 l Cable B/M No. Cable Manufacturer Cable Description D50-04 Three conductor (600V) Non-shielded control cable
- #14 AWG ANACONDA Ins. - 30 mils EPR (flame retardant)
Jacket - 45 mils CPE (flame retardant) OKONITE Ins. - 30 mils EPR with 15 mils Neoprene jacket Jacket - 45 mils Hypalon
-(CSPE) overall ROCKBESTOS Ins. - 30 mils XLPE (Polyolefin) (flame retardant)
Jacket - 45 mils Hypalon (CSPE) overall SAMUEL MOORE Ins. - 30 mils EPR (flame retardant) Jacket - 45 mils (CSPE) overall f l l
APPENDIX C (Cont'd) i EQ CABLE DESCRIPTION ! Cable B/M No. Cable Manufacturer Cable Description 050-05 Five conductor (600V) l Non-shielded control cable
- #14 AWG ANACONDA Ins. - 30 mils EPR (flame .l retardant)
Jacket - 45 mils CPE (flame retardant) .overall 0K0 NITE Ins. - 30 mils EPR with 15 j mils Neoprene jacket > Jacket - 6v mils Hypalon (CSPE) overall SAMUEL MOORE Ins. - 30 mils EPR (flame retardant) Jacket - 45 mils (CSPE) overall w-_-____.---_-__
1 APPENDIX C j (Cont'd) EQ CABLE DESCRIPTION I Cable B/M No. Cable Manufacturer Cable Description I l D50-07 Nine conductor (600V) , Non-shielded control cable l
- #14 AWG )
ANACONDA Ins. - 30 mils EPR (flame i retardant) i Jacket - 60 mils CPE.(flame I retardant) overall {
)
OKONITE Ins. - 30 mils EPR with 15 -li mil Neoprene jacket Jacket - 60 mils Hypalon-(CSPE) overall SAMUEL MOORE Ins. - 30 mils EPR (flame retardant) Jacket - 60 mils (CSPE) overall I
(_ APPENDIX C (Cont'd) EQ CABLE DESCRIPTION Cable B/M No. Cable Manufacturer Cable Description ! 071-01 One twisted pair thermocouple extension wire i (Chromel-Constantan)
- #16 AWG ANACONDA Ins. - 25 mils EPR (flame resistant) ;
Jacket - 45 mils CPE overall SAMUEL MOORE Ins. - 30 mils EPR (flame : resistant) i Jacket - 45 mils (CSPE) overall CERR0 Ins. - 25 & 30 mils XLPE (flame resistant) Jacket - 45 mils Hypalon (CSPE) overall l l l _--___----.m___ _ _ _ _ _ _ _
APPENDIX C (Cont'd) EQ CABLE DESCRIPTION Cable B/M ho. Cable Manufacturer Cable Description 082-01 OKONITE Single conductor (600V) Non-shielded power cable 250 MCM Ins. - 65 mils EPR Jacket - 65 mils CSPE overall D83-01 Shielded two conductor (300V) instrumentation cable
- #14 AWG ANACONDA Ins. - 30 mils EPR Jacket - 45 mils CPE overall ROCKBESTOS Ins. - 30 mils XLPE (Polyolefin) (flame retardant)
Jacket - 45 mils (CSPE) overall SAMUEL MOORE Ins. - 30 mils EPR with or without 15 mil CSPE jacket Jacket - 45 mil (CSPE) overall
APPENDIX C (Cont' d ) EQ CABLE DESCRIPTION Cable B/M No. Cable Description Cable Manufacturer D83-03 Shielded four conductor Samuel Moore (300V) instrumentation cable - #14 AWG Ins. - 30 mils EPR with or without 15 mil CPE
-jacket Jacket - 45 or 60 mils CPE overall D99-82 Quadrax Cable Rockbestos Ins. - Cross-linked Supplied By Polyolefin Combustion Engineering Jacket - Hypalon D99-83 Mineral insulated (MI) Nuclear instrument Triaxial cable encased cable and sealed in metal Supplied by outer jacket Combustion Engineering l
APPENDIX D i 1 1 4 I i l l v l WATERFORD 3 CABLE SUBMERGENCE EVALUATION I i I Prepared by: Martin L. Raines May, 1989
TABLE OF CONTENTS l SECTION TITLE ' l I Introduction .j II Anaconda Cable III Rockbestos & Cerro IV Samuel Moore V Okonite VI Tefzel i VII Comparison of Submergence Test Results to LOCA Final Temperature Dwell VIII LOCA Sump Chemistry IX Justification for Water Immersion Tests on Unaged Insulation X Operability of Cables Under LOCA
-Conditions XI Radiation Exposure of Cables Submerged in Containment XII Industry Experience with Submerged Cables XIII Conclusion ATTACHMENTS TITLE 1 Anaconda LOCA Test Results 2 Rockbestos & Cerro LOCA Test Results 3 Samuel Moore LOCA Test Results 4 Okonite LOCA Test Results 5 Tefzel LOCA Test Results 6 Cable Water Immersion Test i
Results l l 7 Industry Experience with l Submerged Cables l
I. INTRODUCTION The purpose of tLis evaluation is to adequately demonstrate that cables which are subject to submergence are capable of perfonning their intended safety related function after full exposure to LOCA conditions followed by subsequent submergence conditions for a minimum of 120 days plus 10% margin. In order to adequately demonstrate that a cable is qualified for long term submergence after full exposure to a LOCA, two questions must satisfactorily be i answered:
- 1) Af ter 40 years of service does exposure to a LOCA degrade the insuleting characteristics of a cable to such a degree that any l further degradation after submergence may be reason to doubt the continued operability of that cable?
l
- 2) Does the cable insulation in a submerged condition continue to degrade to such a degree that the operability after 132 days is in serious question?
In order to adequately address these questions, there are three cable insulating parameters that are used to evaluete the stated concerns and are frequently used in determinat.ing cable insulation performance characteristics. These parameters are defined as: IR = INSULATION RESISTANCE. This is a measure of an insulating material's capacity to maintain its resistance to current flow to the outside medium such as water, cable trays and conduit and is measured in Ohms. Damaged insulation or insulation which has had significant moisture intrusion or diffWn will result ir, low or i zero values in Ohms. Temperature will no cause a variation in IR ) values. ' SIC = SPECIFIC INDUCTIVE CAPACITY. This value is a measure of an insulating material's ability to store a charge and has a significant bearing on insulation moisture resistance. Cable I l insulation can be considered as a capacitor since it is a dielectric between two or more conducting surfaces and the introduction of water into this dielectric will significantly effect the capacitance resulting in a change in SIC. The SIC of a vacuum is 1, therefore, the greater the SIC, the better the . dielectric properties of the insulating material. Lower values of ' SIC indicate moisture within the dielectric. PF = POWER FACTOR. This is a measure of the power losses that occur in a cable as the result of the storage of energy in the cable insulation and is sensitive to the moisture content of the insulation. The higher the PF the greater the moisture content of the insulation which must nn+ significantly increase with water immersion time. A PF value of 2% or less is typically considered acceptable.
- - - _ - - - - - - - - - _ _ _ - - - - - - - _ - - - - _ - - - - - - - - - - - - - - - - . - - - . - - - - l
This evaluation has considered the effects of submergence on the cables j supplied by the following manufacturers: l Anaconda Rockbestos and Cerro Samuel Moore Okonite Okozel, which is manufactured by Okonite, is their trade name for Tefzel, ; which was also evaluated for post-LOCA submergence effects. Tefzel is used on the Barton transmitter lead wires and also subject to potential submergence at the splice. An evaluation for these splices is performed in Appendix E. , 1 The Rockbestos triax and mineral insulated cable associated with the excore ] detector system does not require submergence qualification per this evaluation 4 since qualification of these cables was performed by type testing for LOCA ) followed by submergence. Additionally, the evaluation for Rockbestos is ! intended to address any concerns for the triax cable which is constructed of 1 XLPE insulation and a CSPE (hypalon) jacket. ) i i 1 l i I 1 l _2
.__-__-__a
II. ANACONDA The Anaconda Report No. F-C4350-2 dated July, 1976, performed environmental qualification testing on 5 samples of low voltage power cable, 2 samples of low voltage control cable and 4 samples of low voltage instrumentation cable. For this evaluation a total of 6 representative cable samples were evaluated. The cable samples under consideration are: Sample # Description 18.32 Low Voltage Power Cable 1/C #12 AWG 30 mils FREP 18.35 Low Voltage Power Cable 1/C #14 AWG 30 mils FREP 20.12 Control Cable 7/C #12 30 mils FREP Asbestos-Mylar Tape, 60 mils CPE (hypalon) Jacket 20.16 Same as Sample 20.12 20.25 Instrumentation Cable 2/C #16 AWG 25 mils FREP, Silicone Glass Tape, Tinned Copper Drain Wire, Aluminum / Mylar Tape, 30 mils CPE (hypalon) Jacket 20.26 Same as Samp1r 20.25 All of the cable samples were a minimum 10.5 foot length and aged for 7 days at 150*C to simulate 40 years at 90 C followed by irradiation to 200 MRADS. The cables were then subjected to a 30 day LOCA test while mounted on a mandrel in which a peak temperature and pressure of 346 F/113 psig was obtained and held for 8 hours. After four days the temperature and pressure was maintained at 212 F/0 to 5 psig for the remainder of the test. The chemical spray consisted of 3000 ppm boron as boric acid, 0.064 molar sodium , thiosulfate with the pH = 9 to 11 ntrolled with sodium hydroxide. The spray ! rate was maintained at 0.15 gpm/ft ' The chemical spray commenced at the start of the test and remained for the - entire 30 day duration. IR measurements were taken periodically during the LOCA test with 500VDC applied for one minute upon completion of the LOCA test, the cables were removed from the mandrel, straightened and recoiled around a mandrel 40X cable diameter and immersed in a tank of tap water at room tempera ture. The cables were then subjected to a high potential withstand test of 80V/ mil for 5 minutes. The results of the insulation resistance tests in KMegohms perfonned on the representative sample of the tested cables are: SAMPLE # 18.32 18.35 20.12 20.16 20.25 20.26 Prior to Aging 4000 6000 1300 1200 8200 4000 After Aging 22 9.4 2.4 2.6 7.6 10 Post-LOCA 29 3.0 3.0 3.5 8.8 8.6
I l J These IR values were obtained at 500 VDC for one minute. It should be noted ! that thermal aging did result in a significant decrease in IR values; however, l the 30 day LOCA test did not result in a significant impact, 2 samples I experienced a decrease and 4 samples experienced an increase. It can be concluded that the LOCA test did not degrade the cable insulating characteristics through moisture intrusion or diffusion and will not degrade the electrical characteristics of Anaconda cable to a point where operability of the cable will be of concern after exposure to a LOCA. The results of the IR measurements taken during the LOCA test indicate that the IR values vary inversely with the temperature of the chamber (increase with decreasing temperature) as opposed to being strongly influenced by the presence of the chemical spray, since post aging and post LOCA test IR values were essentially unchanged. See Attachment 1 for plots of the test results. The conclusion that the chemical spray under pressure has no significant , impact is based from the fact that as pessure and temperature were lowered simultaneously, the measured IR values increased and responded rapidly due to the immediate loss of imparted energy to the cable insulation. If there had 1 been significant deep moisture penetration, IR values would have recovered in a lagging fashion to much lower values, if any significant recovery was to occur. A separate unaged sample (unrelated to the LOCA test, which can be found in Anaconda document #8211211) of a 1/C #12 AWG Anaconda conductor with 30 mils of FREP insulation was subjected to a 26 week water immersion test at 90 C with 600 VAC continuously applied. The reported values for SIC, PF and IR at the start and conclusion of the test are as follows: I IR(Megohms per l SIC 1000 ft.) P_F at 40V/ mil Start of Imersion 2.43 0.59% 4050 { i End of Immersion 2.71 0.43% 3303 j Lowest Value Obtained 2.37 0.43% 3303 Highest Value Obtained 2.71 0.62% 4697 The stability of the SIC, PF and IR values indicate that Anaconda is l insensitive to moisture absorption. See Attachment 6 for the water immersior. IR test results. The abser:ce of chemicals in the immersion water is not considered significant since the LOCA test has demonstrated that these chemicals do not attack the cable insulation. Since the Anaconda cable passed the LOCA with sufficient margin and no significant loss of insulating properties, the 26 week immersion test at 90 C demonstrated submerged cables will not be subjected to continual degradation. This provides a high degree of assurance that cable which has been exposed to a LOCA can operate satisfactorily under conditions of subsequent submergence. It is also concluded that Anaconda cable is capable of withstanding a LOCA followed by subsequent submergence for a period in excess of the Waterford 3 120 day post accident requirement plus 10% margin. III. ROCKBESTOS and CERR0 Thermocouple cable supplied by Cerro is qualified by the Rockbestos report QR-5804. Qualification is based on similarity because:
- 1. From 1974 to 1983, Cerro Wire and Cable and Rockbestos were divisions of the Cerro Corporation. Purchase orders for Cerro Wire and Cable were executed during this time frame for Waterford 3.
Today they have merged into one company known as Cerock.
- 2. The purchase orders to Cerro Wire and Cable identify the cable insulation as KXL-760 and the jacket material as hypalon.
- 3. Thermocouple cable with KXL-760 insulation and hypalon jacket was qualified in Report QR-5804.
Rockbestos Report No. QR-5804 performed environmental qualification testing on a variety of KXL-760D chemically cured cross-linked polyethylene (XLPE) control, instrument and power cables. Similarly, report No. QR-5805 performed environmental qualification testing on KXL-760G irradiated XLPE in almost an identical manner. The cable samples selected for evaluation are summarized as follows: QR-5804 Chemically XLPE GROUP DESCRIPTION A 7/C #14AWG control cable 0.030 inch KXL-7600 chemically XLPE insulation Nomex Binder Tape 0.045 inch Neoprene jacket B Same as Sample A except with hypalon jacket i I E 1/C #6AWG power cable ! 0.045 inch KXL-7600 chemically XLPE insulation ! QR-5805 Irradiated XLPE j The three tested groups were the same as in QR-5804 except that KXL-760G irradiated XLPE was used. In both qualification reports the same number of samples were tested for each of the described groups as summarized below: GROUP SAMPLE NO. DESIGNATIONS TOTAL NO. OF SAMPLES i A Al A2 A3 A4 A5 A6 6 B B1 B2 B3 B4 4 I E El E2 E3 E4 4
I i The aging performed was to simulate 40 years at 90 C with thermal aging ; performed prior to irradiation: i SAMPLE # THERMAL AGING IRRADIATION Al 168 hours @ 121"C 200 MRADS A2 168 hours @ 121 C 200 MRADS A3 941 hours @ 150 C plus 12 hours @ 148 C 200 MRADS l A4 941 hours @ 150 C plus 12 hours @ 148 C 200 MRADS A5 UNAGED 200 MRADS 1 A6 UNAGED 200 MRADS j B1 941 hours @ 150 C plus 12 hours @ 148 C 200 MRADS j B2 941 hours @ 150 C plus 12 hours @ 148'C 200 MRADS i B3 UNAGED 200 MRADS 1 B4 UNAGED 200 MRADS l El 941 hours @ 150 C plus 12 hours @ 148 C 200 MRADS 1 E2 941 hours @ 150 C plus-12 hours @ 148 C 200 MRADS l E3 UNAGED 200 MRADS E4 UNAGED 200 MRADS ;
.. \
The thermal and irradiation aging performed in QR-5805 was the same except that 909.5 hours @ 150 C was performed in place of the designated 941 hours @ 150 C plus 12 hours @ 148 C. The 909.5 hours at 150 C simulates in excess of 40 years at 90 C. Samples A1, A3, AS, B1, B4, El and E3 were selected for LOCA testing in QR-5804 whereas the remaining samples were kept for spares. The test samples were arrcnged in loose coils on a' perforated steel grating within the LOCA chamber approximately one foot from the bottom. The 600VAC energized samples were subjected to a 105 day LOCA test in which , two initial transients were obtained at 340 F. The peak pressure obtained { during the test was 107 psig with the minimum temperature and pressure 1 maintained at 235 F @ 3 psig for 101 days. When the maximum temperature of i the first transient was obtained the chemical spray was started and continued l for 24 hours. The chemical spray consisted of 3000 ppm boric acid, 0.064 J molar sodium thiosulfate with a pH = 10.5 with sodium hydroxide. The flow rate was adjusted to 1.4 gpm. See Attachment 2 for the plot of the IR test values. Upon completion of the LOCA test, the samples were wrapped around a 40X mandrel, then removed and placed in room temperature tap water and tested 1 fcr IR at 80V/ mil after the designated soak times.
]
The results of the IR testing performed at 500VDC for 1 minute in QR-5804 for ; the chemically XLPE is as follows in MEG 0HMS - 1000 ft.: SAMPLE # PRIOR TO AGING PRE-LOCA POST-LOCA SOAK TIME (hrs) Al 36,000 1200 3,900 18 A2 37,000 940 NOT TESTED ; A3 26,000 2200 10,000 18 A4 23,000 2000 NOT TESTED A5 27,000 600 7,000 18 A6 22,000 620 NOT TESTED B1 21,000 1700 7,600 6 B2 23,000 1300 NOT TESTED B3 18,000 450 NOT TESTED B4 23,000 500 5,600 6 El 37,000 940 5,600 18 E2 37,000 960 NOT TESTED E3 38,000 440 2,640 18 E4 35,000 440 NOT TESTED Samples A2, A3, A6, B1, B4, El and E4 were selected for LOCA testing in QR-5805 with the remaining samples kept for. spares. The. test was conducted in the same fashion as discussed in QR-5804 except that the samples were mounted on a mandrel, the total test- duration was'111 days, the puak pressure was 117 psig, and the flow rate was 1.6 gpm. See Attachment 2 for plots of .the IR test results. Upon completion of the LOCA test,. the samples were removed from the mandrel straightened and rewrapped around a 40X mandrel, removed and placed in room temperature tap water. The samples were tested for IR at 80V/ mil for 5 minutes after a 24 hour soak time. - The results of the IR testing performed at 500VDC for .1 minute in QR-5805'for the irradiated XLPE is.as follows in MEG 0HMS - 1000 ft.: SAMPLE # PRIOR TO AGING PRE-LOCA POST-LOCA Al 124,000 2100 NOT. TESTED. A2 '70,500 2600 50,000 A3 331,000 5000 25,000 A4 207,000 3000 NOT TESTED-A5 99,400 2800 NOT TESTED A6 124,000 1300 100,000-81 248,000 3000 24,000 B2 331,000 1900 NOT TESTED'. B3 41,000 1700 NOT TESTED B4 82,000 2200 90,000 El 49,700 1900 30,000 E2 62,100 1200 NOT TESTED E3 145,000 1200 NOT TESTED E4 116,000 1300 56,000 The results of these two tests demonstrate that the IR values of the tested cables are not adversely affected by the long term LOCA exposure. Although the test results do indicate that IR varies inversely with the chamber temperature which would suggest that during exposure to a LOCA environment temperature plays the predominate role in IR variatins and not the presence of chemical spray. This' position' and conclus6 h ei same as addressed for the Anaconda cables. At the conclusion of a FI. ev/uure the cables' will not be significantly degraded. Three separate and unaged samples (unrelated to the'LOCA tests) of a 1/C'
#12AWG conductor with 30 mils of chemically XLPE and three samples of the same size conductor with irradiated XLPE were subjected to a 12 month 90 C water immersion test while constantly energized with 600V and monitored.for SIC ~and PF, No IR reading were taken during this series of. immersion' testing. The results of the water immersion test were as follows:
CHEMICALLY XLPE IRRADIATED XLPE SIC PF SIC PF START OF IMMERSION E7 T4 2.6 T0 END OF IMMERSION 4.1 1.9 3.0 0.9 LOWEST VALUE OBTAINED 2.7 0.7 2.6 0.8 HIGHEST VALUE OBTAINED 4.1 1.9 3.0 1.1 l- AFTER 5 MONTHS 3.6 0.8 2.8 0.9
The' stability of the SIC and PF values indicate that the chemically and irradiated XLPE is insensitive to moisture intrusion. The absence of-chemicals in the imersion water is not considered significant since the LOCA tests have demonstrated these~ chemicals do not attack the cable insulation. Since the Rockbestos cable passed the LOCA tests with more than sufficient margin and the water immersion test demonstrates that no significant degradation will occur, it can be concluded that the cable is capable of withstanding a LOCA followed by subsequent submersion for a period of 120 days plus 10% margin. Rockbestos Report QR-1E09, Rev.1, dated June 30, 1981, provided information - regarding' submergence testing on a 10 foot sample of 1/C #12 AWG insulated with 30 mils:of irradiated XLPE which was subjected to and passed the IEEE-323-1974 and IEEE-383-1974 LOCA qualification requirements. The sample
- onductor was subjected to 200 MRADS of gamma radiation and. aging of 850 hours at 150 C to simulate 40 years at 90 C prior to the LOCA test.
Following the LOCA exposure the sample was immersed for 8 weeks at 200 F and 0 psig in 'a chemical bath which consisted of 1800 ppm boric acid, 50 ppm hydrazine with the pH = 7 to 7.5 controlled ~with trisodium phosphate. While submerged, IR readings were taken to determine the effects of the chemical bath on the cable insulation with the following results: ELAPSE TIME (Weeks) MEG 0HMS MEG 0HMS - 1000 feet 1 12,000 120 2 15,000 150 3 17,000 170 4 17,000 170' 5 18,000 180 6 14,500 145 7 18,400 184 8 15,000 150-Over the 8 week submergence test, the range of the IR values was 120 to 184 Megohms - 1000 ft. with an average reading of 158 Megohms - 1000 ft. . See Attachment 6 for a plot of the IR test results. If these values are compared 1-with sample B1 of Appendix 2 during the final temperature dwell after the-fourth day of testing where an average temperature and pressure of 225 F/3psig was maintained the following results ar e obtained. . QR-1809 SAMPLE QR-5805 B1 SAMPLE l ! Tempera ture 200 F 225 F Pressure O psig 3'psig High IR (Megohms -'1000 ft.) 184 24 Low IR-(Megohms - 1000 ft.) 120 5.25 Average IR (Megohms - 1000 ft.)- 158 10.7-At 200*F (Appdx 2, IR vs. Temp) - 27-
= _ _ _ ____-_- _ _ _ _
From the Appendix 2 IR vs. Temperature for the aged B1 sample, the expected IR at 200 F would be 27 Megohms - 1000 ft. These values demonstrate that ; submerged Rockbestos cable has better performance characteristics than cable i under LOCA conditions for approximately the same conditions. The test results of QR-1809 when compared to the '5ults and conclusions of QR-5805 demonstrate and provide additional assuran- that Rockbestos cable is capable of withstanding a LOCA followed by submergence and that the conclusions obtained q from the other qualification repcrts with regards to the approach and { methodology are valid. ) i 1 i i l I 4 l
-9_
IV. SAMUEL MOORE The qualification report for Samuel Moore cable, dated June 1978, subjected 9 samples of ethylene proplyene insulated ccbles to a complete environmental qualification test program of which 5 representative cable samples are discussed in this evaluation. The cable samples under consideration are described below along with the aging performed to simulate 37 years at 90 C for sample 10. Sample # Description Thermal Aging Radiation 9 2/C #16AWG 30 mils FR-EPDM 7 days @ 150 C 200 MRADS 16 Gauge drain & shield plus 45 mils hypalon jacket 7 days @ 121 C 10 2/C #16AWG 30 mils FR-EPDM 7 days @ 163 C 200 MRADS 16 Gauge drain & shield plus 45 mils hypalon jacket. 7 days @ 121 C 12 2/C #10AWG 30 mils FR-EPDM 7 days @ 121 C 200 MRADS 15 mils hypalon jacket 10 Gauge drain & shield 45 mils hypalon jacket 14 2/C #16AWG 20 mils EPDM 7 days @ 121'C 200 MRADS 10 mils hypalon jacket 16 Gauge drain & shield 45 mils hypalon jacket 16 1/C #16AWG 30 mils FR-EPDM 7 days @ 150 C 50 MRADS The 35 foot mandrel mounted cables were subjected to a 30 day LOCA which consisted of 2 initial 340 F/105 psig transients each lasting three hours with deviations in the temperature and pressure never less than 200 F/10 psig. During the LOCA tests, the cables were energized with 600VAC and 0.5 amps. The chemical spray consisted of 3000 ppm boric acid, 0.064 molar sodium thiosulfate with the pH = 9 to 11 controlled with sodium hydroxide. The spray was applied at a rate of 0.15 gpm/fta. The chemical spray commenced at 20 seconds into the test and remained throughout the 30 day exposure. At the conclusion of the LOCA test, the cables were removed from the mandrel straightened and recoiled around a 40X mandrel, Per IEEE-383, the cable and mandrel were immersed in tap water and subjected to a 80V/ mil withstand test for 5 minutes. The results of the IR measurements taken at 500VDC during the LOCA test ! indicate that the IR values vary inversely with the temperature of the chamber (increase with decreasing temperature) as opposed to being strongly influenced by the presence of the chemical spray. Tnis position and conclusion is the , same as addressed for the Anaconda cable. See Attachment 3 for a graph of the IR test results of a representative sample. I i
The results of the insulation resistance. tests in KMEG0HMS performed on the representative sample of the tested cables are: SAMPLE # 9 10 12 14 16 CONDUCTOR 1 2 1 - 2 1- 2 1- 2 ~T-PRIOR TO AGING 1400 1000 1600 1200- 260 260 210 20 0 5000 AFTER AGING 500 500 500 500 500 500 500 500 9000 POST LOCA 0.440 1.1 340 430- 190 190 310 150 0.019. These IR values were obtained at 500VDC applied for one minute. It should be noted that the thermal and irradiation aging did not result in a significant impact on-the_IR values obtained. In two of the samples the IR values decreased whereas in three of the samples the IR values increased. From the obtained results it can be concluded that the impact'of aging and subsequent LOCA testing will not degrade the' electrical characteristics of ! Samuel Moore cable to a degree where the operability of the cable will be of ' concern. l -A separate uniged sample (unrelated to the LOCA test) of 1/C #16AWG FR-EPDM insulated Samuel Moore conductor was subjected to a 30 week immersion test in 90 C water. The values for SIC and PF at the start and conclusion of the test are as follows: SIC PF at 40V/ mil PF at 80V/ mil J START OF IMMERSION E3 1.2 1.2 END OF IMMERSION 2.5 1.3 1.4 LOWEST VALUE OBTAINED 2. 2 0.68 0.75 HIGHEST VALUE OBTAINED 2.5 1.5 1.6 The stability of the SIC and PF values indicate that Samuel Moore cable is insensitive to moisture intrusion. Attachment 6 contains a 52 week water immersion test performed on another Samuel Moore FR-EPDM cable sample which provided IR results for comparison purposes to other tested cables. The~ test results indicate that no significant IR degradation occurred over the required Waterford 3132 day post accident operability requirement. The absence of -hemicals in the immersion water is not considered significant since LOCA isting has demonstrated that these chemicals do not attack the cable nsulation. Since the tested Samuel Moore cable passed the LOCA test with sufficient margin and the separate 30 and 52 week water immersion test at 90 C - demonstrated that no insulation degradation will occur if subjected to submergence, it can be concluded that the Samuel Moore cable is ' capable of-withstanding a _LOCA followed by subsequent submergence for a period in excess of the 120 day post acciJent requirement plus 10% margin.
1 l I l V. OKONITE
]
The Okonite Report No. NQRN-IA, Rev. 5 dated October 24, 1988, performed environmental qualification testing on two groups of Okonite cable. Each group had two samples tested, one which was fully aged prior to LOCA and one ! which was unaged for a total of four cables in the test program. Cable group A was a 600 volt #12 AWG insulated with 0.030 inch Okonite EPR insulation with a 90 C rating and 130 C overload rating.. Cable group B was a 2KV #6 AWG insulated with 0.055 inch Okonite EPR insulation and a 0.030 inch Okolon chlorosulfonated polyethylene jacket. The two Cable A samples were 25 feet-in ! length and the two Cable B camples were 22 feet in length. The aged sample from each group was first thermally aged for 21 days at 150 C to simulate 60 years at 135 F followed by exposure to 200 MRADS of i rradiati on. All four cable samples were subjected to a 130 day LOCA test while mounted on a mandrel in which a peak temperature and pressure of i 345 F/110 psig minimum were obtained in two initial transients. During the LOCA test, the tested cables were energized with 600VAC and 18 AMPS. A chemical spray was used during the first 30 days of the test followed by -team and water spray throughout the remaining 100 days. The chemical spray , const;ted of 3000 ppm boric acid (0.28 molar), 0.064 sodium thiosulfate and pH l
= 10.5 controlled with sodium hydroxide. IR measurements were taken periodically during the LOCA test by applying 500 VDC for one minute. See Attachment 4 for a plot of the IR test results.
4 After 30 and 130 days into the test program, the cable samples were removed ! from the LOCA chamber uncoiled from the mandrel, straightened and recoiled I around a 40X mandrel and immersed in tap water. The samples were then subjected to an 80V/ mil high potential withstand test for 5 minutes. The pre and post-LOCA insulation resistance results in Megohms - 1000 ft. obtained , were: GROUP A GROUP B Unaged Aged Unaged Aged Pre-LOCA 2,300 3,500 22,000 18,000. i Post-LOCA 3,300 2,900 3,300 3,100 This information obtained for the pre-aad post-LOCA IR measurements indicates that regardless of whether the cable is aged or not, the IR electrical characteristics are not significantly affected and that new and aged cable will essentially perform the same. Therefore, for the purposes of post-LOCA submergence testing and demonstrating operability, aging is not a significant factor. This can be attributed to the fact that during the cable l manufacturing process, the curing of the insulation is typically performed by heat or radiation. l f. ______-_____A
The~ IR values obtained from the tested cable indicates that the ' temperature of , the cable insulation is more of a significant factor in the variation of the insulation resistance than to the other environmental parameters around the cable insulation. This is attributed to the IR plot of the aged and unaged Group A cabic. which did not-contain a jacket. The IR values did not change significantly at the start and end of the LOCA test; however, significant variations did occur with major temperature variations. This. data strongly-suggests that IR values are more temperature dependent, increasing rapidly with decreasing temperature and pressure rather than chemical ' spray dependent. This position and conclusion is the same 'as addressed.for' the Anaconda cable. As a consequence such conditions as submergence will not alter the-cable IR characteristics. 1 This position .is substantiated by. Table 1, Appendix' 6 of NQRN-1A in which a-1/C #14 AWG 0.047 inch EPR insulated conductor was subjected to a~36 month 90 C water immersion test. See Attachment 6.for a plot of the IR test results. The results of this test were: SIR (Megohms per-SIC PF at 80V/ mil 1000 ft. l Start of Immersion 3.09 2.89 1300 End of Immersion 3.44 0.70 5200 Lowest Value Obtained 3.05 0.70 1300 Highest Value Obtained 3.44 2.89 5200 These results indicate that no moisture diffusion occurs within the. tested cable and that the IR values- actually improve. This information also suggests that submergence has less effect than the LOCA environment and is substantiated by the SIC and PF values. These smal Pchanges in SIC and PF indicate there was negligible moisture diffusion into the insulation as the result of submergence in 90 C (194 F) water over a 36 month period. It can be concluded that Okonite cable is capable of withstanding the effects of a LOCA 3 followed by submergence for a period of 120 days plus 10% margin. ,) i l L 3 1 i
VI. TEFZEL The capability of Tefzel insulated conductors to withstand the effects of a LOCA followed by subsequent submergence is demonstrated in Okonite Report No. NQRN-4A, Rev. 1, dated March 31, 1987. This report was unique in that j submergence testing was performed on aged and unaged samples after being j subjected to LOCA testing, i The tested configuration was a 25 foot 600 volt,1/C '!16 AWG conductor insulated with 0.015 inch of Okozel (Tefzel). A total of four samples were I subjected to qualification testing with two samples thermally aged and - irradiated and the other two samples subjected only to irradiation. The two j thermally aged samples were subjected to 7 days at 180 C to simulate 40 years " at 90 C and all four samples were then subjected to a minimum of 200 MRADS of gamma radiation. The four mandrel mounted samples were then subjected to a 130 day LOCA test in which a peak temperature and pressure of 345 F/116 psig was obtained in two initial transients. The chemistry of the spray solution consisted of 3000 ppm boric acid (0.28 mo'ar) 0.064 molar sodjum thiosulfate with the pH = 10.5 ! applied at the rate J 0.15 gpm per ft. for tM first 30 days of the test l followed by steam and water spray for the remaining 100 days. All conductors j were continuously energized with 600 volts and 20 amperes throughout the 130 l day LOCA exposure, j l After 30 days of testing, the samples were subjected to a high potential j withstand test of 80V/ mil applied for 5 minutes with the LOCA vessel flooded. i Upon completion of the LOCA test, the samples were removed from the test { mandrel and recoiled around a 40X mandrel and immersed in tap water. The ; samples were then subjected to another high potential withstand test at 1 80V/ mil fcr 5 minutes. One sample of 'he unaged conductor failed early in the pre-LOCA exposure and , upon inspecti , was attributed to mechanical damage. The results of the l qualification tests are as follows with IR values in Megohms-1000 ft. l determined at 500 VDC. I l UNAGE0 AGED j Sample 1 Sample 2 Pre-Irradiation 700,000 600,000 650,000 1 Pre-LOCA 1700 1900 2100 i Post-LOCA 1700 2050 2500 An evaluation of the results taken during the LOCA exposure indicate that IR variations are primarily influenced by temperature as opposed to the effects of the chemical spray under pressure. This position and er ,clusion is the same as addressed for the Anaconda cable. The data demonstrates that Tefzel insulated conductors are unaffected by exposure to a long duration LOCA simulation with sufficient ability upon completion of a LOCA exposure to perform its safety function. See Attachment 5 for IR test results.
Upon completion of the LOCA simulation test, the unaged and one aged conductor was subjected to a 5 month and 8 month (respectively) long term water immersion test at 90 C. See Attachment 6 for the water immersion IR test results. The fact that no chemicals were present in the immersion water is not considered significant since the LOCA test demonstrated that these chemicals do not attack the Tefzel insulation. The results of the water innersion test with a continuous 600VAC applied to the two test samples is as follows with IR given in Megohms - 1000 ft: UNAGED AGED SIC PF IR SIC PF IE Prior to Immersion 3.77 2.04 508,654 3.58 :.18 508,654 Er.d of Immersion 4.53 1.16 23 3.55 1.51 46 Highest Value Obtained 4.53 2.04 11,057 3.61 1.51 21,194 Lowest Value Obtained 3.77 0.89 23 3.25 0.90 46 Value After 5 Months 4.53 1.16 23 3.53 0.90 1272 Value After 8 Months - - - 3.55 1.51 46 The unaged sample failed a 1.65KVAC (110V/ mil) one minute withstand test at 5 months and the aged sample failed the same withstand test after eight months. It should be noted that no failures were noted in the water immersion test while the { conductors were energized with 60CVAC but only failed after repeated 1.65KVAC ] one minute withstand tests which are known to be destructive to the tested 1 insulating materials after it has been applied several times. It can be l concluded from the Okonite test ind ormation that Tefzel insulated conductors 1 are capable of withstanding 130 days of LOCA exposure and 120 days plus 10% j margin of post-LOCA submergence. i l 1 1
5 . a 9 n l 1 f oo E - o t 5 d 4 s ge t ns csia eitb fh s fT ee 6 l e t r 5 2 l 0 a 2 1 e e .A - w 4 h s0s 0 2 D 2 ttC e 2 4 - rL u e 4 5 no l r B 2 apea u h eh v t 5 t rt a R r enfI e roo p ei e m vtl h e S eal t l 5 T T sce l 0 L i wf e 0 l U sfd o w 6 1 a 4 S si D 1 - n 2 EL el el . 7 i - RL l arl e 0 . F 1 5 E uuA r 2 6 t B 2 TW e qt u f F SD r a t 7 5 E aer . a 9 0 7 TE h e) r 6 0 2 4 R et pF e 4 0 2 8 DU c m p - 1 1 ET nme4 m 3 - GA eot9 gr e 0 e- - RR 1 T 3 l E EE rfl( 3 ps P 0 MP BM e a l a mm L 2 md nC ah X 1 UE b ei n sog ST ul f0 i 0 d si pe 9 F 2 ae e 0 FL 2 - M t OA mmht F 1 6 m a 1 N rot a . oo i 2 NI ec gs 2 0 2 rt d 6 - OF t 1 2 1 f a A 5 S nnt 2 d r IA geis d e r 4 RC nere - et I AO ob ut nr PL l d a i e s M s n d av o OO f aso n t n t CT oh ei o b o s us c oc e snl r a b 0
. t oae n 5 2 se k I civm A 3 4 gr c 1 I et m - ne o 2 V f aRi 8 3 i w R 3 -
f mI 1 2 d A 5 er r ad oee en 4 efht ra h nt a ti w nh s ot t anro gem. d n g 0 hi art t w pf f ee sl 3 om 2 3 a - 3 el od0 3 6 b n 2 - tl ce0 - i A 7 ao n0 8 7 e rf ni1 1 4 rt 2 t oa ee 1 seit - nhtb wef ot aos s m m m e g e5 e e ,ceh u g dAcso n l 0 n Cf og a a1 a oOnh e R v R TLitM e y e g R Rl g R n I I e n I a t a R n t a R n
# o sm # o R i ei R i e I s t x e I s l
p r o l r A e Ar p A e m C m C p m C m a O m Op a O m S L I L a S L I i
6 5 - 2 7 1 7 B 2 5 5 9 1 l 8 E 6 _ 5 d e g 0 7 5 a 1 l 9 n 1 l 0 5 U - e 1 4 - - 2 w 4 - 1 5 B 4 D B 5 7 5 1 e r 6 u l t l l a l e r e w e 7 4 w 0 D p 8 D 0 m 0 1 0 e e 1 e 1 r T - - r u 1 5 u 5 - t 0 l B 5 0 t a d 0 a 2 a 1 0 r e g 3 n 6 1 r 0 1 e - i e 1 p A 0 F p 2 - m - 5 m 1 5 e A 2 F e . T T 8 t 4 3 o l 2 7 l o a 2 a f n
- 0 n i 0 1 i 5 F 2 - F 3 2 5 5 F A 5 E F a - P 0 2 d 6 L 0 0 m 1 e g
E X 0 7 o 2 P . 2 0 - r a 0 L dt 1 4 f - n 5 X es - 2 U 4 t e 2 d e - y at e e l - 0 l i r n b 3 l 7 dA o i a A 3 a aC o a . C c 0 rO M tt i 1 rL bf e m 3 - i l o t e A 5 r e 0 i h C 5 mo u s0 n o oi m g0 6 rr a n1 k s f p S i O o d - t d a 1 a s e 9 es b e no it 5
- rmh k a 9 2 no c 7 td og o 9 be 6 e R ot 2 dM l 0 c e A - ee rj so 5 at eb b 0 wus d ee s rt en er ue l e wev ab sn e v e eo g d g uc n R a n l a Ih a ae R R vr e nh e e e g R oc g R R w g n I ii n I I n a sh a d a R n o
rw R n o t n R
# e # sa #
e R I i s ms me e R I i s e t e e R I l r il l r l l p A e p p A e Ap p A m a C m em m C m Cm m C O m h a a O m Oa a O S L I Ts S L I Ls S L
A gl Cfni Ooia l L tf t l l s olil al ee Anete atd C zszo o Olf fttem Ll ede ac iT nT eh n n ew e a eu rt eo g tAhthstrm h Ctst oceo oO e pamc tLmtmxfb o oe - un rarAr esi o fCf rh ir O etot redLdt tl u et e ef n pf nnnaors 8 uaioi oe d 3 s aiaedir e g 3 stttcer rebabnspo A 5 ol oroea t
- 2 b u gbd 2 1 easdsr ee l ct testv b lll misi 4 af ul ub et 9 rosusunta _
1, a efeso l _ pnr r iAu _ 1 2 mo oin a rsCm eouOu l - csoohflLc l 6 ritT c e 4 ees nnny w rmrd ooel D ebee .icet e r d e 3 1 3 wumtet s s ccas iencidc bn a u g - et jeihai t A 0 uarbufThf a - 4 l h e u qi i r 1 1 attsel yn e v a sa .eg p m tt awn etqnts u gh i e saees i - T edhbettrt t t t aseo 8 l e d hehn 1 a d RhtaAttt n h a ItahC O yeh r ee - i h F n m thL gc w a 7 h oo c on F 5 c i rdiel efg _ 2 d 5 1 0, i h sfehhogon r t wtdr i 1 e g 2 1 w ed o oest _ 2 - 1 ensohmss a n 5 2 3
- s e
md iuel ettbee eul t U 1 2 l lb pd m s d l e p rc me rl m endaredaa _ z a tol sah n gi f s acu ptaet - T e m w oem ehl otn rn e . o ebsbcronu r h a oioq L. f t ntcspsie - aI aprhs d e tc a m ues af n . it ht tI yf n s i
.oesmfo rc as t ee es .
te e i e gl rl t bt t .l rrbebc o asieeatae A cebwmcarf . eC i ua b waf - rO dl rEuE pe . eL naePsPem - w r i v pL Ltoe oXeXach - so sR h r t ei eI estsaa ur l p l l o blbtgt enae op t . a al asnssihc va ceceie t n : w brbmd e _ Ro ldfkuk oesg e It l ocd crmer g ae o ofrt e . e g n a nd oe rus rsRrR e dfcb oam - n R it ot odtderiu a se fal eteneds R n rj r t e ti p n
# o eb seaabaa i g R i u t p i its n l
e I s r i ms l mndddbeoi ueiaeaol sr p m A C e m ee n st e rmr trrreaad
.bl u a O m he rlli oibc S L I Tb au f ss ensdrdneie hienenahh r *
- Tf rapactTu L!L , l
1 VIII. LOCA SUMP CHEMISTRY The sump chemistry is determined by the chemistry of the Reactor Coolant and Refueling Water Storage Tank. The resulting composition for Watrrford 3 would be approximately 2000 ppm Boron, as boric acid, buffered to a ph of 7.0 by trisodium phosphate dodecahydrate. There is negligible effect on sump chemistry as a result of discharge of the Safety Injection Tanks because of the volume of these units. The other constituents of the sump are (1). ppm quantities of solid fission products and iodine, and (2) materials washed from containment surfaces as a result of the containment spray. The exact concentrations of materials washed from containment surfaces as a result of the spray system could only be determined from an actual accident. The expected primary constituents would be calcium and silicon from uncoated surfaces, dust and dirt from pipe racks- and cable trays, small quantities of greases and oils from the surfaces of rotating equipment, and small quantities , of metals as a result of corrosion of exposed metal surfaces. i There are no significant quantities of acids or bases inside the containment that could affect the sump pH. The boron and trisodium phosphate are by far the largest constituents of the sump water. The other constituents are ) relatively small and it is anticipated that they would have little or no ) effect on the cables. l This general chemical resistance was obtained by literature provided by DuPont, ITT Grinnell, Gates Rubber Co. and other industry sources. This literature clearly shows the ability of cable materials to o continuously at high temperatures (greater than 250 deg. F) and perate their resistance to attack by acids such as boric acid, caustic bases such as trisodium phosphate, inorganic salt solutions such as metal compounds, and oils and greases. The cables which are potentially subjected to submergence are: ; MANUFACTURER INSULATION JACKET j Anaconda EPR CSPE or CPE. Okonite EPR CSPE Samuel Moore EPR 'CSPE i Rockbestos XLPE CSPE Cerro EPR CSPE ! Rockbestos (Triax Cable) XLPE CSPE Combustion Engr. (Supplier) Quartz Stainless Steel
1 1 The age degradable components of the identified cables are ethylene proplyene i rubber, chlorosulfonated polyethylene and cross-linked polyethylene. The ; hypalon (CSPE) jacket material which is used in all cases with the exception i of some Anaconda and the mineral insulated cable will display little or no effect to its material properties when exposed to boric acid, trisodium pNsphate or a wide variety of metal compound solutions. Oils will have a minor to moderate effect on hypalon at full concentration at 158 F. , Additionally, hypalon is resistant to sulfuric acid up to 95%, nitric acid to I 30% and hydrochloric acid up to 37% at 122 F. Chlorinated polyethylene (CPE) I displays many of the same characteristics as CSPE being equal to, or better i than CSPE. i EPR has a broad resistance to chemicals with good electrical properties, but I does not display a good resistance to oils and other hydrocarbon fluids. This I is not considered significant because the concentration of such fluids are j expected to be minimal. The tendency of oil to remain on the surface will minimize contact with any EPR insulation. The hypalon jacket, which does l display good resistance to oils, will eliminate contact with the EPR j insulation. Polyethylene and its cross-linked derivatives demonstrates excellent resistance to motor and fuel oils, as well as gasoline and kerosene although it has a tendency to swell ht higher temperatures when exposed to full concentrations of these oils and fuels. It also demonstrates excellent i i resistance to acids such as sulfuric, nitric and hydrochloric as well as, to fixed alkalies such as sodium hydroxide. Polyethylene also demonstrates good j resistance to ketones, esters such as lacquer thinners and alcohols. j l Neoprene is used on some Okonite 600V nulticonductor control cable as a jacket material for the inner conductors which have an overall CSPE jacket. Neoprene I is highly resistant to petroleum oil since it was originally developed as an ' oil resistant substitute for natural rubber and is widely used for this purpose. Neoprene indicates little if any change in properties or appearance when exposed to diluted mineral acids or inorganic salt solutions. It displays excellent resistance to boric acid solutions and good resistance to a fresh or salt water environment. Neoprene also displays good resistance qualities to fixed alkalies such as sodium hydroxide. In all, neoprene is not i; likely to be significantly affected by the expected containment sump contaminants. , i The metal jacketed mineral insulated cable will be unaffected by submergence I because Quartz, which is primarily a ceramic material, is not affected by j radiation and is not age sensitive. In addition, it is enclosed in a j stainless uel waterproof jacket. 1 The 8 weeks submergence test performed on a sample of Rockbestos irradiated i XLPE which was exposed to a 200 F chemcial bath of 1800 ppm boric acid, 50 ppm j hydrazine and trisodium phosphate for pH control also aids to demonstrate that j the expected sump chemistry will not attack cable insulation. ! The various cable insulation and jacket materials have been shown to be j resistant to expected sump contaminates, as well as t wide variety Of common { industrial agents. It can be concluded that cable insulation and jacket j materials will not be degraded as the result of chemical attacks during : submergence conditions. l
j IX. JUSTIFICATION FOR WATER IMMERSION TESTS ON UNAGED INSULATION The water immersion tests were performed on unaged conductor insulation with j the exception of the Okozel (tefzel) insulated conductors and Rockbestos ' irradiated XLPE. This is not considered-significant with regard to the results obtained during the-water immersion tests because the cable exposure 1' to 90 C water had the effect of accelerated. aging. If.the 90 C water temperature is considered as the aging time and. 50 C is considered as the l service temperature for instrument and unenergized conducti rs, the following - i table may be prepared with the activation energies obtained from the qualification reports: 1 Inunersion Imersion Time Tempera ture , Ea(eV) ta . Ta Ts . J Samuel Moore 1.41 30 Heks 90 9 50 C Rockbestos 1.34 52 weeks 90 C . 50 C l Okonite 1.09 156 weeks 90 C 50 C
]
Anaconda 1.69 26 weeks 90 C 50 C j By converting Ta and Ts into K and solving for ts (service time) from the ) Arrhenius Equation: 1 In(ts) = in(ta) - [Ea (1 - 1)] [RE (Ta Ts)] ; Kb = Boltzmans Constant = 8.617 X 10-SeV R i and ts = exp [in (ts)] Solving for ts using the data from the supplied table, the accelerated aging based upon a service temperature of 50 C results in the. equivalent life: MANUFACTURER EQUIVALENT AGE Samuel Moore 153 years ! Rockbestos 201 years _ Okonite 224 years Anaconda 401 years It can be concluded that aging will have no significant impact on the ability. of the cable insulation to resist moisture absorption. The testing performed on Tefzel and irradiated Rockbestos XLPE cable which, were both aged to an equivalent 40 years at 90 C and then subjected to full LOCA testing also l demonstrates that aging will not adversely impact the ability of cable.
-insulation to perform its intended safety related function. - 21 -
l X. OPERABILITY OF CABLES UNDER LOCA CONDITIONS To address the impact of cable insulation resistance loss on instrument circuits, LP&L calculation EC-I86-009 performed an evaluation of all transmitter circuits within the containment. The calculation identified that I there were 67 transmitters. One of the transmitter field-pulled instrument I cables was supplied by Rockbestos, three by Samuel Moore with the remaining 63 supplied by Anaconda. J The calculation considered the lowest IR value taken from the LOCA test results of the applicable instrument cables. From this data, the calculation l established the actual error in a protective current loop signal under l accident conditions. The calculation assumed that the transmitter cables are i exposed to the environment and did not take credit for the fact that they are ) all routed in a conduit. The results of the calculation determined that the ! low SIR error contribution to the transmitter circuits under worst case I accident conditions did not effect transmitter circuit set point values. Control and power cable is less sensitive to IR fluctuations because I instrument loop accuracies and leakage current are not a concern. For control i and power cable, the IR must be maintained at sufficient levels so that faults do not occur. Operability for these cables are verified by energizing tested cables at rated voltages during LOCA tests to ensure thtt no ground faults occur and by performing high voltage withstand tests. I i 1 i
i l l XI. RADI ATION EXPOSURE OF CABLES SUBMERGED IN CONTAINMENT LP&L performed a calculation (N0SA-RP-CALC 89-001) to determine the total integrated dose that cables and associated splices would be exposed to during submergence conditions after a postulated LOCA. The *eactor core inventories for the individual radionuclides in this calculation are consistant with the FSAR Table 12.3A-1 and was assumed to be: 100% of the core inventories for the noble gasses 50% of the core inventories for the halogens 1% of the core inventories for the other isotopes The results of the submergence calculation for the gamma, beta and total doses ! in Mrads are as follows for 132 days and one year post accident: Period Gamaa Dose Beta Dose Total Dose 132 days 22.8 7.9 30,7 1 Year 29.4 12.2 al.6 The one year total integrated dose of 41.6 Mrads compares favorably with the cable qualified dose of 200 Mrads gamma. Since all the submerged cables are enclosed in conduits the beta contribution would be negligible and the one year gamma total dose would be orly 29.4 Mrads. In conclusion the cables are qualified for the postulated total integrated radiation dose during submergence by substantial margin. 1 l l l
l XII. INDUSTRY EXPERIENCE'WITH SUBMERGED CABLES l Attachment 7 contains documents pertaining to industry experience with . l submerged cables. That data indicates that there is no known experience where 4 water submergence in field conditions has contributed to cable failures due to water absorption. i l .St. Lucie reports that FP&L in the Miami Beach distribution system has installed over 600,000 feet of cable since 1956, ranging in size-from #12 AWG to 1,500 kemil butyl rubber. Many more feet of cable was. installed ' prior to 1956. The Miami Beach system never attempted to control the water level in ; any underground installation except as. required _ to facilitate personnel access into the underground system and manholes. Cable failure records since 1945 i indicate that there has never been an electrical deterioration failure of any '. kind on these cables or the splices associated with them due to moisture absorption. LP&L has also never experienced any cable failures.due to being immersed in 1 water with some cables which had been submerged for 20 years.. This information was obtained from cables used.in power plants and substations. These highlights of Attachment 7 provide supporting assurance that submergence '! of cables in water under actual field conditions does not result in any l moisture absorption related failures. 1 1 1 l h l j I 1 h
~
i l
1 i XII. INDUSTRY EXPERIENCE WITH SUBMERGED CABLES I
)
LP&L : i
- 1. Waterford 3 - Basemat (RAB)
'1982-1989 conduits filled with ground water (minerals mainly calcium). . Relicensing Issues Task Force Report Sub-Attachment 1.
- 2. Power plants, Substations etc. LP&L letter January 21, 1985 Sub-Attachment 2.
i i OTHER UTILITIES 1 1
- 1. St. Lucie Unit 1 FSAR Appendix 3A, Section D, entitled " Underground l Cable Qualification for Service in Potentially-Submerged )
Environment." l Sub-Attachment 3. l i
- 2. Shippingsport, Indian Point, Peach Bottom et al. Excerpt from paper T74044-4 Sub-Attachment 4.
- 3. Okonite letter of November 24, 1986 Sub-Attachment 5. ]
INDUSTRY l
- 1. Okonite Bulletin Attachment 6.
_ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ a
XIII. CONCLUSION l It has.been demonstrated that the cables supplied by the identified manufacturers can be considered qualified for submergence Lfor a minimum of 120 days plus 10% margin after exposure to full LOCA conditions based upon the following demonstrated facts. .; i
- 1. Insulation resistance was demonstrated to be influenced primarily by j the temperature and pressure of the environmental medium as. opposed !
to the content of the medium such as' steam, water or chemical i compositi on. Attachments 4 and 5 demonstrate that the presence of- l chenicals in the LOCA test does not significantly alter IR LOCA test j values. The effect of. pressure variations at a constant temperature ' will change cable IR values because the imparted energy (BTU's per unit volume) to the cable insulation is varied. These conclusions j are consistent with those of Indian Point Unit 3 Report 890080-3. i l
- 2. Cable insulation resistance values were demonstrated not to be degraded to a significant level upon conclusion of the LOCA exposure to where post accident operability of the cable was a concern prior to the potential of submergence. In some situations the IR values remained essentially unchanged, others improved, and others decreased to lower but more than adequate values. These were acceptable based on the fact that all IR values on several samples were consistent with each other.
- 3. Water immersion tests conducted at 90 C-for various times was in I excess of the 132 day post accident operability requirement.
These tests demonstrated that the tested cables did not degrade with imersion times, but in many instances actually improved when values' of SIC, PF and IR are considered. This cenclusion is consistent with Indian Point Unit 3 Report 890080-3.
- 4. The submergence profiles are less severe than the containment' atmosphere profile which would result in better cable performance characteristics. Due to the similarity in temperatures between the 90 C (194 F) water imersion tests and the LOCA tests with a final temperature dwell of typically 212 F, it can be postulated that cable performance characteristics under submerged conditions will be the same or similar. This conclusion is consistent with Indian Point Unit 3 Report 890080-3.
- 5. Section VII demonstrates that the IR values obtained during the L water immersion and submergence tests are similar or superior to those obtained during LOCA final temperature dwell testing and that continued operability upon submergence following an exposure to a LOCA can be expected. This conclusion is consistent with Indian Point Unit 3 Report 890080-3.
- 6. Immersich testing on the Tefzel conductors following LOCA testing 3 and submergence testing on Rockbestos irradiated XLPE insulated !
conductor following LOCA testing, support the position that cables are capable of withstanding a full LOCA test followed by subsequent submergence. The IR results of the final temperature i dwell during the LOCA test and the subsequent submergence test of the Tefzel and irradiated Rockbestos XLPE as presented in Section VII demonstrate that these cables performed in a comparable fashion 4 to those cables which were subjected to separate testing. This also indicates that the effects of sequential testing are not significantly cumulative to result in common mode failures during submergence. This also supports the position that qualification for ! submergence can be demonstrated by LOCA test results and separate 1 water imersion tests for unaged cable. !
- 7. Product literature and LOCA testing on the cable insulation and ,
jacket materials demonstrate that attack by anticipated sump chemical composition and contaminants is not a concern. 'lq l l l 1 1
ATTACHNEffT 1 AtlAC0llDA LOCA TEST RESULTS I l O
iR V I ** - ohms 0 9 8 0 1 0 0 0 0
+ + + 0 0 E E E 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 i
n g [ y! il : g: - ' ' d ' t e ' s) ' e6 T1 " 0 0 e m A 29 ' 1 i T C ' V/' Oe s _ l ' v L p ) ema 'g/ ' s R l b
\' ]/
r u I aS \ f 0 oe
" h '
1 ( _ C (2
)
i m s T e t R 0 ' n e m P5 ' m i T - E3 ( e . _ 4 1 r s - " u v aC '
' s a .
d - e p nF ' m - o ' M T e c A _ a n
/'j /' j '
C j/ - O A : 1 L 0 0 0 0 0 0 0 0 0 0 4 3 2 1 e* ; e9' 8 p
- 5 3
i n N 0 d ' 0 e \' 3 t s) \ e6 T1 N.N 0
'5 A0 2 2
C Oe l
)
L p 0F l ema ' 02r(e u b aS t a r e C (2
) ' 5m 0p 1 e R 0 T P5 E3
( 0 4 ' 0 aC 1 d - nF o c 0 a ' 5 n - A
- C - 2E~ _ - =! - - - E O
0 9 8 0 0 1 0 0 0 0
+ + + 0 0 E E E 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 T
Va ue* O$s iL l
I i t> l ATTACHMENT 2 1 i i ROCKBESTOS & CERR0 LOCA TEST RESULTS 1 1 e s> m l l
e f - M e goh s/1O00 { 0 1 0 0 1 0 0 0 0 0 A 0 0 0 0 1
- 0. 0 C 1 1 1 1 1 0 0 O0 g $5 2 !- [ $J5 : s5 I 0 O '
' 0 ' 1 L ' ) '
l e3 2 bE a e 0 e Cl p ,'
' 0 m " 0 i )
Em - ' 1 T s Pa
' v LS ' ) R X( '
s r u I 4 y0 0 o 0 (h l a 8 " l 1
" e i
c5 .
' m d e
mR i T d eQ , ,! , :,l ! l l 2';l : e n E h ' m y Cin
/,'
i T a r ( d s' .
" 0 1 s p
sot e f s
/ / j \+ '
v p S m t s e se
/ m / ' '- e h eT .' T C b j/ '
k !,/ - c _h _ - : 1 o 0 0 0 0 0 R 0 0 0 2 0 4 3 1 em . De9I es F
_ 9 j - M e g*h m o0 ft 0 0 0 0 0 0 1 0 0 0 0 . A 1 1 1 1 1 O C _: =g- g[ gE gE g: O = ^ L e1
) ^
w ' l b B 0 0 ae ' 0 1 e Clp ' m
) m ' i T
Ea ' PS s L(
) v s
X5 " 0o 0u r R I d 0
~
1 h t e85 ' ( e a m iR d ' i T s aQ r y' ' j ' t n e m
; e r n i i
( sd i y' f'0 n1 m e r T s u oe t j
/ \ \ s a
v se se t
/ /
N' e p
\' / \ M m e
b T
/
A
\ A T k 1' '
C c j O o _,\
- / 1 L R 0 0 0 0 0 0 0 0 0 4 3 2 1-Te*P - Deg ees F
A "N C O ' 0 0 L 3 1 l e )3 bE a e 0 Cl p ' 5 2 ) Em . P a LS X(4 l l y0 a8 N\ N N 0 )F 0(
~2er t
u a r i c5 ' 5 0ep mR 1 m e eQ T h Ci n ( s N 0 sot e t s d
\X . ' 0 1
seTe 0 b ' 5 k c o R 25 h:: s = } s5_ O 0 0 0 0 . 1 1 t 0 0 0 1 0 0 0 0 0 1 0 0 0 1 0 1 T k _ 9 0 ft R I
n 0 . 5 3 :_
\
A . C
\
O 0 0 : L l b ae
)
e1B ['\ g \' 3 0 Clp ' 5 ) Ea m \N d 2 N PS N ) L( 0 (F 0 - X5 2er u d 0 . N% t - e8 a r t 5 e iR a N
\ '
0p 5m _ d 1 e aQ r \ T - i ( r n i N N' 0 sd ' 0 oe t s' N 1 t se se ' * - b T 0 k ' 5 c o R e _ r
- - gt: gE : $ 5: $5 _ u ._
O t - 0 0 0 1 1 a r 0 0 1 0 - 0 0 ' e : 0 1 p - 1 m e - T - j - y e 9O$ 'OO t s v R I
l i I i i e } i i 1 I i
'l .l I
1 l l l I ATTACHMENT 3 SAMUEL MOORE LOCA TEST RESILTS e e a w________m._
[ Va ue - k m_s 3 2 1 0 9 8 1 1 1 1 0 0
+ + + + + +
E E E E E E 0 0 0 0 0 0 0 0 0 0 0 0 y b 0 1 0 1 0 1 0 1 0 1 000 1
;l a5 : ;5h![5 0 d e )9 t 1 -. i 1
s - ' e0 T1 N ' Ale e Cp i m Om L a 0 T
' 0 s N
1 v e (S
' ' )
l . s r R b 8 ' u a7 ' o C91 h '
' ( ) e Re '
i m s Pn N T t n e Eu q . e m ( J 0 i mT x'f 1 ed /
' e r .
f u s r e v
/
ot ' s oa ' a p x e Md x M m e l R x' ' A T eT \ f C uQ m - N ryr ~i
/4 1
O L a 0 0 0 0 S 0 0 0 4 3 1 - T** . gefe p
5 y 3 b d \ e ' 0 0 t s Y 3 T1)9 e
- \
A1 0
,lx 0 Ce Olp F / ' 5 2 )
Lm F ( a l eS / 0 0 e r ba (8 ' 2 u 7 t C9 1 a r
)
Rne 0 e p
' 5 Pu J 1 m
E ( d N e e T et r a od
\N ' 0 0
1 oR MT Q l e ' 0 5 u m a S g#t - sFt eE : _ gt: ~g 3 FI O 3 2 1 0 9 8 1 1 1 1 0 0
+ + + + + +
E E E E E E 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 iR I Ues - O ms
4 i I f ATTACHMENT 4 ! l OKONITE LOCA TEST RESULTS i i 4 I l l ,
- M e 9$ " (Oo0 1' 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1
, 51::a}g - g:: sl' i
n ' 0 de )A - w " 0 0 0 t sp ' " 0 eu 1
, l ::
I l l
,[
l i e To r 4 ' i m AG
' T 0 .
Cd 0 s Oe _x
\ '
0 0 )s R v L g v '
' 1 e I
t u e (A l i n ! ba 5 ' 0 e ( m x C v. x'
,/ " 0m d ) e N' / " 0i 1T d e
RR
~' 'r' n s P A " f lf;{ \ j '
e E m y a y d E ( 1 ;\
\
i T a r 0
- . p 3 eN "
0 0 s v S t
=
iR t " 1 p m t o nQ m e 0 koN e h = T C t O i / / I j/l i
~
m o r j - f
- ( j, - 0 -
1 y 0 0 0 0 0 a 0 0 0 0 r p 4 3 2 1 s l a Temp 0g ees y i c m e h C rflllll i
,x \ ' \ 0 A ' 0 p 5 N s \ 3 u v .
x or e
\
x 0 GR - N
' 5 2
d A N I e1 - N' gN ' A R 0)
' 0 (F - Q 2er u
sN t a e n r
% e l
bi 0p ad Cet ,
* ' 5m 1 e T ) s Re PT 0 E A
(
' 0 1
e C - i t O nL k o ' 0 5 O gbF l - Eb_ - - = 2 _ 0 0 0 0 1 - 0 0 1 0 1 0 ' 1 R QO$ S/10 0 f
,l ,
,R - g e Sh $/ jgU -
ft 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 0 g2 : g5 - z5 ~~ ;! - " 0
' 0 0
n) ' 1 iA ' d el e 0 0 s p t 0 Ta em .%p ll 0 0 e m'i1
!1 !
A S i m Cd ' 0 T Oe L a g s' N',
/' "
0) 0s e s v
' " 0t en ~
1 u R l ' i n I ba (U ' ( m ' e C5 .
/ / '
0m 0i
)
v ,N / " 0 T s Re x 1 t n e m PR ' e i E - m T ( A ' e r s e-1 0 u v t ' 0 s iN " 1 a p nR e o ' M m e kQ ' A T ON ' C O
- - - - - - 0 L 1
0 0 0 0 0 0 0 0 - 5 0 5 0 5 0 5 3 3 2 2 1 1
~
Te" . De9 ees F
\
0 n) s N ' 0 3 iA d ele ' N s p t Ta em ' 5 0 2 A S ) F Cd ( s Oe L a g
' 0e ' 0e r 2g l
en .e D ba (U s' e r C5
)
v g\ 0t
' 5a 1
u r e p Re m PR T e E ( e-A 1 S\ ' 0 0 1 t 3 iN : nR o kQ 0 ON ' 5 gt __ g _ - ! _ __ O 0 0 0 0 1 0 0 0 1 0 0 1 0 1 - 1 - I R _ M Qoh j OO ft W
'l t 8 1
1 1 1 i i i I ATTACHMENT 5 l TEFZEL LOCA TEST RESULTS l 1 l
/
I emp e l l r ?
l R Meg hms/3o0O ft 0 1 0 0 0 0 0 1 0 1 1 1 1 0 O _: gE g5- gmI eg E =g i n - d e mlM - t 9 0 s - ' 0 Tle )e i
\a - / ' 0 1
e p ' m Am Ca i T OS L
)
e v d
,f ,/ '\ s 'x // R l
b g e e <' " 0u 0o r I 1 h( aA ' e C( ' m
) A '
i T s e l e4 z-N N
,r' j '
t e n m f '
,/ mT i
TeR / ' 0 e r . ( Q N
< \
1 s u s v eN
\y\ a .
t i [j/ \
' e p m
n f
/\ \ ' M e k
o \
\ / / / ' A T _
_ \.\ '!, C O - - '/ 7 1 O L 0 0 0 0 O 0 0 0 0 4 3 2 1 e*P* De9'ee F ll l t
Il1l,lI
= _
0 5 3 1
- h' _
0 e 0 l 3 p - m aA = S4 - 0 5 _ 2 dN eR - gQ
)
A x N y' \ s ' 0 (F 0 e 2r n ' u l bd ei ae
\ t a
r e t 0p Cs ' x ' 5m -
)l e 1 e T
e T \ fzA eC 0 0 TOL ( t i e A'\ 1 n ' 0 k o ' 5 O e r
= g5 - 5 3E- 5 g5 _
O t u 0 0 0 0 1 1 a r 0 1 0 0 0 1 0 e 1 p m e T I g M 9Oh m s gt s v R I EIlllll l 1l .1 l' '
iR - y* hms/1 o f 1 0 0 0 1 0
' 1 1 $ 0 O
_: ! g5 5:;r: r5 i n - - d e w ' t ) s 1 Te e l p is bi , tf ' 0 0 0 1 e m A m
' i C a '
T O S e/N N s L d
) v I s e e ' 0u r R I
l g " 0o b a ' 1h a n ' (
- C (U #, e i
m -
) s l
T e A x t n e N z4 e m f i e- # mT TN / 0 e ( R N'/, \'L ' 1 r u s s v eO
/ \ '
a p t f
\ e iN n /, \ \
M m
/ '
e o A A T k x\ '// ' C O ' O 1 L 0 0 0 0 0 0 0 0 0 4 3 2 1 emp - Degrees F
0 5 3 h n 4N i 4 0
' 0 d \' 3 te) s1 Tlee p \ %s' ' ' 5 0
2 Am Ca N x OS ( s' x ) L d 0 (F
% ' 0 eeg 2r e l
b CU a an NN \ 0p t u a r e
) ( s ' 5m l
1 e eA N T z 4 f e- \ TN ' 0 0 ( R 1 t eO iNn o 0 k '5 O
- gE- g: - i: gb_
0 t 0 0 0 ' 1 1 0 0 1 0 0 0 1 0 1 T Meg $m) a00 ft R I
f 8 i ATTACHMENT 6 CABLE WATER IMMERSION TEST RESULTS i l 1 l t
w
)
R t s P E
! ; :'d .
s
;.l
( e) -a x' . ' 0 0 e Ts y 3 1 t i n a o
)
k ed O c r ' n2 \ e3 \ ) g1 r(
+ / - j,' /
_M %
)
P d a e s ' E r r me i (
) R bl s F s y ( o u ab ' '
a e t s S d r o e ( C ,. ' ( e o b k m M c A3 i . o E d ' 0T 1 S R Cr o I '
+ '
f mre o
)
P E
)
l r t e f a ' R F f z t sW - ' ( d a T ( e l n e uo n t o i s c a n e ' n k o R A O 2 ee : - .sE - - i 1 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 1 0 1 - 1 f _ Me9o$ s;1o00 ft - , i
1 ATTACHMENT 7 INDUSTRY EXPERIENCE WITH SUBMERGED CABLES
l, INDUSTRY EXPERIENCE _WI_TH SUBMERGED CABLES
- \
LP&L
- 1. Waterforo 3 - Basemat (RAB) 1982-1989 conduits filgd with ground water (minerals mainly calcium). Relicensing Issues Task Force Report !
Sub-Attachment 1. i d
- 2. Power plants, Substations etc. LP&L letter January 21, 1985 Sub-Attachment 2.
i l OTHER UTILITIES
)
- 1. St. Lucie Unit 1 FSAR Appendix 3A, Section D, entitled " Underground Cable Qualification for Service in Potentially-Submerged Envi ronment."
Sub-Attachment 3. i i
- 2. Shippingsport, Indian Point, Peach Bottom et al. Excerpt from paper .{
T74044-4 i Sub-Attachment 4. I
- 3. Okonite letter of November 24, 1986 Sub-Attachment 5.
j i i INDUSTRY l i
- 1. Okonite Bulletin l JDB-Attachment 6.
l 1 i _ ___.._._______f
?
EUB~/97TAcHmepr j_' Issue #19 Page 1 of 2 PRILICENSING ISSUES TASK TORCE REPORT Issue fl9: Water in Easemat Instrumentarien Conduit
- 1. NRC Recommended Action NRC instructed LP&L to review all conduits that penetrate the basecat and terminate above the tcp of the basemat to ensure that these potential direct access paths of water are properly sealed.
- 2. Task Force Evaluation
- a. The Task Force Support Group (TFSG) has. performed a walkdown of the Reactor Auxiliary Building (RAB), Fuel Handling Building (FHB), and cooling tower. As a result of this walkdown inspection, the TFSG has concluded that the quantity of water resulting from the seepage through l some of the seals of the permanent conduits embedded in the. basemat I
area (EL. -35') is insignificant and should not cause a fidoding concern. The quantity of water seeping through the conduits is very small, and most of it evaporates before reaching a floor drain. The i TTSG has verified that LP&L has pressure grouted the piezometer . . standpipe, thereby eliminating it as a potential leak path. The TFSG l has also verified that LP&L has pressure grouted the temporary blockout j pits for temporary conduits. Further details of this issue and Task j Force Validation results are discussed below. The Task Force considers 4 the LP&L response to Issue 19 to be appropriate to the NRC j instructions,
- b. The TFSG has perfermed an independent walkdown of the basemac area l (EL. -35) and identifi ed approximately 543 permanent cenduit locations.
These conduits are entirely within the basemat and the building structures. The TFSG'c walkdown identified 40 places where watness due to seepage from the conduit seals or asidence of past seepage from the conduit seals was found. Some of these locations were different from the locations identified in the LP&L walkdown performed in May 1984 The TTSG has concluded that the seepage problem is not limited only to the areas identified in the LP&L walkdown but could affect the other embedded cenduit locations in the future. The TTSG has verified that LP&L has pressure grouted the piezometer standpipe, thereby eliminating it as a potential leak path. The TFSG walkdowr has also verified that LP&L has pressure grouted the temporary blockout pits for the temporary conduits. l The TFSG has observed that the per=anent conduits do not leak grcund I water in a sufficient quantity to cause flooding cencerns. The floor drain and sump pump systems are mere than adequate to handle the present seepage. The LP&L decision to replace the leakine seals at a future cine with new seals consisting of light density silicene elastener based upcn maintenance convenience should be effectua' in further reducing the leakage. The results ef the TFSG walkdown in this regard.are found in Appendix XIX, Section A, to this repert.
..- ~ 1
/ /
j- . Issue #19 Page 2 of 2
- c. To understand what might happen if the conduits were to leak at some future times, the TFSG reviewed the operators log sheets and the alarm
. response procedure for the high-level sump alarm. The TFSG concluded that plant personnel visit the basemat area (EL. -35') of RAB, THB, and i
cooling tower on a daily basis. A leakage in a quantity sufficient enough to cause concerns would attract the attention of the plant i personnel during their daily visits to these areas. In the event a j large leak develops, the sump high level alarm would alert the plant personnel to a possible flooding concern. See, Appendix XIX, Section B. t ! d. The TFSG has reviewed the EBASCO specifications for power and control cables, Class IE equipment and vendor's qualification documentation and ; cencluded that the cables are qualified for direct burial or dry or wet ' conditions, but it could not be substantiated that*the degradation of i these cables vould not occur over the life of the plant under the l continual submergence condition. See Appendix XIX, Section C. The Task Force recommends that LP&L take appropriate action to obtain environmental testing data on the submerged cables or institute a , surveillance program which will check for cable degradation so that appropriate action can be taken, if necessary, to replace cables during I convenient scheduled maintenance periods. l
- 3. Summary -
In summary, the TFSG concludes that the present seepage through the existing silicone foam seals do not pose a flood hazard. Pressure greuting 4 of temporary blockout pits and pieremeter standpipe eliminates these potential leak paths. The observed seepage through all of the conduits is l estimated to be less than ten gallons per day, and most of this evaporates ' before it reaches a floor drain. Moreover, the floor drain and sump pump systems are more than adequate to Fandle this insignificant quantity of water. The daily visit by plant personnel and requirements of the alarm response procedure indirectly provide surveillance of the area. 4 Cause, Generic implication, and Safety Significance The cause of this concern was that the silicone foam was installed as a waterstep barrier in the conduits, but does not provide a total vatarstep characteristic against a high hydrostatic head. The TFSG believes that the LP&L walkdown of the entire base =at, the identification of the =agnitude of the seepage, LP&L surveillance, and the decision to replace the seals as necessary and when convenient has showr that there are no generic implications associated with this issue, nor is it now safety significant. The TFSG has concluded that it is very unlikely that the seepage through the conduits v111 cause a flooding problem. In the unlikely event cf a considerable increase in the seepage rate, the flocr drain and sump pump systems are more than adequate to preclude a flooding condition. The sunp high-level alare response procedure and operator's daily visit to the basecat area veuld indirectly provide surveillance cf the area. On this basis, it is concluded that that this issue should not constrain fuel icad or power operatien.
()B-f RN AMENT 2 L.OUISIANA P OW E A & LIG H T/ INTER-OFFICE CORRESPONDENCE Eu NIvsYdU January 21, 1965 W3P84-3185 A4.07 3-A104 T0: N.S. Carns FROM: R.F. Burski
SUBJECT:
WATERFORD SES UNIT 3 WATER SEEPAGE FROM CONDUITS AT ELEVATION -35
Reference:
Ebasco memorandum ES-10331-84 dated October 26, 1984 We have reviewed the questions raised by NUS relative to concern #19. This involved damage to cables immersed in water. In the 40 plus years of experience of the electrical engineers in our group, along with consultation from other company engineers, none have seen any damage to cable caused by simply being immersed in water. This expertence is obtained from all types of exposure ranging from power plant, to substations, and involves cables that may have been immersed in water for an excess of 20 years. This experience is backed up by the referenced memorandum.
- We do not feel,that it is necessary to even perform a cost benefit analysis, let alone a qualification program, or to reroute cable.
I If you have any questions, please let us know. ( i
.k, &
R.F. Burski RFB/PAJ/tz cc: D.E. Dobson, R.F. Burski, M.I. Meyer, P.A.' Jackson, Sue Thomas, ; Project Files, Administrative Support (2) l I
- h. ~
OCT291984 MEMORANDUM ES-10331-84 October 26, 1984 1 To: D. Dobson gIj-(/~f55 From: M. K. Yates g' ;
Subject:
LOUISIANA R & LIGHT COMPANY. { WATERFORD SES UNIT NO. 3 WATER SEEPAGE FROM CONDUITS AT ELEVATION -35 FT In respone to questions raised by NUS relative to concern no. 19 Electrical ' Engineering has revieved the available information on the water seepage problem affecting conduits and low-voltage cables at the Waterford Plant.and offer the following comments: 1.. Water ingress into the conduits is most probably taking place at conduit joints. Because there may not be a tight bond between the conduit surfaces j and the concrete surrounding it, underground water which filtered into the i basemat through the hairline cracks* has a natural path along the external surface of the conduits. ,
. 2. Due to water ingress, the cables in these conduits will be immersed in water. The insulation of the cables involved has been qualified by extensive type testing under water immersion for up to 2 years, with water at 90 degree C and 600 V applied potential. These tests have shown the excellent water stability of these insulations (with no f ailures occurring during the up to two-year testing). It is to be noted that the affected cables wil'1 be exposed to substantially less severe conditions than those of these teats, i.e., water temperatures well below 90 degree C and operating voltages of 480 V or less.
- 3. Because of the exec 11ent results of the water immersion tests, it is conservatively estimated that the service life of these cables under the installed conditions will exceed 10 years as compared with their 40 year service life expectancy under normal installed conditions.
- 4. It should be noted that no qualification program presently exists in the industry to predict incipient cable failure of the remaining life of cables installed under water. However it should be pointed out that cables of similar construction and materials have been installed immersed in water in the Gulf area for more than 15 years without a failure.
l l
----1-. - --------x-- - - - - - _ - - -- . . _ _ - - - -u----m
a- _2 Based on the above, it is recommended that a cost benefit analysis be performed to establish the merits of a qualification program versus a cable rerouting program. JTG/ds cc: K. Hancock J. Montalbano M.W. Migliaro P.A. Nobile A. Carolmagno C. Ruiz A. Schildkraut R. Esnes J. Houghtaling R. Burski (LP&L) J. Grillo . e l l
sue = Arr^cuneur 3_ St Lucia Unit 1 FSAR FSAR Appendix 3A Section D " Underground Cable Qualification for" Service in Potentially - Submerged Environment" l19 pages I, i l 4 l' ) l 1
.I i ~i i
f
-_._t--___________________________________________.__
_____.__,_________1,__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ __j
v vv v - ww-vvv v-v*--vv N/ Q O O e UU t/ r0 Q U e %) Ve 8 h D. UNDERGROUND CABLE QUALIFICATION FOR SERVICE IN POTENTIALLY-ESHERCED ENVIRONMENT All plant cables, including those fer Class IE service, are suitable for service in vet or dry locations and are tested in accordance with Insulated Power Cable Engineers Association. (IFCEA), National Electrical Manufacturers Association (NEMA), Association of Edison Illuminating Companies (AEIC) Standards, and purchase order specification (as appli-l . cable) to insure their suitability for normal, (dry, alternately wet and , dry, and subserged) conditions of service in tray, conduit and underground installations. Class II cables are not directly buried or installed for submarine service. These cables are installed indoors in cable trav or conduit and outdoors'in underground duct banks ,The underground duct banks are installed above the normal water table. Concern has been expressed by the Regulatory Staff as to the compatibility of St. Lucie underground cable with the service environment, i,s., have the cables been adequately qualified for the conditions they might experience during their service life. The propriety of utilising the cables for this service has been reevaluated and the suitability for the service environment raaffirmed. Basically, the environmental qualification study consisted of three parts, namely,
- 1. Insitu Cable Experience The underground cable system of Miami Beach closely duplicates the St. Lucie service environment. The performance of cable on this I system was reviewed to det6rmine whether or not the dry, alternately, .
wet and dry, and flooded conditions experienced by the Miami. Beach l cable results in cable failure due to deterioration. This review, which included all failure data since 1945, indicates that there has "never" been an electrical deterioration failure of any kind in the Miami' Beach system.
- 2. Experimental Confirmation l
Four cable manufacturers have performed independent tests to demon. strate that the modern insulations utilised for St. Lucie cable have superior resistance to electrical deterioration than the insu- ' lations that hava performed so well at Miami Beach. Each cable manufacturer has independently demonstrated that the modern insula-tions are superior. Thus, the St. Lucia cable itfs will undoubtedly exceed that of the Miami' Beach cable.
- 3. Expert Opinion IVo nationally recognised cable experts have reviewed the cable for the intended service environment. Both conclude that it is adequately qualified for the service environment.
3A-14
- __-- --_ __ ~_
The qualifiestions of the cable expet es are provided as Tables .3A-D1' and .1A-lo. Details of the Miami gesch service experience and the testing conducted by the' cable manufacturers is provided infra. The conclusion reached is that the cables have been adequately qualified for the service environment'. In this regard it is noteworthy that the new insulationsexhibit stable electrical the propensity of the properties (power f actor), i.e., test data indicate that Because of this,electri-modern cable is to stabilized electrical properties. cal stability tests in 90*C water were terminated af ter 36 months. Insi_tu cab _ie Experienca Confidence in the ability of the underground cables at St. Lucie to operate satisfactorily in a potentially. submerged or. alternately wat and dry env1ron-ment can be demonstrated by extensive Florida Power and Light experience with underground duct and manhole installations in which cables have.actually been subjected to environn:ntal conditions simi.lar to or more severe than those that can be expected at St. Lucia. The liiami Beach underground electri-cal distribution system installation provides an excellent expertened base to establish a confidenco level for the Sts Lucie installation. The ' oases being that the Miami Beach system providest a) An extensive and long established underground distribution system with a systemissd recorded failure data base for the last thirty years. b) A natural environment which includes his' htemperatures, decaying ' ( vegetation, torrential rains, and hurricanes. c) An underground distribution system that sees the potential wet and dry conditions, or combination thereof, that are character-istic of. the St. Lucie installation. Part of the system is con-tinually submergeds part is installed near the normal tide line and hence, is wet twice a days and the remainder is installed above the normal high tide line and hence is subjected to occassion-al submergence f rom rain or hurricane driven sea water. It is noteworthy that in the Miami gesch system no attempt has ever been made to control the water level in any underground installation except as required-to f acilitate personnel access into the underground system and manholes. In addition one of the cable experts, Mr. W. A. Thus has been personally involved in the Miami Beach system. Being directly involved in system development: and normal occurrence evaluation for 25 years, Table 3A-D4 provides a chronology of 600 vole cable installation on Miami Beach by type since 1957. Table 3A-D3 provides the cable feet - years experience record for this cable since 1957. Data on cable feet prior to 1957 is not readily available. However, cable f ailure records since 1945 are available.
~
thus any correlation of failure data and cable feet - years based on Miami geach' data would be conservatively biased. A .. b . ' >,l, , '
. t . ,
3A-15
VWWW QWWOOD UV OWW2 WO OO OUUW OOV OU rp W y l Since 1956 Florida Power and Light has insta,11ed over 600,000 feet of cable ranging in size from (12 AWG to represents This 1500 kemilover butyi rubber insulated 7,000,000 600 volt cable feet-years rated cable on Miant Beach.Many more cable feet of cable were installed prior to 1956. of experience. i Over the facilities service lif e the St. Lucie underground installation 7,000,000 cable feet - represents but a small fraction of this Breater than l years insitu experience base. A review has been completed of all cable failure records since 1945 along with confirmatory interviews of supervisors, splicers, and engineers in the Miami Beach district, The results indicate an extraordinary experience record. There has "never" been an electrical deteriorationIt failure is alsoofnote- any kind on these cables or the splices associated with them. ) worthly that there is no known data to contradict this outstanding reliability record from any other utility that has utilized butyi rubber insulated cable in underground installations. It must be noted however, that f ailures do occur. These have occured due to mechanical damage (e.g. damage during l installation) that are not germane to a discussion of environmental qualifi- i cation of cables. Preservice testing procedures at St. Lucis detect I the presence of such mechanical damage prior to plant operation, thereby obviating the need to consider these failures during inservice operation. l Experimental Confirmation Butyl rubber insulated cables have had excellant in-service experience as indicated by the Hiami gesch underground distribution service record. How- , ever, technological advances has made availabis newer superior insulating ; materials, namely, crosslinked polyethaline (XLPE or CLPE) and ethylene 1 propylene rubber (EPR or EPM). These newer insulations have been utilized widely by the industry for the last five to tan years. They provide superior l performance to that experienced with butyl rubber. , 1 Since an acceptable inservice data base is available for the butyl rubber insulation, qualification of the newer materials is readily accomplished through laboratory testing. The new materials and the butyi rubber must undergo appropriate identical tests. To qualify the new materials ic is l necessary and sufficient that these materials have resistance to electrical deterioration equal to or greater than butyi rubber. The test results clearly indicate the superior characteristics of these new materials. Whatever tests these never insulations are subjected to as a means to deter-mine their electrical stability in wet or dry locationa, the new crosslinked polyethylene or ethylene propelene materiala out perform the old butyl types Florida Power and Light has in service on Miami Beach by 2 to 6 times or mera. Typical evidence in regards to the superior perf ormance to be expected with crosslinked polyethylene or ethylena propelene rubber is provided as f ollows;
- 1. Excerpts concerning moisture resistance from cochnical paper T 74 044-4 presented by Okonite et the IEEE Power Engineering Society Winter Meeting in 1974. (Attachment 1 to Section 3A Part D) 3A-16
- 2. The Okonito Compary (Attachae2t' 2 to 8::etion 3A.Part D) a) Aa Okonite Company letter of November 27, 1974 to sbasco services, Inc.
b) Oraphical representation' indicating stability of various insulations in water at a temperature of 900C.
- 3. General cable Corporation (Attachment 3 to Section 31 Part D) a) Test results of long ters immersion in water at.90cc.
b) Aging evaluation of General Cabis N r inside and outside containment.
- 4. Cyprus Wirs and Cable Company (formerly Rome Cable)
(Attachment 4 to Section 3A Part D) a) Cyprus Wire and Cable company letter of November 26, 1974 to sbasco Services, Inc.
- 5. Raychem Corporation (Attachment 5 to Section 3A Part D) a) Rayches Corporation letter of November 27, 1974 to sbaseo Services, Inc.
The following provides a summary of the cable tests and resulte thereof_: [ 1. Holsture resistance (See Attachment 1) , In the 1950-57 era IPCFA developed a 16 wook test procedure based on . a continuous immersion at 500C while under 600 volts de to provide I a seens of assessing the effect of moisture resistance on cable ! life. Today, modern insulations can be immersed at 750C, under l the same de potential, for 1 1/2 to 2 years, or more. Cables at the j Indian l first generation reactors (1957 vintage), i.e. Shippingsport,due Point and Peach Bottom, have not asperienced insitu problema to ) aoisture.- This experience adds to the large base of successful cable performance, Whereas, the test data provided on modern : insuistions reaffirms the high confidence level associated with l these new asterials.
- 2. The Okonite company (See Attachment 2)
The Okonita Company compared the performance of the older butyi insulation, used so successfully in the Miami Beach distribution l-systes, with the performance of the crosslinked polyethylene and ethylene propelene rubber insulation, used on the cables installed at St. Lucia. The tests were made on small sample cables at 9000, which is the marinum operating temperature for the insulations. Power factor, which is a measure of . cable losses, is used to indicate insulation integrity, and is one of the best methods to deterrine destadation due to soisture. The results indicate that at the start of the test the butyl insutstion has a power fac..ne - sicut .e !=ts. ant. (See Pisure 3A-D1.) After 12 months the power factor has increased to 30 percent at which point the insulation is judged to be unfit for 3A-17 Am. 6-7 /81 ,
FPOM EBASCC STUART c. 3,1939 ggigg P. 6
)
service. In contrast, the " Natural CLPE", the insulation used at St. Lucie Unit 1, has an initisi power factor less than one percent. As the test r .tinued, the power factor decreased slightly to about one half of one p rcant. Af ter 36 sonths the test was discontinued due to the remarkable electrical stability of the insulation. The tests show that for EPH insulation (more generally called EPR) the l initial power factor was 2.7 percent which decreased to about two percent ' and remained stable. Again the test was discontinued at 36 months. These tests indicate that the crosslinked polyethylene and EPA insulations have lower initial power factors than the butyl insuistion. Both the crosslinked polyethylene and the EPR insulations performed without deterioration for 36 months. With the butyl insuistion, the power factor increased continually and reached an intolerable value within 12 months. .In show superior conclusion, the crosslinked polyethylene and EPR insulations performance for at least three times as long as the butyl insulation, which j gave satisfactory service in the Miami Beach distribution system.
- 3. General Cable Corporation (See Attachment 3) l l
)
A test performed by General Ca'ule provides an indication of ions life The cabis samples were 1 espected for crosslinked polyethylene insulation.The life ef *,he sample cable ismersed l energized continuously at 600 volts. in 900C water was 951 days or approximately !. 6 years. It must be noted j This is well l that failure occurred At a test overvoltasa of 1200 volts a-c. above the operating voltage (480V) for i.nis class of cable at St. Lucie. l l An evaluation was performed and documented by Florida Power & Light Power Plant Engineering (EPO-86-805-E-2). Utilising the LOCA profile of the Franklin Institute Test certified by General Cable for the cable supplied for 8t Lucie Unit 1 by their letter of May 3,1973 (found in St Lucie Unit 1 Document Package 8770.A.451 8.0, Section 3), it can be demonstrated that the cable is qualified for 40 years of containment service plus the required Design Basis Event (DBE) service. The same documentation also demonstrates that the cable is qualified for 40. years of Steam Trestle service plus the required Design Basis Event (DBE) service.
- 4. Cyprus Wire and Cable Company (Formerly Rose Cable)
(See Attachment 4) Air oven tests were performed by Cyprus on crosslinked polyethylene insulation. These tests were made at 13500, 1500C and 1750C. The Arrhenius plot of these points indicated life of 50 years (See at 9000,)the Figure 3A-D2 . rated maximum insulation operating temperature. , l Air oven tests and Artic ius plots represent one method used by cable manufacturers to evaluate cable insulations. These tests supplement tests in water, examples of which have been previously discussed. The objective l is a balance of electrical and physical properties which will insure long life under operating conditione. 5.. Roychen Corporation (Attachment 5) t Mt- Raychem has tested sample cables continuously immerned in water at 7500 in excess of twenty sonths with no failures. This data confirms the test j results obtained by the other cable vendors. l
.)
SA-18 Am. 6-7/87 l
.- __ _. - .- - n , 'I Qng.lusion In summary, the past performance of similar cables in an even more severe environment has chown their service reliability to be unaf fected by any form of water for over 25 years. The newer insulations have been evaluated All accelerated tests indicate that they outperform those cables that have been used at Miami Beach. ]
1 The foregoing experience and testing demonstrate ipso facto that the cables j installed at St. Lucia will have greater than a 40 yeaf service life in a i dry, alternately wet and dry, or potentially submerged environment. TA3LE 3A-D1 QUALIFICATIONS W. A. Thue William A. Thue is the System operations Engineer - Underground for Florida Power & Light Company. He has been employed thete since 1946 in varions pos-itions in the Engineering, Construction and Operating Departments. His present assignment involves the responsibility for all underground transmission. 3 ( distribution and power plant cables and associated squipment as staff to the l Group Vice President of Operatione. , He is presently Vice Chairman of the Insulated Conductors Committee of the Power Engineering Society of the Institute of Electrical and Electronic l Engineers. He is the immediate Past Chairman of the the Cable Engineering i Section of the Association of Edison Illuminating Companies which is the source ; of cable standards for all paper insulated cables in the United States; He also serves as their Chairman of Extruded Dielectric Cable Standards. In this field, he also serves as the Chairman of the Joint Association of Edison 111u-minating Companies-Insulated Power Cable Engineers Association Committee for Extruded Dialectric Power Cables. He is a member of the U. S. National Committee for High Voltage Cables of CICRE (International Confarance of Large Electric Systems), Committee C-8 (Electric Cabiss) of American National Standards Institute, of the Electric Power Research Institute's Rossarch Project 78 for High Voltage Cables, and the Task Force for Power Cable Ampacities. He is the author or coauthor of technical papers such as the " Shielding Performance of Power Cables", " Thermal Backfill for Transmission cables". and " Improved Low Volta"gebliirect Burie'd Cables". 3A-19 _ - - _ - _ - - - - -- ___ )
3 06- ATTACHMNT Y Excerpt from paper T74 044-4, " Class IE Cables for Nuclear Power Canerating Stations", by E. E. McIlveen. V. L. Garrison,' C. T. Dobrowski. Hoisture Resistance Moisture resistance is a major factor in determining the normal life of a - solid dielectric insulated conductor. It has become traditional to gain assurance of long life performance by totally immersing a #12 or 14 con-ductor insulated with a 45 mil wall of dielectric in water at an elevated temperature to accelerate the deteriorating effects of moisture. Moni-toring. the electrical properties then provides an indication of long term behavior. In the 1950-57 era with service sained experience that negative de potential presented the most severe condition, IPCEA developed a 16' veek test procedure along these lines based on a continuous immersion at 50' C while under 600 volts dc. At this time, more than sixteen years later, new generation moisture resisting insulations of similar geometry can be continuously immersed at 75* C while under the same de potential, and survive from 1-1/2 to 2 years, or more. This 1s at least 5 times longer and at an effective camperature acceleration rate of 6 times greater-than anticipated by the IPCEA procedure. Since insulated conductors of the 1957 vintage dielectrics installed at Shippingsport, Indian Point and Peach Bottom 'among others, have not. experienced distress due to moisture,
~
it can be reasoned that control cable insulations now specified which have the capability of withstanding total immersion at 75' C under 600 V'dc as discussed herein should develop the designed life of the cable plant. Fig. 1 presents data for a 45 mil wall of an ethylene-propylene base insulation conductor, and Fig. 2 illustrates the electrical behavior of a - composite wall' composed of 30 mils EP base plus 15 mils neoprene compound.
~
Reference to Table I discloses similar data for an ethylene-propylene base dielectric and also a flame resistant cross-linked polyethylene compound (FR-CLPE), but at 90*C continuous water immersion while under 600 V ac . potential except when percent power factor (I PF).and the specific induc-tive espacity (SIC) are being measured at 40 and 30 V/ mil ac. Following each test measurement the specimens were subjected to a 5 minute withstand test at 110 V/ mil. The specific insulation resistance (SIR) were made at 500 V de while at 90*C. The difficulty of predicting long term performance based on the customary 2 week test data is obvious. It may be of interest that the time to failure for a particular specimen is a complex f unction of several variables, one of which. is 'the degree of mechanical perfection of the dialectric wall. Failure is often sudden with little or-no forewarning, and occurs when the cable is undergoing 60 cycle' power f actor 110 and capacity measurements, or during the subsequent withstand at V/ mil. Tig. 3 not only shows the SIC values for an ethylene-propylens base insula-tion during 'a long term continuous water immersion study, but also the accel-ersting effect of temperature as manifested by a change in the 60 cycle cap. acity. The 142*C/42 psig steam autocleve exposure further accelerates the increase event, in the SIC value but could change the reaction mechanism. In any time scale.if plotted on Fig. 3 the end point is still some two years out on the 3,A-23
- _ _ - - _ _ _ _ - - ~~ ~ '
S U B - p n M in EN T 5 TH _, Pe:t Omo Cu 340 g Romney, New Jor emy 07448 201-82t5 0300/Cabte: Okanna COMPANY no w .r_2 4 C Mr. Dantry Carter Louisiana Power 6 Light Waterford 3 SBS Highway 18 Trailer 95 Einona, Louisiana 70066
Dear Mr. Carter:
1 W3B 86-0495 l A4.10/P 2.58 Okonito Okolon Cable Waterford 3 laai_siana Power 6 Light Your letter of Novmber 10, 1986 requests infomation on 01onite/0kolon 600 vnit 1 power and control cabics with reference to possible water submersion. , ) It is well known and established that water has deteriorating effects on all in-gulations. Theso deteriorating effects are evidenced by stxa things as loss ; of dielectric strength, rise in power factor and dielectric constant and lower- ' ing of insulation resistance. The rate at which these deteriorating effects occur is the key question. The rate is dependent on temperature, i.e., higher water temperatures increase the rate. , The rate is also dependent on other things such as voltage stress, wall thickness, l and the presence of discharge. High stress increases the rato; thinner walls at J equal stress deteriorate faster than heavier walls; in the case of 600 volt cable the discharge question is moot, uver the years as insulation technology has inproved more resistance to water has been built into insulations. Attached is a chart showing the behavior of 4 power factor vs. time of various insulations in 90*C water. ! As can be seen modern Chonite EP out performs butyl and oil base natural rubber by significant margins. Both butyl sai oil baso natural rubber insulations have been in service in excess of 20'-30 years in wet locations and submarine applica-tions. 'Ihe suporior behavior of EP in highly accelerated testing leads logically to the conclusion that it will behave far better in wet locations or submarine envirornnents than the butyi or oil base insulations. We, therefore, concludo that EP/Hypalon cables at Waterferd will perform satisfactorily if submerged in water in underground ducts during tho 40 year design life of Waterford Unit 3. Very truly yours, THE OgCNITE CO@ANY M JRC/ row ,T/ E. Cancelosi Attachment for Staff Electrical Engineer cc : Dr. J) S. Lasky
S UB- &Trn cr/meur (.; 0KOGUARD ' Okoguard has Excellent () Stability in Water - Assurance of long term moisture stability is essential indication of long term characteristics. Based upon ex-when selecting an insulation for many applications. It perience with cables in service over many years, it is is fairly common that a power cable be required to oper- determined that insulations which withstand total water ate in an attemately wet and dry e1vironment. The immersion at 90C for an extended period of time will method used to determine the long term water stability have a life in excess of a generating station's designed of a cable is one in which a sample, insulated with a thin life in an environment of 100% humidity. As shown in wall dielectnc, is immersed in water at an elevated tem- the chart below Okoguard's long term water stability perature to accelerate the deteriorating effects of mois- closely approximates that of cross-linked polyethylene, ture. Monitoring the electrical properties provides an both of which are far superior to other solid dielectrics, . ) I
- ., j ! {
ELECTRICAL STABILITY ' l O 90' C WATER 14 AWG. WIRE .047" WALL 80V/ MIL ELECTRICAL STRESS i I 40 40 ;
. . 1 )
0 3A
,,ppar ssMM"=
pser ssM""####' SIC OKOGUARD , I i w O O SIC XLP k mag.wmagasmMacmMMMMcMMMMMMM , c 2.0 2.0 $ , e 10 k 1.0 P.F. OKOGUARD P.F. XLP
- v. O 2 4 6 8 10 12 i
MONTHS f 5
APPENDIX E
~
i l 1 1 I I i j EVALUATION OF OK0 NITE V-TYPE SPLICES CONSTRUCTED AT WATERFORD 3 ; i Prepared By: Martin L. Raines April, 1989 i
TABLE OF. CONTENTS l SECTION TITLE I Introduction II Similarity Analysis III Failure Modes & Effects Analysis IV Test Summaries ; V Justification of Imersion Test ! Results on Unaged Sample l 1 VI Test Data Summary i VII Analysis of Test Results ; VIII Cable and Splice Material ! Compatibility ! l IX Radiation Exposure of Splices Submerged in Containment X Conclusion l XI Conductors Used In Okonite Parallel V-Type Splices ATTACHMENTS TITLE 1 Okonite Inline Splice LOCA Test Results 2 Okonite Inline Splice Immersion Test Results l l
)
- 1. INTRODUCTION
'This evaluation is intended to demonstrate that the Okonite parallel V-type splices without the T-95 insulating tape applied between the two spliced conductors, as presently constructed at Waterford 3 SES'(W3),
will perform their safety related function under design basis accident conditions including long term submergence. The parallel V-type splice is constructed by crimping or bolting terminal lugs back-to-back, such that the joined conductors form a "V". A nuclear grade cement is applied to the joined conductors followed by wrapping T-95 insulating tape in half-lapped layers. The " taped V-type '. splice" is constructed by wrapping the individual conductors prior to wrapping both-conductors. The "untaped V-type splice" is constructed by applying the T-95 tape over the two conductors without any tape applied over the individt.a1 conductors. Af ter the application of the T-95 insulating tape, both. type splices are then covered with No. 35 jacket tape in the same fashion. Okonite splicing tapes, T-95 and No. 35, manufactured by the Okonite-Company have been extensively used in the nuclear . industry as insulation. for field splices. When this type of splict is used' to terminate a safety-related circuit that is located in- a harsh environment, it _ falls under the requirements of 10CFR50.49 and must be environmentally qualified. Okonite instituted a qualification program to demonstrate the adequacy of their product to the industry. Their tapes were tested-in a single, straight-line splice configuration. This program was successful in-demonstrating the qualification of the Okonite tape _ splice materials and the tested configuration. The report for this test program has been reviewed and accepted as adequate to demonstrate qualification for the accident conditions at W3. j I Okonite did not test a "V-type" splice configuration. The V-type splice ; is used in areas where there is not enough room to' fit a straight-line ! splice. V-type splices have been tested, however, by various utilities ! to demonstrate adequacy of their installed configurations. - l Five test reports have been reviewed and accepted as adequate (either , alone or combined with other tests) to demonstrate qualification for the accident conditions at W3. These tests are documented in the following 4 reports: l
1 I
- 1. " FINAL TEST REPORT ON OK0 NITE TAPE SPLICE INSULATION FOR POWER ,
I AND CONTROL CABLES AND ROCKBESTOS PYROTROL III AND FIREWALL III CABLES"; REPORT NO. PEI-TR-842900-1 FOR NEW YORK POWER AUTHORITY.
- 2. " FINAL REPORT ON THE QUALIFICATION OF PARALLEL SPLICE INSULATION SYSTEMS USED ON POWER AND CONTROL CABLES IN THE JAMES A. FITZPATRICK NUCLEAR POWER PLANT"; REPORT N0.
PEI-TR-84-704-1 FOR NEW YORK POWER AUTHORITY.
- 3. " QUALIFICATION TEST PROGRAM ON SPLICES FABRICATED WITH 3M l
SCOTCH SUPER 33+ VINYL PLASTIC ELECTRICAL TAPE AND OKONITE l SPLICING TAPES NO. 35 AND T-95 FOR THE ALABAMA POWER COMPANY FOR USE IN THE FARLEY NUCLEAR GENERATING STATION"; WYLE REPORT NO. 17947-01.
- 4. "LOCA QUALIFICATION REPORT FOR OK0 GUARD INSULATED CABLES AND T-95 & NO. 35 SPLICING TAPES"; REPORT NO. NQRN-3.
- 5. OKONITE LONG TERM WATER IMMERSION TEST, DATED JANUARY 10, 1983, i SUPPLEMENTED WITH REPORT EXCERPT FROM ILLINOIS POWER COMPANY, TRANSMITTED MARCH 7,1989.
The reports of items 1, 2 and 3 are primarily used to demonstrate that the tested splices do not possess failure mechanisms (i.e., leakage path) by which moisture intrusion can cause loss of operability. They are not used as the sole means of environmental qualification. ' As a result of a review of device location in relation to the flooding level inside containment at W3, it was determined that there is a potential for submergence of some devices and their sp' ices. In order to ensure adequate demonstration of qualification, the above test reports were re-reviewed with specific attention given to submergence qualification. This approach to demonstrating qualification for long term submergence will be performed in the following sequence:
- 1. Establish the similarity between the "untaped V-type splice" as they exist at W3 and the qualification test reports identified in Table 1 to the taped V-type and in-line Okonite splice configurations.
- 2. Perform a Failure Mode and Effects Analysis on Okonite splices with particular emphasis given to the design basis accident conditions of LOCA and long term submergence.
- 3. Summarize and evaluate the tested splice configurations and the environmental parameters under which the splices were tested for each of the test reports under consideration.
- 4. Compile all the significant test and configuration data into a single table for analysis.
- 5. Use the data within the table to demonstrate that the Okonite untaped V-type splice is capable of maintaining its safety related electrical function under design basis accident conditions including long term submergence.
1 II; SIMILARITY ANALYSIS' Tape T-95 is described:in Okonite product data as "...an ethylene-propylene based thermosetting compound..." which is known as A !'self-fusing" tape. This- means that it fuses with itself and molds to its substrate and cures as it ages. The T-95 insulation tape" cures to form a mechanical seal around the conductors similar in concept to that of a static pressure retaining 0-ring.where the seal does not rely on chemical bonding with the contact surfaces but compresses against them j to form a watertight and moisture proof barrier. Application'of the T-95 ; tape also aids in the creation of the mechanical seal, in that the 1 installation wrapping requirement stretches the tape so that a tight seal is formed around the conductors. The splice configuration .under consideration is referred to as the ur, taped V-type splice where the conductors are' connected- by a crimp .. connection or a bolted pair of ring tongue terminals. A cement is used to cover the conductor connection prior to the' application of the T-95 insulating tape. The difference between the untaped and taped V-type splice is that the taped V-type splice has the half-lapped layers of the
- T-95 insulating tape *tpplied to each of the individual conductors prior to the application of.the I-95 tape over both of the conductors. The-purpose of taping both conductors is to ensure that a good mechanical seal exists around the conductors and to eliminate voids or pathways where moisture intrusion could potentially penetrate between the conductors. Upon completion of the insulating tape, both splices are .then-covered with the Okonite No. 35 jacket tape in the same half-lapped. .
fashion until a minimum of two layers are obtained. ] It is postulated that the configuration objections that the untaped' V ) splice may have, as compared to the taped V splice, is' that the application of tape only around both conductors will result in a. void from the joined conductors to the outside. environment running parallel and adjacent to where both conductors touch. 51milarity can be established between the untaped and taped V splice configurations if the self-fusing tape properties of the T-95 insulating , and No. 35 tape jacket results in the elimination of voids by which moisture intrusion may result in splice failure. To demonstrate that the taped and untaped V-type splice configurations l are similar in their capacity to provide a moisture seal and prevent the intrusion of steam or water, four splice samples were constructed using the Okonite T-95/35 tape with two splices constructed as a taped V and the other two splices constructed as an untaped V. . The simulated field conductor was a #12 AWG single Okonite conductor and the simulated instrument pigtail was a #22 AWG tefzel conductor obtained , from a Barton 763 pressure transmitter. This splice configuration can be i considered as conservative since the simulated field conductor used in both constructed splice models was one size larger than the plant , configuration which uses a #14 AWG conductor. The larger conductor in ' the untaped V-type splice will make it more difficult to fill all voids ; within the splice if the insulating tape self-fusing and molding i properties are less than adequate. No model of the inline splice was constructed since; its interior construction characteristics consist of only a conductor surrounded by a ring of T-95' insulating. tape followed by i No. 35-Jacket tape. .;
)
i i _ - - _ - _ - _ - _ _ - _ _ _ - _ _ _ _ _ _ - l
The taped and untaped V-splices were next cut into 3 sections at 90 degree angles or perpendicular to the conductors to examine the splices for the potential presence of voids. The interior of each splice was examined in sufficient detail to determine if the untaped splice could be considered similar to the' taped splice.- The physical evidence of the two splice sample cutaways indicated that neither configuration contained voids or physical paths by which moisture intrusion could occur. All conductor surface areas of the untaped V splice were sufficiently covered and sealed as in the taper' . #ce sample to establish that the untaped V splice sample coulo be considered > similar in construction to the taped V splice sample. The inline splice is very similar to the taped V-type splice in that the l angle in the V can be considered as 180 degrees. The inline splice is ' constructed by crimping the conductors in a butt splice connection or bolting the terminal lugs together in such a manner that no V at the joint is formed. The application of the cement and T-95/-35 .. insulating / jacket tape is the same as the V-type splices and is also applied in such a manner as to eliminate voids and paths for moisture intrusion. i 4 It is therefore concluded that through physical evidence that the untaped V splice is similar in construction to the taped V splice for the ; following reasons: Materials used in both splices are the same. There was adequate T-95 coverage around all conductors. ! l There was adequate mechanical sealing around all conductors.
]
There were no voids on conductor surfaces or within the I splice material. 1 It also can be concluded that the taped V splice is similar to the inline splice in that the only difference between the two is the angle at the V, but this is not impoi. ant to the splice material and splice sealing capabilities for the properly constructed splice. As a consequence, the i untaped V splice can be considered similar to the inline splice as long i as the materials are the same, there is adequate coverage and mechanical sealing and there are no voids within the splice'for moisture intrusion to occur. Section IV, with primary emphasis on the first three reports, will demonstrate that in addition to the physical characteristics being the same for the untaped and taped V type splice, test evidence will also demonstrate that the untaped V splice has comparable or superior insulating characteristics to that of the taped V splice. 1
I III. FAILURE MODES AND EFFECTS ANALYSIS i This section addresses the Failure Mode and Effects Analysis for splices. j Since splices and cable' insulation perform the same function and'are " exposed to the same environmental . conditions,- the FMEA for cable splices is also applicable for cable insulation. Splices are used in Nuclear Power Plants to join two. conductors .together l to establish circuits between field routed cables and electrical l equipment or instruments.- Improper circuit' operation .at the splice-occurs when inadvertent contact is made between the connected conductors and ground. This will result in a ground fault for a power circuit or inaccurate instrument readings due to leakage current. These identified l circuit anomalies will be caused when: improper conductor splices are constructed q l splices a'e damaged after installation j splices are exposed to moisture which ' penetrates and makes ) contact with the conductor and ground. The first two circuit anomalies can be avoided by following proper installation procedures and are not'the result of a design basis accident. The remaining anomaly for splices is caused by exposure to moisture in the form of' a vapor or a liquid. These potential failure - modes are:
- 1. Moisture intrusion through ' voids .in the splice which permits a circuit to be formed between the conductors within .the splice l and ground.
- 2. Moisture diffusion (absorption) through the surface of a splice which results in unacceptable values of insulation-resistance.
There are three parameters by which moisture intrusion or diffusion into a cable / splice sample configuration can be. detected by testing, which are: 1 Insulation Resistance (IR) i Specific Inductive Capacity (SIC)
]
Power Factor (PF) ' IR measurements can detect moisture intrusion through voids in insulating materials (cracks or open pathwayt) which results in catastrophic ~1oss of I IR values. It should be noted that repeated high voltage withstand tests on an insulating material can lower IR values or..cause an insulation failure within itself due to high potential stresses which weaken or break the chemical bonds. ' Moisture diffusion within a cable insulation can also be detected with decreasing IR values since there is an increase of easier and shorter conduction paths for current flow under the applied voltage stress.
SIC is used to determine if moisture diffusion or absorption into the insulation has taken place since this value is essentially a dielectric-constant. 'If moisture diffusion has taken place, the ability of an insulator to store a charge has been impaired and the dielectric constant 1 or SIC will lower. Since this constant is based on a comparison with- i air, the SIC of air is therefore 1. If water is used as a dielectric between two charged plates, this would result in a short circuit with no ability to store a charge and the SIC value would be zero. Therefore, the higher the SIC, the better the dielectric and the presence of moisture is J less. Power factor (PF) is expressed as a percent and is a measure of the power l losses that occur in a cable as the result of the storage of energy in ! the cable insulation. This value is sensitive to the moisture content of { the cable insulation and will increase as the moisture content of the ; insulation increases, with 2% considered as the maximum acceptable value. .] The effect of moisture diffusion or absorption into a cable splice would be: Gradual losses in IR Lower SIC . Higher PF The effect of moisture intrusion into a cable splice would be a , significant loss of IR values. Significant losses in IR with adequate
)
values of SIC and PF can indicate damaged insulation due to repeated high j potential withstand tests. 1 The ability of a splice to resist moisture intrusion and diffusion'can d be demonstrated by testing. Representative splices subjected to a high temperature and pressure moisture spray can detect the presence of voids within the splice. A long term submergence test in water can determine-the splice material's ability to withstand moisture diffusion. Long term submergence will also demonstrate the capability of the splice to form a mechanical seal and resist the intrusion of moisture. It is' considered that a LOCA/MSLB at a high temperature and pressure is an adequate measure of a splice's ability to withstand moisture intrusion and diffusion whether the moisture is in the form of a vapor or a liquid. lhe rational is that steam at a given pressure is considered much more penetrating and carries with it the potential for inflicting much more , damage than water at the same pressure.
I IV. TEST SUtHARIES The samples tested in both Patel test reports PEI-TR-842900-1 and ! PEI-TR-840704-1 included taped "V-type" configurations constructed with 1 Okonite tapes T-95 and No. 35 in half-lapped layers in accordance with the j Okonite taping procedure. The only exception is that the tape overlay on ! some of the samples was less than that recommended by Okonite. This was j done in order to proof test less conservative constructions than those i tested by Okonite. There was a tape seal in the " crotch" (between the legs of the "V") of the splice. The samples tested in the WYLE test report 17947-01 included "V-type" configurations constructed with Okonite tapes T-95 and No. 35 in i half-lapped layers in accordance with the Okonite taping procedure. The ! only exception is that the tape overlay on the samples was as short as 1/4" (less than the Okonite recommendation and less than the Patel tests) and there was no insulation in the " crotch" of the splice. This was also done to proof Test less conservative constructions. The Okonite Report NQRN-3 performed an environmental test qualification program on the T-95/35 splice materials with a butt splice compression connector. The two tested splice samples were the inline configuration. Okonite identified in a January 10, 1983 letter to W3 and supplemented with an update from Illinois Power Company, transmitted March 7,1989, a long term submergence test on three in-line Okonite splices using only the T-95 insulating tape on Okonite cables. This test is considered a conservative representation since the T-95 splice was not afforded the protection of the No. 35 jacket tape. The following five summaries extract the significant features of each test identified in Table 1-Report #1 - PEI-TR-842900-1 The report of PEI-TR-842900-1 performed environmental qualification testing on taped V-type Okonite splices. The splices consisted of both crimped and bolted lug type connections using the Okonite T-95 insulating j tape and the No. 35 jacket tape. Additionally, the splices were i constructed with and without the splice cement and 2 variations of I overl ap. The conductor used to construct the splices was #12 AWG Rockbestos XI fE. A summary of the test splice configurations is ! presented in Table I. ; I i I i 4
I
- l 1
Table I Test Splice Configurations in PEI-TR-842900-1 i Cement No Cement Splice Overlap 1" 1" 1-1" 1-1" Crimped dutt sample numbers 1-1 2-1 3-1 4-1 Bolted Lug sample numbers 1-2 2-2 3-2 4-Z The constructed splices were subjected to 107 MRADS minimum of irradiation followed by thermal aging of 600 hours at 275 F to simulate i 40 years 0 149.8"F. The test samples we e mounted in a NEMA-4 junction ' box with a weep hole to expose the samphs to the LOCA environment, but protects them from direct impingement. The samples were energized continuously throughout the LOCA test with 80% of rated current at 696 VAC. The profile of the 0.15 gpm/ft 2water spray used during the conduct of-the test is listed in Table 2. ' Table 2 Water Spray Profile For PEI-TR-842900-1 > Time Temperature 10 to 2000 seconds 360 F 2000 seconds to 4.5 days 232 F Time Pressure
]
10 to 100 seconds 50 psig , 100 to 1000 seconds 50 to 8 psig ' 1000 seconds - 4.5 days 8 psig Upon completion of the simulated LOCA test, the splice samples were subjected to insulation resistance tests prior to removal from the test chamber. The IR tests were conducted at 1000 VDC for one minute. A sunmary of the baseline and post LOCA IR tests are summarized as in ' Table 3. 1 l i i
i i Table 3 i Baseline And Post LOCA IR Values For PEI-TR-842900-1 BASELINE FUNCTIONAL POST AGING SAMPLE NUMBERS (IR) (IR) POST LOCA (IR) l 1-1 4000 KMEG0HMS 4000 KMEG0HMS Greater Than'1 l i l KMEG0HM i 1-2 4000 KMEG0HMS 4000 KMEG0HMS Greater Than 1 i i KMEG0HM- l 2-1 4000 KMEG0HMS 4000 KMEG0HMS Greater Than 1 2 KMEG0HM. l 2-2 4000 KMEG0HMS 4000 KMEG0HMS Greater Than 1 I KMEG0HM l 3-1 4000 KMEG0HMS 4000 KMEG0HMS Greater Than 1 KMEG0HM 3-2 4000 KMEG0HMS 4000 KMEG0HMS Greater Than 1 KMEG0HM 4-1 4000 KMEG0HMS 4000 KMEG0HMS Greater Than 1 KMEG0HM 4-2 4000 KMEG0HMS 4000 KMEG0HMS Greater Than 1 KMEG0HM i A dielectric voltage withstand test was conducted on the cable splice samples at 2400VAC while immersed in water and mounted on a mandrel. The resultant dielectric current leakage on all test splice samples was less than one milliampere. One sample identifieu as 2-3 was constructed as a crimped inline butt splice. This splice had a post-LOCA IR value of 16 Megohms as compared I to the post aging value of 4000 Megohms. Although the post-LOCA value was significantly lower than the rest of the samples, it still remained above the acceptance criteria of 1 Megohm. Although the taped V type splice did not display any significant IR decrease after aging, the post-LOCA IR values did display a significant drop but maintained sufficiently high values in excess of 1000 megohms to perform its safety function. Report #2 PEI-TR-840704-1 j The qualification report PEI-TR-840704-1 subjected two taped V-type samples to a wide range of testing. Both splices were #12 AWG Rockbestos conductor with one splice constructed as a crimp connection and the other splice constructed as a bolted ring tongue terminal. ! Pre-aging consisted of 100 hours at 115 C thermal aging to simulate 40 years at 99.1 F followed by 11.7 MRADS of irradiation. During the LOCA ; test, both samples were continuously energized to 80% of rated load or 24 ' amps. Insulation resistance was determined with 1,000 VDC for one minute before and after the LOCA test. The summary of the LOCA profile, with no chemicals used in the pressurized spray, is listed in Table 4. i l l
Table 4 LOCA Profile for PEI-TR-840704-1 Time Temperature (F) ; Pressure (psig) 0 sec. 270 21 12 sec. 270 14 , 20 sec. 255 12 40 sec. 260 1 60 sec. 255 1 100 sec. 250 1 I 2 min. 240 1 3 to 10 min. 210 to 220 1 11 min, to 8 days 206 to 210 0 The effects of insulation degradation prior to thermal aging and after i the LOCA simulation is summarized in Table 5. Table 5 Effects of Insulation Degradation . Sample Baseline Functional (IR) Post Aging IR Post LOCA (IR) B 2000 KMEG0HM 2000 KMegohm 320 KMEG0HM C 2000 KMEG0HM 2000 KMegohm 260 KMEG0HM There was no post LOCA submergence dielectric withstand test performed on j the cable splice samples. It is interesting to note that in the same i fashion as Report #1, no decrease was noted in IR values before and after age conditioning although there was a magnitude drop in IR values after LOCA testing. This is not an indication of significant moisture intrusion into the splice since IR values still remained above acceptable i values and were comparable to each other. 1 i Report #3 - WYLE Report 17947-01 The ability of the Okonite untaped V type splice configuration used at W3 1 to resist moisture intrusion through the absence of paths and voids is ! demonstrated by WYLE Report 17947-01 dated October 8,1987. The report indicated that a total of 5 test samples were constructed using the T-95 insulating tape and No. 35 jacket tape. Additionally, four V type splices were tested using only the T-95 insulating tape. Sample 12.1 was aged for 110 hours at 110 C with the remaining samples aged for 112 hours ! at 110*C to simulate an equivalent 15 years at 108.4*F and exposed to 22 MRADS of radiation prior to LOCA testing. the splice samples were placed in the enclosures indicated in Table 7 which protected them from direct impingement of chcmical spray, but were open to the chamber atmosphere. The chemical spray consisted of 2500 ppm boron and sgdium hydroxide to maintain a pH of 10.7 at a flow rate of 0,15 gpm/ft within the test chamber. The peak temperature obtained during the LOCA test was 425 F at 55 psig, which decreased to 244'F and 25 psig after 167 minutes. The temperature remained constant at 245 F throughout the remainder of the 45 hour test with the minimum pressure profile listed in Table 6.
Table 6 Minimum Pressure Duration Minimum Pressure (psigl Time (hrs.) 1 ! 25 7.9 20 7.9 15 7.2 10 22 Table 7 WYLE Report 17947-01 Splice Configuration T-95 No. 35 Sample No. Overlap Overlap Lead 1 Lead 2 Enclosure Loading 1.1 1/2" 1/2" 1/0 1/0 NEMA 1 1 2.1 3/4" 1/2" 1/0 2/0 NEMA 1 1 3.1 1" 1/2" 1/0 8 AWG NEMA 1 1 10.1A 1/4" 1/2" 18 AWG ASCO Condulet 2 10.1 B 1/4" 1/2" 18 AWG ASCO Condulet 2 12.1 1/4" N/A 12 AWG ASCO Condulet 3 7.1 1/2" N/A 1/0 12 AWG Limitorque 4 8.1 3/4" N/A 8 AWG 8 AWG Limitorque 4 9.1 1" N/A 8 AWG 12 AWG Limitorque 4 The T-95 overlap indicates the distance the insulating tape extends past the conductor insulation. The No. 35 overlap indicates the distance the jacket tape extends past the insulating tape. The loading indications during LOCA testing are as follows:
- 1. Energized continuously throughout LOCA test with 632 VAC and 27 amperes
- 2. Energized continuously throughout LOCA test with 137.5 VDC and 200 mil 11 amperes
- 3. Energized for the first 60 minutes of the LOCA test with 137.5 VDC and 200 mil 11 amperes
- 4. Energized for the first 65 minutes of the LOCA test and then reenergized at the end of the LOCA test for 2 minutes at 633 VAC and 20.5 Amperes.
The splices were energized as indicated on Table 7 and were subjected to a post LOCA insulation resistance test. The results of the baseline functional, post aging and post LOCA 1R tests are listed in Table B in KMegohms.
- II -
Table 8 IR Comparison For WYLE Report 17947-01 i Insulation Resistance Post Aging Insulation Resistance ! Sample No. Baseline Functional IR IR Post LOCA IR j 1.1 2200 1900 150 ; 2.1 2800 710 150 l 3.1 4500 620 160 10.1A 2500 510 110 10.1B 2800 33 0 150 12.1 11,000 380 150 1 7.1 2500 200 160 8.1 1500 180 180 l 9.1 2000 190 180 The cables and splices all started the qualification test with similar IR values and completed the LOCA test with similar values. Although the l post LOCA functional test indicated that insulation resistance had I decreased, they remained sufficiently high to indicate that the function j of the splice was not impaired. This also indicates that-for this test ; the thermal and irradiation aging is responsible for the significant decrease in IR values and the LOCA exposure is much less of a factor, ; i thus indicating that LOCA exposure does not significantly degrade the j l splice material to an extent where continued operability is questionable. , l It is interesting to note that the last four splices which had no No. 35 1 jacket tape remained essentially unchanged as the result of a LOCA l exposure, whereas the splices with the No. 35 jacket displayed more of an IR loss between the post-aging and post-LOCA IR values but had post-LOCA IR values of the same magnitude as those four splices without the T-35 ; jacket tape. It can be concluded that the presence of the jacket tape ' will not affect the post accident operability of the splice. It also ; should be noted that the untaped V configuration performed better than Report No.1 and approximately equal to that of Report No. 2 even though l the untaped V splice was exposed to greater LOCA temperatures and p ressure, i l Report #4 - NQRN-3 Okonite Report NQRN-3, Revision 4, dated October 24, 1988, performed environmental qualification testing on two 15 foot cable sections with a T95/35 inline splice configuration on each cable. One cable was aged and ' one was unaged. The cable was a SKV rated #6 AWG with 0.090 inch Okoguard EPR insulation and no jacket. The T95/35 splice configuration used half lapped layers of the insulation tape which was 5/16 inch thick and jacket tape on three half lapped layers. The aged sample was subjected to 21 days at 150 C to simulate 40 years at 165.2 F with both samples receiving 200 MRADS of irradiation after thermal aging. 1
^
i
-i i
1 The samples were then LOCA tested while mounted on a mandrel with two initial transients of 345 F/112 psig for three hours. The remaining LOCA l profile over the 130 day test was: l Time Tempera ture Pressure ) 3 hours 335 F 95 psig 4 hours 315 F 69 psig 3 days + 9 hours 265 F 24 psig 126 days 212 F 0 The chemical composition of the spray was 3000 ppm boric acid, 0.064 l molar sodium thiosulfate with a pH = 10.5 controlled with sodium hydroxide with a spray rate of 0.15 gpm per ft . The chemical spray commenced at the start of the test and continued for 30 days with the remaining 100 days employing steam and water spray. The cables were constantly energized with 5KVAC and 80 amperes. The cables were also given a 80V/ mil withstand test at pre-LOCA, 30 days into the test and at 130 days. The tests performed at 30 days and 130 days were accomplished by immersing the 40X mandrel and cable in tap water at room temperature j and applying the 80V/ mil for 5 minutes. Both samples passed the i withstand tests. l j The results of the testing with IR measurements taken at 500 VDC were as j follows in Megohms - 1000 ft. (See Attachment 1 for the LOCA IR test i results): l UNAGED AGED Baseline 21,000 Not provided Post Thermal N/A 30,000 (Pre-Irradiation) Post Irradiation 14,000 15,000 (Pre-LOCA) Final Temperature Dwell (212 F) 57 to 99 84 to 130 Post-LOCA 5,300 4,800 ' The data provided in the test report indicated that IR values decreased with increasing temperature then increased with decreasing temperature. The aged sample provided extremely low IR values after the first day to the 29th day, at which time the test was stopped. It was determined that ' the source of the low IR values was a terminal failure unrelated to the splice or cable sample. The sample was reterminated and testing was restarted. The results of the test demonstrated that the tested splice was capable of withstanding a 130 day LOCA and that the cable and splice sample was not degr6ded to a degree that post-accident operability following a LOCA is of concern.
1 The IR data also indicates that the effects of thermal aging are insignificant and that 200 MRADS of irradiation does not degrade the ' cable / splice sample to an appreciable degree regardless of thermal aging. Also, during the LOCA test, the IR range was greater for the sample which had been thermally aged and irradiated as epposed to the sample which ) only had been irradiated. The results of the post-LOCA tests were comparable to both test samples which demonstrates that +he effects of ; testing pertaining to degradation were not significant ;umulative to i result in common mode failures. ) Report #5 - Okonite Letter, Dated January 10, 1983 Okonite perfonned a long term water immersion test as identified in their I January 10, 1983 letter to W3 and supplemented with a report excerpt from Illinois Power Company, transmitted March 7,1989. The test
]
was conducted on three samples of a 1/C #4/0 AWG Okonite EPR insulated i cable which was immersed in 90 C water at a minimum depth of 2.5 feet. ! Each of the three test samples contained a single insulation splice in the form of an insulation repair constructed of cement and T-95 insulation splicing tape. No No. 35 jacket tape was used in the tested ' i samples. The samples were subjected to periodic 40 and 80 volt per mil stress tests at the indicated immersion times. The specific insulation resistance (SIR) measurements were conducted at 40 V/ mil with SIC and PF taken at 40 and 80V/ mil. The tested cables were subjectad to additional l withstand tests of 250 V/ mil AC (42.5KV) and 750V/ mil DC (131KV) for one minute at the times indicated in Table 9. The values of SIC, PF , and SIR were provided as the average of the three cables except as noted I in Table 9. l Table 9 l Inmersion Time vs. IR Values i IR (MEG 0HMS - Immersion Time SIC 0 80V/ mil %PF 0 80V/mil 1000 ft.) i l 1 day 2.78 1.35 639 l 1 week 2.79 1.35 613 2 weeks 2.79 1.14 608 3 weeks 2.79 1.03 612 l 1 month 2.79 0.96 660 2 months 2.62 1.16 969 3 months 2.76 1.15 683 *1 , 4 months 2.79 0.88 1530 5 months 2.79 0.78 1845 *2 1 6 months 2.80 1.19 1575 9 months 2.86 0.68 2018-10 months 2.88 0.67 2037 11 months 3.03 1.14 2440
*1 One sample failed the 131KVDC one minute withstand test, however, the fault occurred 1 inch outside the splice area. This sample was removed from testing and the remaining data was based on the two intact samples. *2 The remaining two samples failed the 131KVDC one minute withstand test, however, as in the first sample the faults occurred outside the T-95 splices. The fault areas were repaired and testing continued on the two remaining samples to eleven months.
The stability of the PF, and 'the fact it did not exceed 20 along with a stable but gradual increase in SIC provides'assurar.re that moisture is not entering the splice or cable. insulation. This ale. aids to demonstrate that the cumulative effects of submergence does not degradate the cable / splice insulating' material. It can be concluded that a cable / splice configuration, which. concludes its LOCA exposure with sufficient operability margin, will not be subjected to.any significant ) cumulative degradation which wil1 result.in common mode. failures. See-Attachment 2 for the plot of water imersion IR test results. . The' presented data provides a high degree of assurance that moisture , intrusion or diffusion is not a source of common mode failures for T-95 Okonite insulating material. It should be noted'that the eleven month water imersion test at 90"C was performed without any No. 35 insulation tape protection which is applied to all V type splices at W3. Also the. failures which occurred on the samples at the 3 month and 6 month 131-KVDC one minute withstand test did not occur at the' splice but at the : i cable insulation. Since the splices in question and their cables are not subjected to such extreme voltages, the breakdown failures which occurred in testing will not be experienced by field installed cable. The cable .' did however, continue to. pass the 42.5 KVAC withstand test in all cases which is in excess of 'the rated 4.16 KV for safety related loads. It is also anticipated that the repeated AC and DC withstand tests weakened the cable insulation to such a degree that failure was inevitable; however, no failure occurred at the splice. The water immersion test also demonstrates the ability of.T-95 Okonite insulation tape and cement to provide a moisture and water resistant seal when proper contact is made to the cable conductor and original insulating material under adverse temperature extremes. l l l ; i i l l i i E_ _ _ _ _ _ _ _ _ _ . _ _ _ . _ . . _ _ _
V. JUSTIFICATION OF IMMERSION TEST RESULTS ON UNAGED SAMPLE The water immersion test was performed cn an unaged splice / cable assembly. This is not considered significant with regard to the results obtained during the water inmersion test because the splice exposure to 90 C had the effect of accelerated aging and the fact that the aged LOCA tested splice assembly of Report #4 displayed higher IR values than the new splice assembly at the start of the immersion test. If the 90 C water temperature is considered as the aging temperature, the immersion time as the aging time and 50 C is taken as the service temperature for an instrument splice, then the equivalent age of the splice at the conclusion of the immersion test can be calculated with the the known activation energy of 1.07 eV. The Arrhenius equation may be rearranged as: EA (1 - 1)] In(ts) = in(ta) -[ [K5 (Ta Ts)] where: ta = aging time = 11 months or 47.67 weeks Ea = activation energy = 1.07eV Kb = Boltzmans Constant = 8.617 X 10 ~5 eV/ K Ta = aging temperature = 90 C = 363 K is = service temperature = 50 C = 323 K Solving for ts we have 63.4 years at 50 C. It can be concluded that aging will have no significant impact on the ability of T-95 splice material to resist moisture absorption.
VI. TEST DATA
SUMMARY
l The results of the post LOCA insulation resistance tests and the long l term submergence test identified in section I are summarized as follows, .j in addition to significant test. data: Table 10 Report No. I' 2 3 4 5 , Baseline IR 4000 2000 1500-11',000 30,000 639 MEG 0HMS- ! KMEG0HMS KMEG0HMS KMEG0HMS MEG 0HMS- 1000 ft. I' 1000 ft. Post Aging 4000 2000 180-1900L 15,000 Not Applicable - , IR KMEG0HMS KMEG0HMS KMEG0HMS MEG 0HMS- ! 1000 ft. ~i Final 1R .0.016-1 260-320 110-180. 4,800 2440 MEG 0HMS-KMEG0HMS KMEG0HMS KMEG0HMS' MEG 0HMS- 1000 ft. 1000 ft. Type Splice Taped V Taped V Untaped V Inline Inline Splice T-95 T-95 T-95 T-95 T Insulati on- J Splice No. 35 No. 35 No. 35 No. 35 None Jacket Cement Yes & No No No Yes Yes DBA Test LOCA 1.0CA LOCA LOCA Submergence Duration 4.5 days 8 days 45 hours 130 days '11 months Peak Temp. 360"F 270 F 425 F 345 F 194'F Peak Pressure 50 psig 21 psig 55 psig 112 psig 26~ft. wtr.-. Chernicals No No Yes Yes No y Conductors / XLPE XLPE EPR and EPR EPR Insulation Silicone Rubber Aging 600 hrs. 100 hrs. 112 hrs. 21 days @ No l
@ 275'F @ 115 C @ 110 C 150 C '
Equivalent 40 yrs @ 40 yrs @ 15 yrs @ 40 yrs @ N/A l Life 149.8 F 99.1*F 108.4 F 165.3*F. I i Radiation 107 MRADS 11.7 MRADS 22 MRAUS' 200 MRA0S No
'l REPORT NO. 1: PEI-TR-842900 ;
REPORT NO. 2: PEI-TR-840-704-1 REPORT NO. 3: WYLE REPORT NO. 17947-01 i REPORT NO. 4: OKONITE NQRN-3 REPORT NO. 5: 0K0 NITE LONG TERM WATER IMMERSION. TEST - 1/10/83 l i i I i
VII. ANALYSIS OF TEST RESULTS The summary of the five examined test reports identified in Table 10 provides a strong demonstration of the Okonite splice ability to withstand long term submergence and maintain its design basis function following exposure to LOCA conditions. The test results of Reports #1 and 2 indicate that the taped V-type -) splice configuration substained a greater or' comparable loss insulation resistance as compared to the untaped V-splice configuration of. Report
#3. The taped V-type splice of Report.1 indicated a maximum insulation resistance of greater than.1KMEG0HM whereas the taped V splices of' Report #2 had a post-LOCA functional IR test range of 260 to 320 KMEG0HMS. . The untaped V splices of Report #3 had a post LOCA functional IR range of 110 to 180 KMEG0HMS which suggests that the untaped V splices performed better than or-comparable to the taped V splices.
The test results of Report #1, 2,'and 3 provide demonstration that the V-type splice configuration contains no voids or paths by.which moisture-under pressure and elevated temperatures can penetrate to the~ energized conductors within a splice and cause a loss of performance or a degradation of circuit performance which substantiates the. splice' model cutaways discussed in Section II. The results of Report #4 indicate that the T95/35 inline . splice configuration can withstand a long duration LOCA exposure without a serious loss of insulating function. The post-LOCA IR values of Report
#4 maintained higher IR values than that of a new submerged splice in Report #5. The data of Report #5 for the inline splice water immersion test indicates that the IR values tend to increase over time rather than decrease. The eleven month submergence test of the T-95 Okonite in-line splice indicates that there was no degradation of circuit performance while continually exposed to 194 F water at a minimum depth of 21 feet.
The data provided indicated that insulation resistance increased almost 4 ! times as compared with the start of the test. The tested splice was also conservative in that it was not afforded the protection of the No. 35 jacket tape. This long term submergence test and subsequent results provides a conclusive demonstration of the Okonite insulating materials. ability to resist moisture intrusion around its mechanical seal with the cable conductors and insulation and to resist moisture diffusion through the splice surface and into the splice material. Additionally, the SIC and PF values did not indicate any moisture diffusion or absorption into the insulating material. The SIC' continued to climb indicating that good dielectric propertie.s were maintained and .) the PF dropped or remained stable within a narrow range and never ! exceeded the maximum value of 2%. This is also substantiating evidence that no moisture absorption occurred. i i 1
i l VIII. CABLE AND SPLICE MATERIAL COMPATIBILITY The T-95 insulation tape is constructed of EPR, and the No. 35 jacket tape is constructed of neoprene. The T95/35 splice configuration has been tested with XLPE insulated cables in Reports
#1 and #2 and with EPR insulated cables in Reports #3, 4 and 5. The j data contained within these reports have no noted failures or -
anomalies due to adverse reactions. i The constructed T95/35 splice samples, which contain a tefzel insulated conductor from e. Barton model 763 transmitter, also do not display any immediate degradation as the result of any j incompatibilities with the T-95 insulating tape. Product literature i concerning tefzel states that it possesses excellent weather resistance and is inert to most solvents and chemicals that often cause rapid deterioration of other plastic materials. Tefzel is inert to strong mineral acids,-inorganic bases, halogens and metal salt solutions. Carboxylic acids, anhydrides, aromatic and aliphatic hydrocarbons, alcohols, aldehydes, ketones, esters, chlorocarbons and polymer solvents have little effect on tefzel. Tefzel will be affected by very strong oxidizing agents such as nitric acid, organic bases and sulfuric acid at high concentrations near their boiling points. Section VIII, Appendix D of the Cable Submergence Evaluation discusses the chemical resistance properties of the cable insulation and jacket materials. This section ind! cates that these materials are highly resistant and inert to a wide variety of potentially reactive agents. Based upon the information supplied in product literature and qualification reports discussed in this evaluation it is considered highly unlikely that the splice configurations that could be formed with the T95/35 tape, XLPE, EPR, neoprene and tefzel insulation would degrade due to any material incompatibilities. Cable j insulating and jacket materials are chosen because of their ability to resist attack by most chemical agents which are found in industrial applications while maintaining excellent electrical ! characteristics. With these materials possessing such chemical resistant characteristics, it is considered improbable that they j would attack each other. The various qualification test reports have not indicated any anomalies as the result of any material ; incompatibilities. I
IX. RADIATION EXPOSURE OF SPLICES SUBMERGED IN CONTAINMENT , LP&L performed a calculation (N0SA-RP-CALC 89-001) to determine the total integrated dose that cables and associated splices would be exposed to during submergence conditions after a postulated LOCA. The reactor core inventories for the individual radionuclides in this calculation are consistant with the FSAR Table 12.3A-1 and was assumed to be , 1 100% of the core inventories for the noble gasses 50% of the core inventories for the halogens 1% of the core inventories for the other isotopes The results of the submergence calculation for the gamma, beta, and total i doses in Mrads are as follows for 132 days and one year post-accident: 1 PERIOD GAMMA DOSE BETA DOSE TOTAL DOSE 132 days 22.8 7.9 30.7 1 year 29.4 12.2 41.6 The one year total integrated dose of 41.6 Mrads compares favorably with the Okonite T-95/35 splices qualified dose of 200 Mrads gamma. Since all the subnerged splices are enclosed in splice boxes or condulets, the beta contribution would be negligible and the one year ganma total dose would a be only 29.4 Mrads. { In conclusion, the Okonite T-95/35 splices are qualified for the total anticipated integrated radiation dose during submergence by substantial a rg i n.
\
l I l J L
= }
X. CONCLUSION It has been adequately demonstrated that there exists a high degree of confidence that the untaped V-type Okonite splice will perform its { safety related function under design basis accident conditions t including long tenn submergence. This position is based upon the : following facts and conclusions presented within this evaluation.
- 1. The untaped V-type splice was demonstrated to be sufficiently similar to the taped V-type and inline splice by constructing models ]
of the taped and untaped Y splices for the examination of voids'by .l disection. This similarity is based on the following: j Materials used in both type splices was the same There was adequate T-95 coverage around all' conductors l There was adequate mechanical sealing around all conductors ! There were no voids on conductors' surfaces or within the , splice material If these four criteria are met .for any constructed T-95/35 splice, similarity is established regardless of splice configuration. l
- 2. The test data provided in Reports #1, 2 and 3 indicates that no voids or paths exist within the taped and untaped V splices due to the fact that the untaped V splice of Report #3 had baseline and post accident LOCA insulation resistance values comparable to or superior to the taped V splice of Reports #1 and 2.
- 3. The test data of Report #4 for the inline splice indicates that after exposure to a LOCA, sufficient IR remained in the splice sample to maintain operability with post LOCA IR values comparable in percentage retention as compared to post aging IR for the other tested V type splices. The results of Report #4 also indicate that the effects of sequential testing are not significantly cumulative to result in common mode failures upon conclusion to a LOCA exposure.
- 4. The test data of Report #5 indicates that an Okonite splice constructed only from the T-95 insulating tape possesses sufficient resistance to moisture diffusion due to long term submergence in water at 90 C (194 F) for 11 months. The test also provided substantiating data that a properly constructed Okonite splice with all mect.anical seals intact will resist moisture intrusion. Since the constructed models demonstrated that there are no voids in the untaped V type splice, it can be concluded that this splice configuration will remain functional under post-DBA conditions of long term submergence. The results of Report #5 also indicate that-the cumulative effects of long term submergence do not result in degradation which would result in common mode failures for cable / splice configurations submerged with adequate operability margin.
- 5. The measured IR range for the two splices during the final temperature dwell at 212 F was 57 to 130 Megohms - 1000 ft. This compares with the water immersion IR range of 608 to 2440 Megohm -
1000 ft. at 194 F. Based upon thi; data it can be concluded that the LOCA final temperature dwell test is comparable to and more severe than submergence conditions. It can be expected that upon submergence that cable / splice IR will increase or be unaffected since: Submergence profile is less severe than the LOCA profile IR' values are primarily temperature dependent Splice insulating material is relatively unaffected by moisture in any fonn
- 6. The results of Repcrts #3 and 4 indicate that exposure to chemicals such as boric acid, sodium hydroxide and. sodium thiosulfate do not attack the T-95 insulating and No. 35 jacket tape.
- 7. Aging of splice materials as addressed in Section V will- not impact the ability of the T-95 to resist water absorption.
1
XI. CONDUCTORS USED IN OKONITE' PARALLEL V-TYPE SPLICES l A review of the cable and conduit list indicates that the eight instrument splices which are potentially exposed to submergence conditions in the event of a' design basis accident are all 2/C #14 AWG cable from the following manufacturers. Anaconda 1 Samuel Moore J Rockbestos & Cerro The. insulating material used by Anaconda and Samuel Moore is Ethylene- l Propylene Rubber (EPR) and the insulating material used by Rockbestos-is _ cross-linked polyethylene (XLPE). .The jacket material is considered to be of no consequence because it is stripped back and removed to allow sufficient ~ conductor exposure in which to construct the splice. As a consequence, the weak link exposed to submergence will be the conductor insulation and no credit can be taken for any protection that 4 the jacket may be able to afford. Submergence for the cables is ) addressed in the cable evaluation 'for. submergence. l The insulating material used on the Barton transmitter lead wires .is tefzel. The ability of these insulating materials to withstand a.LOCA followed by submergence is provided in the submergence evaluation for cables. i l i l ll l
- i
1 iI l l j l d 1 i a i i i
)
ATTACHMENT 1 I OKONITE INLINE SPLICE l LOCA TEST RESULTS l l l 1 a l
1 llI !I,l W IR _ (9 hm o0O ft C 0 u 0 0 A 0 0 0 0 0 0 0 0 10 1 1 1 1 1 00 A s=: g =g- 0
- = E I" '
0 0 d ' 1 t e ' s 2 , ' 0 Te e) g,' " 0 0 . l " 0 - Ap ." ' 0 1 e m Cm i T Oa L S N 0 s x 0 )s v e d \
" 0e e " 0t R i
c g 1 u i n I l pA ' ( m S( ' 0e 3 \ 0m e-n s,s 0T i t s e iN --s'
" 1 n m
R e l '
.c // ' i nO <y* ' mT i e . _
N r _
\ '
s e
\ .
u v t i n
/. / / \' "
0 0 s a p ni / " 1 e o m
/ '
M e k / ' T a O /,,/ A = C
/ O - - 0 L -
1 0 0 0 0 O - 0 0 0 0 4 w 4 3 2 1 - Te*P. De9re s g - n - a_
!ll .
0 ~ 5 - 3 -- 0 _ d ' 0 t e 3 _ s e T ' 0
) '
5 Ae l 2 _ Cp ) F m Oa L S ( s e 0e 0r ed ' ig l ce 2g D e ._ p (A e _ r S3 0t u _ 5r a eN ' 1 e i nR p l O m n N T e i n 0 ei ' 0 t 1 i n o k . O ' 0 5 e r 3:: - sj-a 3- - a,p: - 5- ~ 5e : _ O t u 0 0 0 0 0 0 0 1 1 a r 0 0 1 0 0 0 0 1 e 0 0 1 p 0 1 m 1 e T IR - ue9 ohms /10O0 ft s v R I
i i l l l
-E smoorEmsroco ~~
O j O O ' O O O "o-O O O ~O o r r r r V OO l 11 i liiiiill i illilit i i 11111 : : : i silitii l i 111111 1 in O
- O D
o r i q) (/) m ! O i q Q) >5 O ; p) 4 _ O ) C. E O l e
<E O' r g o en - ~
OD
' ~ - l ]V a) / ,/ ' =
hg Oe f O) ty) ,' 2 O p *3 E Q 1- - m
*- 2 .E o, C -
E --
~
(f) *D
/ -1 OO l G) co / / \ :
O.E OF
- v3 g
i I
.C g - r r c \ -z I /
N'p - e .E_ CC
- E F l e ;> .$ za C C '(/' s N x =
O 0 e E m e a O '- N. I 2 E 2 ~
'N <
O s _ o , N x O ' i
;f e i i x_ o ,,,,3 O O O O O" !
O O O O ? 9 O N r 1 l Feen- o m ow e e m- a t
.)
j J i l l i 1 ATTACHMENT 2 l OKONITE INLINE SPLICE IMMERSION TEST RESULTS l I
.I i
J i
_ 0 - 6 - 1 0
' 4 1
s)s
/l t - '.,! ! . -,- l l l , ' t Il , ' , ' ' , l
- l uy sa ' 2 0 ed 1 R 2 t s3 1 0 e T (t ' 0 1 e se ) s y i l cT 0 (d a pn ' 8 e So i m esr i T ie n 0 l
- m ' 6 im n i
e t r ie ' 0 nt 4 oa kW O - 0
' 2 = - _ - - - , b- - - O 0 0 G 0 0 0 0 0 1 0 1 i
1 T I R V" ueS - Me0[m [10OO ft R I}}