ML20128H535

Rev 0 to, Review of River Bend Station Fire Barrier Ampacity Assessments, Ltr Rept

ML20128H535
Person / Time
Site:	River Bend
Issue date:	06/07/1996
From:	Nowlen S SANDIA NATIONAL LABORATORIES
To:	Ronaldo Jenkins NRC (Affiliation Not Assigned)
Shared Package
ML20128H541	List: ML20128H535 ML20129A051
References
GL-92-08, GL-92-8, NUDOCS 9610090379
Download: ML20128H535 (26)

Text

.. . --

A Review of the River Bend Station Fire Barrier Ampacity Assessments l

l -

A Letter Report to the USNRC r .

4 Revision 0 f

4 June 7,1996 t

Prepaied by:

• Steve Nowlen Sandia National Laboratories

. Albuquerque, New Mexico 87185-0737 4 (505)S45-9850 Prepared for:

4 Ronaldo Jenkins Electrical Engineering Branch i

Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington, DC 20555 4

EllCLOSURE 2 96100$037T ^^

TABLE OF CONTENTS:

I Section East 4

FORWARD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii

1.0 INTRODUCTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 l 1.1 Obj ective . . . '. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 l.2 Overview of the Utility Ampacity Derating Approach ...........I 1.3 Organization of Report .....'...........................2 2.0 OVERVIEW OF THE UTILITY APPROACH . . . . . . . . . . . . . . . . . . . . . 3 '

2.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.2 Base Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . .........3 2.3 He RBS General Analysis Methodology . . . . . . . . . . . . . . . . . . . . 4 2.4 Calculation l'-218 General Methodology . . . . . . . ............ 5 1  ;

4 3.0 REVIEW OF UTILITY ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 '

3.1 Overview . . . . . . . . . . . . ......................... ...7 i 3.2 Development of Customized General Ampacity Tables . . . . . . . . . . 7 4

3.2.1 The Customized Conduit Ampacity Tables . ........... 7 3.2.2 De Customized Ampacity Tables for Type "L" and "K"

)

Cable Trays . . l . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 i

3.2.3 Estimation of Ampacity Limits for Control Cables . . . . . . . . I1 3.3 Calculation E218 Attuhment 9 Compressor Loads . . . . . . . . . . . . . I1 3.4 Review of Attachment 12 to Calculation E218 . . . . . . . . . . . . . . . 15 4

17 l

4.0

SUMMARY

OF REVIEW FINDINGS . . . . . . . . . . . . . . . . . . . . . . . . .

Summary of Findings on the General Utility Approach and Scope . 17

. 4.1 Summary of Findings on the Overall E218 Analysis Method . . . . . 17 1

4.2 i 4.3 Summary of Finding on Utility Analyses for Nominally j Overloaded Cables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 4.4 He IEEE 242 Standard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 4.5 Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 23

5.0 REFERENCES

i 7'

1 I

a 4

l 1

i I a

1 1

4 t -

H

FORWARD he United States Nuclear Regulatory Commission (USNRC) has solicited the support of Sandia National Laboratories (SNL) in the review of utility submittals associated with fire protection and electrical engineering. His letter report documents the results of a SNL review of a set of submittals from the River Bend Station (RBS) nuclear plant. Rose submittals deal ~with the assessment of ampacity loads for cable trays and conduits protected by Hermo-Lag 330-1 fire barriers. In particular, the submittals document analyses performed by RBS to support an assessment of actual in-plant cable ampacities. These documents were submitted by the utility in response to USNRC Generic Letter 92-08. His work was performed as Task Order 8, Subtask 3 of USNRC JCN J2017.

4 i .

i

?

l i

4 I

+

1 .

I 1

i kkk i

4

, o .

i

1.0 INTRODUCTION

i.

i 1.1 Objective i

l In response to USNRC Generic Letter 92-08, the River Bend Station (RBS) nuclear l l

l plant provided documentation of the utility position regarding ampacity derating I factors associ.W with its installed Hermo-Lag 330-1 fire barrier systems. The objective of%A subtask was to review and evaluate these utility submittals. In l j

particular, tha r.ubmittals included documentation of analyses intended to demonstrate i that the cables at RBS are operating within acceptable ampacity limits. The relevant l- documents reviewed are

- Letter, November 9,1995, J. J. Fisicaro, Entergy Operations /RBS to the l

USNRC Document Control Desk, item RBG-42159, RBFl.95-0265 with i e~h=ents/ enclosures as follows:

- Attachment 1: Utility response to USNRC Request for Additional l Information (RAI) of February 9,1994.

- Attachment 2: Utility response to USNRC Request for Additional Information (RAI) of December 28,1994.

S

- Enclosure 1: Utility Calculation E-218 with Supplements A-C and l Attachments 1-13, "Ampacity Verification of Cables Within Raceways

~

Wrapped with Appendix R Fire Protection Barrier", various dates. ,

i l SNL was requested te review the ampacity derating aspects of these submittals under the terms of the general technicsl support contract JCN J-2017, Task Order 8, Subtask 3. His letter report documents the initial results of this review. De intent of this review was to provide support to the USNRC in determining the adequacy of the

' utility submittals, and in the potential development of a supplemental RAI. Based on

~

the results of this review, it is recommended that such a request be pursued.

1.2 Overview of the Utility Ampacity Derating Approach i

The consideration of ampacity derating factors for fire barriers at RBS, as currently

documented by the utility in the above referenced submittals, is based on an analytical assessment using available test data on the derating impact of the fire barrier systems.

The bulk of the assessment is based on fairly simple, and generally conservative, 3 calculations which begin with an assessment of the nominal ampacity limits for the cables installed at RBS including such factors as the ambient temperature, grouping of conduits, grouping of cables, and the ampacity impact of the fi're barrier itself. These nominal ampacity limits are then compared to the actual in-plant cable loads. The result is an assessment of the acceptability of the in-plant ampacity loads. In general, l

this is considered an appropriate approach to analysis, provided that the analysis itself ,

is properly conducted.

i .

1 ,

4

-+ r-, , - , - . ,

It must, hswever, ba noted that the submittals are incomplete and, as they currently stand, are not sufficient to demonstrate the acceptability of the ampacity loads at RBS.

In particular, the only test results cited by the utility are those performed under the .

sponsorship of the manufacturer Thermal Science Inc. (TSI), all of which have been I discredited. Hence, these tests do not represent an appropriate basis for the utility  ;

ampacity assessment. ne utility does cite that it will reevaluate its ampacity factors once the TU, or potentially other utility tests, become available. A final assessment of the utility ampacity load factors will require that such an assessment be provided.

Another factor which must also be clarified before a final assessment is possible is that the utility states that it is in the process of reviewing all of its installed fire barriers and that many of the barriers may be eithar abandoned in place, removed entirely, or replaced with an alternate material. nis raises certain questions related to cable  ;

which might have once been protected but have since had their fire barriers removed. '

nis issue should be resolved by the utility as will be discussed further below.

1.3 Organization of Report nis review has focussed on a technical review of the utility documentation and the utility ' analysis approach. Section 2 of this report provides a more thorough discussion of the utility approach to ampacity assessments. Section 3 provides a technical review of the actual calculations documented in the utility submittals. Section 4 summarizes l the SNL findings and provides recommendations regarding the need for additional information to support the final assessment of the utility analyses.

i i

I l

l 2

2.0 OVERVIEW OF THE UTILITY APPROACH 2.1 Overview e ne utility submittals are quite extensive, and document ampacity assessments for a very large number ofindividual cable 4, and barriers, ne utility ampacity analysis is based on a relatively straight-forward analysis approach. nat is, the utility -

the baseline limits of its installed cables using the tabulated ampacity values from ICEA P-54-440 [1] or IPCEA P-46-426 [2]. In some specific cases the NEC tables i are also cited [3]. ne nominal tabulated ampacities are derated by the utility to account for a range of service conditions including the local ambient temperature, cabh tray depth of fill, cable Wze, and the fire barrier itself. His results in the deveiopment of customized ampacity tables which are applicable to the RBS cables for a range of specific installation configurations. nese derated cable ampacities (or DCA's) are then compared to the actual in-plant ampacity loads. It the service load is less than the DCA then the cable is assumed to be acceptable. More general aspect of the utility analyses will be reviewed in detail in this Section of the report. A more detailed examination of certain specific aspects of the utility calculations is provided separately in Section 3 below. i

$ 2.2 Base Assumptions 1

There are several basic assumptions made by the utility which will significantly influence the utility analyses, and the final assessment of the adequacy of the utility submittals. De most significant of these factors are as follows:

i ne utility analysis assumes that all of the cables reviewed are rated for i

~

an operating temperature of 90*C. This is a typical value for modern insulation types used in the nuclear industry.

ne utility has used an initial ambient environment temperature of 40*C l for all cable ampacity assessments. However, the individual ampacity values are then adjusted to account for the " maximum ambient design temperature for i each room". While such an assessment is entirely appropriate, it should be noted that the utility assessments will derive little or no conservatism from the j

assumed ambient temperatures. (A common source of conservatism noted in other analyses is use of a single bounding and conservative ambient i

temperature for all areas. The RBS analyses will only derive conservatism in this aspect of the analysis to the extent that the assumed area ambient temperatures are conservative estimates of actual in-service plant conditions.)

- ne utility analysis has not considered Instrument and Control (I&C) cables as a thermal source in its ampacity calculations. Neglecting ofinstrument and non-continuous control cables is consistent with general practice and is

]

i considered appropriate for the assessments at RBS. However, the customized RBS ampacity limit tables include the consideration of ampacity limits for

" continuous duty loading" of control cables. De utility should clarify whether f or not " continuous duty" cont;. a cables have been included in the ampacity

3

-- - - - ,-. , ~ . . --- .

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

( derating assessments. If such cables have not be included, then supplemental analyses should be provided to cover such cables.  !

l i - Ampacity loads for cables are apparently based on actual in plant service loads using the rated power of connected devices (as compared to j breaker settings e.g.), his is generally considered an appropriate practice although the utility will derive no conservatism from this treatment. It was also noted that for the more significant ampacity loads, the calculations j included consideration of at least a 10% degraded voltage operating condition.

l 2.3 The RBS General Analysis Methodology 4

In Attachment 2 of the utility submittal, as identified in Section 1.1 above, RBS has

! documented its general procedure for the assessment of the in-plant ampacity service i loads in cables protected by normo-Lag fire bamer systems. His assessment is

, based on a fairly straight forward approach to the calculation ofin-plant ampacity 1

loads and to the evaluation of allowable ampacity limits. Simple comparison of the actual in-plant ampacity loads to derated cable ampacity (DCA) values provides an

assessment of the adequacy of the service loads.

! I In general, the utility approach to analysis is appropriate. In particular, the approach l

! allows for the assessment of individual cable ampacity loads in comparison to published tables of cable ampacity limits and includes the consideration of important in-plant service conditions, including the fire barrier itself, in these estimates.

l However, there are several points regarding the utility procedure which should be 1

clarified or further justified. In particular:-

l - Attachment 2, Item 1: His item states that the analysis will focus only on " required and abandoned nermo-Lag wrapped raceways." His implies that i cables that were originally enclosed in fire barriers that were subsequently

removed by the utility will not be considered. His is not an adequate basis for i

analysis. He fire barriers have, presumably, been in place at RBS for some i years. If cables in formerly protected rsceways have been operating 1

significantly above allowable ampacity limits, then the aging of these cables l would have been significantly accelerated and the cables may not have little or

no remaining " life expectancy." His may prove to be particularly important

for RBS kam the original utility analyses were based on fire barrier ADF 2 values which are now known to have been highly optimistic, and because even i using these optimistic ADF values certain of the cables at RBS were found to have little or no additional margin available. He utility analysis should

include all cables which are either currently or formerly enclosed in fire barrier wraps. For formerly protected cables, they should be shown to have either

. been operating within acceptable limits during the time they were protected, or alternately, an analysis of the aging of these cables during the period they were

{ wrapped should be provided.

l - Attachment 2, Items 6 and 7: De utility ampacity load calculations have included an "ampacity adjustment factor (value greater than one)" as a 4

4

multiplier on current loads (see utility Attachment 2, items 6 and 7). The basis ,

and intent of this factor is somewhat unclear and should be clarified by the  !

i utility. In particular, the utility should describe all of the effects which are

intended to be addressed by this factor, and should describe how it has been

{ applied in specific cases. More definitive criteria for how this factor is

assessed for specific cases is also needed.

4 Attachment 2, Items 10: It is not clear that the utility has appropriately i accounted for the impact of collocated IAC cables on the performance limits of its cables. In particular, utility Attachment 2, item 10 states that the utility will

) " Calculate the depth of cables in each wrapped tray (other than control cables)"

j

' and will "Use this value to determine an ampacity derating adjustment for cable depth " nese statements imply that in calculating the cable tray depth of fill, i

collocated control cables may not have been included. It appears that the utility might be taking credit for cable load diversity through a modification of

] the cable tray depth of fill calculation.

1 If this is a correct interpretation of the utility practice, this would be

)

inappropriate. He mere presence of collocated cables in either a raceway t f

would impact the ampacity limits of the other cables in the raceway regardless l of their own operating status. Depth of fill calculations should include all l

cables in a cable tray. Even though instrument cables and non-continuous load

! control cables may be assumed to not proWde an additional heating source, they will still act as thermal insulation isolating the other collocated cables l '

from the ambient environment. His important effect must be accounted for, j

' and it is not clear that the utility has done so. Clarification of the utility assumptions in this regard is needed. All of the cables present in a raceway should be included in the calculation of depth of fill, or the utility practice in this regards should be specifically justified and validated in detail. .

2.4 Calculation E-218 General Methodology The utility has also provided Calculations E-218 with three supplements and 13 3

attachments (see citation in Section 1.1 above). His calculation provides the specific l details of the utility methodology, and the actual calculations for the protected cables. ,

i The following identifies points of the general methodology and documentation which l

{ may be inappropriate, require clarification, or require further justification. (Section 3 i j of this report provides a brief review of certain of the actual cable calculations.)

1 l Calculation E-218, Revision 0, Page 2 of 35, item 7: The calculation  ;

cites that RBS " takes credit for the guaranteed average diameters rather than j  !

guaranteed minimum cable diameters for 600 volt 'K&C' cable. His will result l

l in slightly higher DCA's for these cable types " This assumption should be clarified. In particular, what is the difference between the guaranteed minimum diameter and average diameter? How large would the ampacity impact be if i

the minimum diameter is used? In general, it would be considered more 4

appropriate to use the minimum diameter value because this would be more

, conservative, and the manufacturer has apparently indicated that these )

minimum values are not unlikely. If the impact is significant then the utility l

e

! l

.=

J should reassess its ampacity limits using the minimum cable diameter as the basis for analysis.

- Calculation E-218, Revision 0, Page 6 of 35, Item II-a: His item identifies five " types" of cable trays, H J L K and C. These designations should be clearly explained.

Calculation E-218, Revision 0, Page 6 of 35, Item II-a-4: This item states that cables in "K-trays" are based on an assumed depth of fill of 1.5".

This value appears again on page 22 of 35, item la, and in this citation $218-Attachment 3 is cited as the basis for this value. However, E218-Attachment 3 stes that a depth of fill of 2.5" should be used for sizing cables in "K-trays" (see " Conclusion" on page 3 of 3 of E218-Attachment 3). This discrepancy

! should be resolved. In particular does a value of 1.5" bound the upper limit on depth of fill for all such trays? If not, then either an upper bound value or the i actual value associated with a given case should be used.

- Calculation E-218, Revision 0, Page 8 of 35, Item 3: This item cites i the IPCEA derating factors for " cable with maintained spacing". The requirements for " maintained spacing" are quite restrictive and specific (cable-to-cable gaps must be maintained at or above a minimum value, and the cable i must be installed with physical restraints to maintain these gaps). The utility I l

should explicitly justify its use of the " maintained spacing" factors by explicitly verifying that the requirements set forth in the IPCEA standard for use of these factors are in fact met in the RBS installations. Specific cases in which these factors have been applied should be identified by the utility, and each such

' application should be specifically evaluated as to thei,r compliance with the

" maintained spacing" criteria.

Calculation E-218, Revision 0, Page 35 of 35, Item E: The utility has t not provided any detail or results for the calculation of ampacity limits for SkV and 15kV cables. RBS should cite thd tables from which the ampacity limits  !

for these cables are derived, and should describe the appropriate derating

, factors applied to the tabulated ampacities as has been done for all other cable types. Comments associated with the other cable type calculations should also i be considered as applicable for the 5kV and 15kV cables as well (for example, use of the open air tables without consideration of cable tray effects (see section 3.2 below), assumptions regarding maintained spacing, etc.).

l 6

,, y. . . _ . _ _ . _ ._ _ ______

1 3.0 t d REVIEW OF SPECIFIC UTILITY ANALYSES I

3.1 Overview l'

)

{

Included in the utility submittal are the specific calculations for a large number of individual cables. His includes the development of modified or customized tables of j general cable ampacity limits, the estimation ofindividual cable ampacity loads, and the assessment of cable loads in comparison to the " generic" ampacity limits nis effort has not attempted to review all of these individual calculations, but rather, has i

" spot checked" certain of the calculations in an attempt to highlight potential i

shortcomings. He findings of these reviews are documented in this section.

}

j 3.2 Development of Customized General Ampacity Tables 1

j j One aspect of Calculation E218 is the development by the utility of modified generic ampacity limit tables for cables specifically used at RBS under a range of potential i ,

1 application conditions. That is, the utility has taken the base ampacity tables from the  ;

IPCEA or NEC standards, and hu customized those tables to address the specific '

) cables in use at RBS, the placement of those cables in a raceway system (either a j ,

conduit or cable tray), the local ambient temperature of the area, and the type of fire i barrier system installed o'n the raceway (thr or 3hr, conduit or cable tray system).

In general, this practice can be used to provide a more consistent and concise >

assessment of ampacity limits for the cables in use at a given site. These tables can I  !

1 be used as a simple source of estimated ampacity limits to support the cable ampacity assessments. However, there are certain aspects of the utility ampacity tables that

• have not been adequately addressed and that may lead to nonconservative estimates of

, cable ampacity limits. These questions are addressed in the following subsections.

Also noted in the review of these calculations were two potential discrepancies in one j of the customized utility ampacity charts:

d Calculation E-218, Revision 0, Page 29 of 35, Item " Chart 2": H ere

[ appear to be two possible discrepancies in the values cited in this chart; for the s

10AWG 7/C and 12/C cables, and for the 12AWG 7/C and 9/C cables. In general, the ampacity limits should drop with an increase in the number of j

' . con'ductors. For all cases, except the two pairs cited, this expectation is met.

' ne ampacity values cited in column three of this chart should be verified and the apparent discrepancies for these two cable pairs should be resolved.

3.2.1 he Customized Conduit Ampacity Tables 4

ne RBS customized ampacity limits for cables in conduits are presented in E218

. r Charts 1 and 2. In developing these tables, the utility has taken the cited ampacity values from the IPCEA (or ICEA) or NEC tables for a single cable or circuit in i

conduit, and has derated those values to account for the local ambient temperature and for the fire barrier system installed. However, the utility has not fully accounted for 4

4

! 7 i

}

i the additional derating factors which are required when a conduit contains more than i

three power carrying conductors. j i -

nat is, the IPCEA and NEC conduit tables are intended to address only a single cable  :*

}_ (with up to three conductors) housed alone in a conduit. He IPCEA tables do not

• l address any other conduit loadings. For conduits loaded with more than three conductors, typical practice is to apply the NEC cable grouping factors. While the I {'

utility cites these NEC factors in its submittal (see page 8 of 35 of E218), it has not i applied them in a fully appropriate manner. In particular, the NEC handbook states in i

Note 8 to the ampacity tables in Article 310 that

:

"Where the number of current-carrying conductors in a raceway or cable  ;

exceeds three, the allowable ampacities shall be reduced as shown in the  ;

following table:"

j A table of ampacity corrections factors (ACF) is then provided.' For example, if a conduit housed bet.= 4 and 6 current carrying conductors, then an additional 20%

j ADF (or 80% ACF) would be applied to account for the mutual heating eff'octs of one j i

cable on its neighbors within the conduit. His ADF increases as the number of l

{ " conductors meresses. i f i i It was noted that the utility has apparently applied these factors in reducing the j ampacity limits of individual cables with more than three conductors when installed in a conduit (see page 27 of 35 of E218). However, this is not fully consistent with the .

l intent of the NEC design practice, and does not result in the same derating effect as

! that intended by the standard, hat is, the intent of the NEC conductor grouping

[ factors is to account for two situations.

first, for an individual cable which has more conductors than the cables cited in the tables, and 1 -

second, for the installation of multiple cables in a raceway system (either a tray or conduit) where the spacing between cables is not explicitly i -

maintained.

He utility practice has only accounted for the first part of this problem, and has 1

ignored the second aspect, and hence, does not adequately address this factor.

] It was also found that the utility has cited an older version of the NEC tables in its

! assessment (NEC 1984). He newer versions of the NEC tables have updated the

] cable conductor grouping ACF values (all versions issued since 1990). He older values included an assumption of a 50% load diversity, although this was often j overlooked in ampacity calculations. He updated standards allow for their use only j where load diversity can be verified (calculations under engineering supervision as per j Appendix B of the NEC handbook). For general applications, a new set of values has i

I i

' 2 Recall that ADF = 1.0 - ACF

! 8 1

been developed which are more restrictive if the number of conductors exceeds nine.

The utility should be asked to justify its use of the "50% load diversity" based factors

.in lieu of the general values assuming no load diversity consistent with current NEC l l design practice.

l To illustrate, consider the first raceway cited in Attachment I to E218; Raceway i - ICC600RB. His is cited u a raceway protected by a ihr fire barrier, and located in

' an area with a 50*C ambient temperature. A comparison of the ampacity limits cited (

and the utility' customized tables shows that this raceway is, in fact, a conduit.' In all,

) this conduit holds 6 multiconductor cables with a total of 23 individual conductors.

Using the 1996 NEC handbook, an additional ADF of 55% should be applied to the

~

tabulated ampacity limits for each and every cable in this conduit.' Even using the  ;

l i

older NEC values cited by the utility, a uniform ADF of 30% should have been l , applied to all of these cables. In the utility practice,just two of the cables 7 (IRCSARC300 & 301) had individual conductor grouping ADFs (of 20% and 30%

l

- respectively) applied to their ampacity limits in developing the customized ampacity tables.' ne other four cables were either 2/C or 3/C cables, and hence, no conductor j grouping ADF was applied. De overall effect is not at all equivalent. Inthis j particular case, even if the ADF for conductor grouping is applied to the conduit as a i

whole, all of the cited cable loads would still remain within acceptable limits.  ;

l However, the available margins would be significantly reduced.

A second example where this shortcoming may impact the results is discussed in Section 3.4 below. In that section, the supplemental calculations provided for certain j conduits in which the power loads were split between two parallel power cables might 1 also be impacted by the failure to include appropriate conductor grouping adjustment i ' factors.

t j ,

his review has not a-ad all of the utility conduit calculations to determine where this factor might change the utility conclusions regarding individual cable ampacity l

loads. A spot check indicated that most of the conduit installed cables at RBS are i either not impacted by this issue (i.e., are in conduits with 3 or fewer conductors i present), or will likely have sufficient available margin to encompass this additional

! derating effect. Nonetheless, the utility should review all of its conduit assessments to determine how the appropriate treatment of conductor grouping ADFs would impact its l overall assessment conclusions.

i i ,

i

) 2 While the utility nomenclature is not explained, it appears that a "C" in the i . second slot in the raceway identifier would indicate a conduit. Similarly, a "T" in this slot would appear to indicate a cable tray, and an "I" indicates and instrumentation .

i . Pi pe.

- 3 See NEC Handbook,1996, page 70-l%.

i

• Cable x300 is a 5/C cable and x301 is a 9/C cable, hence, different ADF values apply as per NEC i'

9 i

3.2.2  !

4 He Customized Ampacity Tables for Type "L" and "K" Cable Trays ~~

, i In the development ofits customized ampacity tables for type "L" and "K the utility has made certain assumptions regarding the application of the IPCE '

426 open air ampacity table to cable tray installations which appear to be '

inappropriate. In particular: i Calculation E-218, Revision 0, Page 24 of 35, Item " Chart 1"; his i item is described as a table of allowable ampacities for "L-trays". (ne u .

has failed to explain its nomenclature, and it is unknown what the desig "L-tray" implies. nose appear to be associated with lower voltage (<600  ;

light power and control cables.) However, it is clear that the base amp i values are from the ICEA P-46-426 tables for a cable located in no derating factors applied.

Calculation E-218, Revision 0, Pge'6 of 35, Item II-a-4 (and pages 32 '

and 33 of 35, Item VII-c and " Chart 4"): nese items state that for cables with !

" intermittent service" in type "K" trays, ampcity limits are based on IPCEA P.

i 46-426 'without derating for spacing." nat is, the IPCEA P-46-426 table values of the ampacity limits for cables in open air have been used directI assess ampacity load limits for cables in trays with no consideration of the '

impact of placement in a cable tray with other cables on those limits.

He use of open air ampacity values for a general cable tray appears inapprop He basis for the development of these charts requires further explanation and '

justification. He basis for use of open air ampacity limits for cables located in a cable tray is inadequately addressed. On page 6 of 35 of E218 the utility cites a alternate calculation, E-137, as the basis for the cable sizing for "L"-trays. This ,

calculation has not been provided for review. In effect, the utility appears to be

! crediting load diversity in its cable trays through application of the open1 i limits to diversely loaded cables in cable trays. His is a fundamental departu j accepted this approach. ampacity analysis approaches, and the reviewer knows of no precedent f l  !

For example, in the ICEA P-54-440 taNes, ampacity limits are generally establ based on the cable tray depth of fill. 'However, the standard also establi

! bound limit of no more than 80% of the open air values for cables in cab j -

Section 2.2 of the standard). He utility has not justified its use of open a r

limits for cables in cable trays, and the result could be non-conservative asses

) of cable ampacity limits. Even for a cable with intermittent loads, the fact that is i might be collocated with other cables in the same tray could mean that a sig

, mutual heating effect might be observed, or at the least that an insulating effect  ;

impact the allowable cable ampacity limits. It is this type of effect which must b I accounted for in the calculation of cable tray ampacity limits.

i Cable ampacities for cable tray installed cables should be based on appropria o consideration of the installation details. His should include the appropriate

, consideration of cable tray loadings, regardless of the nature of the load placed  !

l 10

, v cables. If the utility is, in fact, crediting load diversity using this practice, then the utility approach must be fully detailed, justified, and validated.- Alterneively, the utility should provide a reassessment of the cable ampacity limits using accepted ampacity. sizing methods which include consideration of the cable tray loading conditions.

3.2.3 Estimation of Ampacity. Limits for Control Cables Control cables are treated in a similar manner to that discussed in Sectio (See E218 Section II.A.5). Dat is, non-continuous load control cable ampacity limits are derived from open air ampacity limits even though these cables might be installed in cable trays. While, in general, intermittent service control cables are excluded from the final fire barrier ampacity derating assessments, the initial sizing of such control cables is an important design consideration. This is particularly true when the cables '

are collocated with other continuously energized control (or power) cables. While in.

the final analysis it is not expected that this will prove to be a point of signi6 cant concern, the utility practice in this regard should be further justified consistent with the discussions provided above.

3.3 Calculation E218 Attachment 9 Compressor Loads i

. Attachment 9 to Calculation E218 documents supplemental menseements performe i

RBS for four specific cables which are nominally identified as operating at least part of the time under ampacity overload conditions. Two of the four cables service l

certain compressor power loads. He purpose of this section is to provide a review of '

i, the calculations for these two cables.

d

\

- ne two compressor power cables considered in this analysis are physically identical j and also carry nominally identical loads (cables 1HVKBBC515 and lHVKDBC506)  ;

Each is a 2/C #10 AWG cable, and each feeds power loads to one of two saparate '

compressors. Each cable carries the power for the crankcase oil heater, and a seal oil

• pump motor for one of the two compressors. (The compressor motors themselves are apparently fed through separate cables.) Both cables are located in a common cable ,

l tray (tray ITC043B) which is wrapped with a 3-hour Hermo-Lag fire barrier.

i

T

' he initial estimate of the full load current for each chble is given as 14.35A, including consideration of a 10% undervoltage condition of operation. The ambient temperature is cited as 40*C, and the "derated cable ampacity" (or DCA) is estimated as 12.1 A including the effects of the 3-br fire barrier (assuming an ADF for the fire i

4 barrier of just 20.5%). Hence, these two cables are nominally cited as overloaded 4 (because 14.35A > 12.l A).

• L Given the actual estimated maximum load and the nominal DCA cited, the utility I calculates that under these load conditions the temperature of the cable may reach

' about 110'C. His calculation was based on a fairly simple correlation which relates the impact of current on operating temperature (this relationship derives from the j

IPCEA tables). The utility then cites the equipment qualification (EQ) results as e

i 1 11 i

, . , r. - -.. ---

' 'f prcvided by the manufacturar, and calculates an cxpected life for this c ble undir i

4 conditions cf continuous ep:ratien at this temperature asjust under seven years. '

Based on these results, the utility cites and initial conclusion that the " cable is inadequate." However, the utility continues its assessment in an attempt to

' demonstrate acceptability. In particular, the utility cites that the oil crankcase heater and the seal oil pump motor " rarely operate simultaneously." nat is, the oil heater shuts down after some period of compressor operation and oil temperature is

) maintained by the excess host of the motor operation, and the seal pump operates onl periodically and then for short periods of time. De utility then states that "it is anticipated that the amount of time that both loads operate simultaneously is far less that 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> per year and less than 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> per occurrence."

j

' Using only the load of the heater, apparently the larger and more frequent of the two

)! loads, a modified ampacity load of 9.17A is calculated. His is within the estimated DCA of 12.l A, and hence, the utility concludes that the " cable is adequate under

!. typical and emergency overload conditions."

i There are several points of concern related to this calculation:

• l 2

~

i -

As has been noted above, the utility DCA values are based on an 2

assumed ADF for a 3-hr wrapped cable tray ofjust 20.5%. This value is j

clearly optimistic (i.e., too low) based on existing test data. A more realistic estimate of the ADF of a 3-hr barrier would be on the order of 40% (this estimate only and is used for illustration purposes, this value is not based on j

any specific test results). Use of this as an estimate of the actual ADF would have a significant impact on the utility calculations:

j -

' Using an ADF of 40% in the utility calculation method, the 1

actual DCA of the cable would be estimated as 9.13A. His would be lower than the value of the full load amperage even with only the

' crankcase heater operating as cited in the second stage of the utility calculation (9.17A) Given this situation, and the fact that both loads can be active for at least part of the time (yielding a maximum load of the original 14.35A), there is a distinct possibility that this cable has been operating for some time at a significant overload condition i

(potentially at more that 150% of the nominal ampacity limit for at least

{

part of its life). '

If the same process used by the utility is again used to estimate

( the operating temperature of the cable under the conditions in which j

both loads are active, a temperature of 163.5'C is found.' This

temperature would be clearly unacceptable even for short term 4 5

- Note that this is only a rough estimate and is probably non-conservative, i.e.,

too low. In particular, the simple temperature calculation cited does not consider the increase in electrical resistance for copper at increasing temperatures. This effect

would increase the cable self heating rate for a given current as temperature increases 12 i

t

i'

" overload" operations. Even very short tenn operation of the cable at these temperatures would severely degrade the cable's " life expectancy."

Recognizing that the 40% ADF value is an estimate only, it is still considered 1 unlikely that the utility can demonstrate that the operating conditions of these two cables are, in fact, acceptable. Operation of the crankcase heater load only might be demonstrated as an adequate operating condition, but clearly, the two i loads operating simultaneously would not be considered acceptable under these conditions. l 1 i 1

ne utility has not provided any assessment of the impact of the stated q  !

conditions of overload operation on the cable's " life expectancy." That is, the  !

utility did provide for an assessment of life impact for the conditions of continuous operation at an overload condition, but did not provide the l l' l corresponding assessment for periodic operation at overload coupled to l

( " normal" operation with a single device operational. Even reisdvely short i periods of operation at a temperature exceeding the qualified life temperature of l 90*C can sigr. ficantly degrade the cables. He utility should provide for an l

' assessment oflife expectancy under the worst case anticipated conditions of operation.' ,

4 I

1 -

l

If these cables have operated under the stated conditions for any significant length of time, then the cables may have already exceeded their rated life expectancy. Given the aging versus temperature chart cited by the  ;

utility, the estimated life expectancy of the cable operating at 165'C would be

j just 850 hours0.00984 days <br />0.236 hours <br />0.00141 weeks <br />3.23425e-4 months <br />, or less than 36 days (this value corresponds to the conditions

! 1 postulated by SNL using an ADF of 40% for the 3hr fire barrier u stated above). Even assuming the operating times cited by the utility, "far less than 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> per year," coupled with near continuous operation of the heater at i

essentially 100% of the cable ampacity limit, the cable would be aging at an i

effective rate of 5.7 years per actual year of plant operation.' Hence, the cables l nominal "40 year at 90*C" life would be exceeded in about seven years.

i, l i

l 1

{

' Note that the~ impact of aging at different conditions of operation can be i

assessed in a fairly simple manner by simply summing the equivalent life impact of l

the various operating conditions. Also note that this assessment must also include the i

aging impact of the ambient envirorment even when the cable is not energized. That  !

is, even at an ambient of 40*C e caole continues to age. A forthcoming DOE report on this subject should soon be available, and the anticipated distribution will include virtually every utility in the U.S. (see reference 4).

' his is an attempt to relate the given operating conditions to'an equivalent

! i

life expectancy of 40 yrs at 90*C. Assuming that the cable would last for 850 hours0.00984 days <br />0.236 hours <br />0.00141 weeks <br />3.23425e-4 months <br />, i

! or 8.5 years at 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> per year overload loading, is an " equivalent aging rate" of l

(40/8.5 = 4.7 years / year) in comparison to its 40 year qualified life at 90*C. Added to this is 1.0 yrs /yr for the normal operation at 100% ampacity with the heater only l

operating at most times. His gives a total equivalent aging rate of 5.7 yrs /yr.  !

13 4

4

___ _ _ . _ _ _ _ _ _ - _ _ _ . . _ . _ _ ~ _ _ _ _ _ _ ._ __- ___ _ - _ _

t 3 4 5 '

l The point of this admittedly course exercise is to illustrate that even if p the utility were to immediately remove the fire barrier for this tray, these two

, cables might still be found to be unserviceable in the context of equipment i qualification, and hence, might still require replacement. His, of course, j' assumes that the conditions cited are a relatively accurate representation of the i

actual ADF impact. A full ansaamment would require application of a well

validated ADF value to this particular case, and should also include a more l accurate estimate of the cable operating temperature under the overload conditions. A more accurate estimate of the actual cable power cycling  !

l behavior should also be included.

The final justification for the acceptability of these two cables operating j conditions is based in part on the manufacturer's stated overload conditions of l operation. Dese ratings are not generally intended to cover anticipated conditions of normal operation, but rather, are intended to cover only i unexpected emergency conditions of operation. For example, it would not be 3 unexpected for a utility to cite overload conditions of operation to allow for j period operation at degraded voltage conditions. However, the reliance on j overload ratings as exercised by RBS is potentially inappropriate. At the least, i this practice represents a fundamental departure from accepted ampacity

assessment practices, and as such, should be thoroughly justified and reviewed j before being accepted as a cable design practice.'

l ,

Further, the utility cites a specific passage in the IEEE 242 standard as

the basis for its assessment of overload conditions (para.11.5.2(3)). A review i of this section of the standard revealed no relevance whatsoever to the inue of ,

j cable overload conditions (attachment I to this report provides the relevant passage). It is expected that either an older version of the standard is being

cited (no specific version is identified by the utility), or that the citation is in j error. In the version reviewed by SNL (242-1986), Section 8.5.2 deals with i overload conditions of operation and 8.5.2.3 states in part that

) "a temperature safely reached during a fault, and maintained for only a

few seconds, could cause severe life reduction if it were maintained for even a few mhutes. IAwar temperatures, above the rated continuous l

{ operating temperatures, can be tolerated for intermediate times."

It is suspected that this is the passage being referred to by the utility.

l However, this section of the standard goes on to state.

i "The continuous current, or ampacity, ratings of cable have been long established and pose no problems for protection. The greatest unknown

in the cable thermal characteristic occurs in the intermediate time zone,

! or the transition from short time to long-time or continuous state."

! Also presented in the standard (section 8.5.2.4) is an approach to the i determination of intermediate overload ratings. At the least, this assessment

.; coupled to an assessment of the impact of overload operations on cable life

[ expectancy should be provided by the utility. (Attachment 2 to this report i provides a copy of these two passages.)

De IPCEA P-46-426 standard also addresses overload conditions of operation. In particular, Appendix III.4 states: .

i l

14 l

^

i "Opirr.rirn at the emergency cvericad t:mperatures tabulated htre shall net exceed 100 hsurs per year. Such 100-hour overload periods shall not exceed five."

j De utility should address this aspect of the IPCEA ampacity standards in its assessment of cable performance. His requirement would appear to preclude a general reliance on the overload ratings as a routine aspect of the anticipated 4 ampacity operating conditions.

De utility has cited a particular equipment qualification test report as 1

the basis for its assessment of cable life expectancy calculations (Okonite rep SWGS-1282-2). De copy of the cable aging chart provided in the submittal is

{

4 not sufficiently clear to verify the acceptability of this calculation as a basis for the assumed cable life expectancy calculations. In particular, the utility has not i

cited the assumed value of the cable insulation's activation energy, nor has it i cited its assumed criteria for the cable endef-life indicator. He chart in 4

in the submittal contains four specific curves, and the utility assessment-it '

based on the most optimistic of these curves. While this may be the '

' appropriate curve for the case under consideration, some explanation and justification for the particular curve chosen is needed. He utility aging calculations should either include the full EQ report, or should at the least provide sufficient information upon which to base a review of tne calculations.

In a more general context, it should be noted that this is the first instance in which a j

citation to the IEEE 242 standard has been encountered by this reviewer in a utility ampacity assessment submittal. Hence, the overall acceptability of this design practice

j. in the context of USNRC requirements has not yet been made. Such an assessment is also considered outside the scope of the review of this specific utility submittal i

because the utility has not yet invoked this particular design practice in its i

calculations, but rather, has simply cited a passage from that standard. Hence, it is 1 strongly recommended that this IEEE design practice should be reviewed in detail b the USNRC and that an assessment ofits acceptability as a general design practice in the context of the USNRC Equipment Qualification Program and other USNRC j requ rements should be made.

3.4 Review'of Attachment 12 to Calculation E218 j Attachment 12 to Calculation E218 documents special calculations performed for 4 certain cables in conduits. In particular, the assessments document calculations of ampacity limits under conditions in which the power loads are split between parallel l runs of cable. His calculation raises one potential concem for this specific calculation which has also been discussed in the broader context of the RBS condu in Section 3.2 above.

ne initial assessment determined the allowable ampacity limit for the subject cables i including the effects of the fire barrier system assuming that three conductors migh i used to carry the required load. Because the ampacity limits were lower than the actual loads, the possibility of paralleling two cables for each load was considered.

{ For the parallel power configuration, the ampacity limit determined for one (out of i

i 15 I

4

, ., .,--r =

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

. e .

three) conductor (s) is simply doubled, or alternately, the same load is assumed to apply to each of the six conductors in the parallel cables.

This practice would not be correct if the paralleled cables (all six conductors) are located in a common conduit. In particular, the IPCEA tables address only the ampacity limits of a single cable of the given type located alone in a conduit (that is for a triplex cable it is assumed that only one such 3-conductor cable is located in the conduit). IPCEA does not address the placement of additional cables in the same conduit.

He NEC handbook includes factors which account for the number of co conduit, and requires additional derating based on the conductor count. Hence, if the cables are collocated in the same conduit, then paralleling does not necessarily doub the total ampacity limit. The.1996 NEC tables would require an additional derating factor of 20% for a conduit with 4-6 current carrying conductors. If these factors are i

included in the calculation of ampacity limits, then the cables considered in this RBS analysis would still be found to be nominally overloaded, even in the paralleled configuration.

Consistent with the recommendations presented in Section 3.2 above, for this specific cable, the utility should be asked to clarify the following point:

1 Are the paralleled cables cited, for example, in Attachment 12 to calculation i E218 enclosed in a common conduit? If so, then direct application of the IPCEA tables to this configuration would be inappropriate. Cable ampacity )

limits should include consideration of multiple conductor derating factors such as those presented in the NEC handbook. i l

i j

l J

. l 1

l 16 l

2

., n- - ~

4 .

4.0 -

l

SUMMARY

OF REVIEW FINDINGS AND RECOMMENDATION 4.1 i Summary of Findings on the General Utility Approach and Scope

The overall ampacity assessment approach taken by the utility was gener

! be an acceptable means of resolving the ampacity derating questions. The ov methodology is basically sound, and can appropriately determine whether or not

- individual cables at RBS are operating within acceptable ampacity limits. He methodology provides for case specific treatment of each protected cable at i compares actual in plant loadings to estimated ampacity limits which include the

{ effects of various environmental factors as well as the fire barrier system itself methodology is fairly straight forward in its overall implementation. i The utility approach is quite similar to a margins analysis in many regards. He

primary difference is that the utility has developed customized tables of

{ limits for its specific cable types to cover a range of potential installations, an i'

these tables of "derated cable ampacity" (DCA) limits that the utility has inclu) '

assumed ADF for the fire barrier system in the calculation. Hence, the utility dr direct comparison between the DCA values in their customized tables, and th in-plant service loads.

i

The acceptability of this approach in the final analysis rests an assessment of

! or not the utility has compared appropriately determined in-plant ampacity service i loads to estimated ampacity limits which are based on appropriate application o

base line ampacity tables, and on appropriate estimates of the fire barrie

impact. The documentation as currently provided by the utility is insufficient this determination. The utility analyses of nominally overloaded cables are also '

considered inadequate as currently documented.

3 i In a very fundamental sense, the utility submittals were viewed as prelim I

and it is in this context that this review has been performed. In particular, calculation ofits DCA limits are currently based on now discredited TSI-spons i tests. In fact, the values used by RBS in its assessment, for cable trays in partic

are known to be highly optimistic in comparison to the most recent test results. H

utility recognizes this limitation and cites plans to review its submittal upon the TU test results.

i

• 4.2 Summary of Findings on the Overall E218 Analysis Method i

Specific areas or weakness or uncertainty as related to the ocall approach to

! ampacity assessments taken by the utility which should be addressed are:

4' RDS Attachment 2, Item I states that the analysis will focus only on

" required and abandoned Hermo-Lag wrapped race ~ ways." This implies th

, cables that were originally enclosed in fire barriers that were subsequent!I

removed by the utility will not be considered. 'If cables in formerly protected - '

raceways have been operating significantly above allowable ampacity limits, then the aging of these cables would have been significantly accelerated and 1'. l 17

the cables may have little er na remaining " life expectancy." For formerly protected cables, RBS should show that they have either been operating within acceptable limits during the time they were protected, or altemately, an analysis of the aging of these cables during the period they were wrapped should be provided.

RBS Attachment 2, Items 6 and 7 indicates that an "ampacity adjustment factor (value greater than one)" has been applied as a multiplier on current loads (see utility Attachment 2, items 6 and 7). De basis and intent of this factor should be clarified as it is applied to the various cases considered.

RBS Attachment 2, Items 10 states that the utility will " Calculate the depth of cables in each wrapped tray (other than control cables)" and will "Use.

this value to determine an ampacitv derating adjustment for cable depth." ' '

nese tastements imply that the depth of fill calculation may not include all cables in a given tray. His aspect of the utility analysis should be clarified, and depth of fill calculations should consider all cablas located in a given cable tray regardless of their function or loading.

~

Utility Calculation E218 and its various attachments and supplements docum'ent the i actual calculations for specific cables, cable trays, and conduits. Two minor points related to the utility nomenclature used in this calculation require clarificatioc:

- De utility uses a plant specific nomenclature to identify various classes of l cable trays (types H, J, L, K, and C). This nomenclature should be explair.ed. i

- The utility submittals fail to explicitly identify raceway systems as e;ther conduits or cable trays, and fails to describe the physical characterides of the '

various raceway systems (such as tray width or conduit diameter). De utility l

should provide supplemental information to characterize the raceways l l considered in its analyses.

Several points relate.' to the overall utility analysis methodology were also noted:

ne utility. analyses are based on optimistic assumptions related to the ampacity derating impact of the fire barrier systems. Use of more realistic values in the subsequent final utility assessments is expected.

Calculation E-218, Revision 0, Page 2 of 35, item 7: his item states that RBS " takes credit for the guaranteed average diameters rather than guaranteed minimum cable diameters for 600 volt 'KAC' cable. His will result in slightly higher DCA's for these cable types." This assumption should be clarified. In particular, what is the difference between the guaranteed minimum diameter and average diameter? How large would the ampacity impact be if the minimum diameter is used? In general, it would be considered more appropriate to use the minimum diameter value because this would be more )

conseivative, and the manufacturer has apparently indicated that these minimum values are not unlikely. If the impact is significant then the utility 18

4

should reassess its ampacity limits using the minimum cable diameter as t 2

basis for analysis.

i Calculation E-218, Revision 0, Page 6 of 35, Item II-a-4: His item l

states that cables in "K-trays" are based on an assumed depth of fill of 1.5".

i This value appears again on page 22 of 35, item la, and in this citation E218

Attachment 3 is cited as the basis for this value. However, E218-Attachment 3 '

i states that a depth of fill of 2.5" should be used for sizing cables in "K-tr (see " Conclusion" on page 3 of 3 of E218 Attachment 3). His discrep

should be resolved. In particular does a value of 1.5" bound the upper l depth of fill for all such trays? If not, then either an upper bound value or th actual value associated with a given case should be used.

! Calculation E-218 Revision 0. Page 8 of 35, Item 3: His item cites

! the IPCEA derating factors for " cable with maintained spacing". He util i should explicitly justify its use of the " maintained spacing" factors by e i verifying that the requirements set forth in the IPCEA standard for use of thes a

factors are in fact met in the RBS installations for which these fa

} been applied. Alternatively, the utility should reassess its analyses using "

i grouping factors for cables without maintained spacing.

4 i Calculation E-218, Revision 0, Page 35 of 35, Item E: ne utility has not provided any detail or results for the calculation of ampacity limits for SkV and 15kV cables.

j RBS should cite the tables from which the ampacity limits i for these cables are derived, and should describe the appropriate deratin

! factors applied to the tabulated ampacities as has been done for all other ca i

types. Comments associated with the other cable type calculations should also

! be considered as applicable for the 5kV and 15kV cables as well (for

! use of the open air tables without consideration of cable tray effects, l assumptions regarding maintained spacing, more appropriate fire barrier ADF I values etc.).

1' In addition, this review identified certain concerns related to the calculation of

{

ampacity limits for cables located in conduits. In particular:

i - The utility application of the NEC conductor grouping ampacity correction factors for more than three conductors in a cable or raceway is considered incomplete. In the case of conduits, the NEC correction factors should be

( applied to the conduit system as a whole whenever the total count of current

carrying conductors exceeds three. In contrast, the utility has only applied these factors to individual multiconductor cables when the co

[ a given cable exceeds three. His is ar. incomplete and nonconservative treatment. The utility analyses should be revised to fully account for the j conductor count adjustment factors for all conduit systems in which the conductor count exceeds three.

- The utility has cited the 1984 version of the NEC handbook as the basis for its assumed conductor count correction factors. However, sir.ce 1990 NEC i

19 i

I

T published an upd:ted listing of correction factors which are more conservative for conductor counts of 10 or more. The older (1984) values in

~

l assumption of 50% or more load diversity in the installed cables. He utilit should either apply the more recent values in its corrections, or should i specifically justify the applicability of the older adjustment factors on the bas

of existing cable load diversity (see Appendix B of the current NEC handbook for further guidance).

i 1

Finally, certain points of potential shortcoming or uncertainty were also noted in the \

i

calculation of general cable tray ampacity limits as follows

i 1 -

i Calculation E-218, Revision 0, Page 24 of 35, Item " Chart 1": his

item is described as a table of allowable ampacities for "L-trays". Howeve is clear that the base ampacity values are from the ICEA P-46-426 tables for a cable located in free air with no derating factors applied (this is stated in the

supporting text). .The use of open air ampacity values for a general cable tra i appears inappropriate and must be either corrected or further justified. In i particular, this practice is not consistent with accepted ampacity design practices, and hence, would require explicit and detailed justification and

! validation. Similar assumptions are also used for interinittent load cable in "K"-trays, and for intermittent load control cables as well. Similarly, these applications should also be explicitlyjustified or corrected.

i Calculation E-218, Revision 0, Page 29 of 35, Item " Chart 2": h ere i appear to be two possible discrepancies in the values cited in this chart; for the i 10AWG 7/C and 12/C cables, and for the 12AWG 7/C and 9/C cables. In general, the ampacity limits should drop with an increase in the number of conductors. For all cases, except the two pairs cited, this expectation is met.

The ampacity values cited in column three of this chart should be verified and

the apparent discrepancies for these two cable pairs should be resolved.

4.3 i

Summary of Finding on Utility Analyses for Nominally Overloaded Cables '

j Attachment 9 to Calculation E218 documents supplemental assessments p i RBS for four specific cables which are nominally identified as operating at le

! of the time ur. der ampacity overload conditions. Two of the four cables service j certain compressor power loads. Here are several points of concem related to the i

supplemental assessments for these two specific cables:

i i -

i As has been noted above, the utility DCA values are based on an i assumed ADF for a 3-hr wrapped cable tray ofjust 20.5%. Using a more realistic ADF value might well indicate that this cable has been operating for l

some time at a significant overload condition (potentially at more that 150% of

the nominal ampacity limit for at least part ofits life). Given the maximum l

, current loads cited by the utility, and ever, nominal estimates of the actual fire

! barrier derating impact, it is considered unlikely that the operating conditions of these cables could be justified by the utility.

i 1

1 1 20 )

i l

If these cables have optrated under the st:ted conditions for any

! significant length of time, then the cables may have already exceeded thei i rated life expectancy. Rough estimates performed as a part of this review 1 indicate that these the cables may exceed their nominal "40 year at- 90*

2 in as little as seven years. He utility should provide an assessment of the impact of past operations on the " life expectancy" of these cables in a

any assessment of existing or future operating conditions.

4

! ne finaljusti'fication for the acceptability of the operating conditions of these two cables is based largely on the manufacturer's stated overload

] conditions of operation. nese overload ratings are not generally intended to

{ cover anticipated conditions of normal operation, but rather, are intended to '

l cover only rarely encountered and unexpected emergency conditions of l operation. Hence, reliance on overload ratings in this case is potentially

! inappropriate. At the least, this practice represents a fundamental depa j from accepted ampacity assessment practices, and as such, should be thoroughly justified and reviewed before being accepted as a cable des

, practice. In particular, the utility must consider the full context of the IPCEA

[ and EEE overload ratings which set severe limits on these overload ratin ,

i Significant additional justification and validation of this design practice shoul

be provided by RBS.

I 3 -

l De utility cites a specific passage in the EEE 242 standard as the basis i for its assessment of overload conditions (para.11.5.2(3)). A review of this !

section of the standard (1986 version) revealed no relevance w j issue of cable overload conditions. It is expected that either an older versio i the standard is being cited (no specific version is identified by the utility that the citation is in error. The utility should clarify its intent in citing the

!- EEE 242 standard.

De utility has cited a particular equipment qualification test report as the basis for its assessment of cable life expectancy calculations (Ok i

SWGS.1282-2). Insufficient information has been provided upon which to bj an assessment of the appropriateness of this correlation to the analysis. The  ;

utility aging calculations should either include the full EQ report, or should at

{. - the least provide sufficient information upon which to base a review of the !'

calculations (for example, identify the materialsnevaluated i' the study and relevance to the cables at RBS, provide the assumed material activat and identify the assumed end oflife assessment criteria used). l

4.4 i

The EEE 242 Standard In a more general centext, the USNRC should note that this is the first instance i which a citation to the EEE 242 s'andard has been encountered by this review

, utility ampacity assessment submit,al. In this case, the utility. has cited on ,

passage from the standard, but a bnef review of this document did reveal that it has potential for a much broader application to these types of ampacity assessments.

l Hence, the overall acceptability of this design practice in the context of USNRC i  :

t 4

21 l t

+

..,e ~

requirements has not yet been assessed. It is strongly recommended that this design practice should be reviewed in detail and that an assessment as a general design practice in the context of the USNRC Equipment Qualification Program and other USNRC requirements should be made.

4.5 I Recommendations '

It is recommended that the USNRC prepare an RAI to the utility. The issues identified above should be addressed in this RAL A final assessment of  ;

submittal is not possible at this' time because (1) the outstanding issues identifiedi above require clarification or other action and (2) because the utility has not utili realistic estimates of the fire barrier ADF impact in its current assessments, a shortcoming acknowledged by RBS. For this reason, the utility submittal should viewed as a preliminary assessment only, a status also acknowledged by is recommended that the utility be asked to address the identified conce its updated final analysis package.

l l

l l

l 22

l

5.0 REFERENCES

l i .'

1 l

Ampacities of Cdles in Open-top Cdle Tmys, ICEA P-54-440, NEMA WC I

51,1986,

\

l

, 2 Power Cale Ampacities Volume 1 - Copper Conductors, ICEA P-46-426, IEEE S-135-1,1%2 \

3 The Natioruaf Elecided Code Hedbook, NFPA. Note that the utility has cited t.

the 1984 editions and this review has also cited the current 1996 edition. .

t

. 4.

Gazdzinski, K. F., et.al., Aging Maugement Guidelinefor Commercid Nuclear

}

Power Plats - Electnca Cale ad Tenninations, SAND 96-0344, Sandia National j Laboratories, Albuquerque, NM, to be published by June 30,1996. (Note: This is a 1 forthcoming publication developed under 'the sponsorship of the U.S. Dept. of E

< as a part of the Plant Life Extension program. ,ne document is in the final stages of the printing and distribution process, and should be available within 30 days or less he distribution list includes representatives of every nuclear utility in the U.S., I i generally those knowledgeable in the area of cable qualification. Hence, this

document should be readily available to RBS by the time the results of this review are !

forwarded to the utility for consideration.)  !

i 4

n I

1 l

l l

I i

i i

i l

23 1

.n- -., - , n