ML20202B389
| ML20202B389 | |
| Person / Time | |
|---|---|
| Site: | Duane Arnold  |
| Issue date: | 09/15/1997 |
| From: | Nowlen S SANDIA NATIONAL LABORATORIES |
| To: | Ronaldo Jenkins NRC (Affiliation Not Assigned) |
| Shared Package | |
| ML20202B369 | List: |
| References | |
| CON-FIN-J-2503 NUDOCS 9901290068 | |
| Download: ML20202B389 (10) | |
Text
.- . .
A Technical Evaluation of the Duane Arnold Energy Center Analysis of Ampacity Loads for Fire Barrier Clad Power Cables ALetter Report to the USNRC -
. Revision 1 September 15,1997 Prepared by: l Steve Nowlen !
- SandiaNationalLaboratories Albuquerque,New Mexico 87185-0737 (505)S45-9850 Prepared for:
- Electrical Engineering Branch
! Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington,DC 20555 _.
USNRC JCN J-2503 i
( ATTACHMENT
! 9901290068 990126 i PDR ADOCK 05000331 ,
! P PDR g
-m- - . . ,.n-_,.....n,....,.. . -. --
O TABLE OF CONTENTS:
Section g FORWARD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
1.0 INTRODUCTION
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I 1.1 Background and Objective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I 1.2 Organization of Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 2.0 RESOLUTION OF PREVIOUSLY IDEN 11ritu ISSUES . . . . . . . . . . . . . . . . 2 3.0 THE LICENSEE CALCULATIONS FOR DARMATT BARRIERS . . . . . . . . . 4 3.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3.2 The Thermal Calculation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3.3 Results and Reviewer Observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3.4 Findings and Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 4.0
SUMMARY
FINDINGS AND RECOMMENDATIONS . . . . . . . . . . . . . . . . . . 7 4.1 Resolution of Previously Identified Concerns . . . . . . . . . . . . . . . . . . . . . . . 7 4.2 The New Darmatt Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1
l l
l I
i l
l l
l 11
~7.._,.. '~' ~
j
FORWARD The United States Nuclear Regulatory Commission (USNRC) has solicited the suppon of Sandia National Laboratories (SNL) in the review of utility submittals associated with fire protection and electrical engineering. This letter report represents the second in a series of SNL review repons associated with Duane Arnold Energy Center (DAEC). The submittats being reviewed by SNL deal with the assessment of ampacity loads for fire barrier clad power cables. The first repon in this series was issued by SNL on April 5, 1996 and was generated under the auspices of USNRC JCN J-2017. As a result, the l
USNRC forwarded an Request for Additional Information (RAI) to the licensee on October 16,1996. The objective of the current effort was to review and assess the licensee's response to this RAI. This work was performed as Task Order 3, Subtask 5 of
, USNRC JCN J-2503.
1 i
i i
j
3 .
i-
1.0 INTRODUCTION
i 1.1 Background and Objective In response to USNRC Generic Letter 92-08, the Duane Amold Energy Center (DAEC) l provided documentation of the utility position regarding both the fire endurance rating and i
- ampacity derating factors associated with its installed fire barrier systems. The licensee :
l response to this request and to a subsequent USNRC Request for AdditionalInformation !
- (RAI) was provided in the form of two letters
- l
- Letter, February 14,1994 (item NG-94-0563), J. F. Franz, DAEC, to L. J.
Callan, USNRC,(with enclosures).
- Letter, June 2,1995 (item NG-95-1223), J. F. Franz, DAEC to W. T. Russell, i USNRC (with one attachment).
1 SNL reviewed these submittals under the terms of the general technical support contract '
! JCN J-2017, Task Order 9, Subtask 1. A letter report documenting an SNL review of this
} set of submittals was completed on April 5,1996.' As a result of this review a number of technical concerns were identified. A subsequent USNRC RAI was forwarded to the i licensee on October 16,1996 requesting resolution of the identified concems The i licensee has now responded to the USNRC RAI via the following document:
a
! - Letter, J. F. Franz, IES Utilities, to F. J. Miraglia, USNRC, December 13,1996 (reference licensee file A-107, P-72a).
In addition, the licensee also submitted an entirely new analysis to assess the ampacity loads for power cables in one set of penetrations wrapped in the fire barrier material Darmatt. This new calculation was forwarded under cover:
- Letter, K. E. Pevelar, IES Utilities to S. J. Collins, USNRC, May 16,1997 (reference licensee file A-107, P-72a).
The objective of the current SNL effort is to review these two new licensee submittals.
This letter report documents the findings and recommendations resulting from this SNL review.
1.2 Organization ofReport Section 2 provides a brief assessment of the licensee's December 1996 RAI response.
Section 3 provides a review of the licensee's new calculations as documented in the May 1997 submittal. Section 4 summarius the findings and recommendations resulting from these reviews
'See "A Review of the Duane Arnold Energy Center Analysis of Fire Barrier Ampacity Derating Factors," A letter report to the USNRC, Revision 0, April 5,1996, forwarded under cover S. Nowlen, SNL, to R. Jenkins, USNRC/NRR/EELB.
1
===; = .- _- ._
. o l 2.0 RESOLUTION OF PREVIOUSLY IDENTIFIED ISSUES i
l The licensee's onginal analyses as reviewed by SNL in April 1996 had been based on an application of the " Watts per foot" methodology. The SNL review had cited that this methodology was inherently inadequate to demonstrate the acceptability of ampacity .
loading factors for the cables installed at DAEC. In particular, the methodology provides for no assessment of the ampacity limits for any given individual cable. This was cited as an inherent limitation of the DAEC method that could not be easily corrected. SNL had also cited a number of additional points of technical concern regarding the specific case .
examples provided by the licensee in its submittal.
The licensee response to the USNRC RAI states that "Thermo-Lag has been removed from all raceway containing continuously energized power cables." The licensee goes on l
to state that only one fire barrier application involving continuously energized power j cables remains at the plant, and this application involves a Darmatt wrapped penetration. ;
I A separate and entirely new calculation has been provided for this one application as i discussed funherin Section 3 below.
As a result, the licensee has not responded to each of the speciSc items raised in the RAI.
Rather, the licensee has provided a general response to the overall RAI. This is based on the assumption that because the calculations are, in effect, being abandoned, the NRC staff will not funher pursue the concerns. The highlights of this response include the following:
- The licensee cited that it had recognized that the data upon which the chosen
! methodology was based was suspect, and cites further that it had planned to update the methodology as more data became available. The use ofinappropriate data as the basis for validation was one of the major concerns identified by SNL in its review.
- The licensee clarified that its definition of" continuously energized" included any circuit that could be continuously energized, including control circuits, or had a "long duty cycle". They funher state that all pumps, compressors, fans, heaters "and the like" were treated as continuos loads. This clarification resolves a second significant point of concern raised by SNL. This panicular point is also of relevance to the new calculations discussed in Section 3 below. SNL finds that the licensee response to this item is adequate to resolve the identified concern, including as applied to the new calculations.
- The licensee acknowledges that its methodology was taking credit for cable load i diversity, and that crediting diversity has been a point of debate within the power industry.
- The licensee also cites that in addition to the nominal ampacity loads dGived from the Esteves " Watts /ft" method, the licensee also imposed a maximum ampacity limit corresponding to the ICEA tray limits for a 30% fill. This constraint was not
- identified in the original submittal and would significantly impact the ampacity assessments. This additional constraint is of particular significance because it does I
2
,-w,n, .. .. . .
_ _ _ - _ _ _ g.y 7
indicate that the licensee had imposed an important check on the potential non-
. conservative results which can be obtained using the Watts /ft method.
- The licensee cites that all ofits conduit ampacity loads were sized in accordance with the NEC limits.
Based on these responses, SNL concurs with the licensee that further staff review of the identified concerns is not necessary. SNL finds that most of the concerns previously identified by SNL have been rendered moot by the licensee's abandonment ofits previously submitted calculations. A satisfactory response to the one items which remain ofrelevance to the new calculations has been provided satisfactory responses; namely, how continuously energized cables were defined and identified. SNL recommends that the USNRC "close out" the previously identified technical concems without further actions.
I l
i n
1 3
t
(
3.0 THE LICENSEE CALCULATIONS FOR DARMATT B ARRIERS 3.1 Overview In the May 1997 submittal, the licensee has provided an entirely new calculation to assess the operating temperatures of energized power cables in a Darmatt fire barrier system. !
The objective of this section is to review and assess the acceptability of this calculation. It !
is noted that the new calculation has no direct correspondence to the calculations l previously submitted by the licensee; hence, none of the review findings regarding the !
earlier licensee calculations are directly relevant. i Only one barder system is now being analyzed. This is a " boxed" enclosurejointly I surrounding a pair ofreactor building to drywall electdcal penetration assemblies. The l enclosure includes a number of conduits that enter into or pass through the fire barder. l I
One side of the " boxed" enclosure is represented by the concrete wall separating the drywell from the balance of the reactor building, and the other sides are all compdsed of Darmatt. The objective of the analysis is to conservatively estimate the operating j temperature of the energized power cables that pass through this barrier.
l 3.2 The Thermal Calculation , ;
j The calculation was performed under contract by engineers at Sargent and Lundy (S&L).
l The bulk of the calculation, including the general desedption of the analysis methodology and the specific mathematical implementation of the core thermal relationships, is cited by the licensee as the proprietary property of S&L, but the full calculation has been provided for review. The SNL discussions that follow have been wdtten in such a way as to avoid the disclosure of any of this proprietary information. This has necessitated a somewhat less detailed discussion of the methodology and implementation than might normally be
)
anticipated in a SNL review of this type. l The calculation itselfis quite complex in that it involves a rather complex thermal geometry, and a number of potential heating sources and effects. SNL has reviewed the model in some detail. In general, the model is intended to calculate the operating temperature of the clad power cables which pass through the fire barrier envelope. The l scope of the calculation appears to have properly accounted for all relevant factors including:
- Heat transfer via radiation, conduction, and convection are all appropriately credited and analyzed using an appropriate solver for simultaneous non-linear equations.
- Heat generation both within the cables leading up to the penetration assemblies and that generated within the penetration assemblies themselves has been conservatively estimated.
- The flow of heat through the concrete wall that forms one side of the bErier into the :
! protected envelope has been conservatively estimated and included in the calculation as appropriate.
l 4
'~
m_.__ - _ _ __ _ _ __._._ _._.
.. . l The licensee calculation is based en well founded and modern principals of heat transfer i analysis. The descriptions of the methodology, underlying assumptions, and chosen i correlations are all well written and clear. The cited heat transfer correlations were )
referenced appropriately and verified by SNL.
The actual implementation of the thermal model is also presented in a clear and understandable manner. While SNL has not attempted a direct implementation of the thennal model, it is quite clear that sufficient information has been provided to allow for such an implementation. The thermal model has been implemented using a commonly I available commercial software package called "MathCad" (a trademark product of the !
Mathsoft Inc.). MathCad is a well known analysis package that allows for the entry of mathematical relationships in a familiar equation format rather than in a cryptic programming language format. This has allowed for a direct evaluation of the thennal relationships implemented in the model, and no discrepancies in this regard were noted.
SNL found no discrepancies in the thermal model implementation, and was able to spot- 1 check a number ofintermediate results to verify the calculation process.
]
One factor ofparticular note is the S&L treatment of heat transfer within the electrical 1 penetration itself. This aspect of the problem is potentially quite complex and would be !
very difficult to predict analytically by a direct implementation of a heat transfer model.
S&L has taken a rather unique approach to this aspect of the problem that is based on electrical performance test data made available directly by the manufacturer. This ;
approach has significantly simplified this part of the analysis, and yet appears to have been handled in a fully appropriate manner. A particular point of strength in this aspect of the model in that S&L included an " extrapolation check" to confirm the modeling ,
assumptions. This has provided an adequate validation of this critical aspect of the thermal model. I l
3.3 Results and Reviewer Observations '
I The licensee calculation has estimated the maximum conductor temperature for all of the cables in this fire barrier system to be 56.5'C in comparison to a maximum allowable value of 90'C. This indicates a very high level of margin available for all cables, and the licensee has hence concluded that the cable ampacity loads are acceptable.
As noted above, SNL did find the licensee model to have been implemented in a thorough and well thought out manner. However, given the complexity of the thermal model, it is difficult to assure with 100% reliability that all aspects of the model are leading towards a j realistic or conservative result. SNL has made certain supplemental observations that lend i some additional qualitative support to the final conclusion regarding acceptability. These l I
include:
- The fire barrier system is comprised of a rather large physical box confi5 ration, and hence, has a very large surface area. At the same time the heat load intemal to the barrier system is rather modest. This leads to a very low heat load to-surface area ratio; hence, one should anticipate a lower nominal ADF for such a barrier than, for 5
- - - - - , w - - , - -
..__7.,. _7 . _ .
example, a typical cable tray fire barrier system would be expected to display. This
, is consistent with the licensee calculation.
The area housing the barrier is cited as having a maximum ambient temperature of 90'F. This is relatively low in comparison to typical values assumed in a cable initial :
ampacity design. Hence, the licensee likely realized some excess ampacity margin for the cables in this area.
l
- All of the cited cables have very high levels of margin available in comparison to l
I general conduit ampacity limits. SNL noted that all of the cited cables had margins of 50% or more available in comparison to the ICEA conduit ampacity limits. Given this margin, it is not surprising that the cable temperatures predicted were well below the nominal limits, even in the presence of the fire barrier system.
3.4 Findings and Recommendations SNL finds that the licensee has demonstrated that the cables within the subject Darmatt fire barrier envelope ofinterest are operating within acceptable ampacity limits. No l further actions on this calculation are recommended.
I l
i l
l l
6
4.0
SUMMARY
FINDINGS AND RECOMhENDATIONS 4.1 Resolution ofPreviously Identified Concerns SNL finds that, given that the previously analyzed fire barriers have all been removed, most of the previously identified concerns are rendered moot. For the one RAI item that might have impacted the new calculations, related to how the licensee defined
" continuously energized power cables," a satisfactory response was provided. SNL recommends that no further actions on the previously identified concerns be pursued.
. 4.2 The ew Darmatt Calculations SNL has reviewed the licensee thermal model in some detail. The thermal model was found to have been well conceived and well documented. No mistakes in application or implementation were identified. Hence, SNL finds that the licensee has demonstrated that the cables within the subject Darmatt fire barrier envelope ofinterest are operating within acceptable ampacity limits. It is recommended that the new calculation be accepted as an adequate demonstration of ampacity margin for this specific case.
7
- - . . . . . . _