Revision as of 06:16, 1 July 2018

Text

Stress Re-Evaluation of Nine Mile Point Unit 2 Steam Dryer at 115% CLTP, CDI Report No. 14-08NP, Revision 0, Non-proprietary Version
	ML15022A661
Person / Time
Site:	Nine Mile Point
Issue date:	07/31/2014
From:	Boschitsch A H; Continuum Dynamics
To:	; Office of Nuclear Reactor Regulation
Shared Package
ML15023A070	List: ML15022A661; ML15022A662; ML15022A663; ML15023A030; ML15023A031;
References
	NMP2L 2566 14-08NP
	Download: ML15022A661 (145)
	v • d • e

ATTACHMENT 2NONPROPRIETARY VERSION OF CDI REPORT NO.14-08P This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information CDI Report No. 14-08NPStress Re-Evaluation of Nine Mile Point Unit 2 Steam Dryerat 115% CLTPRevision 0Prepared byContinuum

Dynamics, Inc.34 Lexington AvenueEwing, NJ 08618Prepared under Purchase Order No. 7736902 forConstellation Energy GroupNine Mile Point Nuclear Station, LLCP.O. Box 63Lycoming, NY 13093Prepared byAlexander".

Boschitsch Approved byAlan J. BilaninJuly 2014This report complies with Continuum

Dynamics, Inc. Nuclear Quality Assurance Programcurrently in effect.

This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Executive SummaryThe stresses resulting from acoustic loads at the 115% CLTP operating condition (alsoreferred to herein as the extended power uprate or EPU condition) are re-evaluated for the NineMile Point Unit 2 (NMP2) steam dryer using a finite element model and frequency-based analysis methodology.

The re-evaluation is performed following identification

[1] andcorrection

[2] of an inconsistency in the acoustics loads predictions that resulted in a new set ofloads requiring a stress analysis.

The re-evaluation produced a new stress state that required amore detailed examination of the stress state at a limited number of locations and, for onelocation, necessitated installation of a modification to limit vibration of the inner side plateconnecting the inner vane banks. In all other respects, the finite element model of the steamdryer (including the modifications

[3-6] to achieve an alternating stress ratio at EPU operation ofSR-a>2) are essentially identical to those previously described in [5]. Similarly the stressevaluation is consistent with those carried out in the U.S. for prior dryer qualification to EPUconditions.

The resulting stresses are assessed for compliance with the ASME B&PV Code2007 [7],Section III, subsection NG, for the load combination corresponding to normaloperation (the Level A Service Condition).

The acoustic loads are prepared using the acoustic circuit model (ACM) version 4.1 R [2, 8,9]. This version represents the analysis that has been corrected to address an inconsistency

[1] inthe representation of the acoustic solution in the gap between the above-water skirt and reactorvessel wall, and also a minor deviation associated with the use of single rather than doubleprecision in the Helmholtz solver used to procure acoustic loads. Three load conditions areconsidered:

Baseline:

This is the normal operating load associated with EPU steam dryer operation.

Prior to power ascension this was the only relevant load at 115% CLTP power.Drain Trap Out-of-Service:

This off-normal, but not infrequent operating condition wasidentified during power ascension and is associated with a noticeable 92.5 Hz signaloccurring when the drain trap in the reactor core isolation cooling (RCIC) system is isolated.

RCIC Valve Closed: This off-normal operating load occurs when the RCIC valve isintentionally closed and is characterized by an 89.3 Hz signal.The present analysis evaluates the complete dryer using the Baseline load processed using ACM4.1 R. The identified limiting locations are then used to construct a node list that in turn isprovided as input to real time stress evaluations at the other off-normal conditions.

In a real timestress evaluation, stresses are only calculated at the nodes in this list rather than for the entiredryer. As described in Section 5.7, the real time evaluations use conservative MSL entrancesignals developed by retaining the higher of the total uncertainty identified with the ACM 4.1and ACM 4.1 R loads models over each frequency interval.

It is found that application of the baseline load to the steam dryer results in several locations requiring additional refined modeling to accurately define the alternating stress ratio. Theselocations have been addressed through a combination of high resolution

modeling, repairing poorquality mesh elements and improved modeling of connections to reproduce the as-builtconfiguration.

For one location involving the common junction between the inner side plate, topi This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information and side plates of the inner vane banks and tie bar, it was shown that the recommended margin of2.0 could be achieved using a combination of embedded modeling and stress interpolation alongthe weld lines. However, uncertainty regarding the applicability of the interpolation procedure atthis complex geometry junction involving multiple welds motivated consideration of alternate concepts for meeting margin. The long-term resolution ultimately adopted was to install the U-stiffener modification.

This modification-based approach was considered appropriate as the highstress estimate was attributed to a vibration mode of the end plate and was further increased bythe frequency shifting required by the analysis.

With this channel in place the real time stress evaluation shows that all nodes have a peakstress ratio, SR-P, of 1.3 or higher at all load combinations thus meeting the required margin forthis stress type. With regard to alternating

stresses, all of the nodes on the steam dryer have analternating stress ratio of 2.0 or higher under the baseline and drain trap out-of service loads sothat the dryer qualifies for these conditions.

For the RCIC valve closed, certain locations havealternating stress ratios below 2.0 with the minimum value being SR-a=I.5.

However, since thiscondition occurs infrequently, it is appropriate to assess fatigue using cycle counting.

Thecumulative usage factors (CUFs) for these locations are calculated in [10] and shown to all liewell below 1.0. Taken in their entirety, these results show that the dryer qualifies for level Aservice operation with the U-section stiffener installed.

In producing these results refined estimates of the linearized stresses at selected high stresslocations were obtained using (3)ii This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table of ContentsSection PageExecutive Sum m ary .........................................................................................................................

iTable of Contents

...........................................................................................................................

iiiNom enclature

.................................................................................................................................

iv1. Introduction and Purpose ............................................................................................................

11.1 Summ ary of Overall Evaluation M ethodology

.................................................................

32. M ethodology

& Evaluation Procedures

.................................................................................

52.1 Overview

...............................................................................................................................

52 .2 [[ ......................................................

(3)]] ..............................................................

72.3 Computational Considerations

..........................................................................................

83. Finite Element M odel Description

........................................................................................

113.1 Steam Dryer Geometry

...................................................................................................

113.2 M aterial Properties

..............................................................................................................

163.3 M odel Simplifications

......................................................................................................

163.4 Perforated Plate M odel ...................................................................................................

173.5 Vane Bank M odel ...............................................................................................................

193.6 W ater Inertia Effect on Submerged Panels ......................................................................

203.7 Structural Damping ........................................................................................................

203.8 M esh Details and Element Types ...................................................................................

203.9 Connections between Structural Components

..............................................................

213.10 Pressure Loading ...............................................................................................................

333.11 [[ ....................................................

( ].......................

364. Structural Analysis

....................................................................................................................

374.1 Static Analysis

....................................................................................................................

374.2 Harm onic Analysis

..............................................................................................................

374.3 Post-Processing

...................................................................................................................

434.4 Computation of Stress Ratios for Structural Assessment

...............................................

43[[4.5 .............................................

(3)]] .................................................................

465 .R e su lts .......................................................................................................................................

5 35.1. Preliminary Stress Assessment of Normal EPU Load + Base Dryer Model .................

545.2 Construction of Real Tim e Node List .............................................................................

655.3 Examination of Low Stress Ratio Locations

.................................................................

695.4 General Stress Distribution and High Stress Locations

.................................................

815.5 Load Combinations and Allowable Stress Intensities

...................................................

905.6 Frequency Content and Filtering of the Stress Signals .....................................................

1085.7 Real Time Analysis W ithout U-Stiffener

........................................................................

1175.8 Real Time Analysis Adjusted with U-Section Stiffener Included

...................................

1216. Conclusions

.............................................................................................................................

1237. References

...............................................................................................................................

125[[ .................................................

]]..... ............................

128Appendix B. U-Section Stiffener

...............................................................................................

132iii This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Nomenclature ACM acoustic circuit modelCDI Continuum

Dynamics, Inc.CLTP current licensed thermal powerCUF cumulative usage factorDOF degree of freedomDRF design record fileFFT fast Fourier transform FSI fluid-structure interaction EPU extended power uprateFEA finite element analysisIGSCC inter-granular stress corrosion crackingMSL main steam lineNRC Nuclear Regulatory Commission ooS out-of-service Pm membrane stress intensity Pb bending stress intensity PT pressure transducer QC Quad CitiesRCIC reactor core isolation coolingRFO refueling outageRPS reduced point set (described in Section 5)RPV reactor pressure vesselSa service limit for alternating stress intensity Salt alternating stress intensity SER safety evaluation reportSm service limit for membrane stress intensity SR-a alternating stress ratioSR-P peak stress ratioSRF stress reduction factorSS stainless steelNMP2 Nine Mile Point Unit 2USR upper support ringWEC Westinghouse Electric Corporation WF weld factoriv This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information

1. Introduction and PurposeCurrent licensing procedures to qualify the Nine Mile Point Unit 2 (NMP2) nuclear plant foroperation at Extended Power Uprate (EPU) operating condition require a stress assessment of thesteam dryer to ensure adequate stress margins under the increased loads. The steam dryer loadsdue to acoustic pressure fluctuations in the main steam lines (MSLs) are potentially damagingand the cyclic stresses from these loads can produce fatigue cracking if loads are sufficiently high. The industry has addressed this problem with physical modifications to the dryers, as wellas a program to define steam dryer loads and their resulting stresses.

The EPU qualification process requires a stress evaluation for the NMP2 steam dryer using acoustic loads acquired atEPU operation.

The present report documents this stress evaluation by calculating the maximumand alternating stresses generated using strain gage MSL pressure measurements acquired atEPU operation.

It updates a previous stress evaluation carried out in [5] to account foridentification

[1] and correction

[2] of an inconsistency in the acoustics loads predictions whichresulted in a new set of loads requiring a stress analysis.

The load combination considered here corresponds to normal operation (the Level A ServiceCondition) and includes fluctuating pressure loads developed from NMP2 main steam line data,and weight. The fluctuating pressure loads, induced by the flowing steam, are predicted using aseparate acoustic circuit analysis of the steam dome and main steam lines [13]. Level B serviceconditions, which include seismic loads, are not included in this evaluation.

Stress ratios areobtained by comparing these stresses (appropriately adjusted at welds) against allowable valuesand used to ensure compliance with the ASME Code (ASME B&PV Code,Section III,subsection NG).The stress analysis is carried out in the frequency domain, which confers a number of usefulcomputational advantages over a time-accurate transient analysis including the ability to assessthe effects of frequency scaling in the loads without the need for additional finite elementcalculations.

The analysis develops a series of unit stress solutions corresponding to theapplication of a unit pressure at a MSL at specified frequency,

f. Each unit solution is obtainedby first calculating the associated acoustic pressure field using a separate analysis that solves theHelmholtz equation within the steam dryer [14]. This pressure field is then applied to a finiteelement structural model of the steam dryer and the harmonic stress response at frequency, f, iscalculated using the commercial ANSYS 10.0 finite element analysis software.

This stressresponse constitutes the unit solution and is stored as a file for subsequent processing.

Once allunit solutions have been computed, the stress response for any combination of MSL pressurespectrums (obtained by Fast Fourier Transform of the pressure histories in the MSLs) isdetermined by a simple matrix multiplication of these spectrums with the unit solutions.

Detailsof the frequency-based stress evaluation methodology are contained in Section 2.0.EPU Load Conditions The stress evaluation is performed for three different EPU load conditions described in [2].In addition to the normal baseline EPU load, two other loads were identified during powerascension.

The first occurs when the drain trap is out of service (ooS) and primarily differs fromthe baseline load by the presence of a distinct 92.5 Hz signal. This condition occurs sufficiently often that it is treated as an alternative

'base' EPU load. It mostly, but not always, producesI This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information slightly higher stresses and so is considered along with (rather then replacing) the baseline EPUload. The third condition occurs when the reactor core isolation cooling (RCIC) valve is closedand is characterized by an 89.3 Hz signal. This signal occurs infrequently and therefore isamenable to fatigue assessment based on cycle counting methods to calculate the cumulative usage factor.The signal files identified with these three EPU loads are documented in [2] as:20120721142636 (baseline EPU); 20130524092644 (drain trap out-of service);

and20120904092600 (RCIC valve closed).Acoustic Loads Estimation The current stress evaluation of the NMP2 steam dryer is performed using acoustic loadsgenerated using a revised Acoustic Circuit Model (ACM) Rev. 4.1 R [2, 8, 9]. The development of this revision was motivated primarily by a requirement for consistent usage of noise filtering strategies during both model calibration against available data and application of the model toplants. Other than the removal of known non-acoustic discrete frequencies (e.g., electrical noiseat multiples of 60 Hz) and the application of coherence filtering (which was also invoked whenprocessing the Quad Cities data) no other filtering methods are used. In particular, no noisesubtraction using low power data is performed.

Further details of the ACM Rev. 4.1 Rcalibration activity and its application to obtain NMP2 steam dryer acoustic loads are detailed in[2, 8, 9]. As described in [9] re-benchmarking the ACM against available Quad Cities dataproduced updated estimates of the acoustic speed and damping in the acoustics description andalso revised biases and uncertainties due to changes in the model, coherence-based noise filtering and comparison method. The biases and uncertainties used for the present load estimates arebased on the comparison with QC data at 790MWe using 16 sensors.

For each frequency

interval, the biases and uncertainties obtained in ACM 4.1 R are similar to those in ACM 4.1.While it is technically consistent to use the ACM 4. 1R values in the evaluations, the higher ofthe ACM 4.1 and ACM 4.1 R total uncertainties (bias + uncertainty) over each frequency intervalis also defined to be consistent with the NMP2 SER. The real time stress assessments are carriedout using these conservative signals consistent with the NMP2 SER[ 15].Stress Processing The frequency-based harmonic stress evaluation methodology, finite element model andpost-processing procedures

((3)) arefully identical to those described in the previous NMP2 steam dryer stress evaluation at EPUconditions

[5] and are described in Sections 2-4. In order to qualify the NMP2 steam dryer fornormal EPU operation it is required that the limiting alternating stress ratio be above a targetlevel of 2.0. To meet margin several modifications to the dryer were implemented prior to powerascension (in 2012). The modifications made to the dryer are described in Section 5 of [5] andalso detailed in [6]. These modifications are fully accounted for in the current stress evaluation including the stiffened closure plate, the masses added to the inner and middle hoods, and the1/8" thick reinforcement plate placed over the middle hood section outboard of the closure plate.Based on the present stress evaluation an additional modification consisting of a U-section stiffener bolted onto the inner side plate spanning the inner vane banks, was added. Thismodification is described in Appendix B.2 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information (3)]]1.1 Summary of Overall Evaluation Methodology The stress evaluation involves:

3 different collected signals -baseline, drain trap ooS, and RCIC valve closed.* 2 bias and uncertainty combinations

-one pertaining to ACM 4.1R and the othercorresponding to the higher of the total uncertainty in ACM 4.1 and ACM 4.1R ineach frequency interval.

4 different steam dryer models -(i) the baseline model for which the full calculation was performed over the full 0-250 Hz frequency range; (ii) a revised model where theconnections between the lifting rod and upper lifting rod brace are changed to the as-built configuration; (iii) an improved baseline model where a poor quality grid near alifting rod brace was remeshed to improve element quality; and (iv) a modified modelcontaining the U channel stiffener on the inner side plate. The latter models wereonly developed over reduced frequency ranges.Given the large number of analysis permutations, an overall strategy was developed thatconducted a full dryer analysis using the baseline load and available full frequency range unitsolution, and then executed a series of real time evaluations to analyze the other permutations.

This strategy proceeded as follows (sections where the calculations are performed are given inparentheses):

1. Analyze complete dryer using the baseline load and baseline structural model(Section 5.1).2. Identify the limiting locations and assemble a list of nodes for real time processing (Section 5.2).3. For locations with alternating stress ratios, SR-a<2, exercise analytical options toobtain more accurate stress estimates.

For the current analysis these options included:

a. Improving mesh quality about high stress locations.

b. Developing and/or utilizing embedded models at high stress sites.c. Changing the connection between the lifting rod and top brace from welded(indicated in drawings) to non-welded (as-built).

d. Interpolating along the welds to distinct hotspots

-this is only done at the topsof the welds connecting the closure plates and curved (inner and middle)hoods.(Section 5.3)3 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information

4. Using the results from step 3, repeat step 1. Also generate stress ratio tables andPSDs (Sections 5.4-5.6).

5. Perform real time stress evaluations for the other loads and total uncertainties usingthe node list generated in step 2 (Section 5.7).6. Consider modifications for locations with alternating stress ratios that remain below2.0 in step 5. For the current stress evaluation, the resulted in the attachment of a U-shaped cross-section stiffener to the inner side plate.7. Conduct real time analyses using a FEA model that implements the modifications instep 6.8. For locations that still have alternating stress ratios, SR-a<2.0 either: (i) consideradditional modifications and repeat real time analysis (this option proved unnecessary in the current evaluation),

or (ii) for stress ratios below 2.0 occurring at the RCICvalve closed EPU condition conduct cycle counting to confirm cumulative fatigueusage below 1.0 for this infrequent load condition.

The results from the execution of this overall strategy are presented in the following sections:

Section 5 presents a full stress assessment of the baseline structural model subjected to thebaseline load (steps I & 2). The results from this analysis motivate Section 6 which summarizes the measures carried out to address the locations with alternating stress ratios below 2.0. Theseinclude both analytical approaches (step 4) and the addition of physical modification (the U-section stiffener

-step 6). Section 7 presents a reanalysis of the full steam dryer under thebaseline load, but with the analytical corrections in step 4 now applied (part of step 5). It alsoincludes a series of real time analyses at the other load conditions both without (other part of step5) and with (step 7) the U-section stiffener.

This stress evaluation reports that the limiting alternating stress ratio on the dryer at thebaseline EPU load is SR-a=2.0 and at the drain trap ooS condition, SR-a=2.0.

These results arebased on the most conservative combinations of the bias and uncertainties from ACM 4.1 andACM 4.1 R, and assume that the U-section stiffeners are installed.

For the RCIC valve closedcondition, the limiting alternating stress ratio, SR-a=1.52.

This value meets the ASME stressmargin, but not the NRC staff recommended value of 2.0. However, because this load condition is infrequent it is addressed using cycle counting in [10]. The limiting peak stress ratio for anyof the three load conditions due to maximum membrane and bending stresses including staticcontributions is SR-P=I.3.

These values show that the present modified steam dryer meets therecommended stress margin at EPU operation.

Flaw evaluations for flaws identified in the original dryer are summarized in [16] and inSection 2.5 of [5]; evaluations of flaws identified during 2014 refueling outage are documented in [17].4 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information

2. Methodology

& Evaluation Procedures 2.1 OverviewBased on previous analysis undertaken at Quad Cities Units 1 and 2, the steam dryer canexperience strong acoustic loads due to the fluctuating pressures in the MSLs connected to thesteam dome containing the dryer. C.D.I. has developed an acoustic circuit model (ACM) that,given a collection of strain gage measurements of the fluctuating pressures in the MSLs, predictsthe acoustic pressure field anywhere inside the steam dome and on the steam dryer [8, 9, 13, 14,18]. The ACM is formulated in frequency space and contains two major components that aredirectly relevant to the ensuing stress analysis of concern here. (3)5 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information (3)]]6 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information (3)]]2.2 [[I(3)(3)]]7 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information (3)112.3 Computational Considerations Focusing on the structural computational aspects of the overall approach, there are a numberof numerical and computational considerations requiring attention.

The first concerns thetransfer of the acoustic forces onto the structure, particularly the spatial and frequency resolutions.

The ANSYS finite element program inputs general distributed pressure differences using a table format. This consists of regular 3D rectangular (i.e., block) nxxnyxnz mesh wheren(, is the number of mesh points in the i-th Cartesian direction and the pressure difference isprovided at each mesh point (see Section 3.10). These tables are generated separately using aprogram that reads the loads provided from the ACM software, distributes these loads onto thefinite element mesh using a combination of interpolation procedures on the surface and simplediffusion schemes off the surface (off-surface loads are required by ANSYS to ensure properinterpolation of forces),

and written to ASCII files for input to ANSYS. A separate load file iswritten at each frequency for the real and imaginary component of the complex force.The acoustic field is stored at 5 Hz intervals from 0 to 250 Hz. While a 5 Hz resolution issufficient to capture frequency dependence of the acoustic field (i.e., the pressure at a pointvaries gradually with frequency),

it is too coarse for representing the structural responseespecially at low frequencies.

For 1% critical structural

damping, one can show that thefrequency spacing needed to resolve a damped resonant peak at natural frequency, fn, to within5% accuracy is Af=0.0064xfn.

Thus for fn=10 Hz where the lowest structural response modesoccur, a frequency interval of 0.064 Hz or less is required.

In our calculations we require that5% maximum error be maintained over the range from fn=5 Hz to 250 Hz resulting in a finestfrequency interval of 0.0321 Hz at the low frequency end (this adequately resolves all structural modes up to 250 Hz). Since there are no structural modes between 0 to 5 Hz, a 0.5 Hz spacing isused over this range with minimal (less than 5%) error. The unit load, fn(O),R),

at anyfrequency, Ok, is obtained by linear interpolation of the acoustic solutions at the two nearestfrequencies, coi and coi+1, spaced 5 Hz apart. Linear interpolation is sufficient since the pressureload varies slowly over the 5 Hz range (linear interpolation of the structural response would notbe acceptable over this range since it varies much more rapidly over the same interval).

Detailsregarding the frequency resolution have been provided in [20].Solution Management

(3)8 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information (3)]]Structural DampingIn harmonic analysis one has a broader selection of damping models than in transient simulations.

A damping factor, z, of 1% critical damping is used in the structural analysis.

Intransient simulations, this damping can only be enforced exactly at two frequencies (where thedamping model is "pinned").

Between these two frequencies the damping factor can byconsiderably

smaller, for example 0.5% or less depending on the pinning frequencies.

Outsidethe pinning frequencies, damping is higher. With harmonic analysis it is straightforward toenforce very close to 1% damping over the entire frequency range. In this damping model, thedamping matrix, D, is set toD=2zK (7)()where K is the stiffness matrix and co the forcing frequency.

When comparing the responseobtained with this model against that for a constant damping ratio, the maximum difference atany frequency is less than 0.5%, which is far smaller than the 100% or higher response variation obtained when using the pinned model required in transient simulation.

Load Frequency Rescaling One way to evaluate the sensitivity of the stress results to approximations in the structural modeling and applied loads is to rescale the frequency content of the applied loads. In thisprocedure the nominal frequencies, cok, are shifted to (l+X)cok, where the frequency shift, X,ranges between +10%, and the response recomputed for the shifted loads. The objective of thefrequency shifting can be explained by way of example.

Suppose that in the actual dryer a strongstructural-acoustic coupling exists at a particular frequency, o*. This means that the following conditionshold simultaneously:

(i) the acoustic signal contains a significant signal at co*; (ii) thestructural model contains a resonant mode of natural frequency, con, that is near co*'; and (iii) theassociated structural mode shape is strongly coupled to the acoustic load (i.e., integrating theproduct of the mode shape and the surface pressure over the steam dryer surface produces asignificant modal force). Suppose now that because of discretization errors and modelingidealizations that the predicted resonance frequency differs from co* by a small amount (e.g.,1.5%). Then condition (ii) will be violated and the response amplitude therefore significantly diminished.

By shifting the load frequencies one re-establishes condition (ii) when (1+ X)"o* is9 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information near (On- The other two requirements also hold and a strong structural acoustic interaction isrestored.

[[I(3)]Evaluation of Maximum and Alternating Stress Intensities Once the unit solutions have been obtained, the most intensive computational steps in thegeneration of stress intensities are: (i) the FFTs to evaluate stress time histories from (5); and(ii) the calculation of alternating stress intensities.

(3)The high computational penalty incurred in calculating the alternating stress intensities is dueto the fact that this calculation involves comparing the stress tensors at every pair of points in thestress history.

This comparison is necessary since in general the principal stress directions canvary during the response, thus for N samples in the stress history, there will be (N-1)N/2 suchpairs or, for N=64K (the number required to accurately resolve the spectrum up to 250 Hz in0.01 Hz intervals),

2.1 x 109 calculations per node each requiring the determination of the roots toa cubic polynomial.

(3)10 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information

3. Finite Element Model Description A description of the ANSYS model of the Nine Mile Point Unit 2 steam dryer follows.3.1 Steam Dryer GeometryA geometric representation of the Nine Mile Point Unit 2 steam dryer was developed fromavailable drawings (provided by Constellation Energy Group and included in the design recordfile, DRF-C-279C) within the Workbench module of ANSYS. The completed model is shown inFigure 1. This model includes on-site modifications to the Nine Mile Point Unit 2 steam dryer.These are as follows.On-Site Modifications (Before 2012 and EPU operation)

(i) The top tie rods are replaced with thicker ones.(ii) Inner side plates are replaced with thicker ones.(iii) Middle hoods are reinforced with additional strips.(iv) Lifting rods are reinforced with additional gussets.(v) Per FDDR KG 1-0265 the support conditions are adjusted to ensure that the dryeris supported 100% on the seismic blocks.These additional modifications have been incorporated into the NMP2 steam dryer modeland are reflected in the results presented in this report. The affected areas are shown in Figure 2.Modifications Implemented for EPU Operation In [23] several modifications were proposed to meet target EPU stress margins using aprevious acoustic loads model (ACM Rev. 4.0) without noise subtraction.

These modifications are now superseded here by the ones below and detailed in Section 5 that are obtained by on thebasis of acoustic loads processed using the ACM Rev. 4.1R analysis.

These plannedmodifications include:(vi) Reinforcement strips are added to the closure plates.(vii) Increase the attachment weld size of the lower-most lifting rod brace from 1/4" to1/22.(viii) Reinforcements to the upper-most and middle lifting rod braces are made in theform of additional strengthening plates.(ix) A 1/8th in curved plate is placed over the middle hood section lying outboard ofthe closure plate.(x) Four 15 lb masses are added to the central inner hood panels.(xi) Stress relief cut-outs are added to the outer hood/hood support/base platejunctions to alleviate local stresses.

(xii) A wrap-around weld is added to the bottom of the drain channel/skirt weld.(xiii) Four 10 lb masses are added to the central middle hood panels.All of the modifications summarized here and detailed in [4], [6] and Section 5 of [24] areimplemented in the full steam dryer evaluation produced in Section 5.11 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Modification Implemented During 2014 Refiueling OutageFollowing the revision to the acoustic loads analysis from ACM 4.1 to ACM 4.IR and theidentification of additional signals during power ascension and subsequent collection of data [2],an additional modification was made to the dryer to suppress vibration of the inner side platespanning the two inner vane banks. The modification consists of a:(xiv) Stiffener with a U-shaped cross-section that is bolted to the inner side of the innerside plate spanning the two inner vane banks. The channel whose upper edge islocated 11" below the upper edge of the inner side plate, is described in AppendixB. The evaluation of the bolt stresses is described in [25].A summary of these modifications, references for additional details and how they areimplemented in the finite element analysis is given in Table 1.Reference FrameThe spatial coordinates used herein to describe the geometry and identify limiting stresslocations are expressed in a reference frame whose origin is located at the intersection of thesteam dryer centerline and the plane containing the base plates (this plane also contains the top ofthe upper support ring and the bottom edges of the hoods). The y-axis is parallel to the hoods,the x-axis is normal to the hoods pointing from MSL C/D to MSL A/B, and the z-axis is vertical, positive up.12 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 1. Summary of modifications made to the NMP2 steam dryer.Reinforcement

/ Modification Details FEA Implementation

1. Add reinforcement ribs to all (8) closure plates. Section 5.5 in [5] Closure plates are thickened to obtain dynamically equivalent structure a described in Section 3.52. Increase weld of the lowest lifting rod brace/vertical Section 5.4 in [5]; [[plate welds to 0.5" Section 3.5 in [12] (3)]]3. Reinforce middle and upper lifting rod braces to Section 5.1 in [5] Reduce stresses by 0.18 at this location based oneliminate stress concentration on weld to vertical plate. FEA reductions shown for Concept 2 in Table 11 of[26])4. Add 1/8" thick plate over the middle hood section Section 5.2 in [5] Thicken the existing plate by 1/8".lying between the closure plate and existingreinforcement strip.5. Add total of four 15 lb masses to the central sections Section 5.3 in [5] Place 15 lb point masses on the inner hoods at theof the inner hoods. mass centers.6. Add stress relief cut-out at the bottom edge of the Section 5.4 in [5]; [[outer hood supports.

Section 3.4 in [12] (3)]]7. Reinforce the bottom of the drain channel/skirt weld Section 5.4 in [5]; [[with thickened wrap-around weld. Section 3.1 in [12] (3)]]8. Add total of four 10 lb masses to the central sections Section 5.4 in [5] Place 10 lb point masses on the middle hoods at theof the middle hoods. mass centers.9. Add U-section stiffener to inner side plate Appendix B Add shell element model of stiffening using beamconnecting inner vane banks. elements to represent attachment bolts.13 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information ANSYS1ln%

FBENCH/azY50.00Figure 1. Overall geometry of the Nine Mile Point Unit 2 steam dryer model.14 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 2. Existing on-site pre-EPU modifications accounted for in the model and associated geometrical details.15 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information 3.2 Material Properties The steam dryer is constructed from Type 304 stainless steel and has an operating temperature of 550'F. Properties used in the analysis are summarized below in Table 2.Table 2. Material properties.

Young's Modulus Density Poisson(106 psi) (Ibm/in 3) Ratiostainless steel 25.55 0.284 0.3structural steel with added water 25.55 0.856 0.3inertia effectThe structural steel modulus is taken from Appendix A of the ASME Code for Type 304Stainless Steel at an operating temperature 550'F. The effective properties of perforated platesand submerged parts are discussed in Sections 3.4 and 3.6. Note that the increased effective density for submerged components is only used in the harmonic analysis.

When calculating thestress distribution due to the static dead weight load, the unmodified density of steel(0.284 lbmr/in3) is used throughout.

Inspections of the NMP Unit 2 dryer have revealed IGSCC cracks in the upper support ring(USR) and skirt. A separate analysis of these cracks [27] has been performed to determine whether:

(i) they will propagate further into the structure and (ii) their influence upon structural response frequencies and modes must be explicitly accounted for. To establish condition (i) thestress calculated in the global stress analysis is used in conjunction with the crack geometry tocalculate the stress intensity factor which is then compared to the threshold stress intensity.

Forthe USR and skirt cracks the highest stress intensity factors are 1.47 ksi-in05 and 2.75 ksi-in°5respectively; both values are below the threshold value (3 ksi-in05) implying that fatigue crackgrowth will not occur.To determine (ii) the change in modal response frequencies due to the presence of a flaw ispredicted by analytical means (in the case of the USR) or using finite element analysis (for theskirt). In each case, the flaw size used in these calculations is increased to ensure conservative estimates (for example, in the case of the skirt flaws extending up to 1/22 the panel width areconsidered).

For the USR, the change in modal frequencies due to the presence of the cracks isless than 0.5%. For the skirt, using a conservative estimate for the crack to panel width of 0.3(the measured value is less than 0.17) the change in modal frequency is also less than 0.5%. Inboth cases such small changes in modal frequencies are considered negligible and are readilyaccounted for when performing frequency shifting.

3.3 Model Simplifications The following simplifications were made to achieve reasonable model size while maintaining good modeling fidelity for key structural properties:

Perforated plates were approximated as continuous plates using modified elasticproperties designed to match the static and modal behaviors of the perforated plates. The16 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information perforated plate structural modeling is summarized in Section 3.4 and Appendix C of[28]." The drying vanes were replaced by point masses attached to the corresponding troughbottom plates and vane bank top covers (Figure 4). The bounding perforated plates, vanebank end plates, and vane bank top covers were explicitly modeled (see Section 3.5)." The added mass properties of the lower part of the skirt below the reactor water levelwere obtained using a separate hydrodynamic analysis (see Section 3.6).(3)1]" Four steam dryer support brackets that are located on the reactor vessel and spaced at 900intervals were explicitly modeled (see Section 3.9)." Most welds were replaced by node-to-node connections; interconnected parts sharecommon nodes along the welds. In other locations the constraint equations betweennodal degrees of freedom were introduced as described in Section 3.9.3.4 Perforated Plate ModelThe perforated plates were modeled as solid plates with adjusted elastic and dynamicproperties.

Properties of the perforated plates were assigned according to the type and size ofperforation.

Based on [29], for an equilateral square pattern with given hole size and spacing,the effective moduli of elasticity were found.The adjusted properties for the perforated plates are shown in Table 3 as ratios to materialproperties of structural steel, provided in Table 2. Locations of perforated plates are classified by steam entry / exit vane bank side and vertical position.

Tests were carried out to verify that this representation of perforated plates by continuous ones with modified elastic properties preserves the modal properties of the structure.

These testsare summarized in Appendix C of [28] and compare the predicted first modal frequency for acantilevered perforated plate against an experimentally measured value. The prediction wasobtained for 40% and 13% open area plates (these are representative of the largest and lowestopen area ratios of the perforated plates at NMP2, as seen in Table 3) using the analytical formula for a cantilevered plate and the modified Young's modulus and Poisson's ratio given byO'Donnell

[29]. The measured and predicted frequencies are in close agreement, differing byless than 3%.(3)17 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information

[1(3)]][1(3)]]Figure 3. I(3)18 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 3. Material properties of perforated plates.I(3)3.5 Vane Bank ModelThe vane bank assemblies consist of many vertical angled plates that are computationally expensive to model explicitly, since a prohibitive number of elements would be required.

Theseparts have significant weight, which is transmitted through the surrounding structure, so it isimportant to capture their gross inertial properties.

Here the vane banks are modeled as acollection of point masses located at the center of mass for each vane bank section (Figure 4).The following masses were used for the vane bank sections, based on data found on provideddrawings:

inner banks, 1618 Ibm, 4 sections per bank;middle banks, 1485 Ibm, total 4 sections per bank; andouter banks, 1550 Ibm, 3 sections per bank.These masses were applied to the base plates and vane top covers using the standard ANSYSpoint mass modeling option, element MASS2 1. ANSYS automatically distributes the point massinertial loads to the nodes of the selected structure.

The distribution algorithm minimizes thesum of the squares of the nodal inertial forces, while ensuring that the net forces and momentsare conserved.

Vane banks are not exposed to main steam lines directly, but rather shielded bythe hoods.The collective stiffness of the vane banks is expected to be small compared to thesurrounding support structure and is neglected in the model. In the static case it is reasonable toexpect that this constitutes a conservative

approach, since neglecting the stiffness of the vanebanks implies that the entire weight is transmitted through the adjacent vane bank walls andsupports.

In the dynamic case the vane banks exhibit only a weak response since (i) they havelarge inertia so that the characteristic acoustically-induced forces divided by the vane massesand inertias yield small amplitude

motions, velocities and accelerations; and (ii) they areshielded from acoustic loads by the hoods, which transfer dynamic loads to the rest of thestructure.

Thus, compared to the hoods, less motion is anticipated on the vane banks so that19 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information approximating their inertial properties with equivalent point masses is justified.

Nevertheless, the bounding parts, such as perforated plates, side panels, and top covers, are retained in themodel. Errors associated with the point mass representation of the vane banks are compensated for by frequency shifting of the applied loads.3.6 Water Inertia Effect on Submerged PanelsWater inertia was modeled by an increase in density of the submerged structure to accountfor the added hydrodynamic mass. This added mass was found by a separate hydrodynamic analysis (included in DRF-C-279C supporting this report) to be 0.143 Ibm/in2 on the submerged skirt area. This is modeled by effectively increasing the material density for the submerged portions of the skirt. Since the skirt is 0.25 inches thick, the added mass is equivalent to adensity increase by 0.572 lbm/in3.This added water mass was included in the ANSYS model byappropriately modifying the density of the submerged structural elements when computing harmonic response.

For the static stresses, the unmodified density of steel is used throughout.

3.7 Structural DampingStructural damping was defined as 1% of critical damping for all frequencies.

This dampingis consistent with guidance given on pg. 10 of NRC RG-1.20 [33].3.8 Mesh Details and Element TypesShell elements were employed to model the skirt, hoods, perforated plates, side and endplates, trough bottom plates, reinforcements, base plates and cover plates. Specifically, the four-node, Shell Element SHELL63, was selected to model these structural components.

Thiselement models bending and membrane

stresses, but omits transverse shear. The use of shellelements is appropriate for most of the structure where the characteristic thickness is smallcompared to the other plate dimensions.

For thicker structures, such as the upper and lowersupport rings, solid brick elements were used to provide the full 3D stress. The elementsSURF154 are used to assure proper application of pressure loading to the structure.

Mesh detailsand element types are shown Table 4 and Table 5.The mesh is generated automatically by ANSYS with refinement near edges. The maximumallowable mesh spacing is specified by the user. Here a 2.5 inch maximum allowable spacing isspecified with refinement up to 1.5 inch in the following areas: drain pipes, tie rods, the curvedportions of the drain channels and the hoods. Details of the finite element mesh for the baselinemodel (i.e., without the U-section stiffeners) are shown in Figure 5. Numerical experiments carried out using the ANSYS code applied to simple analytically tractable plate structures withdimensions and mesh spacings similar to the ones used for the steam dryer, confirm that thenatural frequencies are accurately recovered (less than I % errors for the first modes). Theseerrors are compensated for by the use of frequency shifting.

The baseline analysis is carried out without the U-section stiffener.

A supplemental model isalso developed with the U-section stiffener represented using shell elements and connected to theinner side plate as shown in Figure 25 using beam elements to represent the bolts. This model isdescribed further in Appendix B.20 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information 3.9 Connections between Structural Components Most connections between parts are modeled as node-to-node connections.

This is thecorrect manner (i.e., within the finite element framework) of joining elements away fromdiscontinuities.

At joints between shells, this approach omits the additional stiffness provided bythe extra weld material.

Also, locally 3D effects are more pronounced.

The latter effect isaccounted for using weld factors.

The deviation in stiffness due to weld material is negligible, since weld dimensions are on the order of the shell thickness.

The consequences upon modalfrequencies and amplitude are, to first order, proportional to t/L where t is the thickness and L acharacteristic shell length. The errors committed by ignoring additional weld stiffness are thussmall and readily compensated for by performing frequency shifts.When joining shell and solid elements,

however, the problem arises of properly constraining the rotations, since shell element nodes contain both displacement and rotational degrees offreedom at every node whereas solid elements model only the translations.

A node-to-node connection would effectively appear to the shell element as a simply supported, rather than (thecorrect) cantilevered restraint and significantly alter the dynamic response of the shell structure.

To address this problem, constraint equations are used to properly connect adjacent shell- andsolid-element modeled structures.

Basically, all such constraints express the deflection (androtation for shell elements) of a node, R1, on one structural component in terms of thedeflections/rotations of the corresponding point, P2, on the other connected component.

Specifically, the element containing P2 is identified and the deformations at P2 determined byinterpolation between the element nodes. The following types of shell-solid element connections are used in the steam dryer model including the following:

1. Connections of shell faces to solid faces (Figure 6a). While only displacement degrees offreedom are explicitly constrained, this approach also implicitly constrains the rotational degrees of freedom when multiple shell nodes on a sufficiently dense grid are connected to the same solid face.2. Connections of shell edges to solids (e.g., connection of the bottom of closure plates withthe upper ring). Since solid elements do not have rotational degrees of freedom, thecoupling approach consisted of having the shell penetrate into the solid by one shellthickness and then constraining both the embedded shell element nodes (inside the solid)and the ones located on the surface of the solid structure (see Figure 6b). Numerical testsinvolving simple structures showed that this approach and penetration depth reproduce both the deflections and stresses of the same structure modeled using only solid elementsor ANSYS' bonded contact technology.

Continuity of rotations and displacements isachieved.

The use of constraint conditions rather than the bonded contacts advocated by ANSYS forconnecting independently meshed structural components confers better accuracy and usefulnumerical advantages to the structural analysis of the steam dryer including better conditioned and smaller matrices.

The smaller size results from the fact that equations and degrees offreedom are eliminated rather than augmented (in Lagrange multiplier-based methods) byadditional degrees of freedom.

Also, the implementation of contact elements relies on the use ofvery high stiffness elements (in penalty function-based implementations) or results in indefinite 21 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information matrices (Lagrange multiplier implementations) with poorer convergence behavior compared topositive definite matrices.

The steam dryer rests on four support blocks which resist vertical and lateral displacement.

The support blocks contact the seismic blocks welded to the USR so that 100% of the dryerweight is transmitted through the seismic blocks per the FDDR KG1 -265. Because the contactregion between the blocks and steam dryer is small, the seismic blocks are considered free torotate about the radial axis. Specifically nodal constraints (zero relative displacement) areimposed over the contact area between the seismic blocks and the support blocks. Two nodes oneach support block are fixed as indicated in Figure 7. One node is at the center of the supportblock surface facing the vessel and the other node is 0.5" offset inside the block towards thesteam dryer, half way to the nearest upper support ring node. This arrangement approximates thenonlinear contact condition where the ring can tip about the block.For the finite element model with the U-section stiffeners on the inner side plates, 5 shortbeam elements per beam are added to represent the bolts that attach it to the side plate. Inaddition constraints to prevent separation and relative rotation between the side plate and U-section stiffener surfaces that are in common contact, are imposed.

The U-channel itself isrepresented using shell elements whose nodal positions and element connectivities are 'lofted'(copied and translated) from the underlying side plate nodes to ensure a direct one-to-one correspondence between the nodes in the channel and side plate. This facilitates the imposition of aforementioned constraints between the channel and side plate shell elements.

22 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information CEMasses areconnected totop and bottomsupportsPoint massesGussets to lifting'rods connections Skirt to support-, rings connections Simplysupported restraints A,/Figure 4. Point masses representing the vanes. The pink shading represents where constraint equations between nodes are applied (generally between solid and shell elements, point massesand nodes and (3)).23 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 4. FE Model Summary.Description Description QuantityTotal Nodes I baseline model 159,793with U-section stiffener 189,862Total Elements baseline model 124,496with U-section stiffener 155,6801. Not including additional damper nodes and elements.

Table 5. Listing of Element Types.Generic Element Type Name Element Name20-Node Quadratic Hexahedron SOLID 18610-Node Quadratic Tetrahedron SOLID1874-Node Elastic Shell SHELL632-Node Beam element(only used for U-section stiffener BEAM188bolts)Mass Element MASS21Pressure Surface Definition SURF154Damper element COMBIN1424 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 5a. Mesh overview.

25 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 5b. Close up of mesh showing on-site modifications.

26 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 5c. Close up of mesh showing drain pipes and hood supports.

27 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 5d. Close up of mesh showing node-to-node connections between various plates.28 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 5e. Close up of mesh showing node-to-node connections between the skirt and drainchannels; hood supports and hoods; and other parts.29 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 5f. Close up view of tie bars.30 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Shell nodes DOF are related to solid element shape functions Surface of solid elementFigure 6a. Face-to-face shell to solid connection.

Shell nodes DOF are related to solid element shape functions Surface of solid elementFigure 6b. Shell edge-to-solid face connection.

31 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 7. Boundary conditions.

Inside node is half way between outer surface of support blockand upper support ring.32 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information 3.10 Pressure LoadingThe harmonic loads are produced by the pressures acting on the exposed surfaces of thesteam dryer. At every frequency and for each MSL, the pressure distribution corresponding to aunit pressure at the MSL inlet is represented on a three-inch grid lattice grid (i.e., a mesh whoselines are aligned with the x-, y- and z-directions) that is superimposed over the steam dryersurface.

This grid is compatible with the 'Table' format used by ANSYS to 'paint' generalpressure distributions upon structural surfaces.

The pressures are obtained from the Helmholtz solver routine in the acoustic analysis

[14].In general, the lattice nodes do not lie on the surface, so that to obtain the pressuredifferences at the surface it is necessary to interpolate the pressure differences stored at thelattice nodes. This is done using simple linear interpolation between the 8 forming nodes of thelattice cell containing the surface point of interest.

Inspection of the resulting pressures atselected nodes shows that these pressures vary in a well-behaved manner between the nodes withprescribed pressures.

Graphical depictions of the resulting pressures and comparisons betweenthe peak pressures in the original nodal histories and those in the final surface load distributions produced in ANSYS, all confirm that the load data are interpolated accurately and transferred correctly to ANSYS.The harmonic pressure loads are only applied to surfaces above the water level, as indicated in Figure 8. In addition to the pressure load, the static loading induced by the weight of thesteam dryer is analyzed separately.

The resulting static and harmonic stresses are linearlycombined to obtain total values which are then processed to calculate maximum and altemating stress intensities for assessment in Section 5.I(3) This is useful since revisions in theloads model do not necessitate recalculation of the unit stresses.

33 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information NODE S NF WWPRE S-NORK.349671 .749872-.212228 .612429 .887315Figure 8a. Real part of unit pressure loading MSL A (in psid) on the steam dryer at 50.1 Hz. Noloading is applied to the submerged surface and lifting rods.34 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information ODZ 6 UF U WPRE S-NORM-. 490178 28 .831766Figure 8b. Real part of unit pressure loading MSL A (in psid) on the steam dryer at 200.45 Hz.No loading is applied to the submerged surface and lifting rods.35 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information 3.11 , Subsection NG-3216.2 the following procedure was established to calculate alternating, stresses.

For every node, thestress difference

tensors, Fmnm = (5n -Um, are considered for all possible pairs of the stresses caand cym at different time levels, t,, and tmn. Note that all possible pairs require consideration sincethere are no "obvious" extrema in the stress responses.

However, in order to containcomputational cost, extensive screening of the pairs takes place (see Section 2.3) so that pairsknown to produce alternating stress intensities less than 250 psi are rejected.

For each remaining stress difference tensor, the principal stresses S1, S2, S3 are computed and the maximum absolutevalue among principal stress differences, Snm=max{ISI-S 21, SI-S31,1S2-S31}, obtained.

Thealternating stress at the node is then one-half the maximum value of Snm taken over allcombinations (n,m), i.e., salt = max IS. }. This alternating stress is compared against allowable n,mvalues, depending on the node location with respect to welds.37 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information NODAL SOLUTIONSTEP=1SUB =1TIME=1USUM (AVG)RSYS=ODMX =.068847SMN =.505E-03 SMX =.068847SO50 E-03 .0 8 9 0 15 jll ý6067 .0 36 061254 .0 8 4.008099*056.687 Figure 9a. Overview of static calculations showing displacements (in inches).

Maximumdisplacement (DMX) is 0.069". Note that displacements are amplified for visualization.

38 This Document Does Not ContainIContain uu Dynamics Inc. Proprietary Informnation

.Lmwm6 1 ANFigure 9b. O verview ofm smtC l ltointen sty ) is 9,598 psi. N o0Ss] ote tha t disp Sho~we ing stress intensities (in psi). StresPlacement are amplified for Psli), um stressSualization 39 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information 7Figure 9c. Stress singularities.

Model is shown in wireframe mode for clarity.

NSHotSpots represents the node selection of the stress singularities.

40 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information NODAL SOLUTIONSTEPw377SUBD -1FREQ-50.097 REAL ONLYSINT (VCG)DUX -.083772SWN -.299472SX =)11002'ii 3 '4444555. 556 )3889 5000Figure 10a. Overview of harmonic calculations showing real part of stress intensities (in psi)along with displacements.

Unit loading MSL A at 50.1 Hz (oriented to show high stress locations at the hoods).41 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information ANNODAL SOLUTIONISTEP-337SUB =1FREQ-200.446 REAL ONLYSINT (AVG)DEC =.014L39SN u.219769SAX =130063000Figure 10b. Overview of harmonic calculations showing real part of stress intensities (in psi)along with displacements.

Unit loading MSL A at 200.5 Hz.42 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information 4.3 Post-Processing The static and transient stresses computed at every node with ANSYS were exported intofiles for subsequent post-processing.

These files were then read into separate customized software to compute the maximum and alternating stresses at every node. The maximum stresswas defined for each node as the largest stress intensity occurring during the time history.Alternating stresses were calculated according to the ASME standard described above. For shellelements the maximum stresses were calculated separately at the mid-plane, where onlymembrane stress is present, and at top/bottom of the shell, where bending stresses are alsopresent.For nodes that are shared between several structural components or lie on junctions, themaximum and alternating stress intensities are calculated as follows.

First, the nodal stresstensor is computed separately for each individual component by averaging over all finiteelements meeting at the node and belonging to the same structural component.

The timehistories of these stress tensors are then processed to deduce the maximum and alternating stressintensities for each structural component.

Finally for nodes shared across multiple components the highest of the component-wise maximum and alternating stresses is recorded as the "nodal"stress. This approach prevents averaging of stresses across components and thus yieldsconservative estimates for nodal stresses at the weld locations where several components arejoined together.

The maximum stresses are compared against allowable values which depend upon the stresstype (membrane, membrane+bending, alternating

-Pm, Pm+Pb, Salt) and location (at a weld oraway from welds). These allowables are specified in the following section.

For solid elementsthe most conservative allowable for membrane stress, Pm, is used, although bending stresses arenearly always present also. The structure is then assessed in terms of stress ratios formed bydividing allowables by the computed stresses at every node. Stress ratios less than unity implythat the associated maximum and/or alternating stress intensities exceed the allowable levels.Post-processing tools calculate the stress ratios, identifying the nodes with low stress ratios andgenerating files formatted for input to the 3D graphics

program, TecPlot, which provides moregeneral and sophisticated plotting options than currently available in ANSYS.4.4 Computation of Stress Ratios for Structural Assessment The ASME B&PV Code,Section III, subsection NG provides different allowable stresses fordifferent load combinations and plant conditions.

The stress levels of interest in this analysis arefor the normal operating condition, which is the Level A service condition.

The loadcombination for this condition is:Normal Operating Load Combination

= Weight + Pressure

+ ThermalThe weight and fluctuating pressure contributions have been calculated in this analysis and areincluded in the stress results.

The static pressure differences and thermal expansion stresses aresmall, since the entire steam dryer is suspended inside the reactor vessel and all surfaces areexposed to the same conditions.

Seismic loads only occur in Level B and C cases, and are notconsidered in this analysis.

43 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Allowable Stress Intensities The ASME B&PV Code,Section III, subsection NG shows the following (Table 6) for themaximum allowable stress intensity (Sm) and alternating stress intensity (Sa) for the Level Aservice condition.

The allowable stress intensity values for type 304 stainless steel at operating temperature 550'F are taken from Table 1-1.2 and Fig. 1-9.2.2 of Appendix I of Section III, in theASME B&PV Code. The calculation for different stress categories is performed in accordance with Fig. NG-3221-1 of Division I, Section I1, subsection NG. The allowable value foralternating stress is taken from curve C of Fig. 1-9.2.2 in Appendix I in Section III of the ASMEB&PV Code.Table 6. Maximum Allowable Stress Intensity and Alternating Stress Intensity for all areasother than welds. The notation Pm represents membrane stress; Pb represents stressdue to bending; Q represents secondary stresses (from thermal effects and grossstructural discontinuities, for example);

and F represents additional stress increments (due to local structural discontinuities, for example).

Type Notation Service Limit Allowable Value (ksi)Maximum Stress Allowables:

General Membrane Pm Sm 16.9Membrane

+ Bending Pm + Pb 1.5 Sm 25.35Primary + Secondary Pm + Pb + Q 3.0 Sm 50.7Alternating Stress Allowable:

Peak = Primary + Secondary

+ F Salt Sa 13.6When evaluating welds, either the calculated or allowable stress was adjusted, to accountfor stress concentration factor and weld quality.

Specifically:

" For maximum allowable stress intensity, the allowable value is decreased by multiplying its value in Table 6 by 0.55.* For alternating stress intensity, the calculated weld stress intensity is multiplied by a weldstress intensity (fatigue) factor of 1.8 for a fillet weld and 1.4 for a full penetration weld,before comparison to the Sa value given above.The weld factors of 0.55 and 1.4 (full penetration weld) or 1.8 (fillet weld) were selectedbased on the observable quality of the shop welds and liquid penetrant NDE testing of all welds(excluding tack and intermittent welds, which were subject to 5X visual inspection) duringfabrication.

These factors are consistent with fatigue strength reduction factors recommended bythe Welding Research

Council,

[34], and stress concentration factors at welds, provided in [35]and [36]. In addition, critical welds are subject to periodical visual inspections in accordance with the requirements of GE SIL 644 SIL and BWR VIP-139 [37]. Therefore, for weld stressintensities, the allowable values are shown in Table 7. These factors (0.55 and 1.4 or 1.8) alsoconservatively presume that the structure is joined using fillet welds unless specified otherwise.

Since fillet welds correspond to larger stress concentration factors than other types of welds, thisassumption is a conservative one.44 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 7. Weld Stress Intensities.

Type Notation Service Limit Allowable Value (ksi)Maximum Stress Allowables:

General Membrane Pm 0.55 Sm 9.30Membrane

+ Bending Pm + Pb 0.825 Sm 13.94Primary + Secondary Pm + Pb + Q 1.65 Sm 27.89Alternating Stress Allowables:

Peak = Primary + Secondary

+ F Salt Sa 13.6Comparison of Calculated and Allowable Stress Intensities The classification of stresses into general membrane or membrane

+ bending types was madeaccording to the exact location, where the stress intensity was calculated; namely, generalmembrane, Pm, for middle surface of shell element, and membrane

+ bending, Pm + Pb, forother locations.

For solid elements the most conservative, general membrane, Pm, allowable isused.The structural assessment is carried out by computing stress ratios between the computedmaximum and alternating stress intensities, and the allowable levels. Locations where any of thestresses exceed allowable levels will have stress ratios less than unity. Since computation ofstress ratios and related quantities within ANSYS is time-consuming and awkward, a separateFORTRAN code was developed to compute the necessary maximum and alternating stressintensities, Pm, Pm+Pb, and Salt, and then compare it to allowables.

Specifically, the following quantities were computed at every node:1. The maximum membrane stress intensity, Pm (evaluated at the mid-thickness location forshells),2. The maximum membrane+bending stress intensity, Pm+Pb, (taken as the largest of themaximum stress intensity values at the bottom, top, and mid thickness locations, forshells),3. The alternating stress, Salt, (the maximum value over the three thickness locations istaken).4. The stress ratio due to a maximum stress intensity assuming the node lies at a non-weldlocation (note that this is the minimum ratio obtained considering both membrane stressesand membrane+bending stresses):

SR-P(nw)

= min{ Sm/Pm, 1.5

Sm/(Pm+Pb)

}.5. The alternating stress ratio assuming the node lies at a non-weld

location, SR-a(nw)

= Sa/(I.1

Salt),6. The same as 4, but assuming the node lies on a weld,SR-P(w)=SR-P(nw)

0.557. The same as 5, but assuming the node lies on a weld,SR-a(w)=SR-a(nw)

/ fsw.45 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Note that in steps 4 and 6, the minimum of the stress ratios based on Pm and Pm+Pb, is taken.The allowables listed in Table 7, Sm=16,900 psi and Sa=13,600 psi. The factors, 0.55 and fsw,are the weld factors discussed above with fsw=1.8 being appropriate for a fillet weld and fsw=1.4for a full penetration weld. The factor of 1.1 accounts for the differences in Young's moduli forthe steel used in the steam dryer and the values assumed in alternating stress allowable.

According to NG-3222.4 in subsection NG of Section III of the ASME Code [7], the effect ofelastic modulus upon alternating stresses is taken into account by multiplying alternating stressSalt at all locations by the ratio, E/Emodel=

1.1, where:E = 28.3 106 psi, as shown on Fig. 1-9.2.2.

ASME BP&V CodeEmodel = 25.55 106 psi (Table 2)The appropriate maximum and alternating stress ratios, SR-P and SR-a, are thus determined anda final listing of nodes having the smallest stress ratios is generated.

The nodes with stress ratioslower than 4 are plotted in TecPlot (a 3D graphics plotting program widely used in engineering communities

[38]). These nodes are tabulated and depicted in the following Results Section.(3)]](*3)j46 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information (3)]]47 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information (3)]]48 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information (3)]]49 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information (3)]]50 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information

(3)51 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information (3)]]52 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information

5. ResultsThe stress intensities and associated stress ratios resulting from the Rev. 4.1 Racoustic/hydrodynamic loads [2, 8, 9] with associated biases and uncertainties factored in, arepresented below. The acoustic loads applied to the steam dryer are obtained using the mostrecent and complete strain gage signals [2] and processed using the ACM Rev. 4. 1R analysiswith associated biases and uncertainties updated to reflect the new revision as described in [2, 8,9]. For the FEM structural model there are three main contributors to the bias and uncertainty.

The first is an uncertainty (21.5%) that accounts for modeling idealizations (e.g., vane bank massmodel), geometrical approximations and other discrepancies between the modeled and actualdryer such as neglecting of weld mass and stiffness in the FEA. The second contributor is a biasof 9.53% accounting for discretization errors associated with using a finite size mesh, uponcomputed stresses.

The third contributor is also a bias and compensates for the use of a finitediscretization schedule in the construction of the unit solutions.

The frequencies are spaced suchthat at 1% damping the maximum (worst case) error in a resonance peak is 5%. The averageerror for this frequency schedule is 1.72%.Results are presented with frequency shifting included.

Unless specified otherwise, frequency shifts are generally performed at 2.5% increments.

The tabulated stresses and stressratios are obtained using a 'blanking' procedure that is designed to prevent reporting a largenumber of high stress nodes from essentially the same location on the structure.

In the case ofstress intensities this procedure is as follows.

The relevant stress intensities are first computed atevery node and then nodes sorted according to stress level.. The highest stress node is noted andall neighboring nodes within 10 inches of the highest stress node and its symmetric images (i.e.,reflections across the x=O and y=O planes) are "blanked" (i.e., excluded from the search forsubsequent high stress locations).

Of the remaining nodes, the next highest stress node isidentified and its neighbors (closer than 10 inches) blanked.

The third highest stress node issimilarly located and the search continued in this fashion until all nodes are either blanked orhave stresses less than half the highest value on the structure.

For stress ratios, an analogous blanking procedure is applied.

Thus the lowest stress ratio of a particular type in a 10"neighborhood and its symmetric images is identified and all other nodes in these regionsexcluded from listing in the table. Of the remaining nodes, the one with the lowest stress ratio isreported and its neighboring points similarly

excluded, and so on until all nodes are eitherblanked or have a stress ratio higher than 5. The set of points thus obtained is referred to as areduced point set (RPS).As described in Section 1.1, the large number of combinations of loads, bias anduncertainties selections, and dryer models necessitates a tailored analysis combining full and realtime stress analyses.

The following sections are structured and ordered as follows.

Section 5.1lists the limiting stress ratios obtained from a full steam dryer evaluation using the baseline EPUload (i.e., using the bias and uncertainty values pertaining to ACM 4.1R and without considering the drain trap out of service or RCIC valve closed conditions) applied to the baseline dryergeometry (i.e., without the numerical corrections developed in Section 5.3). This calculation isused to develop a list of nodes for real time analysis in Section 5.2 and to motivate analytical refinements for reanalysis in Section 5.3. The full stress evaluation is repeated using the samebaseline EPU load and the refined structural model; the results from this evaluation are given in53 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Sections 5.4 to 5.6. The highest maximum and alternating stress intensities indicating whichpoints on the dryer experience significant stress concentration and/or modal response, arepresented in Section 5.4. The lowest stress ratios obtained by comparing the stresses againstallowable values, accounting for stress type (maximum and alternating) and location (on or awayfrom a weld), are reported Section 5.5. The frequency dependence of the stresses at nodesexperiencing the lowest stress ratios is depicted in the form of accumulative PSDs in Section 5.6.Real time assessments of the other loads and total uncertainty combinations using the listdeveloped in Section 5.2 are summarized in Section 5.7. These assessments are carried outwithout the U-section stiffener which allows reuse of existing unit solutions and post-processing utilities (since the mesh indexing did not change).

While a finite element model of the dryerwith the U-section stiffener added was developed, schedule requirements did not permitgenerating the associated unit solutions over the entire 0-250 Hz frequency range or developing the mesh specific post-processing tools (e.g., assignment of SRFs, handling the added beamelements representing the U-section stiffener attachment bolts, etc.). The stress evaluation withthe U-section stiffener included is therefore carried out using real time analysis as described inSection 5.8.5.1. Preliminary Stress Assessment of Normal EPU Load + Base Dryer ModelThe first stress evaluation performed serves as a preliminary assessment and screening of lowstress ratio locations and processed all steam dryer nodes using the baseline or normal EPU load(i.e., drain trap out-of-service and RCIC valve closed loads were not considered) adjusted inaccordance with the ACM 4.1 R bias and uncertainty values [2]. The steam dryer model does notinclude the U-section stiffener described in Appendix B nor the reanalysis options in Section 5.3.It also employs a stress reduction factor of 0.77 at locations 2, 8 and 14 in Table 9c. This SRF isadopted from a similar location on the outer vane bank. In the subsequent analysis in Section 5.3and also the real time evaluations it is replaced by the SRFs of 0.66 and 0.80 developed inSection 4.5 under embedded model 6 which pertains to this exact junction location.

Note toothat junctions involving the lifting rods and upper lifting rod braces were excluded from theevaluation in this Section 5.1 while awaiting resolution as to whether or not these junctions arewelded. It ultimately became clear that while the original drawings indicated a weld, the as-builtconfiguration does not, so that no weld factor is applicable at this particular junction.

Again, inthe detailed evaluation of Sections 5.4 to 5.6 and the real time assessments, these locations areretained.

The purpose of this evaluation is mainly to identify locations that have alternating stress ratios, SR-a<2 or are likely to drop below 2.0 at other load conditions.

Locations with lowstress ratios are used to assemble the list of nodes in Section 5.2 for use in subsequent real timeprocessing in Sections 5.7 and 5.8. Additionally scrutiny of the lowest stress ratio sites led to theanalytical analysis methods in Section 5.3, the construction of the embedded model 6 andassociated SRFs in Section 4.5 and Table 8, and the U-section stiffener in Appendix B.At zero frequency shift the limiting peak and alternating stress intensities are SR-P=I. 18 andSR-a=l.87.

The effects of frequency shifts are conservatively accounted for by identifying theminimum stress ratio at every node, where the minimum is taken over all the frequency shiftsconsidered (including the nominal or 0% shift case). The stress ratios computed for EPU withfrequency shifting included are listed in Table 9. The stress ratios are grouped according to type54 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information (SR-P for maximum membrane and membrane+bending stress, SR-a for alternating stress) andlocation (away from welds or on a weld). The tabulated nodes with the smallest alternating stress ratios in Table 9c are also depicted in Figure 13 together with all nodes with SR-a<5 (forbrevity, comparable plots of the other stress ratio types are not listed here as they essentially reappear in the full analysis summarized in Section 5.5).With frequency shifting

included, the lowest stress ratio, SR-P=1.14, occurs at the samelocation as in the nominal case and is only slightly lower. This value is dominated by the staticcontribution and so is not very sensitive to frequency shifting.

Three locations in Table 9c arefound to have stress ratios below 2.0. Close up views of these locations are depicted in Figure14. The lowest alternating stress ratio occurs at the top of the closure plate/middle hood weld(see location 1 in Figure 13a, and Figure 14a) and assumes a value of SR-a=l.68.

Based on thelargest Fourier coefficient, the dominant frequency contributing to this stress is 86.8 Hz. Thesecond location involves the common junction of the inner vane bank top and side plates, theinner side plate spanning the inner vane banks, and the inner vane bank perforated plate (seeFigure 14b). Its stress ratio is 1.75 after application of an SRF of 0.77. This stress ratio isrevised in the subsequent sections using the methods in Section 5.3. While the recommended margin can be achieved using a combination of the methods summarized in Sections 5.3.1 and5.3.2, uncertainty over the applicability of the stress interpolation procedure along the weld linesat this complex geometry junction involving multiple welds motivated consideration of alternate concepts for meeting margin. This resulted in the implementation of the U-section stiffener todirectly suppress the resonant response of the inner side plate. The third location with SR-a lessthan 2 occurs near the end of the upper lifting rod brace on the outer vane bank side plate/closure plate junction and has SR-a=I.80 (see Figure 14c). Comparison with the other three upperbraces at the reflected locations shows that their alternating stress ratios are all well above 2.0and also that the grid is more regular.

This motivated a re-evaluation of the local stress using abetter quality mesh as described in Section 5.3. Virtually all of the limiting stress ratios occur ateither the +7.5% or +10% shifts.55 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 9a. Limiting non-weld locations with at EPU conditions with frequency shifts. Stress ratios are grouped according to stresstype (maximum

-SR-P; or alternating

-SR-a).Stress Location Location (in.) node Stress Intensity (psi) Stress Ratio % Freq. Dom.Ratio x y z Pm Pm+Pb Salt SR-P SR-a Shift Freq. (Hz)SR-P 1. Inner Side Plate 3.1 119 0.5 37229 6685 8548 2281 2.53 5.42 5 123.52. Support/Seismic Block 10.2 123.8 -9.5 113286 6210 6210 4314 2.72 2.87 10 13.73. Thin Vane Bank Plate -15.6 -118.4 0.6 2558 4311 4727 833 3.92 14.84 10 16.0SR-a 1. Support/Seismic Block 10.2 123.8 -9.5 113286 6210 6210 4314 2.72 2.87 10 13.72. Inner Side Plate 14.4 -119 88 37592 764 5055 4235 5.02 2.92 7.5 78.73. Side Plate -79.4 -85.2 76.8 10819 436 3463 3122 7.32 3.96 10 17.94. Hood Support 89 -28.4 0 14474 5050 5210 3068 3.35 4.03 5 82.056 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 9b. Limiting peak stress ratios, SR-P, on welds at EPU conditions with frequency shifts. Bold text indicates minimum stressratio on the structure.

Location SRF(a) Location (in.) node Stress Intensity (psi) Stress Ratio % Freq. Doam.x y z Pm Pm+Pb Salt SR-P SR-a Shift Freq. (Hz)1. USR/Support/Seismic Block -6.9 -122.3 -9.5 113554 8133 8133 3048 1.14 2.25 5 13.72. Middle Base Plate/Inner Backing Bar Out/Inner 39.9 108.6 0 85631 1478 10031 2750 1.39 2.5 10 82.6Backing Bar/Inner Hood3. Side Plate Ext/Inner Base Plate 16.3 119 0 94143 6396 9246 1525 1.45 4.51 10 39.64. Hood Support/Outer Base Plate/Middle

-71.3 0 0 95428 6159 6227 2634 1.51 2.61 7.5 14.4Backing Bar5. Inner Side Plate/Inner Base Plate -2.3 -119 0 99200 4026 8778 3167 1.59 2.17 7.5 123.56. Tie Bar(2) 0.77 -49.3 -108.1 88 143795 5644 5644 1773 1.65 3.87 10 82.67. Thin Vane Bank Plate/Hood Support/Inner 24.1 -59.5 0 85191 5245 5330 2030 1.77 3.38 10 13.3Base Plate8. Thin Vane Bank Plate/Hood Support/Middle 55.6 -54.3 0 98968 5241 5318 2232 1.77 3.08 10 82Base Plate9. Hood Support/Outer Cover Plate/Outer Hood(4) 0.8 -102.8 28.4 0 95267 5054 5081 2996 1.84 2.29 10 13.710. Hood Support/Middle Base Plate/Inner

-39.9 59.5 0 90468 5030 5165 1843 1.85 3.73 10 82.6Backing Bar/Inner Hood(b)11. Upper Support Ring/Seismic Block/Support

-122.1 10.2 -9.5 113508 4769 4769 2060 1.95 3.33 10 16.112. Outer Cover Plate/Outer Hood 102.8 -58.1 0 94498 1252 6440 1422 2.17 4.83 7.5 8213. Closure Plate/Middle Hood 60.2 -85.2 87 89317 1459 6286 4085 2.22 1.68 10 86.8Notes: (a) [[(b) Full penetration weld so that weld factor, WF=1.4.(1-5) Number referring to the EPU conditions with frequency shifts. Locations are depicted in Figure13.Location SRF(a) Location (in.) node Stress Intensity (psi) Stress Ratio % Freq. Dom.x y z Pm Pm+Pb Salt SR-P SR-a Shift Freq. (Hz)1. Closure Plate/Middle Hood 60.2 -85.2 87 89317 1459 6286 4085 2.22 1.68 10 86.82. Top Thick Plate/Side Plate/Exit Top 0.77 -15.6 -119 86.5 101861 550 4677 3931 2.98 1.75 7.5 78.7Perf/Inner Side Plate(2) 13. Side Plate/Closure Plate/Exit Top Perf/Exit

-78.5 -85.2 74.5 87784 1875 3907 3824 3.57 1.80 10 17.9Mid Top Perf4. Outer End Plate/Outer Hood 101.9 -63.3 24.6 94509 795 4368 3233 3.19 2.12 10 82.05. Thin Vane Bank Plate/Inner Base Plate 15.6 114.4 0 99635 3531 5532 3194 2.52 2.15 7.5 82.06. Inner Side Plate/Inner Base Plate -2.3 -119 0 99200 4026 8778 3167 1.59 2.17 7.5 123.57. Upper Support Ring/Support/Seismic Block -6.9 -122.3 -9.5 113554 8133 8133 3048 1.14 2.25 7.5 13.78. Side Plate/Top Plate(2) 0.77 49.6 -108.6 88 103080 1406 5009 3046 2.78 2.25 10 82.69. Seismic Block/Support

-123.8 10.2 -9.5 113400 3577 3577 3001 2.60 2.29 10 16.010. Hood Support/Outer Cover Plate/Outer 0.8 -102.8 28.4 0 95267 5054 5081 2996 1.84 2.29 10 13.7Hood(4)11. Thick Vane Bank Plate/Thin Vane Bank 87 -85.2 11.6 90786 907 10028 2941 1.39 2.34 10 82.0Plate/Side Plate/Side Plate Ext/Outer End Plate12. Tie Bar 17.6 59.8 88 137575 4549 4549 2906 2.04 2.36 10 82.613. Closure Plate/Inner Hood 28.8 -108.6 87 95172 2025 5675 2822 2.46 2.43 10 82.614. Side Plate/Top Plate(2) 0.77 81.1 -85.2 88 91055 858 3611 2811 3.86 2.44 10 86.815. Submerged Drain Channel/Submerged Skirt 76.7 -100 93488 498 4963 2811 2.81 2.44 10 123.5Notes: (a) EPU operation with frequency shifts. The recorded stress ratio at a node is the minimum value taken over allfrequency shifts. Numbers refer to the enumerated locations for SR-a values at welds in Table9c. This view shows locations 1-4, 6-8, 11, 13 and 14.59 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 13b. Locations of minimum alternating stress ratios, SR-a_<5, at welds for EPU operation with frequency shifts. The recorded stress ratio at a node is the minimum value taken over allfrequency shifts. Numbers refer to the enumerated locations for SR-a values at welds in Table9c. View showing locations 5 and 15.60 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information xFigure 13c. Locations of minimum alternating stress ratios, SR-a.<5, at welds for EPU operation with frequency shifts. The recorded stress ratio at a node is the minimum value taken over allfrequency shifts. Numbers refer to the enumerated locations for SR-a values at welds in Table9c. View around locations 9, 10, 12 and 15.61 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 14a. Close up view of Entry 1 in 9c at the top of the closure plate/middle hood weld.62 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 14b. Close up view of Entry 2 Table 9c involving the components near the top of theinner vane bank and tie bar.63 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 14c. Close up view of Entry 3 Table 9c involving the perimeter of the reinforcement plate added to the lifting rod restraint bracket.64 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information 5.2 Construction of Real Time Node ListAs discussed in [2] other load conditions occurring when the drain trap is out of service andwhen the RCIC valve is closed, also require attention.

Moreover, it has been noted [2] that whileusing the bias and uncertainties developed for ACM 4. 1R are technically appropriate, it is also ofvalue to consider a hybrid and conservative set of biases and uncertainties that retains the higherof the ACM 4.1 and ACM 4.IR total uncertainties over each frequency interval.

To expediteanalysis of these nodes a subset of nodes with smallest stress ratios is selected and stresses re-evaluated at those locations only. The particular set of nodes is selected in a manner similar tothat used for real time stress evaluation during power ascension

[24]. Specifically the list iscomprised of the following nodes:" All nodes with an EPU alternating stress ratio, SR-a<2.0.

There are 12 such nodes on aweld.* The RPS sub-set of nodes with SR-a<3 at EPU. There are 30 such nodes on a weld andanother 2 away from a weld." The RPS set of nodes on a weld with SR-P<2 at EPU. There are 16 such nodes.* The RPS set of nodes away from a weld with SR-P<3 at EPU. There are 2 such nodes.After eliminating redundant nodes (i.e., ones that appear in more than one of the above sets) thetotal number of nodes is 53. These are listed below in Table 10. Locations with stresses that areaffected by the installation of the U-section stiffener are separated out in a separate group tofacilitate subsequent presentation of results.

In real time stress analysis, the stress intensities arecalculated in the same manner as when analyzing the entire dryer and the limiting stresses withfrequency shifting included are reported.

The resulting stress ratios are summarized in Table 17together with the dominant frequencies.

65 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 10a. List of nodes used for real time evaluation.

Nodes 1-18.Index Description Node Coordinates

[in ] Reason for Otherx y z SRF Selection (1)1 Inner Side Plate 37229 3.1 119.0 0.5 1 12 Support/Seismic Block 113286 10.2 123.8 -9.5 1 2 (2)3 Inner Side Plate 37592 14.4 -119.0 88.0 1 24 Side Plate Ext/Inner Base Plate 94143 16.3 119.0 0.0 1 35 Thin Vane Bank Plate/Hood Support/Inner Base Plate 85191 24.1 -59.5 0.0 1 36 Tie Bar 143795 -49.3 -108.1 88.0 0.77 37 Upper Support Ring/Seismic Block/Support 113508 -122.1 10.2 -9.5 1 38 Thin Vane Bank Plate/Hood Support/Middle Base Plate 98968 55.6 -54.3 0.0 1 39 Outer Cover Plate/Outer Hood 94498 102.8 -58.1 0.0 1 310 Hood Support/Middle Base Plate/Inner Backing Bar/Inner Hood 90468 -39.9 59.5 0.0 0.78 311 Side Plate/Top Plate 101600 -17.6 -119.0 88.0 0.66 312 Hood Support/Middle Base Plate/Inner Backing Bar/Inner Hood 88639 39.9 0.0 0.0 0.78 313 Thin Vane Bank Plate/Hood Support/Inner Base Plate 92995 24.1 0.0 0.0 1 314 Tie Bar 141237 25.0 108.1 88.0 1 315 Closure Plate/Middle Hood 89317 60.2 -85.2 87.0 1 4 (X)16 Side Plate/Closure Plate/Exit Top Perf/Exit Mid Top Perf 87784 -78.5 -85.2 74.5 1 4 (3)17 Outer End Plate/Outer Hood 94509 101.9 -63.3 24.6 1 418 Thin Vane Bank Plate/Inner Base Plate 99635 15.6 114.4 0.0 1 4Notes: 1. Reasons for selection:

1 -small SR-P on no-weld node; 2 -small SR-a on non-weld node; 3 -SR-P<2; 4 -node on weld with SR-a<3 taken from RPS. 14 -remaining nodes on welds with SR-a<2.2. Stress corrected to account for 0.75 extension of support lug under USR (see Section 5.3.5).3. Compensates for poor mesh quality near top lifting rod brace (see Section 5.3.3).X. Extrapolation used to estimate stress at top of closure plate/curved hood weld (see Section 5.3.1).node on weld with66 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 10b. List of nodes used for real time evaluation.

Nodes 19-36.Index Description Node Coordinates

[in ] Reason for Otherx y z SRF Selection (1)19 Inner Side Plate/Inner Base Plate 99200 -2.3 -119.0 0.0 1 420 USR/Support/Seismic Block 113554 -6.9 -122.3 -9.5 1 4 (2)21 Side Plate/Top Plate 103080 49.6 -108.6 88.0 0.77 422 Seismic Block/Support 113400 -123.8 10.2 -9.5 1 423 Hood Support/Outer Cover Plate/Outer Hood 95267 -102.8 28.4 0.0 0.8 424 Thick Vane Bank Plate/Thin Vane Bank Plate/Side 90786 87.0 -85.2 11.6 1 4Plate/Side Plate Ext/Outer End Plate25 Tie Bar 137575 17.6 59.8 88.0 1 426 Closure Plate/Inner Hood 95172 28.8 -108.6 87.0 1 4 (X)27 Side Plate/Top Plate 91055 81.1 -85.2 88.0 0.77 428 Submerged Drain Channel/Submerged Skirt 93488 -91.0 -76.7 -100.0 1 429 Thin Vane Bank Plate/Hood Support/Outer Base Plate 98956 -87.0 28.4 0.0 0.58 430 Middle Base Plate/Inner Backing Bar Out/Inner 85631 39.9 108.6 0.0 1 4Backing Bar/Inner Hood31 Entry Bottom Perf/Side Plate/Outer End Plate 101818 -87.0 85.2 29.3 1 432 Thin Vane Bank Plate/Side Plate Ext/Outer Base Plate 98624 78.5 85.2 0.0 1 433 Outer End Plate/Outer Hood 94514 100.8 -64.9 36.8 1 434 Hood Support/Outer Base Plate/Middle Backing Bar 95428 -71.3 0.0 0.0 1 435 Outer Cover Plate/Outer End Plate/Outer Hood/Outer End Plate Ext 84090 102.8 -62.0 0.0 1 436 Thin Vane Bank Plate/Hood Support/Middle Base Plate 99451 55.6 54.3 0.0 1 4Notes: 1. Reasons for selection:

1 -small SR-P on no-weld node; 2 -small SR-a on non-weld node; 3SR-P<2; 4 -node on weld with SR-a<3 taken from RPS. 14 -remaining nodes on welds with SR-a<2.2. Stress corrected to account for 0.75 extension of support lug under USR (see Section 5.3.5).3. Compensates for poor mesh quality near top lifting rod brace (see Section 5.3.3).X. Extrapolation used to estimate stress at top of closure plate/curved hood weld (see Section 5.3.1).-node on weld with67 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 10c. List of nodes used for real time evaluation.

Nodes 37-53.Index Description Node Coordinates

[in I Reason for Otherx y z SRF Selection (1)37 Submerged Drain Channel/Skirt 93451 -11.5 -118.4 -101.5 0.56 438 Hood Support/Outer Base Plate/Middle Backing Bar 98172 -71.3 54.3 0.0 0.78 439 Submerged Drain Channel/Submerged Skirt 90924 91.0 76.7 -101.5 0.56 440 Side Plate/Exit Top Perf/Inner Side Plate 100989 15.6 -119.0 85.3 1 441 Submerged Drain Channel/Submerged Skirt 90926 11.5 118.4 -100.0 1 442 Skirt/Skirt USR overlap 99931 54.2 105.9 -9.5 1 443 Thick Vane Bank Plate/Thin Vane Bank Plate/Side 91091 24.1 119.0 11.6 1 4Plate/Side Plate Ext/End Plate44 Closure Plate/Middle Hood 88702 -60.2 85.2 87.0 1 14 (X)Nodes Addressed Using U-Section Stiffener 45 Top Thick Plate/Side Plate/Exit Top Perf/Inner Side Plate 101861 -15.6 -119.0 86.5 0.8 4 (M)46 Top Thick Plate/Side Plate/Exit Top Perf/Inner Side Plate 95197 15.6 -119.0 86.5 0.8 14 (M)47 Side Plate/Inner Side Plate/Top Plate 99407 -16.6 -119.0 88.0 0.66 14 (M)48 Top Thick Plate/Side Plate/Exit Top Perf/Inner Side Plate 98442 15.6 119.0 86.5 0.8 14 (M)49 Top Thick Plate/Side Plate/Inner Side Plate/Top Plate 98444 15.6 119.0 88.0 0.8 14 (M)50 Top Thick Plate/Side Plate/Exit Top Perf/Inner Side Plate 98451 -15.6 119.0 86.5 0.8 14 (M)51 Top Thick Plate/Side Plate/Inner Side Plate/Top Plate 98452 -15.6 119.0 88.0 0.8 14 (M)52 Top Thick Plate/Side Plate/Inner Side Plate/Top Plate 99408 -15.6 -119.0 88.0 0.8 14 (M)53 Top Thick Plate/Side Plate/Inner Side Plate/Top Plate 91240 15.6 -119.0 88.0 0.8 14 (M)Notes: 1. Reasons for selection:

1 -small SR-P on no-weld node; 2 -small SR-a on non-weld node; 3 -node on weld withSR-P<2; 4 -node on weld with SR-a<3 taken from RPS. 14 -remaining nodes on welds with SR-a<2.2. Stress corrected to account for 0.75 extension of support lug under USR (see Section 5.3.5).3. Compensates for poor mesh quality near top lifting rod brace (see Section 5.3.3).X. Extrapolation used to estimate stress at top of closure plate/curved hood weld (see Section 5.3.1).M. Nodes to be addressed in Section 5.8 using U-section stiffener.

68 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information 5.3 Examination of Low Stress Ratio Locations Section 5.1 identifies three distinct locations with alternating stress ratios below 2.0. Theseare addressed below.5.3.1 Analysis of Nodes Associated with Entry I in Table 9cThe limiting alternating stress node corresponds to the top of the weld joining the closureplate to the curved surface of the middle hood. This location had previously been troublesome due to vibrations of the unmodified closure plate. Before modification it was noted that theorigin of high stresses in the weld were caused by the presence of a 2-1 plate mode (2n-order mode in the vertical direction and 1st order mode in the horizontal direction).

This led to highweld stresses both near the top of the weld and further down near where the second peak of themodal response occurs. Because the stress was high at interior weld nodes (i.e., not only at theend) it was deemed physically correct and was evaluated accordingly.

At first, an investigation was undertaken to increase weld size, but this was found insufficient for meeting the target stressratios. Therefore, a more aggressive option was implemented of installing reinforcement ribsthat completely suppress closure plate vibration at frequencies below 250 Hz.With the reinforced plate present, the top node on the closure plate/middle hood attachment weld has a computed alternating stress ratio of SR-a=l.68 which is below the target of 2.0. Asecond node, node 88702, at the 1800 rotated location has an alternating stress ratio of SR=1.98,which is also barely below the target level. In each case, unlike the weld stress distribution observed in the pre-modification dryer (i.e., before ribs were installed) the current stress peak ishighly localized, confined to a single node and occurs at the end (rather than within the run of) aweld. The high stress occurs in the middle hood and is mainly due to bending of the shellelement.

Examination of the weld node immediately below the limiting node location revealsthat its stress intensity is less than 20% of the limiting value -i.e., the stress grows by more thana factor of 5 over the 0.9" distance separating the limiting node and the one immediately belowit. Such rapid rises in stress are characteristic of a structural discontinuity and are notrealistically captured in finite element modeling.

Moreover, using the stresses at suchdiscontinuities is not consistent with proper application of the ASME code, which relies on'nominal' stresses

-i.e., the gradually varying stresses one would normally obtain in beam orplate analysis

-and concentration factors to account for stress rise at structural discontinuities.

This complication is well known -see [39] for a layman's review -and is particularly problematic at the end of a weld since even the linearized stress can become singular at suchlocations (which also makes these locations more difficult to analyze using embedded modelingand explicit representation of the welds). Thus instead of utilizing a numerical code to estimate astress that increases without bound as mesh resolution is reduced, it is preferred to refer tostresses that are converged and then account for the stress increase through the use of adjustment factors obtained through years of field and testing experience of welded structures.

Weld factorsare a familiar example where rather than conducting a detailed geometry weld analysis, resulting in stresses that grow without bound as grid spacing is reduced, it is more accurate (in terms ofagreement with actually observed weld failure) to multiply a converged nominal stress with aweld factor that accounts for stress concentration, weld quality and plastic deformations at toesand roots.69 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information It is sometimes the practice to simply dismiss such locations on the basis that they are clearlysingularities and thus not compatible with ASME code stress definitions.

Instead, in the presentcase, a realistic estimate of the stress at the end of the weld is obtained by extrapolating theconverged stresses from weld nodes located below the limiting point. In order to obtain aconservative estimate of the end-of-weld stress intensity suitable for ASME code evaluation, thefollowing procedure is followed.

First the succession of nodes along the y-direction is identified.

The first adjacent node is denoted as node 'A', the second by 'B' and the third by 'C'. Next,various extrapolations of the stresses at these nodes to the limiting location are performed.

These include (for uniform spacing):

Linear extrapolation through nodes A & B: GLEI =2*YA -GBLinear extrapolation through nodes A & C: GL2 = 1.5"cA -0.5yCQuadratic extrapolation through nodes A, B & C: crQ = 3*aA -3*aB + GCwhere 0A, caB and crC are the stresses at nodes, A, B and C, respectively.

These formulae presumeuniform spacing of the involved nodes along the edge. More general formulae are easilydeveloped using Lagrangian interpolation functions so that:Caext = WA*GA + WB*aB + Wc*cycwhereLin. extrap. (A & B): wA_= d = d AdA dB dA dBdc -dALin. extrap. (A & C): wA dc , dwdA dC dA Cd~dC dcdA dAdBQuadratic extrap.:

w A = d , w B ==c ,WCd(dA dB)(dA dc) (dB dc)(dB dA)' WC (dc dA)(dc dB)and dA, dB and dC are the respective distances of nodes A, B and C from the singular location.

The last step is to define the limiting stress as:,7* = max{ UL1, YL2, aQ aA, A B, }which is simply the maximum stress obtained from any interpolation method or the threeadjacent nodes along the respective edge.The results of this extrapolation are summarized in Table 11. In this case, the quadratic orderfit through the alternating stress intensities of the three nodes immediately below the limitingnode is found to be the most conservative of the six extrapolation options (1 quadratic, 2 linear,and 3 zero-th order fits). Even so, for the limiting node the stress at the weld end obtained usingthe quadratic interpolation is less than 25% of the value extracted from the finite element code,further confirming that this FEA stress estimate is fictitious and not representative of the stressactually existing in the steam dryer. If one uses the weld factor of 1.8 used at other fillet welds,the limiting alternating stress ratio obtained with quadratic extrapolation is SR-a=6.83 confirming that the target alternating stress ratio of 2.0 margin is amply met.In the subsequent analyses in Sections 5.4-5.8 this extrapolation is employed at the tops ofthe welds connecting the closure plates and curved hoods (inner and middle).

This is a total of 8welds each exhibiting the same kind of singular behavior near the top end of the weld.70 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 11. Alternating stress intensities distribution along the middle hood/closure plate weldnear the limiting stress locations with FEA computed stress ratios, SR-a<2. Stress valuescorrespond to limiting values with frequency shifting.

The top of the weld occurs at z=87 in.End of weld stress estimates obtained by quadratic extrapolation are also shown together with thequadratic fit parameters.

Node 89317 88702(x,y) (in) (+60.2, -85.2) (-60.2,+85.2) z (in) Salt (psi)82.2258 (node C) 655.4 555.984.1692 (node B) 585.9 486.286.1098 (node A) 807.1 757.687 4183111 3459Lin. extrap. (nodes A, B) 908.6 882.1Lin. extrap. (nodes A, C) 841.9 803.9Quad. extrap. (nodes A, B, C) 1006 996.2Max {extrap, A, B, C 1 1006 996.2Extrapolated end SR-a 6.83 6.89Note: 1. The stress differs slightly from Table 9c since they correspond to results obtained withthe lifting rod/upper brace junctions modified to reflect as-built constraints (i.e., no weld at thesejunctions) 71 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Stresses Near Node 89317! -I--,' .[..5000-1J40000.U)U)a)L.300020000 Nodal Stresses-Linear Extrap. (A,B).........

Linear Extrap. (A,C)-Quadratic Extrap.1000000123 4 5 6 7s (= 87 -z) [in ]Stresses Near Node 887023500.-.II--....-I o0U)300025002000150010005000* Nodal Stresses-Linear Extrap. (A,B).........

Linear Extrap. (A,C)-Quadratic Extrap....................................

0012 3 4 5 6 7s (= 87 -z) [in ]Figure 15. Nodal FEA stresses and various extrapolations to the limiting stress location at nodes89317 and 88702 located on the lower and upper edges respectively of the top thick plate. Thevariable s is measured from the limiting stress location (end of horizontal weld or edge).72 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information 5.3.2 Analysis of Nodes Associated with Entr, 2 in Table 9cThis location involves a complex junction between multiple, orthogonally oriented elementsand experiences high stresses that are readily identifiable with the modal response of the innerside plate spanning between the inner vane banks. Several options were considered foraddressing the stress at this location including:

(i) Developing an embedded model where thelocal welds are modeled explicitly and linearized stresses extracted from various paths. (ii)Utilizing the stress interpolation method of Section 5.3.1. (iii) Various concepts for suppressing vibration of the inner side plate, culminating in the U-section stiffener that has been installed during the 2014 outage.Development of the embedded model (option (i) above), which was carried out after thepreliminary evaluation in Section 5.1, is summarized in Section 4.5. During the preliminary evaluation a stress reduction factor of SRF=0.77 was used which corresponds to the valuecalculated for a geometrically similar junction referenced as Embedded Model 3 in [12] and inTable 8. For the final evaluation carried out below an embedded model for the actual locationwas developed that more accurately represents the local joined components and characteristic local loading.

Two weld-specific SRFs are calculated using the paths indicated in Figure 12 andcomparing the linearized embedded model stresses against those in the baseline global model atthe same locations over the 60-120 Hz frequency interval.

The first is associated with the weldconnecting the top thick plate to the inner side plate for which the SRF=0.80, which is slightlyhigher than the 0.77 value previously imputed to this location in the preliminary screening.

Thesecond stress reduction factor pertains to the weld connecting the tie bar to the top plate and iscalculated to be SRF=0.66.

The revised stress reduction factors for these welds are used in theensuing results below. By themselves, these values are still insufficient to qualify Entry 2 inTable 9c.Therefore the option (ii) of utilizing the stress interpolation method in Section 5.3.1 wasconsidered for the present location also. It was determined that if stresses along the horizontal welds connecting the top thick plate to: (a) the top plate of the vane bank and (b) the perforated plate, were extrapolated to the high stress locations and the revised SRF applied, then therecommended margin could be achieved at all load conditions.

However, because the presentlocation involves multiple parts and weld lines that connect in a complex manner, there wassome uncertainty as to whether the stress extrapolation method could be applied.

Also, for oneof the load conditions involving closure of the RCIC valve the limiting alternating stress ratio atthis location was slightly below the recommended value thus necessitating cycle counting toestablish adequate fatigue margin for this occasional operating condition.

For these reasons it was decided that physical modification of the dryer at this locationshould be considered as an option to meet the recommended alternating stress ratio of SR-a>2.The dominant stress contribution at Entry 2 is attributed to vibration of the side plate so thatsuppressing this vibration reduces the stress. This can be accomplished by shifting the responsefrequency away from the dominant signal peaks, and/or strengthening the structure so that theresponse amplitude is reduced.

The reinforcement strip accomplishes both goals. The addedstiffness and mass are such that the fundamental frequency is shifted upward. This is beneficial in the present case since the limiting stresses are obtained at positive frequency shifts of +7.5 to73 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information

+10%. However, even if the modal frequency were unchanged, the reinforcement strip wouldreduce the response amplitude leading to a lower stress.As summarized in Appendix B the final design consists of a U-section stiffener bolted to theside plate. Bolts are preferred over welding to limit diver exposure to radiation and also to avoidheat-induced stresses in the materials.

The final design bolts the U-cross-section stiffening beamon the interior side (exposed to the steam path) of the side plate using 6 bolts. Further details ofthe final design and its FEA implementation are given in Appendix B. Unit solutions with thebolted stiffening channel attached to the dryer were generated over 60-130 Hz. Since the meshfor this dryer model differs from that used in previous global dryer evaluations it is not possibleto simply 'mix-and-match' the associated unit solutions.

Instead, the dryer stresses with the U-section stiffener attached are only evaluated at the current Entry 2, at the ends of the stiffening channel (to ensure there are no local stress increases that may exceed margin) and at the middleof the channel for the purpose of bolt design [25]. Section 5.8 shows that with the stiffening channel in place all nodes at Entry 2 meet the recommended SR-a>2 margin for all load and totaluncertainty combinations.

These results in Section 5.8 with the channel in place still make use the SRFs (0.80 and 0.66as calculated under option (i) above). However, it has also been shown that even if these SRFsare not invoked the alternating stress ratios at these locations remain above 2.0 except for theoccasional loads associated with RCIC valve closure.

For the RCIC line closed cases, it isappropriate to use cycle counting and evaluate the cumulative usage factors (CUFs) to assessfatigue.

Based on the CUFs calculated for other nodes, it is expected that the resulting CUFs atthese locations for these load cases would be well below 1.5.3.3 Analysis of Nodes Associated with Entry 3 in Table 9cThis location lies on the perimeter of the thickened plate used to reinforce the upper liftingrod bracket.

When considering all other nodes in the neighborhood of the limiting node (87784)and its reflected images, it is found (see Table 9c and Figure 14c) that this location only involvesa single node with an alternating stress ratio, SR-a=I.80.

It occurs at the common elementjunction of the outer vane bank side plate, closure plate and the perforated plates. The side plateexperiences the highest stress even though it is also the thickest of the connected members.

Thisis because the forces induced by rod motion are equilibrated by the support structure

-in thiscase, primarily the bracket itself and the vane bank side plate. The modifications prior to theEPU operation reinforced this location using reinforcement plates. However, the stress becomeshigh at the perimeter of the reinforcement where the effective thickness transitions back to theoriginal 3/8" value.Examination of the stress ratios at the corresponding locations at the three othercorresponding upper lifting rod braces reveals that these ratios are considerably higher than 2.0(2.45 at node 87592, 2.18 at node 89598 and 2.79 at node 101870).

Also, the mesh at thelimiting location is highly skewed whereas at the other three locations, it is more isotropic.

Thehigh stress at the limiting location is therefore attributed to numerical overestimation associated with low mesh quality.

To address this location the unit solutions were recomputed over the 13-26 Hz frequency range with a locally improved and refined (using a 1" spacing rather than the 2"used before) mesh that also implements the revised upper brace/lifting rod constraint (see next).74 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information The respective grids are compared in Figure 16. The specified frequency range covers onedecade (as was done in embedded models) and completely encompasses the dominant responsepeak. The unit solution stresses near the response peak frequency 19.3 Hz on the old and newgrids are compared in Figure 17.A structural evaluation, with frequency shifting

included, is then carried out over this limited13-26 Hz using the original and updated unit solutions using the normal operation EPU entrancepressures (appropriately zeroed outside the 13-26 Hz). The limiting alternating stress ratiosobtained in this manner on the original and revised grids are SR-a=2.16 and 3.53 respectively sothat over this frequency range only one can reduce the computed stress by a factor ofSRF=2.16/3.53=0.61.

A conservative estimate of the limiting stress can then be obtained from:a = a(0,13) + SRF*a(13,26)

+ a(26,250) where a(f1 ,f2) is the stress contribution over the frequency interval

[f ,f2] on the original mesh.Using this formula, the limiting stress ratio is SR-a=2.51 for' the baseline EPU load.A summary of the stress ratios at this location for the other loads is given below.Load SR-aoutN55 (base EPU operation):

2.510outN59 (drain trap out of service -92.5Hz signal):

2.636outN58 (RCIC line close -89.3 Hz signal):

2.090outN61 (base EPU load with conservative bias+unc.):

2.482outN62 (same as outN59, but with conservative bias+unc.):

2.599outN63 (same as outN58, but with conservative bias+unc.):

2.07275 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information~Node 87784, overstressedNew nodesto analyzeFigure 16. Original (top) and refined (bottom) meshes and the nodes accessed for stresscomparison.

Solid elements representing the lifting rod are omitted here for better viewing of theshell element mesh.76 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information ANSYS M0G MAJAN 31 201409 :51:45NODAL SOLUTIONSTEP-ISUB -1FREQ=19.

334REAL ONLYSINT (AVo)PoverOraphics EFACET-1AVRES-Xat DUX -.258722SUN -.627013SUx -6103303333666710000133331666720000233332666730000ANSYS 10. OA1JAN 31 201409:52:41NODAL SOLUTIONSTEP-13SUB =1 334REAL ONLYSINT (AVG)PoverOraphics EFACET=1AVRES-Mat DUX -.28S587SUN -.738192SUX -5683103333666710000133331666720000233332666730000Figure 17. Nodal stresses obtained on the original (top) and refined (bottom) grids for the unitsolution at 19.3 Hz. The stress at the limiting node is 17597 psi, whereas on the new mesh thelimiting stress in the same neighborhood (see Figure 16) is 13818 psi, which corresponds to a22% reduction in stress. The solid elements associated with the lifting rod are omitted here forbetter viewing of the high stress region.77 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information 5.3.4 Evaluation of Nodes on Upper Brace/Lifting Rod Junctions In previous evaluations, the upper lifting rod brace and the lifting rod were assumedconnected by a weld. This reflected the best interpretation of the combined understanding of thedrawings (particularly steam dryer modification drawing number 0016010001504) and as-builtconfiguration, which are known to differ (for example, the drawings originally indicated theinstallation of four braces and the addition of straps to the upper braces).

Several nodes locatedon the upper lifting rod brace were observed to have stress ratios below 2.0 under the assumption that the upper brace and lifting rod are welded together, thus requiring imposition of a weldfactor of 1.8 on the alternating stress intensity.

However, careful review and photographic evidence have since confirmed that the uppermost brace/lifting rod junctions are not welded (the lower two braces however, are). Application of the weld factor at the upper brace/lifting rod junction is therefore no longer warranted.

Also,lack of a welded connection between the upper brace and lifting rod means that the lifting rodonly imparts lateral forces to the top brace; no transmission of vertical forces or any momentsoccurs between the rod and remaining steam dryer structure.

Therefore it is appropriate to modelthese connections as displacement constraints in the horizontal directions only (i.e., require thatat the lines of connection the bracket and lifting rod experience identical deflections in thehorizontal directions);

no constraints are imposed on displacements in the vertical direction or onany rotational degrees of freedom.

Modifying the constraints in this manner does not alter themesh structure or node indexing so that one retains the option of generating new unit solutions over the frequency ranges of concern and then overwriting existing unit solutions over theseranges. Despite this change in constraints conditions, one finds that the modal frequencies associated with the lifting rod do not change significantly.

This is because the structural reactionmoments imposed by the support bracket upon the significantly stiffer and more massive liftingrod are small compared to the deformation stresses and inertial forces of the lifting rod. Themain change is that the support structure no longer experiences moments and vertical forcesimparted from the lifting rod leading to an overall reduction in local stresses.

The naturalfrequency of the fundamental mode changes from 19.62 Hz (fully welded) to 19.28 Hz (laterally constrained),

which is amply covered by the +/-10% frequency shifting required in the stressevaluation.

In the real time stress evaluations performed below, the unit solutions are regenerated overthe 17-22 Hz frequency interval with the boundary conditions at the lifting rod/upper braceconnection modified as described above. All subsequent post-processing in Sections 5.4 to 5.8 isperformed with these connections treated as non-welded.

Specifically, no weld factor is applied.This results in a limiting alternating stress ratio of SR-a=2.09 occurring on the contact edgebetween the upper brace and lifting rod. Several high stress locations also emerge on the bracethat are on the junction with the vane side plate. These are located approximately 0.35" to 0.6"from the vertical face representing the outer vane bank wall. Since the reinforcement extends outto 1.63" from the outer vane bank wall (drawing 10082C94 in [6]), these nodes are actuallywithin the reinforcement structure.

Stresses at these nodes are thus substantially reduced and theassociated stress ratios well above 2.0 (SR-a=6.44 or higher).78 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information 5.3.5 Correction of Upper Support Ring/Support Interface GeometryThe limiting peak stress occurs on the weld connecting the upper support ring to theearthquake block. This high stress was discovered to be due to an approximate and overlyconservative representation of the USR/support interface contact area. Specifically in previousmodels the support lugs did not contact the USR so that all of the dryer weight is transferred through the weld attaching the earthquake block to the USR. However, based on drawing197R624, I of 3 (transmitted by E-mail from Exelon on 03-03-2014),

the support lugs actuallyextend under the support ring by 0.75". This means that a considerable portion of the steamdryer weight is now transmitted though this contact area, thus relieving the stresses in theearthquake block/USR weld. This relief was quantified by loading the USR with the deadweight of the steam dryer and comparing the limiting static stresses in the USR/earthquake block weldsobtained with and without the 0.75" support lug projection underneath the USR. The twomodels are depicted in Figure 18 and are both meshed with the same settings, namely, 2" spacingeverywhere, with 0.75" spacing near support lugs. The results from the two cases are comparedin Table 12 and show that a conservative correction factor of 0.64 can be used to account for theactual 0.75" contact surface between the USR and support lug, and thus more accurately estimatethe stresses in the USR/earthquake attachment weld.Table 12. Stresses in the USR/earthquake attachment welds at each of 4 support locations due todryer deadweight load. Variations in stress intensities are due to mesh differences and dryerasymmetry.

Support Location Weld stress [ ksi] Weld stress [ ksi] RatioOriginal support lug Extended (0.75") support lug1 19.9 12.3 0.622 21.5 13.7 0.643 21.8 13.0 0.604 21.6 12.7 0.5979 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information ANSYS'(311.

zýAWSJSQLOWWANY~ "~z)I-r XFigure 18. Geometry of the USR/support lug location involving the earthquake block (top, with0.75" overlap included) and meshes with (left) and without (right) the 0.75" overlap.80 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information 5.4 General Stress Distribution and High Stress Locations The stress evaluation performed in Section 5.1 using the normal operating load is repeatedhere in greater detail and using the following analysis modifications listed in Section 5.3: (i) Theextrapolation procedures described in Section 5.3.1 are applied to the tops of the weldsconnecting the closure plates to the curved hoods. (ii) The stress reduction factors developed inSection 4.5 for embedded model 6 and further summarized in Section 5.3.2 are appropriately applied instead of the 0.77 value used in the preliminary evaluation of Section 5.1. (iii) Real timeresults are used to manually adjust the stress ratio near the lifting rod brace as indicated inSection 5.3.3 to compensate for poor mesh quality.

(iv) No weld factor is applied at the junctionbetween the lifting rod and upper restraint brace as described in Section 5.3.4. (v) Nodes on theside plate found to under the brace reinforcement plate (see Section 5.3.4) are adjusted in thesame manner as other nodes on the brace/side plate junction.

(vi) The stresses in theUSR/earthquake block attachment welds are adjusted as described in Section 5.3.5. As beforethe base EPU load with ACM 4. 1R bias and uncertainty values, is applied.

The full steam dryerevaluations in Sections 5.4 to 5.6 do not account for the U-section stiffener and retain a fullconnection between the lifting rod/upper brace junction (i.e., all degrees of freedom are coupled).

The maximum stress intensities obtained by post-processing the ANSYS stress histories forEPU at nominal frequency and with frequency shift operating conditions are listed in Table 13.Contour plots of the maximum stress intensities with all frequency shifts included are shown inFigure 19. The figures are oriented to emphasize the high stress regions.

Note that these stressintensities do not account for weld factors but do include end-to-end bias and uncertainty.

Further, it should be noted that since the allowable stresses vary with location, stress intensities do not necessarily correspond to regions of primary structural concern.

Instead, structural evaluation is more accurately made in terms of the stress ratios which compare the computedstresses to allowable levels with due account made for stress type and weld. Comparisons on thebasis of stress ratios are made in Section 5.5 as well as the real time analysis Sections 5.7 -5.8.From Figure 19a and Table 13 the maximum stress intensities in most areas are low (lessthan 1000 psi). For the membrane stresses (Pm) the high stress regions tend to occur: (i) on thewelds joining the seismic blocks and upper support ring (USR); (ii) the portion of the inner hoodlocated outboard of the closure plate connecting the inner and middle vane banks; (iii) thebottom of the central vertical side plate that joins the innermost vane banks (stress concentrations occur where this plate is welded to the inner base plates resting on the upper support ring); (iv)the welds joining the tie bars to the top cover plates on the vane banks; and (v) the bottoms of theinner vane bank side plates where they connect to the USR.The membrane

+ bending stress (Pm+Pb) distributions evidence a more pronounced modalresponse especially on the inner and middle hood structures, and on the inner closure plates.High stress concentrations are recorded on the bottom edge of the inner hood outboard of theclosure plate where it joins to the base plate and also near the dryer support locations.

Otherareas with high Pm+Pb stress concentrations include:

(i) the welded junctions between the tiebars and the top plates of the vane banks; (ii) tops of the closure plates where they are welded toa hood or vane bank end plates; (iii) the skirt/drain channel welds; and (iv) the outer hood sideplates and their welded connection to each outer hood (see Figure 19b-c).81 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information The alternating stress, Salt, distributions are most pronounced on the hoods and theirconnected side plates. Though not exposed directly to the MSL acoustic

sources, the inner andmiddle hoods are thinner than the outer ones and their responses are driven mainly by structural coupling rather than direct forcing.

The highest stress intensity at any frequency shift occurs atthe bottom of the inner hood where it meets the middle base plate. Significant response is alsoobserved on: (i) the welds connecting the tie bars to the vane bank top plates; (ii) parts involving the inner side plate; (iii) the bottoms of drain channels and the junctions between the hoods, hoodsupports and base plates; (iv) the welds joining the closure plates to the hoods and vane banks;and (v) parts connecting to the lifting rods. These locations are characterized by localized stressconcentrations as indicated in Figure 19e and have emerged as high stress locations in othersteam-dryers also.82 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 13a. Locations with highest predicted stress intensities for EPU conditions with no frequency shift.Stress Location Weld SRF(a) Location (in) node Stress Intensities (psi) Dom.Category x y z Pm Pm+Pb Salt Freq. (Hz)Pm Inner Side Plate No 3.1 119 0.5 37229 6648 8168 1761 83.3Side Plate Ext/Inner Base Plate Yes 16.3 119 0 94143 6208 9017 1275 19.8" USR/Support/Seismic Block(c)

Yes -6.9 -122.3 -9.5 113554 5060 5060 1852 15.7" Thin Vane Bank Plate/Hood Support/Inner Base Plate Yes -24.1 59.5 0 99487 4882 4895 1677 15.7" Hood Support/Outer Cover Plate/Outer Hood(4) Yes 0.8 -102.8 28.4 0 95267 4774 4820 2602 15.7Pm+PbMiddle Base Plate/Inner Backing Bar Out/Inner Backing Bar/Inner HoodYes-39.9-108.608419712939339179915.7Side Plate Ext/Inner Base Plate Yes 16.3 119 0 94143 6208 9017 1275 19.8Inner Side Plate No 3.1 119 0.5 37229 6648 8168 1761 83.3Side Plate/Top Plate(2) Yes 0.77 49.6 108.6 88 93256 1995 6963 1465 19.7Outer Cover Plate/Outer Hood Yes 102.8 -58.1 0 94498 1061 6065 1014 15.7Collar/Collar Contact No -79.2 -87.5 75.8 91651 929 5394 5192 19.8Brace No -79.6 -85.5 53.5 37693 4069 4312 3403 19.8Inner Side Plate No 14.4 -119 88 37592 657 3831 3126 82.6Top Thick Plate/Side Plate/Exit Top Perf/Inner Yes 0.8 -15.6 -119 86.5 101861 535 3935 3107 82.6Side Plate(6)Side Plate/Closure Plate/Exit Top Perf/Exit Yes -78.5 -85.2 74.5 87784 1171 2627 2624 19.8Mid Top Perf(d) I I INotes:(a) [I ((b) Full penetration weld so that weld factor, WF=1.4.(c) Corrected for support lug contact area per Section 5.3.5.(d) Compensated for mesh quality per Section 5.3.3.(1-6) Number referring to the 3) Entry is empty if no SRF is applied.(3)j]83 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 13b. Locations with highest predicted stress intensities taken over all frequency shifts at EPU conditions.

Stress Location Weld SRF(a) Location (in) node Stress Intensities (psi) % Freq. Dom.Category x y z Pm Pm+Pb Salt Shift Freq. (Hz)Pm Inner Side Plate No 3.1 119 0.5 37229 6685 8548 2281 5 123.5" Side Plate Ext/Inner Base Plate Yes 16.3 119 0 94143 6396 9246 1525 10 39.6Tie Bar(2) Yes 0.77 -49.3 -108.1 88 143795 5644 5644 1773 10 82.6Thin Vane Bank Plate/Hood Support/Inner Yes 24.1 -59.5 0 85191 5245 5330 2030 10 13.3Base PlateUSR/Support/Seismic Block(c)

Yes -6.9 -122.3 -9.5 113554 5205 5205 1951 5 13.7Pm+Pb Middle Base Plate/Inner Backing Bar Yes 39.9 108.6 0 85631 1478 10031 2750 10 82.6Out/Inner Backing Bar/Inner Hood III" Side Plate Ext/Inner Base Plate Yes 16.3 119 0 94143 6396 9246 1525 10 39.6" Inner Side Plate/Inner Base Plate Yes -2.3 -119 0 99200 4026 8778 3167 7.5 123.5Side Plate/Top Plate(2) Yes 0.77 49.6 108.6 88 93256 2154 7999 2563 10 82.6" Outer Cover Plate/Outer Hood Yes 102.8 -58.1 0 94498 1252 6440 1422 7.5 82.0Salt Collar/Collar Contact No -79.2 -87.5 75.8 91651 968 6176 5922 5 18.8Inner Side Plate No 14.4 -119 88 37592 764 5055 4235 7.5 78.7Top Thick Plate/Side Plate/Exit Yes 0.8 -15.6 -119 86.5 101861 571 4859 4084 7.5 78.7Top Perf/Inner Side Plate(6)Brace No -79.6 -85.5 53.5 37693 5100 5328 3913 7.5 18.7Outer End Plate/Outer Hood Yes 101.9 -63.3 24.6 94509 795 4368 3233 10 82.0Notes: (a) (3) Entry is empty if no SRF is applied.(b) Full penetration weld so that weld factor, WF= 1.4.(c) Corrected for support lug contact area per Section 5.3.5.(d) Compensated for mesh quality per Section 5.3.3.(1-6) Number referring to the (3)84 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information YPm (psi]70006500600055005000450040003500300025002000150010005000Figure 19a. Contour plot of maximum membrane stress intensity, Pm, for EPU operation withfrequency shifts. The recorded stress at a node is the maximum value taken over allfrequency shifts. The maximum stress intensity is 6685 psi.85 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information ZYPm+Pb [psi]100009000800070006000500040003000200010000Figure 19b. Contour plot of maximum membrane+bending stress intensity, Pm+Pb, for EPUoperation with frequency shifts. The recorded stress at a node is the maximumvalue taken over all frequency shifts. The maximum stress intensity is 10031 psi.First view.86 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 19c. Contour plot of maximum membrane+bending stress intensity, Pm+Pb, for EPUoperation with frequency shifts. Second view from beneath.87 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information ZSal t[psi)600055005000450040003500300025002000150010005000Figure 19d. Contour plot of alternating stress intensity, Salt, for EPU operation with frequency shifts. The recorded stress at a node is the maximum value taken over all frequency shifts. The maximum alternating stress intensity is 5922 psi. First view.88 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information

,, YSalt [psi]600055005000450040003500300025002000150010005000Figure 19e. Contour plot of alternating stress intensity, Salt, for EPU operation with frequency shifts. The recorded stress at a node is the maximum value taken over all frequency shifts. Second view from below.89 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information 5.5 Load Combinations and Allowable Stress Intensities The stress ratios computed for EPU at nominal frequency and with frequency shifting arelisted in Table 14 (without frequency shifting) and Table 15 (with frequency shifting).

The stressratios are grouped according to type (SR-P for maximum membrane and membrane+bending stress, SR-a for alternating stress) and location (away from welds or on a weld). The tabulated nodes with frequency shifting in Table 15 are also depicted in Figure 20. The plotscorresponding to maximum stress intensities depict all nodes with stress ratios less than 5 or lessthan 4 as indicated.

For EPU operation at nominal frequency (no frequency shift) the minimum stress ratio isidentified as a maximum stress, SR-P=l.49, and is recorded at the bottom of the inner hoodwhere it meets the middle base plate. In previous evaluations

[5] the weld joining the uppersupport ring to the earthquake block had the lowest stress ratio, but this location is no longerlimiting following the correction for support lug extension under the USR as described in Section5.3.5. The minimum alternating stress ratio at zero frequency shift, SR-a=2.21, occurs at the topof the inner vane bank where it meets the inner side plate and vane bank end plate (location 1 inFigure 20h).The effects of frequency shifts are conservatively accounted for by identifying the minimumstress ratio at every node, where the minimum is taken over all the frequency shifts considered (including the nominal or 0% shift case). The results are summarized in Table 15 and show thatthe lowest stress ratio, SR-P=l.39, occurs at the same location as in the nominal case (rotated by1800) and is only slightly lower. The next three lowest SR-P locations in Table 15b are the sameas in Table 14b or at a reflected image location (for the second entry). With frequency shiftingthe lowest alternating stress ratio also occurs at the same location (see location 1 in Figure 20h)and assumes a value of SR-a=l.68.

Based on the largest Fourier coefficient, the dominantfrequency in the signal contributing to this stress is 78.7 Hz; with the +7.5% shift accounted forit induces a structural response at 84.6 Hz. More details of the corresponding stress responsespectra are provided in the following section.

Other than this one location, all other entries inTable 15c have alternating stress ratios above 2.0. The alternating stress ratio below 2.0 isaddressed in Section 5.8 by adding a U-section stiffener.

When this channel is added to the sideplate, all stress ratios about this location achieve stress ratios that are well above 2.0 for all of theload and total uncertainty combinations.

The next lowest alternating stress ratio is in Table 15aand occurs where the upper brace contacts the lifting rod with SR-a=2.09.

In the CLTP-based stress evaluation

[11] the limiting node location was node 95267, which appears here as the 6thentry in Table 15c. Virtually all of the limiting stress ratios occur at either the +7.5% or +10%shifts.90 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 14a. Limiting non-weld locations with at EPU conditions with no frequency shift. Stress ratios are grouped according to stresstype (maximum

-SR-P; or alternating

-SR-a).Stress Location Location (in.) node Stress Intensity (psi) Stress Ratio Dom.Ratio x y z Pm Pm+Pb Salt SR-P SR-a Freq. (Hz)SR-P 1. Inner Side Plate 3.1 119 0.5 37229 6648 8168 1761 2.54 7.02 83.32. Hood Support 89 -28.4 0 14474 4491 4545 2562 3.76 4.83 15.7" 3. Thin Vane Bank Plate -15.6 -118.4 0.6 2558 4204 4617 565 4.02 21.89 19.7SR-a 1. Collar/Collar Contact -79.2 -87.5 75.8 91651 929 5394 5192 4.70 2.38 19.8" 2. Brace -79.6 -85.5 53.5 37693 4069 4312 3403 4.15 3.63 19.8" 3. Inner Side Plate 14.4 -119 88 37592 657 3831 3126 6.62 3.96 82.691 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 14b. Limiting peak stress ratios, SR-P, on welds at EPU conditions with no frequency shift. Bold text indicates minimum stressratio on the structure.

Location SRF(a) Location (in.) node Stress Intensity (psi) Stress Ratio Dom.X y z Pm Pm+Pb Salt SR-P SR-a Freq. (Hz)

Plate/Inner Backing Bar Out/Inner,--...

39.9 -108.6 0 84197 1293 9339-. 1799 .1.49 3,82 .1517..... ... ... ......

H... =. .. .. .-. ...... .....*. ....... ......_ _ :. .._. ...... : ... ._ : /7 ..: o ..2. Side Plate Ext/Inner Base Plate 16.3 119 0 94143 6208 9017 1275 1.50 5.39 19.83. Hood Support/Outer Base Plate/Middle Backing Bar -71.3 0 0 95428 5691 5803 2216 1.63 3.10 15.74. Inner Side Plate/Inner Base Plate -2.3 -119 0 99200 3972 7741 2363 1.80 2.91 82.65. Upper Support Ring/Support/Seismic Block(c)

-6.9 -122.3 -9.5 113554 5060 5060 1852 1.84 3.71 15.76. Hood Support/Middle Base Plate/Inner Backing -39.9 0 0 85723 5008 5242 1500 1.86 4.58 73.0Bar/Inner Hood(b)7. Tie Bar(2) 0.77 49.3 108.1 88 141275 4911 4911 1042 1.89 6.59 19.78. Thin Vane Bank Plate/Hood Support/Inner Base Plate -24.1 59.5 0 99487 4882 4895 1677 1.90 4.10 15.79. Hood Support/Outer Cover Plate/Outer Hood(4) 0.8 -102.8 28.4 0 95267 4774 4820 2602 1.95 2.64 15.710. Thin Vane Bank Plate/Hood Support/Middle Base Plate 55.6 -54.3 0 98968 4542 4625 1716 2.05 4.00 60.411. Hood Support/Middle Base Plate/Inner Backing 39.9 -59.5 0 101435 4525 4918 1621 2.05 4.24 60.4Bar/Inner Hood(b)12. Outer Cover Plate/Outer Hood 102.8 -58.1 0 94498 1061 6065 1014 2.30 6.77 15.7Notes: (a) [[(b) Full penetration weld so that weld factor, WF=1.4.(c) Corrected for support lug contact area per Section 5.(d) Compensated for mesh quality per Section 5.3.3.(1-6) Number referring to the EPU conditions with no frequency shift.Location SRF(a) Location (in.) node Stress Intensity (psi) Stress Ratio Dom.x y z Pm Pm+Pb Salt SR-P SR-a Freq. (Hz)1. Top Thick Plate/Side Plate/Exit Top Perf/Inner 0.8 -15.6 -119 86.5 101861 535 3935 3107 3.54 2.21 82.6Side Plate(6)2. Side Plate/Closure Plate/Exit Top Perf/Exit Mid -78.5 -85.2 74.5 87784 1171 2627 2624 5.31 2.62 19.8Top Perf(d)3. Hood Support/Outer Cover Plate/Outer Hood(4) 0.8 -102.8 28.4 0 95267 4774 4820 2602 1.95 2.64 15.74. Inner Base Plate 23.1 113.2 0 66988 862 6254 2473 2.23 2.78 19.75. Inner Side Plate/Inner Base Plate -2.3 -119 0 99200 3972 7741 2363 1.80 2.91 82.66. Side Plate/Brace(d) 79.7 -85.2 31.2 87633 3034 3175 2319 3.06 2.96 19.7Notes: (a) [I(b) Full penetration weld so that weld factor, WF=1.4.(c) Corrected for support lug contact area per Section 5.(d) Compensated for mesh quality per Section 5.3.3.(1-6) Number referring to the (3) Entry is empty if no SRF is applied.(3)]]93 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 15a. Limiting non-weld locations with at EPU conditions with frequency shifts. Stress ratios are grouped according to stresstype (maximum

-SR-P; or alternating

-SR-a). Locations are depicted in Figure 20.Stress Location Location (in.) node Stress Intensity (psi) Stress Ratio % Freq. Dom.Ratio x y z Pm Pm+Pb Salt SR-P SR-a Shift Freq. (Hz)SR-P 1. Inner Side Plate 3.1 119 0.5 37229 6685 8548 2281 2.53 5.42 5 123.5" 2. Hood Support 89 -28.4 0 14474 5050 5210 3068 3.35 4.03 10 82.0" 3. Inner Hood 26.8 108.2 88 72608 1780 6838 1743 3.71 7.09 10 82.64. Thin Vane Bank Plate -15.6 -118.4 0.6 2558 4311 4727 833 3.92 14.84 10 16.0SR-a 1. Collar/Collar Contact -79.2 -87.5 75.8 91651 968 6176 5922 4.10 2.09 5 18.82. Inner Side Plate 14.4 -119 88 37592 764 5055 4235 5.02 2.92 7.5 78.7" 3. Brace -79.6 -85.5 53.5 37693 5100 5328 3913 3.31 3.16 7.5 18.7" 4. Side Plate/Brace(a)

-79.7 -85.2 75.8 90307 2363 3327 3232 7.15 3.83 2.5 19.5Note: (a) Adjusted according to Table 11 of [26]94 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 15b. Limiting peak stress ratios, SR-P, on welds at EPU conditions with frequency shifts. Bold text indicates minimum stressratio on the structure.

Locations are depicted in Figure 20.Location SRF(a) Location (in.) node Stress Intensity (psi) Stress Ratio % Freq. Dom.x y z Pm Pm+Pb Salt SR-P SR-a Shift Freq. (Hz)1. Middle Base Plate/Inner Backing Bar 39.9 108.6 0 85631 1478 10031 2750 1.39 2.50 10 82.6Out/Inner Backing Bar/Inner Hood2. Side Plate Ext/Inner Base Plate 16.3 119 0 94143 6396 9246 1525 1.45 4.51 10 39.63. Hood Support/Outer Base Plate/Middle Backing -71.3 0 0 95428 6159 6227 2634 1.51 2.61 7.5 14.4Bar4. Inner Side Plate/Inner Base Plate -2.3 -119 0 99200 4026 8778 3167 1.59 2.17 7.5 123.55. Tie Bar(2) 0.77 -49.3 -108.1 88 143795 5644 5644 1773 1.65 3.87 10 82.66. Thin Vane Bank Plate/Hood Support/Inner Base 24.1 -59.5 0 85191 5245 5330 2030 1.77 3.38 10 13.3Plate7. Thin Vane Bank Plate/Hood Support/Middle 55.6 -54.3 0 98968 5241 5318 2232 1.77 3.08 10 82.0Base Plate8. USR/Support/Seismic Block(c)

-6.9 -122.3 -9.5 113554 5205 5205 1951 1.79 3.52 5 13.79. Hood Support/Middle Base Plate/Inner Backing 39.9 0 0 88639 5159 5558 1926 1.80 3.57 10 69.5Bar/Inner Hood(b)10. Hood Support/Outer Cover Plate/Outer Hood(4) 0.8 -102.8 28.4 0 95267 5054 5081 2996 1.84 2.29 10 13.711. Hood Support/Middle Base Plate/Inner Backing -39.9 59.5 0 90468 5044 5180 1849 1.84 3.72 10 82.6Bar/Inner Hood(b)Notes: (a) (3) Entry is empty if no SRF is applied.(b) Full penetration weld so that weld factor, WF=1.4.(c) Corrected for support lug contact area per Section 5.3.5.(d) Compensated for mesh quality per Section 5.3.3.(1-6) Number referring to the (3)95 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 15c. Limiting alternating stress ratios, SR-a, on welds at EPU conditions with frequency shifts. Locations are depicted inFigure 20.Location SRF(a) Location (in.) node Stress Intensity (psi) Stress Ratio % Freq. Dom.x y z Pm Pm+Pb Salt SR-P SR-a Shift Freq. (Hz)1. Top Thick Plate/Side Plate/Exit Top Perf/Inner 0.8 -15.6 -119 86.5 101861 571 4859 4084 2.87 1.68 7.5 78.7Side Plate(6)2. Outer End Plate/Outer Hood 101.9 -63.3 24.6 94509 795 4368 3233 3.19 2.12 10 82.03. Thin Vane Bank Plate/Inner Base Plate 15.6 114.4 0 99635 3531 5532 3194 2.52 2.15 7.5 82.04. Inner Side Plate/Inner Base Plate -2.3 -119 0 99200 4026 8778 3167 1.59 2.17 7.5 123.55. Side Plate/Top Plate(2) 0.77 49.6 -108.6 88 103080 1406 5009 3046 2.78 2.25 10 82.66. Hood Support/Outer Cover 0.8 -102.8 28.4 0 95267 5054 5081 2996 1.84 2.29 10 13.7Plate/Outer Hood(6)7. Side Plate/Brace(

4) 0.64 79.7 -85.2 31.2 87633 3526 3734 2976 2.64 2.31 10 96.38. Thick Vane Bank Plate/Thin Vane Bank 87 -85.2 11.6 90786 907 10028 2941 1.39 2.34 10 82.0Plate/Side Plate/Side Plate Ext/Outer End Plate9. Tie Bar 17.6 59.8 88 137575 4549 4549 2906 2.04 2.36 10 82.610. Side Plate/Top Plate(2) 0.77 81.1 -85.2 88 91055 858 3611 2811 3.86 2.44 10 86.811. Submerged Drain Channel/Submerged Skirt 76.7 -100 93488 498 4963 2811 2.81 2.44 10 123.512. Side Plate/Brace 85.7 85.2 31.2 89614 1762 3957 2798 3.52 2.45 10 82.6Notes: (a) 1[(3)]] Entry is empty if no SRF is applied.(b)(c)(d)Full penetration weld so that weld factor, WF=1.4.Corrected for support lug contact area per Section 5.3.5.Compensated for mesh quality per Section 5.3.3.(1-6) Number referring to the (3)96 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information SR-P4.94.64.343.73.43.12.82.5Figure 20a. Locations of minimum stress ratios, SR-P<5, associated with maximum stresses atnon-welds for EPU operation with frequency shifts. The recorded stress ratio is the minimumvalue taken over all frequency shifts. The numbers refers to the enumerated location for SR-Pvalues at non-welds in Table 15a. This view shows locations 1 and 3.97 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 20b. Locations of minimum stress ratios, SR-P<5, associated with maximum stresses atnon-welds for EPU operation with frequency shifts. The recorded stress ratio is the minimumvalue taken over all frequency shifts. The numbers refers to the enumerated location for SR-Pvalues at non-welds in Table 15a. This view shows locations 1 and 2.98 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information zSR-P4.94.64.343.73.43.12.82.5Figure 20c. Locations of minimum stress ratios, SR-P<5, associated with maximum stresses atnon-welds for EPU operation with frequency shifts. The recorded stress ratio is the minimumvalue taken over all frequency shifts. The numbers refers to the enumerated location for SR-Pvalues at non-welds in Table 15a. This view shows location 4.99 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information K4~xSR-a54.84.64.44.243.83.63.43.22.82.62.42.22Figure 20d. Locations of minimum alternating stress ratios, SR-a<5, at non-welds for EPUoperation with frequency shifts. The recorded stress ratio at a node is the minimum value takenover all frequency shifts. Numbers refer to the enumerated locations for SR-a values at non-welds in Table 15a. View showing locations 1-4.100 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information YSR-P43.83.43.23282.42.221.81.61.4Figure 20e. Locations of minimum stress ratios, SR-P<4, associated with maximum stresses atwelds for EPU operation with frequency shifts. The recorded stress ratio at a node is theminimum value taken over all frequency shifts. Numbers refer to the enumerated locations forSR-P values at welds in Table 15b. This view shows locations 1, 3, 6, 7 and 9-11.101 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information z¥Figure 20f. Locations of minimum stress ratios, SR-P<4, associated with maximum stresses atwelds for EPU operation with frequency shifts. The recorded stress ratio at a node is theminimum value taken over all frequency shifts. Numbers refer to the enumerated locations forSR-P values at welds in Table 15b. This view shows locations 2, 4, 5 and 10.102 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information zY ý xFigure 20g. Locations of minimum stress ratios, SR-P___4, at welds for EPU operation withfrequency shifts. The recorded stress ratio at a node is the minimum value taken over allfrequency shifts. Numbers refer to the enumerated locations for SR-P values at welds in Table15b. This view from below shows locations 4, 5, 8 and 10.103 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 20h. Locations of minimum alternating stress ratios, SR-a_<4, at welds for EPU operation with frequency shifts. The recorded stress ratio at a node is the minimum value taken over allfrequency shifts. Numbers refer to the enumerated locations for SR-a values at welds in Table15c. This view shows locations 1, 2, 4, 5 and 7-10.104 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 20i. Locations of minimum alternating stress ratios, SR-a_<4, at welds for EPU operation with frequency shifts. The recorded stress ratio at a node is the minimum value taken over allfrequency shifts. Numbers refer to the enumerated locations for SR-a values at welds in Table15c. View showing locations 1, 4-6, 8, 9 and 11.105 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information YSR-a43.63.43.23Figure 20j. Locations of minimum alternating stress ratios, SR-a.4, at welds for EPU operation with frequency shifts. The recorded stress ratio at a node is the minimum value taken over allfrequency shifts. Numbers refer to the enumerated locations for SR-a values at welds in Table15c. View around locations 3, 9 and 11.106 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information I34.6.2.6Figure 20k. Locations of minimum alternating stress ratios, SR-a_<4, at welds for EPU operation with frequency shifts. The recorded stress ratio at a node is the minimum value taken over allfrequency shifts. Numbers refer to the enumerated locations for SR-a values at welds in Table15c. View around locations 1, 2, 5, 9, 10 and 12.107 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information 5.6 Frequency Content and Filtering of the Stress SignalsThe frequency contribution to the stresses can be investigated by examining the powerspectral density (PSD) curves and accumulative PSDs for selected nodes having low alternating stress ratios. The accumulative PSDs are computed directly from the Fourier coefficients as1(con) = Yl(cOk2k=lwhere 8(ok) is the complex stress harmonic at frequency, (ok. Accumulative PSD plots areuseful for determining the frequency components and frequency ranges that make the largestcontributions to the fluctuating stress. Unlike PSD plots, no "binning" or smoothing offrequency components is needed to obtain smooth curves. Steep step-like rises in X(c1) indicatethe presence of a strong component at a discrete frequency whereas gradual increases in thecurve imply significant content over a broader frequency range. From Parsival's theorem,equality between E(O)N) (where N is the total number of frequency components) and the RMS ofthe stress signal in the time domain is established.

The selected nodes are the ones having the lowest alternating stress ratios (at a weld) in Table15. These are:Node 101861 -Entry 1 in Table 15c located on the weld joining the tie bar and inner vanebank top plate. The associated PSDs are shown in Figure 21a.Node 91651 -Entry 1 in Table 15a located where the lifting rod contacts the upper liftingrod brace. The associated PSDs are shown in Figure 21b.Node 94509 -Entry 2 in Table 15c located on the weld connecting the outer hood and its endplate. The associated PSDs are shown in Figure 21c.Node 99200 -Entry 4 in Table 15c located on the weld joining the inner side and base plates.The associated PSDs are shown in Figure 21d.Node 95267 -Entry 6 in Table 15c located on the welded common junction between theouter hood, hood support and outer cover plate. The associated PSDs are shown inFigure 21 e.These nodes are respectively labeled as 1 in Figure 20h, 1 in Figure 20d, and 2, 4 and 6 in Figure20h-k.In each case, since there are six stress components and up to three different section locations for shells (the top, mid and bottom surfaces),

there are a total of 18 stress histories percomponent.

Moreover, at junctions there are at least two components that meet at the junction.

The particular stress component that is plotted is chosen as follows.

First, the component andsection location (top/mid/bottom) is taken as the one that has the highest alternating stress. Thisnarrows the selection to six components.

Of these, the component having the highest Root MeanSquare (RMS) is selected.

For comparison the PSDs and cumulative PSDs are also shown forthe CLTP load examined in [11] at the shifts producing the highest stress intensity at that load.108 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information The first node (101861) is dominated by several peaks, with the dominant one centered at84.5 Hz for the +7.5% shifted case. From the accumulative PSD it is evident that frequency shifting increases this peak, but does not shift its frequency.

This is indicative of a peak in thesignal moving closer to a structural resonance.

The second (91651) and fifth (95267) nodes areboth dominated by low frequency peaks that do not change significantly with frequency shift.These are associated with lifting rod vibrations that are transmitted to the upper brace andsupporting structure on the outer vane bank. For node 94509, a broad peak at 90 Hz and a morenarrow one at 106 Hz dominate the response and both differ significantly from the non-shifted curve. Finally, for node 99200 multiple peaks are present that do not shift significantly withfrequency shift.Another way to characterize the dominant frequencies is to plot the dominant frequency overthe dryer surface.

For each finite element node the frequency associated with the largest stressharmonic (at any frequency shift) is recorded.

A contour map of this dominant frequency isshown in Figure 22. This map is useful in a qualitative sense for identifying what dryercomponents appear most responsive to particular frequencies.

Low frequency responses dominate the vane bank side plates, hood supports, lifting rods and parts of the outer coverplates. For most of the dryer hoods, the dominant frequencies are near 70 Hz as indicated inFigure 22b. The skirt response is dominated by higher frequency responses in the 80-130 Hzrange, as are the vane bank top plates, hood side plates, and inner closure plates. It should benoted that loadings above 200 Hz play no role in the dynamic response of this dryer.109 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Node 101861, awCLW.E:310008006004002000050 100 150 200Frequency

[ Hz ]Node 101861, a2500.6510610510410001001010 50 100 150 200 250Frequency

[ Hz ]Figure 21a. Accumulative PSD and PSD curves of the cyy stress response at node 101861.110 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Node 91651, aYYC-ciU)C-75E:3U)0.U)U)I £uu1000800600400200010710810110410001001010-~----------

*.... iý .. ... ...... .. ~ ... ... ... ... # .... .- P.~.. ....i! ..-~ No shiJI---% shf50100 150Frequency

[ Hz]Node 91651, a2002500 50 100 150 200 250Frequency

[ Hz ]Figure 21b. Accumulative PSD and PSD of the (Yyy stress response at node 91651.111 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Node 94509, a500.i;E~:340030020010001011040 50 100 150 200Frequency

[ Hz]Node 94509, aYY250"I"N.10000a.100100 50 100 150 200 250Frequency

[ Hz]Figure 21c. Accumulative PSD and PSD of the c;yy stress response at node 94509.112 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Node 99200, ad(0a-75EU)U)9N6600500400300200100010610410001000100101-No5shift l_ J_ .~ +.... s .. ..500 50 100 150 200Frequency

[ Hz ]Node 99200, a20 50 100 150 200 250Frequency

[ Hz ]Figure 21d. Accumulative PSD and PSD of the yzz stress response at node 99200.113 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Node 96267, a.E.E6005004003002001000TNo shift+10% Sqhift50 100 150 200 2500Frequency

[ HzNode 96267, %No shift10010100010050A-100150200250Frequency

[ Hz ]Figure 21e. Accumulative PSD and PSD of the axx stress response at node 95267.114 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information zYDom. Freq. [Hz]*5040*3020100Figure 22a. Contour map showing the dominant frequencies (i.e., the frequency with the largeststress harmonic).

This shows locations with dominant frequencies in the range 0-50 Hz.115 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information 1 /2Dorm. Freq. [Hz]~JQ1 807060I 50Figure 22b. Contour map showing the dominant frequencies (i.e., the frequency with the largeststress harmonic).

This shows locations with dominant frequencies in the range 50-80 Hz.116 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 22c. Contour map showing the dominant frequencies (i.e., the frequency with the largeststress harmonic).

This shows locations with dominant frequencies in the range 80-200 Hz.5.7 Real Time Analysis Without U-Stiffener As discussed in [2] it was established that during power ascension the MSL B loads wereaffected by the reactor core isolation cooling (RCIC) steam line configuration.

As a result,additional tests and data collections were conducted to define the loads with the RCIC lineisolated as described in Appendix C of [2]. In this section the stresses resulting from twoadditional loads corresponding to different RCIC line configurations are calculated andcompared against the baseline EPU loads used in the full steam dryer stress analysis.

The first load corresponds to an off-normal condition corresponding to the drain trap out ofservice.

This condition gives rise to a 92.5 Hz peak that was first observed in the MSL B datacollected at 110% CLTP. This load was not immediately reproducible, but has occurrednumerous times since. The second load definition was collected at 115% CLTP with the inboard117 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information RCIC valve temporarily closed. This resulted in a narrow peak on MSL B at 89.3 Hz. This alsois considered an off-normal condition corresponding to a short-duration technical specification limiting condition or to a transient loading associated with an unusual steam line drainconfiguration.

In addition to these two additional loads, the reanalysis using the ACM 4.1 R model resultedin new bias and uncertainty values resulting from benchmarking of this model. While it isreasonable to use these values, there has also been interest in calculating the acoustic loads byapplying the more conservative of the bias and uncertainty values associated with the ACM 4.1and ACM 4.1R models in each frequency interval.

Hence there are a total of six different operating condition and bias+uncertainty combinations.

These six loads are summarized inTable 16 together with the limiting stress ratios obtained with the real time evaluation summarized below. The Load Names (corresponding to MSL entrance signal files generated byCDI) are referenced in the subsequent tables (Table 17 and Table 18).In the present section real time stress evaluations are performed for the nodes listed in Table10 of Section 5.2 using each of the six load combinations.

In the real time assessment the stressintensities are calculated in the same manner as when analyzing the entire dryer and the limitingstresses with frequency shifting included are reported.

The complete list of resulting stress ratiosis given in Table 17. Note that one additional node (91651, entry 54) has been added. Thislocation had the limiting alternating stress ratio at a non-weld location (see Table 15a) which issufficiently close to the 2.0 margin that it was deemed of important to include it.For all load cases, the minimum alternating stress ratio is below 2.0 with the limiting stresslocations associated with the {45-53} entry set located near the top of the inner, vane bank/side plate junction.

These locations are addressed in Section 5.8 using the U-section stiffener.

Forconvenience, Table 17 lists separately the minimum stress ratios for this group (entries 45-53)and remaining points outside this group (entries 1-44 and the added entry, 54). This shows thatwhen entries 45-53 are omitted the alternating stress ratios are acceptable (above 2) for all loadcombinations except the RCIC valve closed conditions.

In Section 5.8 it is shown that addingthe U-section stiffener raises the alternating stress ratios of all entries 45-53 to 2.5 or higher sothat set of entries 1-44 and 54 becomes the limiting set.Table 16. Load & Summary for Dryer Without Addition of U-section Stiffener Load Condition Bias & Limiting Stress RatiosUncertainty Peak, SR-P Alternating, SR-aN55 Baseline (Normal)

ACM 4.1R 1.39 1.68N59 Drain Trap ooS 1.43 1.74N58 RCIC closed 1.34 1.35N61 Baseline (Normal)

Max {ACM 4.1, ACM 4.1R} 1.37 1.61N62 Drain Trap ooS " 1.42 1.67N63 RCIC closed " 1.32 1.29118 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 17. Alternating stress ratios obtained at 53 locations using: (i) different load andbias+uncertainty combinations.

Entry Node Load N55 Load N59 Load N58 Load N61 Load N62 Load N63SR-P SR-a SR-P SR-a SR-P SR-a SR-P SR-a SR-P SR-a SR-P SR-a1 37229 2.5 5.4 2.5 5.9 2.4 4.1 2.5 5.3 2.5 5.7 2.4 4.02 113286 4.2 4.5 3.3 3.1 4.0 4.0 4.2 4.5 3.3 3.1 4.0 4.03 37592 5.0 2.9 4.9 3.0 4.4 2.5 4.8 2.8 4.7 2.9 4.3 2.44 94143 1.4 4.5 1.5 4.9 1.4 3.6 1.4 4.4 1.5 4.8 1.4 3.55 85191 1.8 3.3 1.8 3.5 1.7 2.9 1.7 3.2 1.8 3.4 1.7 2.86 143795 1.6 3.9 1.7 4.0 1.6 3.0 1.6 3.7 1.7 3.9 1.5 2.97 113508 1.9 3.2 2.2 4.0 2.1 4.2 1.9 3.2 2.1 3.9 2.1 4.28 98968 1,8 3.1 1.8 3.2 1.6 2.4 1.8 3.0 1.8 3.1 1.6 2.39 94498 2.2 4.8 2.1 4.8 2.1 4.7 2.1 4.6 2.1 4.7 2.1 4.510 90468 1.8 3.8 2.0 3.6 1.9 3.5 1.8 3.7 2.0 3.5 1.9 3.511 101600 2.4 2.5 2.6 2.6 2.1 1.9 2.4 2.4 2.5 2.6 2.1 1.912 88639 1.8 3.6 1.8 3.0 1.7 3.0 1.8 3.5 1.8 2.9 1.7 2.913 92995 2.5 3.3 2.5 3.2 2.5 3.3 2.5 3.3 2.5 3.1 2.5 3.314 141237 2.2 4.0 2.1 3.7 1.8 2.6 2.1 3.9 2.1 3.6 1.8 2.515 89317 2.3 6.8 2.1 6.0 1.7 5.3 2.2 6.8 2.1 6.0 1.6 5.316 87784 4.0 2.5 4.0 2.6 3.2 2.1 3.9 2.5 3.9 2.6 3.2 2.117 94509 3.2 2.1 3.4 2.2 2.7 1.6 3.1 2.0 3.3 2.1 2.6 1.518 99635 2.5 2.2 2.6 2.4 2.4 1.9 2.5 2.1 2.5 2.3 2.4 1.819 99200 1.6 2.2 1.7 2.5 1.6 2.0 1.6 2.1 1.7 2.4 1.5 2.020 113554 1.8 3.5 1.7 2.9 1.7 3.1 1.8 3.5 1.7 2.9 1.7 3.121 103080 2.8 2.3 2.9 2.1 2.4 1.7 2.8 2.2 2.9 2.0 2.3 1.722 113400 2.6 2.3 3.0 3.0 3.1 3.1 2.6 2.3 2.9 3.0 3.1 3.123 95267 1.8 2.3 1.8 2.3 1.7 2.1 1.8 2.2 1.8 2.2 1.7 2.124 90786 1.4 2.3 1.4 2.4 1.3 1.8 1.4 2.3 1.4 2.3 1.3 1.825 137575 2.0 2.4 2.0 2.2 1.6 1.6 2.0 2.3 1.9 2.1 1.6 1.526 95172 2.5 7.9 2.5 6.6 2.2 7.3 2.4 7.9 2.4 6.6 2.2 7.627 91055 3.9 2.4 3.5 2.0 2.8 1.8 3.8 2.3 3.4 2.0 2.8 1.728 93488 2.8 2.4 3.1 2.9 2.8 2.2 2.8 2.4 3.0 2.8 2.7 2.129 98956 2.2 2.5 2.0 2.4 2.0 2.1 2.1 2.4 1.9 2.4 1.9 2.030 85631 1.4 2.5 1.4 2.9 1.3 2.0 1.4 2.4 1.4 2.8 1.3 1.931 101818 2.6 2.5 2.6 2.9 2.6 2.1 2.6 2.4 2.6 2.8 2.5 2.032 98624 2.8 2.5 3.0 2.9 2.6 2.1 2.8 2.4 3.0 2.8 2.5 2.0119 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Table 17 (cont.).

Alternating stress ratios obtained at 53 locations using: (i) different load andbias+uncertainty combinations.

Entry Node Load N55 Load N59 Load N58 Load N61 Load N62 Load N63SR-P SR-a SR-P SR-a SR-P SR-a SR-P SR-a SR-P SR-a SR-P SR-a33 94514 3.6 2.6 4.1 2.7 3.1 1 3.5 2.5 3.9 2.6 3.034 95428 1.5 2.6 1.5 2.5 1.4 2.3 1.5 2.6 1.5 2.5 1.4 2.235 84090 2.2 2.6 2.3 2.9 2.0 2.3 2.1 2.6 2.3 2.8 2.0 2.236 99451 2.3 2.7 2.3 3.1 2.0 2.5 2.2 2.6 2.3 3.1 2.0 2.537 93451 5.1 2.7 5.2 2.8 4.4 2.2 4.9 2.6 5.1 2.7 4.3 2.238 98172 2.1 2.7 2.1 3.1 2.0 2.6 2.1 2.7 2.1 3.0 2.0 2.639 90924 3.8 2.8 3.8 3.1 3.6 2.8 3.8 2.7 3.7 3.0 3.5 2.840 100989 4.8 2.8 4.7 2.8 4.3 2.4 4.6 2.7 4.5 2.7 4.1 2.341 90926 4.9 2.8 5.2 3.3 4.1 2.5 4.8 2.7 5.1 3.2 4.1 2.442 99931 5.1 2.8 5.7 3.1 4.7 2.5 4.9 2.7 5.5 3.0 4.5 2.443 91091 1.9 3.0 1.9 3.3 1.8 2.5 1.9 2.9 1.9 3.2 1.8 2.544 88702 2.6 6.9 2.5 6.4 2.0 7.0 2.5 6.8 2.5 6.3 2.0 7.1MIN (1-44) 1.4 2.1 1.4 2.0 1.3 1.6 1.4 2.0 1.4 2.0 1.3 1.5Locations near top of inner vane bank/side plate/tie bar junctionResults WITHOUT U-Section Stiffener installed 45 101861 2.9 3.1 2.4 2.8 L 2.9 2.346 95197 3.3 3.2 2.7 3.2 1 3.1 2.547 99407 2.4 2.1 2.8 2.4 2.2 2.4 2.1 2.7 2.3 2.148 98442 3.1 3.1 2.5 3.0 3.0 2.449 98444 3.1 2.9 2.5 3.0 2.8 2.450 98451 3.6 3.3 2.6 3.4 1 3.2 2.551 98452 2.6 2.5 I 2.1 2.5 1 2.5 2.152 99408 3.1 3.3 2.0 2.8 3.0 L 3.2 2.753 91240 2.6 2.6 2.1 2.4 2.5 I 2.5 2.0 2.4MIN (45-53) 2.4 1.7 2.5 1.7 2.1 1.4 2.4 1.6 2.5 1.7 2.1 1.3Added Real Time Node (Non Weld)544.1 2.2 4.1 2.1 4.6 2.4 12.2Load N55 Load N59 Load N58 Load N61 Load N62 Load N63MIN (overall)

SR-P SR-a SR-P SR-a SR-P SR-a SR-P SR-a SR-P SR-a SR-P SR-a_______1.4 1.7 1.4 I1.7 1.3 1.4 1.4 1.6 1.4 1.7 1.3 1.3120 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information 5.8 Real Time Analysis Adjusted with U-Section Stiffener IncludedSection 5.7 shows that for all load combinations the alternating stress ratio is below the 2.0target level. The limiting stresses all occur on the set of entries,

{45-53 1, associated with the topof the inner vane bank/side plate junction also involving the inner tie bar. To address theselocations a U-section stiffener is installed 11" from the top edge of the side plate using six bolts.To analyze the stresses with the channel installed a new steam dryer model was developed withthe channel attached and unit stress solutions generated over the 60-130 Hz frequency range.The solutions are combined with the six loads in Table 16 and the stresses evaluated at thelimiting locations using the real time analysis.

Specifically, the nodes in the new FEA modelthat are near the set of entries,

{45-53},

are identified, and the stresses calculated andappropriately adjusted in accordance with the SRFs for embedded model 6 as described inSection 4.5. For nodes on welds the weld factors are also applied.

Finally the stresses and stressratios are adjusted to account for contributions from the remaining frequencies outside the 60-130 Hz range,Since the steam dryer meshes with and without the U-section stiffener are different it isdifficult to develop stresses at the exact same node locations.

However, by taking the maximumstress over all nodes in the same neighborhoods it is straightforward to infer the limiting stress(for each load combination) in the region represented by entries.

The associated limiting stressratios are listed in Table 18 showing that for all but the RCIC valve closed conditions, thelimiting alternating stress ratios are above 3.0. For the RCIC valve closed, these stress ratios are2.4 or higher thus confirming that the U-section stiffener is effective in restoring adequatemargin to these locations.

When compared against the minimum stress ratios from the remaining

entries, (1-44, 54 1, obtained from Table 17 it is clear that these latter locations now constitute thelimiting set. The minimum stress ratios for each load combination with the U-section stiffeners installed are listed in the last row of Table 18. Note that only the alternating stress ratios arelisted since the peak stresses are already demonstrated to meet margin (SR-P>l for all loadconditions) in Table 17. These show that for the normal (baseline) and drain trap out-of-service conditions the target alternating stress ratios are met. For the RCIC valve closed cases however,the limiting alternating stress ratios remain below 2.0. The nodes where the alternating stressratios are below 2.0 are identified in Table 17 by the greyed cells associated with entries 1-44.These nodes are evaluated separately in [10] using a cumulative fatigue usage analysis.

Thisanalysis is justified since RCIC valve closure is infrequent so that cycle counts are low andfatigue assessment using cycle counting to derive a cumulative usage factor (CUF) isappropriate.

As shown in [10] the CUFs for all these nodes are demonstrated to be well below1.0.To summarize, these results show that the dryer meets the target alternating stress ratio,SR-a _2.0, for both normal EPU operation and off-normal operation with the drain trap ooS. Forthe off-normal RCIC valve closed case, the target alternating stress ratio of 2.0 is not met, butsince this condition is infrequent a separate cycle counting analysis is appropriate.

This analysisis performed and shows that the CUF<I. Taken in their entirety, these results show that thesteam dryer qualifies for fatigue under all the load cases considered.

Note that this conclusion presumes that the installation of the U-section stiffener does not introduce new high stresslocations.

This is confirmed by calculating the stresses at the ends of the channels and showingthat these remain well above margin (SR-a>6.5 for all end stresses) for all load conditions.

It121 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information also assumes that installation of the channel does not significantly alter the stress ratios at otherlocations (i.e., associated with entries 1-44 and 54). This assumption is reasonable since thechannel acts mainly to suppress the vibration amplitude of the side plate rather than altering thefrequencies.

Hence the dominant effect of the U-section stiffener is to reduce stresses in thevicinity of the inner side plate, with a lesser reduction farther away.Table 18. Limiting alternating stress ratios with addition of U-section stiffener Entries Load N55 Load N59 Load N58 Load N61 Load N62 Load N63MIN (Locations 1-44, 54) 2.1 2.0 1.6 2.0 2.0 1.5Limiting Stress Ratios at Locations 45-53 with U-Section Stiffener Installed MIN (Locations 45-53) 3.4 3.4 2.5 3.3 3.3 2.4MIN (overall) 2.1 2.0 1.6 2.0 2.0 1.5122 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information

6. Conclusions A re-evaluation of the stresses on the NMP2 steam dryer due to acoustic loads, deadweight and static pressure (the Level A service condition) at EPU (115% EPU) conditions has beenperformed.

This re-evaluation was necessary due to an identified inconsistency in the previousacoustic loads predictions

[2] and also the detection of additional signals in recently collected data on the MSLSs whose impact on the dryer required assessment.

The acoustic loads areinferred from strain gage measurements on the MSLs and a calibrated acoustic circuit model(ACM, Rev. 4.1 R) that processes these measurements to define acoustic pressure distributions onthe steam dryer surfaces.

The ANSYS FEA package is then used to obtain the dryer stressresponse resulting from these acoustic loads and the stress results post-processed to obtain thelimiting alternating stress ratios. The results account for all biases and uncertainties identified for both the ACM Rev. 4. 1R and the FEA harmonic analysis.

The acoustic loads are prepared using the acoustic circuit model (ACM) version 4.1 R [2, 8,9] for three load conditions:

Baseline:

The normal EPU operating load.Drain Trap Out-of-Service:

An off-normal condition occurring when the drain trap in theRCIC system is isolated.

RCIC Valve Closed: An infrequent off-normal operating load occurs when the RCIC valveis intentionally closed.For each load two bias and uncertainty combinations were considered.

The first uses the biasesand uncertainties of the ACM 4. 1R model and the other uses a conservative combination of thetotal uncertainties from the ACM 4.1 and ACM 4.1 R models.A complete dryer stress evaluation using the Baseline load identified limiting locations withalternating stress ratios that were below the target of 2.0, but above the ASME allowable of 1.0.These locations were examined in more detail by improving local mesh quality, identifying welds at lifting rod/brace junctions, developing an embedded model, and processing stresssingularities at the ends of selected welds. The complete dryer stress evaluation was also used todevelop a list of nodes for real time stress evaluation at the other load combinations.

The stress evaluations with frequency shifting at all load combinations identified that the bestoption for maximizing long term operating margin for the inner side plate connecting the innervane banks was the installation of the U-section stiffener that is bolted on to each of these innerside plates (2 total). With this modification in place, all nodes have a peak stress ratio, SR-P, of1.3 or higher at all load combinations thus meeting the required margin for this stress type. Withregard to alternating

stresses, all of the nodes on the steam dryer have an alternating stress ratioof 2.0 or higher under the baseline and drain trap out-of-service loads so that dryer qualifies forthese conditions.

For the RCIC valve closed, certain locations have alternating stress ratiosbelow 2.0 with the minimum value being SR-a=I.5.

However, since this condition occursinfrequently, it is appropriate to assess fatigue using cycle counting.

The cumulative usagefactors (CUFs) for these locations are calculated in [10] and shown to all lie well below 1.0.123 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Taken in their entirety, these results show that the dryer qualifies for all of the level A serviceoperation with the U-section stiffener installed.

124 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information

7. References

1. Continuum

Dynamics, Inc. (2013) Non-Conformance Report (NCR) 03-43. 23September.

2. Continuum

Dynamics, Inc. (2014) Acoustic and Low Frequency Hydrodynamic Loads at115% CLTP Target Power Level on Nine Mile Point Unit 2 Steam Dryer to 250 Hz UsingACM Rev. 4.1R. C.D.I. Report No.14-09P (Proprietary),

December.

3. Continuum

Dynamics, Inc. (2010) Design and Stress Evaluation of Nine Mile Point Unit2 Steam Dryer Modifications for EPU Operation.

C.D.I. Report No.10-12P(Proprietary),

July.4. Continuum

Dynamics, Inc. (2012) Stress Evaluations of Nine Mile Point Steam DryerModifications.

C.D.I. Technical Memo 12-03, Rev. 0, January.5. Continuum

Dynamics, Inc. (2012) Stress Evaluation of Nine Mile Point Unit 2 SteamDryer at 115% CLTP. C.D.I. Report No.12-18P (Proprietary),

Rev. 0, Oct.6. Westinghouse (2011) NMP2 Steam Dryer Modifications and Repairs.

FCN-MODS-NMP2-11, Rev. 1.7. ASME (2007) Boiler and Pressure Vessel Code,Section III, Subsection NG.8. Continuum

Dynamics, Inc. (2011) ACM Rev. 4.1: Methodology to Predict Full ScaleSteam Dryer Loads from In-Plant Measurements (Rev. 3). C.D.I. Report No.10-09P(Proprietary),

November.

9. Continuum

Dynamics, Inc. (2013) Design Record File DRF-CDI-338 A.10. Continuum Dynamics Inc. (2014) Computation of Cumulative Usage Factor for the115% CLTP Power Level at Nine Mile Point Unit 2 with the Inboard RCIC Valve Closed.C.D.I. Technical Note No.14-04P (Proprietary),

April.11. Continuum

Dynamics, Inc. (2011) Stress Evaluation of Nine Mile Point Unit 2 SteamDryer Using ACM Rev. 4.1 Acoustic Loads. C.D.I. Report No.11-04P (Proprietary),

Rev. 0, May.12. Continuum

Dynamics, Inc. (2011) Sub-Modeling in the Nine Mile Point Unit 2 SteamDryer, Rev. 0. C.D.I. Report No.11-03P (Proprietary),

June.13. Continuum

Dynamics, Inc. (2007) Methodology to Predict Full Scale Steam Dryer Loadsfrom In-Plant Measurements, with the Inclusion of a Low Frequency Hydrodynamic Contribution.

C.D.I. Report No.07-09P (Proprietary).

14. Continuum

Dynamics, Inc. (2005) Methodology to Determine Unsteady PressureLoading on Components in Reactor Steam Domes (Rev. 6). C.D.I. Report No. 04-09(Proprietary).

15. NRC (2012) Safety Evalution by the Office of Nuclear Reactor Regulation Related toAmendment No. 149 to Facility Operating License No. NPF-69 Nine Mile Point NuclearStation, LLC, Nine Mile Point, Unit No. 2 Docket No. 50-410. ML1 13560333.

16. Structural Integrity Associates, Inc. (2010) Flaw Evaluation of Indications in the NineMile Point Unit 2 Steam Dryer Vertical Support Plates Considering Extended PowerUprate Flow Induced Vibration Loading (Rev. 0). SIA Calculation Package No.1000814.401, July.17. Structural Integrity Associates, Inc. (2014) Flaw Evaluation of Indications in the NineMile Point Unit 2 Steam Dryer Considering Current Licensed Thermal Power FlowInduced Vibration Loading.

Report No. 1400467.401 Revision 0, April.125 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information

18. Continuum

Dynamics, Inc. (2010) Acoustic and Low-Frequency Hydrodynamic Loads atCLTP Power Level on Nine Mile Point Unit 2 Steam Dryer to 250 Hz Using ACM Rev.4.1 (Rev. 2). C.D.I. Report No.10-10P (Proprietary),

January.19. ANSYS, Release 10.0 Complete User's Manual Set, (http://www.ansys.com).

20. Continuum

Dynamics, Inc. (2007) Response to NRC Request for Additional Information on the Hope Creek Generating

Station, Extended Power Uprate. RAI No. 14.110.21. Continuum

Dynamics, Inc. (2008) Stress Assessment of Hope Creek Unit 1 Steam DryerBased on Revision 4 Loads Model, Rev. 4. C.D.I. Report No.07-17P (Proprietary).

22. Press, W.H., et al., Numerical Recipes.

2 ed. 1992: Cambridge University Press.23. Continuum

Dynamics, Inc. (2009) Stress Assessment of Nine Mile Point Unit 2 SteamDryer at CLTP and EPU Conditions, Rev. 1. C.D.I. Report No.09-26P (Proprietary),

December.

24. Continuum

Dynamics, Inc. (2012) Real Time Monitoring of the Nine Mile Point Steamduring Power Ascension.

C.D.I. Technical Note No.12-17P (Proprietary).

June.25. Continuum

Dynamics, Inc. (2014) Stiffening Channel Bolt Analysis.

C.D.I. Technical Note No. 14-05, Rev. 1, March.26. Continuum

Dynamics, Inc. (2010) Stress Assessment of Nine Mile Point Unit 2 SteamDryer Using the Acoustic Circuit Model Rev. 4.1. C.D.I. Report No. 10-i1P(Proprietary),

June.27. Structural Integrity Associates, Inc. (2008) Flaw Evaluation and Vibration Assessment ofthe Nine Mile Point Unit 2 Steam Dryer for Extended Power Uprate Operating Conditions.

Report No. 0801273.401.

28. Continuum

Dynamics, Inc. (2008) Stress Assessment of Browns Ferry Nuclear Unit 1Steam Dryer, Rev. 0. C.D.I. Report No.08-06P (Proprietary).

29. O'Donnell, W.J., Effective Elastic Constants For the Bending of Thin Perforated PlatesWith Triangular and Square Penetration Patterns.

ASME Journal of Engineering forIndustry, 1973. 95: p. 121-128.30. de Santo, D.F., Added Mass and Hydrodynamic Damping of Perforated Plates Vibrating In Water. Journal of Pressure Vessel Technology, 1981. 103: p. 175-182.31. Ideichik, I E. and E. Fried, Flow Resistance, a Design Guide for Engineers.

1989,Washington D.C.: Taylor & Francis.

pg. 260.32. Continuum

Dynamics, Inc. (2007) Dynamics of BWR Steam Dryer Components.

C.D.I.Report No.07-11P.33. U.S. Nuclear Regulatory Commission (2007) Comprehensive Vibration Assessment Program for Reactor Internals During Preoperational and Initial Startup Testing.Regulatory Guide 1.20, Rev. 3, March.34. Weld Research Council (1998) Fatigue Strength Reduction and Stress Concentration Factors For Welds In Pressure Vessels and Piping. WRC Bulletin 432.35. Pilkey, W.D., Peterson's Stress Concentration

Factors, 2nd ed. 1997, New York: JohnWiley. pg. 139.36. Lawrence, F.V., N.-J. Ho, and P.K. Mazumdar, Predicting the Fatigue Resistance ofWelds. Ann. Rev. Mater. Sci., 1981. 11: p. 401-425.37. General Electric (GE) Nuclear Energy, Supplement 1 to Service Information Letter (SIL)644, "B WR/3 Steam Dryer Failure, "September

5. 2003.38. Tecplot, Inc. (2004) Documentation:

Tecplot User's Manual Version 10 Tecplot, Inc.,October.126 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information

39. Scholl, Rod, Size Matters, in ANSYS Advantage.

2014, ANSYS. p. 50-53.127 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information

[II (3)]](3)]]128 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information (3)]]129 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information (3)]]130 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information (3)]]131 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Appendix B. U-Section Stiffener The analysis in Section 5.7 shows that the limiting alternating stress ratios at all loadcombinations consistently occur on the junction between the top of the inner vane bank, the innervane bank end plate and the inner side plate connecting the two inner vane banks. The locationis depicted in Figure 14b. The high stress is associated with vibration of the inner side plate atapproximately 85 Hz. To suppress vibration of the side plates joining the inner vane banks, a U-section stiffener is bolted on to these side plates at a distance (top edge to top edge) of 11" fromthe top edge of the side plate. A bolted rather than welded attachment is preferred to minimizediver exposure to (radiation) dose during the 2014 RFO installation and also to avoid heat-induced stresses in the materials.

The present Appendix summarizes the design, finite elementanalysis and stress evaluation of the U-section stiffener.

Design of the U-section stiffener proceeded by first investigating the impact on stress due toaddition of a rectangular cross-section strip welded to the inner side plate. To facilitate analysisand simplify combining of unit solutions, the presence of the strip was first modeled as athickness change of selected elements in the side plate. This leaves the existing finite elementmesh and node indexing unchanged so that: (i) the new unit solutions can be swapped in easily(the old ones being simply overwritten) thus allowing the use of existing post-processing software tailored to the Nine Mile Point evaluation, unaltered; and (ii) the effect of the stiffening on the global model, including the changes in the modal properties, are fully accounted forwithout having to regenerate a complete unit solution sweep using a significantly larger andcomputationally time consuming model. Several combinations of strip thickness and verticallocation were considered.

For each one, a series of unit solutions was generated over afrequency range centered at 85 Hz and the combined with the MSL entrance signals to estimatestresses.

At first it seemed natural to locate the stiffening beam near the top edge of the sideplate since this is the elevation where the high stress occurs. However this location proved lesseffective in suppressing the dominant stress-inducing mode than placing it further down. A 5"high, 0.5" thick strip spanning the horizontal length of the side plate and located 11" below itstop edge was found to be most effective in suppressing the stresses at Entry 2 in Table 9c.Next the optimal thickness and location developed in the preceding design process wereexpressed as an equivalent stiffness and mass distribution that in turn were translated into a U-section stiffening beam design. The final design bolts the U-section stiffener on the interior side(exposed to the steam path) of the side plate using 6 bolts. Bolting the beam to interior side wasnecessary to avoid interference with the vertical guide rail intended to guide the steam dryer inand out of the RPV during an outage. The final design is depicted in Figure 24 and consists of a5"xO.5" vertical leg lying on the inner side plate and a pair of 2"xO.65" horizontal legs providing most of the stiffening.

The part is machined out of 304 SS qualified for usage as a steam dryercomponent.

The differences between the modeled (Figure 24a) and machined (Figure 24b) U-section stiffeners reflect alterations made to the manufactured part during and after analysis tosimplify installation and tooling requirements as well as a preference for using a shell element-based representation (to facilitate connection to the side plate and generation of a good qualitymesh). However, the cross-sectional second area moments controlling bending in the horizontal plane are in close agreement so that the modeled element model accurately represents the as-builtcomponent.

132 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information The final design was analyzed using FEA. The finite element model was developed by firstcreating a clean mesh on the side plate such that it contains straight lines corresponding to thereinforcement strip edges and the elements are regularly shaped. Figure 25a shows this mesh onthe outward side of the inner side plate. Next the channel is modeled using shell elements asshown in Figure 25b. The mesh is designed such that each node in the reinforcement strip isassociated with a node on the inner side plate that is located immediately below it. This one-to-one correspondence ensures accurate imposition of the contact conditions between the channeland side plate. Here the two parts are conservatively (since it minimizes net bending stiffness) assumed to be capable of sliding relative to each other. Thus, associated nodes (i.e., nodes on thestrip and inner side plate that lie on top of each other) are constrained to have the same deflection in the normal direction.

However, the other degrees of freedom on associated nodes are leftunconstrained.

The bolts themselves are represented with 0.5" long beam elements that connectthe side plate and U-section stiffener at the cantilevered ends. The exact cross-sectional dimensions of the bolts are not critical since only the shear and tensile forces in these elementsare of potential interest.

These forces, when added to the pre-loads associated with bolttightening, and divided by the bolt cross-sectional area yield the bolt stresses.

Unit solutions were generated for this finite element model over the 60-130 Hz frequency interval.

This range encompasses most of the contribution to the stress intensity at the highstress location the channel is intended to address (typically 97-98%; occasionally down to 90%for certain nodes). Signal content below 60 Hz is expected to have marginal impact since it liesbelow the fundamental frequency of the unrestrained plate (and, accordingly, also below that ofthe reinforced plate). These unit solutions were combined with the six load combinations defined in Table 16 to calculate the stresses at junction of the top of the inner vane bank and sideplate. Specifically, a list of nodes located within several inches of the high stress locations identified in the original (i.e., without the channels) global model, is developed.

The alternating stresses at these locations are then calculated using the six load combinations and the limitingvalues recorded.

Since these results are developed over a reduced frequency

interval, 60-130Hz, it isnecessary to adjust them to be representative of the entire 0-250 Hz range. This is done byevaluating the stresses in the original global model (i.e., without the U-section stiffener) at thesame locations and comparing the stress estimates obtained when using the full interval vs. usingonly the 60-130 Hz interval.

Specifically, if S(fl,f2) is the stress intensity calculated using onlythe unit solutions in the frequency

interval, fl to f2, then one can calculate an adjustment factor:f= S(60,130)

(B. 1)S(0,250)This factor, calculated separately at each individual node and load condition using the originalglobal model, is then used to adjust the stresses at the same location computed in the model withthe U-section stiffener.

Thus stresses are obtained by dividing by f and stress ratios viamultiplication.

Typically f is near unity (0.98, with occasional values as lows as 0.9) since mostof the stress contribution originates in the 60-130 Hz range.133 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Where appropriate, the stress reduction factors developed for embedded model 6 in Section4.5 (SRF=0.80 and 0.66, depending on the particular weld) are also applied.

The final limitingstress ratios for each load combination are reproduced in Table 19.Table 19. Limiting alternating stress ratios for nodes located on or near the common junctionbetween the tops of the inner vane banks, the side plates and vane bank end plates. Stress ratiosare calculated for the six load combinations defined in Table 16.Load Limiting alternating stress ratiosN55 3.41N59 3.43N58 2.49N61 3.28N62 3.31N63 2.39Evaluation of Stresses at End of BeamThe installation of the U-section stiffeners has the potential of increasing stresses locally.Therefore a real time stress analysis is also performed for a collection of nodes located about theends of the U-section stiffeners where the stresses are most likely to be highest.

The stressevaluation is performed using the most limiting load case, outN63 in Table 16, (this loadgenerally gives rise to the highest alternating stresses and also has the highest biases anduncertainties applied).

The limiting alternating stress ratio (taken over all frequency shifts) aboutthe end of the U-channel is SR-a=7.69.

Since only part of the complete frequency range wasconsidered, this estimate is not conservative.

However, from the original global model the 60-130 Hz frequency range contributes approximately 85% of the total stress. Multiplying the stressratio by 0.85 leads to a good estimate of the full stress ratio as SR-a=7.69x0.85=6.54 which isstill well above the recommended margin. The conclusion is that stresses at the end of the U-channel are very low.134 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information I-. 29.00r 4.63 4.63 7.50 -4.63-- 4.63 ---2.005.00.50.65NOTE: CENTERLINE OF CHANNEL LOCATED13.5 INCHES BELOW TOP EDGE OF PLATECRAWN --Marc J. SibilaYAM ' -----3IM2014Continuum

Dynamics, Inc.APPOVEDTChannel Stiffener SI.1 CM NO. OWG NO. REVSCALE SMEETFigure 24a. Drawing of the U-section stiffener

-as designed and modeled in the FEA.135 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information I I STY I NOTE PART TiTUS WACWPTWN I MArItAL I ~CNOTES:1. CENTERLINE OF CHANNEL LOCATED 13.5 INCHES BELOW TOP EDGE OF PLATE.2. BREAK ALL SHARP EDGES AND CORNERS.3. FOR ADDITIONAL FABRICAT1ON PROCESS REOUIREMENTS.

SEE WESTINGHOUSE SPECIFICATION FS-BWR-ENG-20AI003.

4. FINAL SURFACES TO BE ELECTROPOUSHED AND EVALUATED PER APPENDIX COF FS-BWR-ENG-20AI003.

5. THIS DRAWING IS MADE FROM CONTINUUM DYNAMICS DESIGN DRAWING T1718-001.

DaC/ \IpflLt-r4L, t a mi~m~~ n, qataISECTION A-AIAýBDETAIL C-CL 1BaBISOMETRIC VIEW-----0.81TYPDETAIL Bý :1pIRAEI.ECETAMA4STCE4OM4TTISTTTTAIA 1] ................

II -.4-- ~A*.II...............-.

I -~ I 4444.44444 RI'AA I WESTTTTAICATAPITTPITETAR'IALAAT A0& I-NINI-~~~ ZTSIT M T T.4r~iTI-- A O seNINE IALE POINTNUCLEAR STATION 2CHANNEL STIFFENER I D.05D78: I II41 2 1Figure 24b. Drawing of the U-section stiffener

-as built and installed on the NMP stream dryer.136 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 25a. Finite element mesh for the steam dryer FEA model with the U-section stiffener included.

View from outside of the steam dryer.137 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information Figure 25b. Interior and top views of the U-section stiffener installed on the inward face of theinner side plate.138 This Document Does Not Contain Continuum

Dynamics, Inc. Proprietary Information U-Channel Center Line StressesIn order to design the bolts attaching the U-channel shaped stiffener to the inner side plate,the local stresses in the inner side plate are calculated.

Specifically, the stresses are extracted from a line segment that coincides with the vertical line of symmetry of the inner side plate andextends from the bottom to top of the U-section stiffener, or 11" to 16" from the top edge of theside plate (10" to 15" in the modeled inner side plate). The load used to calculate these stressesis the outN58 signal in Table 16 corresponding to the RCIC valve closed condition.

This loadselection is conservative for the ACM 4.1R-based cases since it generally produces higherstresses than the baseline EPU load (outN55) or the condition when the drain trap is out ofservice (outN59).

A slightly higher load occurs when the higher of the bias and uncertainty estimates from the ACM 4.1 and ACM 4.1R over each frequency interval is used (outN63).

Stress calculations using this signal produce stresses at the high stress location that are less than5% higher than obtained with outN58.The maximum alternating stress intensity on this center line is determined to be 651psi andoccurs on the inner side plate. This value is further increased by dividing by 0.85 whichestimates the effects of using more conservative load (outN63 rather than outN58 -this accountsfor an approximate 5% increase as explained above) and extending the frequency interval fromthe 60-130 Hz range over which unit solutions are available to the full 0-250Hz frequency range.This range extension is conservatively estimated to increase stresses by 10% based oncomparisons of full and partial range calculations carried out using the global model to producethe adjustment

factors, f, in (B.1). Using this 0.85 scaling the maximum stress is estimated as766 psi. This is the value used to size the U-section stiffener bolts in [25].139

v t e Letter Sequence Request
TAC:MF4559, (Open)
Initiation Request, Request, Request Acceptance... Supplement Administration Questions 13 August 2014 3 February 2015 5 August 2015 Answers Results Approval... Other:
MONTHYEAR2014-08-13 [Table View]

ML15022A661: Difference between revisions

Revision as of 06:16, 1 July 2018

Text

Navigation menu

ML15022A661: Difference between revisions

Revision as of 06:16, 1 July 2018

Text

Navigation menu

Search