BSEP 12-0133, Document No. 51-9195058-000, Areva Input to Process Energy: Follow-Up to SAFLIM3D LAR Request for Additional Information 3, Enclosure 3 to BSEP 12-0133

Document No. 51-9195058-000, Areva Input to Process Energy: Follow-Up to SAFLIM3D LAR Request for Additional Information 3, Enclosure 3 to BSEP 12-0133

ML12348A016
Person / Time
Site:	Brunswick
Issue date:	11/27/2012
From:	Arimescu V AREVA NP
To:	Office of Nuclear Reactor Regulation
References
20004-018, BSEP 12-0133 51-9195058-000
Download: ML12348A016 (13)

Text

BSEP 12-0133 Enclosure 3 AREVA Document Number 51-9195058-000, AREVA Input to ProgressEnergy: Follow-Up to SAFLIM3D LAR Request for Additional Information (RAI) 3 (Non-ProprietaryVersion)

A 20004-018 (10/18/2010)

AREVA AREVA NP Inc.

Engineering Information Record (EIR)

Document No.: 51 - 9195058 - 000 AREVA Input to Progress Energy: Follow-Up to SAFLIM3D LAR Request for Additional Information (RAI) 3 (Non-Proprietary Version)

Page 1 of 12

A 20004-018 (10/18/2010)

Document No.: 51-9195058-000 ARE VA AREVA Input to Progress Energy: Follow-Up to SAFLIM3D LAR Request for Additional Information (RAI) 3 (Non-Proprietary Version)

Safety Related? YES FI NO Does this document contain assumptions requiring verification? [] YES NO Does this document contain Customer Required Format? [i] YES NO Signature Block P/LP, RILR, PageslSections Name and A/A-CRF, Prepared/Reviewed/

Title/Discipline Signature AIA-CRI Date Approved or Comments V. loan Arimescu P Engineer FDM-AR ________

V. Korolev R Engineer /1./2 7//

FDM-AR Jeffrey C. Morris A*A-CRI Manager FDM-AR _

A. B. Meginnis Manager */g'--* .

Product Licensing 1/4 Note: P/LP designates Preparer (P), Lead Preparer (LP)

R/LR designates Reviewer (R), Lead Reviewer (LR)

ANA-CRF designates Approver (A), Approver of Customer Requested Format (A-CRF)

A/A-CRI designates Approver (A), Approver - Confirming Reviewer Independence (A-CRI)

Page 2

A 20004-018 (10/18/2010)

Document No.: 51-9195058-000 AREVA AREVA Input to Progress Energy: Follow-Up to SAFLIM3D LAR Request for Additional Information (RAI) 3 (Non-Proprietary Version)

Record of Revision Revision PageslSectionsl No. Paragraphs Changed Brief Description I Change Authorization 000 All This is a new document.

This is the Non-Proprietary version of this document. The Proprietary version is 51-9150440-000.

t +

i +

t +

t -I-

• +

t 1-t -t I +

I +

• +

i +

i +

i +

& .1.

Page 3

A Document No.: 51-9195058-000 AR EVA AREVA Input to Progress Energy: Follow-Up to SAFLIM3D LAR Request for Additional Information (RAI) 3 (Non-Proprietary Version)

Table of Contents Page SIG NATURE BLOCK ................................................................................................................................ 2 RECORD O F REVISION .......................................................................................................................... 3 LIST OF TABLES ..................................................................................................................................... 5 LIST OF FIG URES ................................................................................................................................... 6 NRC FOLLOW -UP QUESTION: ..................................................................................................... 7

1.0 REFERENCES

............................................................................................................................ 12 Page 4

A Document No.: 51-9195058-000 AREVA AREVA Input to Progress Energy: Follow-Up to SAFLIM3D LAR Request for Additional Information (RAI) 3 (Non-Proprietary Version)

List of Tables Page THERE ARE NO TABLES IN THIS DOCUMENT.

Page 5

A Document No.: 51-9195058-000 AR EVA AREVA Input to Progress Energy: Follow-Up to SAFLIM3D LAR Request for Additional Information (RAI) 3 (Non-Proprietary Version)

List of Figures Page FIGURE 1: CHANNEL GROWTH DATABASE AND BEST-ESTIMATE NONLINEAR GROWTH M O D E L ................................................................................................................................... 8 FIGURE 2: BEST-ESTIMATE CHANNEL BOW MODEL COMPARISON TO AREVA C- AND D-LATTICE CHANNEL BOW DATABASE ........................................................................ 10 FIGURE 3: CHANNEL BOW MODEL UNCERTAINTY VS. FAST FLUENCE GRADIENT ............... 11 Page 6

A Document No.: 51-9195058-000 AR EVA AREVA Input to Progress Energy: Follow-Up to SAFLIM3D LAR Request for Additional Information (RAI) 3 (Non-Proprietary Version)

NRC FOLLOW-UP QUESTION:

A small number of predicted fuel channel fluence gradient values exceed the empirical database for the damage model that is used to predict channel bow and associated uncertainty. The RAI response (RAI 3, Page 5 of Enclosure 1, 8/29 letter)indicates why these values were predicted to exceed the empiricaldatabase,but did not justify the analytic treatment of these values in light of the fact that the empiricaldatabase does not bound theirpredictedfluence gradients,other than to indicate that these constitute a small portion of the overall analyzed population of fuel channels. Please provide this justification.

Response

The channel bow model used in SAFLIM3D was approved as part of the channel bow methodology associated with the "Realistic Thermal-Mechanical Fuel Rod Methodology for Boiling Water Reactors,"

BAW-10247PA. It is a mechanistic model, which calculates channel bow based on the fast fluences of the four sides of the channel. A brief summary of the model and the associated experimental datasets is presented below.

Channel bow is caused by the differential axial elongation of opposite channel sides. The axial elongation is the result of stress-free irradiation growth, which is dependent mainly on fast fluence.

Therefore, the basis of the channel bow model is the stress-free irradiation growth of either Zry-2 or Zry-4 in the recrystallized (RXA) metallurgical condition.

To that end, an up-to-date nonlinear model was developed for RXA stress-free irradiation growth. The formulation provided in Reference [1], which consists of an initial transient followed by a second stage of asymptotically saturating irradiation growth, superposed to an accelerating growth after an incubation fast fluence that has a power dependence on fast fluence with an exponent of [ ], was adopted.

The nonlinear model is physically based on the following processes:

• The initial transient growth is the result of interstitial a-loops, which are saturated after a relatively low fast fluence.

• The formation of the c-component vacancy dislocation loops on basal planes is generally accepted as the mechanism responsible for accelerated irradiation growth; it allows interstitials to form a-loops on the prismatic planes and cause axial expansion.

• The initiation of accelerated growth is correlated with solute dispersion into the matrix, which is necessary in order to promote c-loop dislocation formation; therefore, an incubation time is needed to trigger the accelerated growth (also called breakaway growth in some papers).

The relationship describing irradiation growth over the whole range of fast fluence and temperatures is as follows:

Page 7

A Document No.: 51-9195058-000 AREVA AREVA Input to Progress Energy: Follow-Up to SAFLIM3D LAR Request for Additional Information (RAI) 3 (Non-Proprietary Version)

I I (1)

where,

= fast fluence E+21 n/cm 2, E> 1 MeV Eg = irradiation growth axial strain (texture assumed similar for all RXA materials),

[%]

A, B, C, D, E = model parameters The model was initially developed with AREVA [ ] data covering a broad range of fast fluence and two temperatures, thus fully representative of the operating domain in BWR reactors. The model was then applied to the historical database of channel growth with only minor fitting needed.

The result of the final validation of the channel growth model is presented in Figure 1, which also includes the fast fluence ranges for the channel bow database and for Brunswick reactors equilibrium and cycle-specific core designs. It can be observed that the fast fluence range of the database greatly exceeds the Brunswick and contemporary BWR application ranges; the extent of the database fast fluence domain is larger because, in the past, channels have been re-used and thus have been exposed to very high fast fluences (due to currently approved channel exposure limits, the high fast fluences of past re-used channels are no longer achievable).

Figure 1: Channel Growth Database and Best-estimate Nonlinear Growth Model Page 8

A Document No.: 51-9195058-000 AR EVA AREVA Input to Progress Energy: Follow-Up to SAFLIM3D LAR Request for Additional Information (RAI) 3 (Non-Proprietary Version)

The channel elongation calculated according to [

where,

[

This mechanistic channel bow model, consisting of Equations (1) and (2), was benchmarked on the AREVA measured channel bow database, which was obtained by poolside dimensional measurements of channels irradiated in [ ]. The model was demonstrated to be best-estimate, as illustrated in Figure 2. The discrepancy between calculations and measurements is mainly attributed to [

Page 9

A Document No.: 51-9195058-000 AREVA AREVA Input to Progress Energy: Follow-Up to SAFLIM3D LAR Request for Additional Information (RAI) 3 (Non-Proprietary Version)

I I

Figure 2: Best-estimate Channel Bow Model Comparison to AREVA C- and D-lattice Channel Bow Database No credit can be taken for these sources of uncertainty, and the channel bow model uncertainty was quantified based on the benchmarking of the channel bow database, including, therefore, these uncertainties.

The results of the channel bow model uncertainty analysis are illustrated in Figure 3. It can be observed that the (calculated - measured) values lie within a practically constant scatter band, with no noticeable trend with fast fluence gradient. In fact, the discrepancy between calculations and measurements is reduced for large fast fluence gradients [ ]. The main reason for this is that the low fast fluence gradient bow data are more affected by [

Page 10

A Document No.: 51-9195058-000 AREVA AREVA Input to Progress Energy: Follow-Up to SAFLIM3D LAR Request for Additional Information (RAI) 3 (Non-Proprietary Version)

[

I Figure 3: Channel Bow Model Uncertainty vs. Fast Fluence Gradient In conclusion, the channel bow model employed in SAFLIM3D is mechanistic and is based on the channel growth model that is validated on the AREVA channel growth database, whose fast fluence domain greatly exceeds the Brunswick and contemporary BWR application range. The channel bow model uncertainty was quantified based on the comprehensive AREVA channel bow database, which shows that uncertainty does not increase with larger fluence gradients.

Therefore, it is acceptable to apply the channel bow model uncertainty to channels that have a calculated fast fluence gradient outside the range of the bow database because:

• The predicted growth causing bow is a function of fast fluence, not fast fluence gradient.

• The channel bow model is primarily a mechanistic model based on the geometry of a channel with opposite sides having different growth.

• The growth model includes channel growth measurements that bound the application range of fast fluence.

• The channel bow model uncertainty is set conservatively based on low fast fluence gradients where uncertainty is high due to [ ], and data show that uncertainty decreases with increasing fast fluence gradient.

Page 11

A Document No.: 51-9195058-000 AREVA AREVA Input to Progress Energy: Follow-Up to SAFLIM3D LAR Request for Additional Information (RAI) 3 (Non-Proprietary Version)

1.0 REFERENCES

1. M. Griffiths, et al., "Accelerated Irradiation Growth of Zirconium Alloys," ASTM STP1023, p. 658, 1989 Page 12