ML21111A165
| ML21111A165 | |
| Person / Time | |
|---|---|
| Issue date: | 04/20/2021 |
| From: | NRC/OCIO |
| To: | |
| Shared Package | |
| ML21111A163 | List: |
| References | |
| FOIA, NRC-2020-000314 | |
| Download: ML21111A165 (175) | |
Text
From:
To:
Cc:
Subject:
Date:
Attachments:
Vince.
Rollins Jesse Gaddy. Vincent Groom Jeremy RA Response I RIV-2019-A-0065 ***SPISWf§ II LliiPilill PQPiililt,~0 Tuesday, May 12, 2020 3:03:18 PM NRC"s response to Enclosure 4-NPPD Drawings and Enclosure 1-NPPD Calculation 16-003 Rev 0.
will be addressed in the Final Response. Memo-License Response Review was provided in the interim response (p235).
19065 Enclosure 3-MPR Report 0315-0084-RPI*00l Rev O:Indeoendent Review of Underlvina Structural steel for Turbine Buildina Blowout Piloels odf 19065 Enclosure 1-NPPP Drawinas.odf 19065 Enclosure 5-L Pl Ref A16254-LR-OOJ Rev P-Resoonse to NRC Ouestjons for NPPP Calculation No J6-lm.Qdf 19065 Enclosure 6-NPPP Enoineeana Evaluation n-011 Revision 3-Jurt>lne Buildlna BJowout Panels-Metal wan System odf 19065 Inyestjaaaon Reoort odf 19065 Sianed Cover Letter odf 19065 Enclosure 1-NPPP Calculation 16-003 Rev o-Structural Evaluation af the Turbine Buildioa Blowout Panel Steet Suooorts.odf 19065 Enclosure 2-LPI Reoort Ret A16254*LR-002 Rev 0-Resoonse to NPPD Supplied Ouestioos for Calculation No.16-003.pdf Memo - Licensee Response Review.doc jmaae0Q3 pna Attached is the licensee's response to the CNS Turbine Building siding allegation. ACES will reach out soon to coordinate a review process for this one that includes HQ SMEs or some other form of independent review.
V/r Jesse M. Rollins Senior Allegation Coordinator RIV/ORA/ACES 817-200-1431
LS2020018 Page l of l Enclo ure 3 MPR Report 0315-0084-RPT-001, Rev. 0, "Independent Review of Underlying tructural teel for Turbine Building Blowout Panels' (53 pages)
~ MPR 0315-0084-RPT-001 Revision 0 Independent Review of Underlying Structural Steel for Turbine Building Blowout Panels Prepared for: Cooper Nuclear Station Preparer:
Travis Brown
~
E-signed by: Travis Brown
~
on 2020-04-13 20:05:15 Reviewer:
Craig B. Swanner
~
E-signed by: Craig B. Swanner
~
on 2020-04-13 20:11:52 Approver:
David Bergquist
~
E-signed by: David Bergquist
~
on 2020-04-13 20:13:48 QA Statement of Compliance This document has been prepared, reviewed, and approved in accordance with the Quality Assurance requirements of the MPR Standard Quality Program.
Created: 2020-04-13 20:05:15 Project-Task No. 03152001-0084 MPR Associates, Inc.
320 King St.
- Alexandria, VA 22314 (703) 519-02 00
- www.mpr.com
~ MPR Independent Review of Underlying Structural Steel for Turbine Building Blowout Panels RECORD OF REVISIONS Revision Pages /Sections Revision Description Number Revised 0
All Initial Issue 0315-0084-RPT-001, Revision 0 Page 2 of 53
Table of Contents 1.0 Introduction......................................................................................................... 5 1.1.
Background.............................................................................................................. 5 1.2.
Purpose..................................................................................................................... 5 2.0 Summary and Conclusions............................................................................... 7 2.1.
Overall Assessment.................................................................................................. 7 2.2.
Assessment of Specific Issues................................................................................. 8 2.3.
Recomn1endations.................................................................................................... 8 3.0 Overall Assessment of NEDC 16-003.............................................................. 11
- 3. 1.
General Assessment............................................................................................... 11 3.2.
Load Path Asscss1nent........................................................................................... 13 4.0 Evaluation of Detailed Issues.......................................................................... 21 4.1.
Issue I - Bolt Strength........................................................................................... 21 4.2.
Issue 2 - Failure Sequence and Load Sharing....................................................... 23 4.3.
Issue 3 - Effect of Sag Rods.................................................................................. 26 4.4.
Issue 4 - Friction Effect on Bolted Joint............................................................... 32 4.5.
Issue 5 - Required of Blowout Area...................................................................... 33 4.6.
Issue 6 - Effect of Columns on Blowout Area...................................................... 34 4.7.
Issue 7 - Load Application and Effect ofFull Wall.............................................. 38 5.0 References........................................................................................................ 43 A
Issues Identified............................................................................................... 45 B
Calculation of Required Blow Out Flow Area................................................. 50 Tables Table 2-1.
Table 3-1.
Table 4-1.
Table 4-2.
Issue SUJnmary....................................................................................................... 10 Bolt Evaluation...................................................................................................... 20 Location rnformation for Wall Load Distribution Evaluation...............................40 Location Information for Wall Load Distribution Evaluation...............................40 0315-0084-RPT-001, Revision 0 Page 3 of 53
Figures Figure 1-l.
Figure 3-1.
Figure 3-2.
Figure 3-3.
Figure 3-4.
Figure 3-5.
Figure 3-6.
Figure 4-1.
Figure 4-2.
Figure 4-3.
Figure 4-4.
Figure 4-5.
Figure 4-6.
Figure 4-7.
Figure 4-8.
Figure 4-9.
Figure 4-10.
Figure 4-l l.
General Overview of Turbine Building Blowout Panels......................................... 6 Model to Assess Clip to Column Welds................................................................ 15 Predicted Plastic Strain.......................................................................................... 15 Ratio to Strain Limit.............................................................................................. 16 Model to Assess Bolts............................................................................................ 18 Deflection Results for High Friction Case............................................................. 19 Stress Results for High Friction Case.................................................................... 20 Example of Girt Connection to Column................................................................ 24 South Wall Section Showing Sag Rods................................................................. 27 Single Girt Geometry for Sag Rod Evaluation...................................................... 29 Multiple Girt Geometry for Sag Rod Evaluation................................................... 30 Single Girt Displaced Shape for Sag Rod Evaluation........................................... 3 l Multiple Girt Displaced Shape for Sag Rod Evaluation........................................ 31 North Wall Predicted Blowout Area...................................................................... 37 South Wall Predicted Blowout Arca...................................................................... 38 Model for Wall Load Distribution Evaluation....................................................... 41 Locations Considered for Wall Load Distribution Evaluation..............................41 Displaced Shape for Wall Load Distribution Eval................................................ 42 0315-0084-RPT-001, Revision 0 Page 4 of 53
1.0 Introduction 1.1.
Background
A portion of the north and south Turbine Building walls at Cooper uclear Station (CNS) arc designed as blowout panels. Per C S lJFSAR Amendment 25 (Reference I), these panels are designed to fail prior to the pressure inside the turbine building reaching 0.5 psig after a high energy line break (HELB).
Previous evaluations have shown that the panels and their attachment to Lhc slructural steel arc not limiting. Because of this, Nebraska Public Power District (NPPD) Calculation NEDC 16-003 (Reference 2) was performed lo demonstrate that the underlying turbine building structure will fail if a 0.5 psig pressure load is applied to the blowout panels. The turbine building blowout panels are secured to the toe of horizontal C-channel girts, which span the length between turbine building columns (see Figure 1-1). Each girt end is typically connected to the turbine building column flanges via an angle iron clip. The girl is bolted to the angle iron clip, and the angle iron clip is lap welded to the column via two fillet welds. NEDC 16-003 concludes that these connections in a subset of blowout panels will fail when subjected to a 0.5-psig internal pressure load from the HELB event.
Recently, NPPD received an anonymous allegation that raises seven potential issues with the calculation, including several of the assumptions and the conclusions ofNEDC 16-003. NPPD requested that MPR perform an independent review of NEDC 16-003 specifically focusing on the merit, or lack thereof, of the issues raised in the allegation.
1.2.
Purpose This report documents MPR 's independent review of the evaluation of the turbine building blowout panels' stnictural suppo11s documented in EDC 16-003. MPR's independent review included our assessment of the technical approach and the overall validity of the evaluation's conclusions. This report also provides MPR's assessment of each of the seven specific allegations raised against EDC 16-003. The allegations are documented in Appendix A.
0315-0084-RPT-001, Revision 0 Page 5 of 53
TB column Blowout panels Figure 1-1.
0315-0084-RPT-001, Revision O General Overview of Turbine Building Blowout Panels Sag rods Page 6 of 53
2.0 Summary and Conclusions
- 2. 1.
Overall Assessment MPR performed an independent review of EDC 16-003 focusing on the technical approach and overall validity of the report's conclusions. As part of our review, MPR developed a number of scoping evaluations using ANSYS to test key assumptions and behaviors of the underlying steel support structure for the turbine building blowout panels. Note that MPR only considered the same subset of blowout panels as those determined in EDC 16-003 to be most susceptible to failure (Girt Types 133A, 133B, and l 33C).
lt should be noted that standard design codes arc intended to build in margin to faihuc in a design and therefore, the results are not gcn,crally intended for calculating accurate actual failure loads. This was recognized in NEDC 16-003, and MPR considers the overall NEDC 16-003 methodology to be appropriate for estimating failure limits. 1n general, assumptions and analytical techniques are consistent with the physics of the problem and current state-of-the-art.
Structural engineering code guidance, which would build margin into the load or stress limits, is appropriately not followed. MPR notes that the 0.5 psig HELB internal pressure load applied to the structure results in stresses for many of the connections that are well beyond the limits of design codes.
The key components in the strnctural load path are the girt-to-clip bolts and the fillet welds that connect the angle clip to the tw*binc building column. Based on our review of the structural behavior, if either one of these connections fails, then the underlying support structure for a blowout panel will fail. To assess each component, MPR performed a series of three dimensional finite clement models of the bolted and welded connections including pla tieity with actual material properties. MPR's assessment of each of these connections is provided below:
Girt to Clip Bolts Analyses were performed with (a) high friction and high preload and (b) low friction and low preload. In both cases, MPR's model demonstrated that the joint slips through the clearance between the bolt and the boll holes such that the bolts begin carrying load directly through shear. The slipping is caused by prying action and a high shear loading. The model predicts that the combination of shear and tensile loading with the moments that result from the prying action of the clip on the girt significantly exceeds the failure criteria. Therefore, MPR concludes that the girt to clip bolts are likely to fail upon a 0.5 psig pressure loading. This is consistent with the conclusions of NEDC I 6-003.
Clip to Column Welds - This model considered that the bolt remained intact and transferred all the load through their connection without slipping. Results demonstrated high plastic strains in the weld well beyond the material's failure strain. Other effects including sag rods, potential for over-sized welds, bolt slippage were evaluated individually. The strains are sufficiently high in the base case that these other effects are not expected to impact the conclusion that the welds will fail. Therefore, MPR concludes that the welds would likely fail due to the 0.5 psig pressure loading. This is consistent with the conclusions ofNEDC 16-003.
0315-0084-RPT-001, Revision 0 Page 7 of 53
MPR concludes that at least one component in the load path of the turbine building blowout panels w1dcrlying structural steel will fail for the Girt Types 133A, 133B, and 133C. Therefore, MPR agrees with the overall conclusions ofNEDC 16-003 and concludes that the specified blowout panels will fail if subjected to a pressure loading of0.5 psig. However, it is noted that MPR estimated the required blowout area to maintain a 0.5 psig pressure is approximately half of the area predicted lo blowout in NEDC I 6-003. Therefore, MPR concludes that there is even more margin than credited in NEDC 16-003 for uncertainty in the calculation inputs and assumptions. See Section 3.0 for additional details on MPR's overall assessment of NEDC 16-003.
2.2.
Assessment of Specific Issues A summary of MPR 's assessment of each of the individual issues related to NEDC 16-003 is provided in Table 2-1. MPR finds that Issues 2, 5, 6, and 7 are either not technical concerns or are already adequately addressed in NEDC 16-003.
MPR notes that Issue 5 (required blowout area) was not addressed in NEDC 16-003. To address the issue, MPR estimated the required area to maintain the internal pressure from the limiting HELB break at or below 0.5 psig to be approximately 2300 ft2. NEDC 16-003 predicts that turbine panels with an area equal to about 5400 fi2 will fail and blowout. Thus, there is considerable margin between the predicted !blowout area and the required blowout area. This margin provides added confidence that the potential input uncertainties (e.g., material limits) and non-conservatisms in NEDC 16-003 will not impact the conclusion that the turbine building blowout panels will satisfy their design basis function after a HELB event.
rssues I, 3, and 4 identify non-conservatisms in the NEDC 16-003 calculation approach. On aggregate, MPR believes that these non-conservatisms are balanced by other factors not credited in NEDC 16-003 (e.g., dynamic effects of the fast pressure load application are neglected) and the quant.ified margin beyond failure. Further, MPR evaluated each of these effects and it is our opinion that these non-conservatisms will not impact our overall conclusions identified in Section 2.J (i.e., at least one portion of the girt connection load path will fail under the 0.5 psig internal pressure loading from a HELB event).
See Section 4.0 for additional details on MPR 's individual assessments of each issue.
2.3.
Recommendations As part of this assessment, M PR offers the following recommendation to support closure of these issues:
All calculations performed in this assessment are scoping evaluations which are meant to determine how a given effect influences the evaluation conclusion. These results have been reviewed but have not been through MPR's complete quality assurance process for typical calculations. The turbine building blowout panel required area to relieve a 0.5 psig pressure afier a HELB event is a key result which provides added confLdence in the NEDC 16-003 conclusions. MPR recommends that NPPD perform a formal 0315-0084-RPT-001, Revision 0 Page 8 of 53
safety-related calculation to determine and document the required turbine building blowout panel area.
As mentioned above, the calculations perfonned in support of this report are scoping in nature and cannot be used as formal design input. However, they support the conclusions of an existing safety-related calculation NEDC 16-003. lfNPPD requires that evaluations performed to support this report be developed into formal calculations MPR recommends the following to support that effort:
Several non-conservatism in NEDC 16-003 were identified through the anonymous allegations raised and through the course of developing this report. The non-conservatisms arc balanced to some extent by other conservatisms in the document.
MPR recommends that NPPD perform an evaluation which addresses all non-conservatisms related to NEDC 16-003 identified in the anonymous allegations and those identified by MPR in Section 3.0 in a single representative model. As appropriate and practical, conservatisms in the NEDC 16-003 methodology should also be removed.
There is also some uncertainty in the inputs used, which impacts both conservatisms and non-conservatisms in NEDC 16-003. If PPD desires to reduce the uncertai.nty in the analysis inputs, MPR recommends performing the following activities on a sampling basis:
I.
Assess weld configurations to identify weld quality, weld size, and length of weld.
- 2.
Perform bolt testing of installed bolts to determine the bolting material properties.
lt may be possible to perform hardness testing of installed bolts, or several bolts could be removed for testing. If bolts will be removed, consider both tension and shear testing.
- 3.
Measure the actual location of the attachment screws that connect the blowout panels to the girts to determine the location on the girt where the blowout panel will act.
0315-0084-RPT-001, Revision 0 Page 9 of 53
Table 2-1.
Issue Summary Issue MPR Conclusion NEDC 16-003 assumptions regarding bolt strength may potentially be non-conservative. While MPR's opinion is that the assumptions made in NEDC 16-003 to determine the bolt ultimate tensile strength (UTS) are reasonable, there is evidence that supports a higher bolt ultimate tensile 1
Bolt Strength strength by as much as 20%. While, there is uncertainty in the equivalence of the material properties of the bolls tested and those installed in the building, the magnitude of this uncertainty would be bounded by the 20% above, which would put the bolts in the upper range of possible bolt strengths for A307. These uncertainties were considered when making our overall assessment that the bolts will fail.
MPR concludes that the overall failure sequence and the load sharing Failure Sequence and documented in NEDC 16-003 between the girt-to-clip connection bolts 2
and the clip lo column welds is based on the actual physics of the Load Sharing problem and is appropriate for use in the evaluation. See conclusions for Issues #3 and #4 which specifically address the sag rods and bolt friction.
NEDC 16-003 neglects the effect of the sag rods. The sag rods would reduce girt twisting and loading on the girt connections, so this assumption is therefore non-conservative. Scoping analyses performed 3
Effect of Sag Rods by MPR show that this reduces the loading on the girt connections by as much as 10% for each weld and 30% for the bolts. The margins beyond failure in NEDC 16-003 would be reduced by this amount, but would still result in a failure. This effect is considered in MPR's overall assessment.
NEDC 16-003 does not address preload or the friction that could develop between the girt and the clip if the bolted connection were preloaded.
Crediting friction results in load sharing between the bolts and friction in the joints. Load sharing reduces the bolt loads and is non-conservative to Friction Effect in neglect when evaluating failure. Scoping analyses performed by MPR 4
Bolted Joint demonstrate that for the largest friction coefficient and the bolts preloaded to two-thirds yield strength, the girt to clip joint is expected to slip through the clearances such that the bolts will be loaded in shear through contact with the clip and girt. These analyses also confirm that the forces and moments applied to the limiting bolt are beyond its failure capacity based on actual material properties. See Section 3.2.2 for additional discussion.
NEDC 16-003 does not address this issue. Scoping calculations Required Blowout performed by MPR estimate that an open area of only about 2300 ft2 is 5
required to maintain the turbine building general pressure to 0.5 psig or Area below during a HELB event. This is less than half of the predicted blowout area from NEDC 16-003 (5400 ft2).
Evaluations performed by MPR demonstrate that the width of the 6
Effect of Columns on columns does not significantly impact the available flow area. This result Blowout Area combined with the conclusions to Issue #5 led MPR to conclude that available flow area is not a technical concern.
Scoping evaluations performed by MPR of the entire wall showed a Load Application and negligible difference in the load sharing between girts when compared to 7
Effect of Full Wall the simplified single girt models. Therefore, MPR concludes that the tributary method used to apply the pressure load in NEDC 16-003 is acceptable for calculating loads on a single girt.
0315-0084-RPT-001, Revision 0 Page 10 of 53
3.0 Overall Assessment of NEDC 16-003 MPR performed an independent review of NEDC 16-003 focusing on the technical approach and overall appropriateness of the report's conclusions. Our general assessment is documented in Section 3.1.
The key components in the stmctural load path arc the girt-to-clip bolts and the fi llet welds that connect the angle clip to the turbine building column. If either one of these connections fails, then the underlying support structure for a blowout panel will fail. MPR performed scoping evaluations using ANSYS 1 to test key assumptions related to each portion of the load path and assess whether these components would fail under the 0.5 psig HELB event loading. The scoping calculations and our assessment of the NEDC 16-003 conclusions are provided in Section 3.2.
3.1.
General Assessment MPR considers the overall NEDC 16-003 methodology to be appropriate for estimating failure limits. In general, assumptions and analytical techniques are consistent with the physics of the problem and current state-of-the-art. N EDC 16-003 's methodology does not strictly fo I low structural engineering code guidance, which would build margin into the load or stress limits and ma!ke failure prediction potentially non-conservative. MPR notes that the 0.5 psig HELB internal pressure load applied to the structure results in stresses for many of the cotrnections that arc well beyond the limits of design codes.
MPR reviewed the NEDC 16-003 methodology and identified U1e following significant conservatisms:
Dynamic Amplification - After the HELB event, the turbine building pressurizes rapidly.
N EDC 16-003 ignores the dynamic response of the structure and any dynamic amplification from the impulse. This could under-predict the applied pressure load by as much as a factor of two (i.e., the dynamic load experienced by the blowout panels may be twice the magnitude of the building pressure because of the structure's dynamic response to the transient). While the transient is relatively quick, it is not quite an impulse load, so the factor will be less than two, but still will be expected to increase the load experienced by the panels and therefore, the girts.
load Application - NEDC 16-003 applies the HELB load to the girt at its leg's midsection. ln reality, the load would be applied at the location where the screws attach the panels to the girts. Based on photographic evidence, the screws typically appear to be closer to the girt C-channel toe than its midsection. Applying the load at this lower location would increase the twisting moment on the girt thereby increasing the likelihood of failure. If the screws are 1/4-inch lower than the assumed load application location 1 ANSYS Version 2019 RI is a general purpose finite element code commonly used in the nuclear industry to belier understand the behavior of structures.
0315-0084-RPT-001, Revision 0 Page 11 of 53
(which seems reasonable based on photographs), the twisting moment would increase by about 20%.
Load Direction -
EDC 16-003 uses ANSYS models with large deflection effects included. The load is applied gradually and the stiffness matrix is recalculated as the girt twists. The load direction i maintained orthogonal to the girt as it twist. The changing load direction is resisted by the strong axis of the girt. In reality, the load would continue in the horizontal direction parallel to the ground. This would subject tbe girt to a greater percentage of weak axis bending, reduce the structure's capacity, and reduce the pressure that causes the girt to locally buckle.
MPR also notes the following which may potentially be non-conservative regarding the NEDC 16-003 methodology:
Bolt Stress Co11ce11tratio11 Factor - NEDC I 6-003 applies a stress concentration factor to the average bolt stress. The bolt stress is determined by extracting reaction loads from the ANSYS model and dividing by the bolt nominal tensile area. Stress concentration factors are typically applied to determine a peak stress for comparison with fatigue limits.2 Use of a stress concentration factor when postulating a gross section. failure is not typical practice. When assessing bolt failure, MPR did not apply a stress concentration factor.
Bolt Clearance - A significant component of the bolted connection loading is a result of the statically indeterminate problem where a beam is loaded in shear with both ends fixed. In this case, the shear loading causes the beam to flex, and the overall beam length increases. Since both ends are fixed!, this stretching induces an axial load in the beam. In our case, an axial load is induced as the girt is stretched when loaded in sheaT by the internal pressure load. The induced axial load is a significant fraction of the total load on the connections. lf clamping friction in the bolting joint is insufficient, then the joint slips through the clearance between the bolts and the holes in the girt web and clip.
Scoping evaluations indicate about 15% to 20% of the reaction load is relieved if the joint slips. N EDC 16-003 models the bolted joint in a manner that does not allow the joint to slip, which is non-conservative if the joint would slip. MPR notes that the initial joint preload and its current tightness is not known. MPR's scoping evaluations consider both a joint with high friction and high preload and a joint with low friction and low preload to determine if these effects impact the overall conclusions.
Flexibility of Column Flange - NEDC 16-003 assumes that the column flange is rigid relative to the girt. ln general, this is a reasonable assumption. However, any tlcxibility in the column flange will reduce the forces and moments that are reacted through the gi11 connection to the column. We expect this effect is small, but MPR will consider this flexibility when assessing whether each connection will fail.
Weld Size - NEDC 16-003 does not include any allowance for oversized welds. In general, fi llet welds specified to have a leg size of 1/4-inch are made slightly larger to 2 Sec for example Paragraph NB-3222 of the ASME Boiler and Pressure Vessel Code, Section 111, Division I.
0315-0084-RPT-001, Revision 0 Page 12 of 53
ensure that the minimum requirement is satisfied. MPR will consider this when assessing whether the welds will fail.
Bolt Strength - See Section 4.1.
Effect of Sag Rods - See Section 4.3.
Friction Effect on Bolted Joint - See Section 4.4.
3.2.
Load Path Assessment Scoping evaluations using AN SYS are provided for each of the key components in the connection load path.
3.2.1. Clip to Column Welds The 3/8-inch thick angle iron clip is connected to the column by a pair of lapped I /4-inch fillet welds (see Figure 4-1 ). If the girt to angle iron clip bolts do not fail, then all of the load applied to the girt must be reacted through these welds into the column. If these welds fail, then girt support is lost and the girt buckles under the pressure load as demonstrated in NEDC 16-003. To assess whether these welds will fail, MPR developed a scoping finite clement modefl of the welded connection using AN SYS. The model is similar lo the one developed in NEDC 16-003.
Key aspects of the model, including differences from the EDC 16-003 model, are identified below:
Each model includes solid model representations ofhalfofthe girt (Type l33C), the two bolts connecting the girt to the clip, the angle clip, the fillet welds attaching the angle clip lo the turbine building column nangc, and a portion of the turbine building column flange.
The model includes halfofthe girt with a ymmetry condition at its midspan. The column flange is fixed where the flange meets the column web and has a symmetry condition at the upper and lower ends of the column flange portion.
The model is meshed with solid elements, including the entire length of the girt. This is a difference with NEDC 16-003, but it is immaterial. The beam representation for a portion of the NEDC 16-003 model also includes the appropriate beam twisting behavior.
Similar to NEDC I 6-003, the 0.5 psig pressure I IELB load and the deadwcight of the blowout panels is applied as a line load assuming the blowout panel screws attach to the girt at the midpoint of the girt flange.
The load direction of the pressure is unchanging as the girt twists. This is different from NEDC 16-003, removes unnecessary conservatism, and more accurately represents the actual load behavior.
The bolted joint between the girt and the clip is assumed to be tight with no slip.
The model includes large deflection (i.e., the stiffness matrix is adjusted as tbe model deflects) and plasticity. Material properties are consistent with the true stress-strain curve developed in EDC 16-003 which is ba ed on mechanical properties documented in the original material certs.
0315-0084-RPT-001, Revision 0 Page 13 of 53
Unlike NEDC 16-003, the model includes a three dimensional representation of each ft llct weld. This permits an assessment of the predicted strain at each weld to determine if a weld is expected to exceed its elongation limit. MPR believes this is a more accurate way to assess whether the weld will fail.
The effect of the sag rods are neglected. These effects are assessed in Section 4.3.
The model is shown in Figure 3-1. The predicted plastic strains at the weld between the end of the clip and the column are shown in Figure 3-2. As shown, the plastic strains for a portion of the weld are estimated to be above 20% (the minimum elongation for A36 steel). Minimum elongations are determined based on a uniaxial tensile test and only consider tensile strain in one direction. The actual weld has strain in multiple directions, so the failure strain will be less than the uniaxial tensile failure strain.
Tension in multiple directions can be assessed using a parameter called triaxiality factor.
Triaxiality factor is the ratio of the hydrostatic pressure stress (i.e., the average of the three principal stresses) to the equivalent stress (i.e., Von Mises stress) at any location. The triaxiality factor is zero for the case of pure shear, one-third for uniaxial tension, and one for triaxial tension in all three direction. A criterion designed to safeguard against local failure by comparing plastic strains to failure strains using the triaxiality factor has been incorporated into modern editions of Section VIII, Division 2 of the ASME Boiler and Pressure Vessel Code (see Paragraph 5.3.3. 1 of Reference 5).3 Figure 3-3 shows the ratio of the calculated plastic strain to the strain limit per the Section VITI Division 2 of the ASME B&PV Code. Values over one in the figure indicate locations in the weld that will fail. As shown, a significant length of the weld is well beyond the limit and predicted to fai I.
otc that results from this figure ignore any residual plastic strain in the weld. Residual plastic strain is common in welds and would further reduce the failure strain.
Once failure is initiated one of two things will happen:
A flaw will develop in the weld that is unstable and complete brittle failure will occur, or Stresses will re-distribute and the remaining weld with less area wi II have higher stresses and strains and failure will continue to propagate through tearing.
In either case, the welded connection will not be able to support the HELB internal pressure loading. Therefore, MPR concludes that the weld will fail under the 0.5 psig pressure load.
Because of the excessive strains compared to the allowable triaxial strain, MPR considers that these conclusions are still vaJid when considering any of the effects identified in Section 3.1 either separately or concurrently.
1 The basis for this methodology is provided in PVP2009-77122 (Reference 6).
0315-0084-RPT-001, Revision 0 Page 14 of 53
Equivaltnt Plubc Strain Typo: Equovalent Pl11tic Strain Unrt:,n/1n 0.31283 Mtx 0.2 0.175 0,15 0.125 0.1 0.07S 0.05 0.025 OMin Figure 3-1.
Model to Assess Clip to Column Welds Figure 3-2.
Predicted Plastic Strain 0315-0084-RPT-001, Revision 0 Page 15 of 53
Expression: EPPLEQV_RST/(0.3646"2.718282"(-1.n6"'((S1 +S2 +~)/3/Sf.QV')-1/3))
lime: 1 4/11/2020 9:42 AM 7.033 Max 1
0.875 0.75 0.625 0.5 0.375 0.25 0.125 3.2.2. Girt to Clip Bolts Figure 3-3.
Ratio to Strain Limit The girt web is 0.24-inch thick and is connected to the 3/8-inch thick angle iron clip via two 3/4-inch, ASTM A 307 steel bolts with appropriately sized nuts and washers. The through holes in the girt web and angle clip are 13/ I 6-inch diameter. If the fasteners are tightened, the preload in the bolts will provide a clamping force between the girt web and angle clip. Friction developed from the clamping force will transfer some of the load through the interface rather tha111 through the bolts. Thus there is the potential for load sharing between the bolts and friction in the connection. To evaluate this condition, two scoping A SYS models were developed. The models are described below:
Each model included solid model representations of half of the girt (Type I 33C), the two bolts connecting the girt to the clip, the angle clip, and the fillet welds attaching the angle clip to the turbine building column flange.
The bolt and nut are modeled as a single structure. The head dimensions are based on the width across the comers of the hex head/nut, and the shank is assumed to be the nominal bolt diameter.
The models are fixed at the leg of each fillet weld to represent the column.
A symmetric boundary condition is provided at the midspan location of the girt.
0315-0084-RPT-001, Revision 0 Page 16 of 53
Similar to NEDC 16-003, the 0.5 psig pressure HELB load and the deadweight of the blowout panels is applied as a line load where the blowout panel screws attach to the girt at the midpoint of the girt flange.
The models include large deflection (i.e., the stiffness matrix is adjusted as the model deflects).
Material properties are consistent with the true stress-strain curve developed in NEDC 16-003 which is based on mechanical properties documented in the original material certs of the structural members.
The following cases are perfonned:
o High Friction and Preload-One case is performed with a 0.78 friction coef(icicnt underneath the bolts heads and between the girt web and angle clip.
This is representative of an upper bound value of friction between two steel surfaces.4 This case is also perfonned with each bolt preloaded to approximately two-thirds of their minimum yield strength (~I 0,000 lbr per bolt). This is a typically the maximum target pre load value for steel fasteners.5 Results for this load case will maximize the load carried through this joint via friction and minimize the loading through the bolts.
o low Friction and Preload - One case is perfonned with a 0.20 friction coefficient underneath the bolt heads and between the girt web and angle clip. This is representative of a lower bound value of friction between two steel surfaces.6 This case is also perfonned with each bolt preloaded to only 1000 lbr per bolt.
This is representative of a snug tight fastener. Results for this load case will minimize the load carried through this joint via friction and maximize the loading through the bolts. It is noted! that if there is no significant pre load in the as-installed assemblies, then the friction carrying capacity could be less, but this case represents the minimum expected preload conditions.
The model geometry is shown in Figure 3-4. Model results demonstrate that the joint slips in both cases. Plots of the model deflected shape and stresses for the high friction case are shown in Figure 3-5 and Figure 3-6, respectively.
4 This friction coefficient is based on the maximum friction coefficient for carbon steel on carbon steel identified in NEOC 16-003. This upper bound value for unlubricated steel friction coefficient is con istent with MPR 's experience.
s For example, this is the limiting value for reactor vessel head studs (sec ASME Boiler and Pressure Vessel Code, Section 111, Division I, 201 I Edition, Section B-3230) where the limit is two times the stud allowable stress. Per ASME Boiler and Pressure Vessel Code, Section 11. Part D, 2011 Edition, Table 2-!00(c), the allowable stress for Class 1 bolting material is equal to one-third the material yield strength. Therefore, the preload is limited to two-thirds its yield strength.
6 This friction coefficient is a reasonable lower bound value for unlubricated steel friction coefficient and is consistent with MPR's experience.
0315-0084-RPT-001, Revision 0 Page 17 of 53
The limiting bolt reaction loads from the models are provided in Table 3-1. Note that since slip occurs in both cases, the bolt reaction loads arc similar in each case. The bolt shear forces arc generated mostly by the axial load induced in the girt as ir stretches between two fixed points.
The bolt tensile loads and moments are generated by the prying action that occurs due to the girt twisting action and the force balance on the clip.
Stresses in the limiting bolt arc determined from the reaction loads based on section properties detem1ined from the minimum minor diameter. A ratio of shear strength to ultimate strength of 0.6.2 is used based on recommendations and testing for typical structural bolts used in buildings published by the Research Council on Structural Connections (Reference 7). Ultimate strength is determined from the shear test results using the average of the results (excluding the test with a noted fixture issue). Esthnated stresses and interaction ratios are also provided in Table 3-1.
Interaction ratios greater than one predict that failure is likely to occur.
As shuwn, lite bulls an: 11ut prcllktcll tu fail due tu shear a11ll 1cnsilc furl:t:s alu11c. A llt:l:t:ssary prerequisite for bolt failure is the moments resulting from the prying action due to girt twisting and the force balance on the clip. With the moment included, the bolts well exceed the interaction limit by a factor of two or more. Once a single bolt fails, all of the load will be transferred to the other bolt and it will subsequently fail resulting in failure of the girt connection and the blowout panel. MPR 's scoping evaluation results are consistent with the conclusions in NEDC 16-003. Based on the magnitude of the bending load in the bolts and the large interaction ratios, MPR considers that these conclusions are still valid when considering all of the effects ide111tificd in Section 3.1 either separately or concurrently.
Figure 3-4.
Model to Assess Bolts 0315-0084-RPT-001, Revision O Page 18 of 53
Tobi Otfonnltion Type: Tobi oe/onnation Unit:lf'I Tmt: 2 4/912020 1~01 PM l.9191Ml11 7.9287 6.9376 5.9465 4.9555 3.9644 2.9m 1.9822 0.99109 OMln 0.000 7.000(,n) 3.500 Figure 3-5.
Deflection Results for High Friction Case 0315-0084-RPT-001, Revision O Page 19 of 53
EquMlent s.rtn Typt. Equivlltnt (von*Mists) ~u Unit:p,,
Tiffie* l 4/WlalO 11~05 PM Figure 3-6.
Stress Results for High Friction Case Table 3-1.
Bolt Evaluation Reaction Loads Stresses (ksi)
Interaction Ratio Load Case Tensile Shear Moment Tension Tension+
(kips)
(kips)
(in-kips)
Tensile Shear Bending
& Shear Bending &
Shear High Friction 7.9 11.0 3.6 25.5 35.3 148 0.8 2.4 and Preload Low Friction 7.8 7.9 3.0 25.2 25.6 122 0.6 2.0 and Preload 0315-0084-RPT-001, Revision 0 Page 20 of 53
4.0 Evaluation of Detailed Issues The following sections discuss and evaluate each of the seven issues regarding NEDC 16-003 in Appendix A. For each issue, the following is provided:
The comment which raised the issue, MPR's interpretation of the comment, MPR's approach for evaluating the comment, and Results of our evaluation approach and conclusions.
4.1.
Issue 1 - Bolt Strength 4.1.1. Description NEDC 16-003 (Reference 2) uses testing results to estimate the ultimate tensile strength of the girt to angle iron clip bolts. Stress results derived from finite element models are compared to the ultimate tensile strength to assess whether failure will occur. Issue# I is related to the estimated ultimate tensile strength of the girt-to-clip bolts. As documented in Appendix A, the comment is as follows:
The girt bolt ultimate strength and shear strength assumed in Calculation NEDC 16-003 is 11011-co11serl'atil*e and may not be representative of the bolts used i11 the blowout pa11el con.ftruction. Specifically, the calculation improperly uses an a\\lerage ultimate strength instead of maximum tested ultimate strength. Additionally, the average tensile strength is derived from testi11g of warehouse stock bolts which are likely from different het.11 a11d lots than of the bolts used in blowout panel constmction. The average ultimate strength may not bound the ultimate strength of bolts used in construction.
Address the following:
Provide a discussion and evidence of any 1*alidation activities (walkdowns, testing, etc.) pe1fon11ed to ensure that the currently installed girt bolts are of the same grade (ASTM Grade A307) and/or ha11e the same material properties as the bolts that were tested/or Calculation NEDC /6-003.
4.1.2. Evaluation The bolt Ultimate Tensile Strength (UTS) used as a failure criteria in the calculation was derived from testing performed by LPI contemporaneously to the writing of the calculation. Specifically, measured data was obtained from seven bolts from the CNS stock.room inventory. The following testing was perfom,ed on each bolt:
Rockwell B hardness testing; three tests performed on each bolt Double shear testing of each using the ational Aerospace Standard NASM 1312-13 test method 0315-0084-RPT-001, Revision 0 Page 21 of 53
Assumption 5 ofNEDC 16-003 states that the ultimate failure stress criteria was found to be approximately 75.8 ksi based on hardness testing; however, this data was not used as part of the failure criteria. The bolt UTS used in EDC 16-003 was derived from the shear testing.7 NEDC 16-003 used double shear failure testing per NASM 1312-13 to establish the ulli mate failure strength. The double shear failure load was measured and halved to find the single shear failure load. From this the bolt shear section was used to calculate the shear strength. Ultimate tensile strength was then derived by assuming that shear strength was 75% of tensile strength, and the shear area was assumed to be equal to the thread root area ba ed on the external thread minimum minor diameter.
In general, MPR agrees witl1 this methodology. It is similar to methods described in Section 4.2 of Reference 7 to determine criteria for structural bolt failure for use in building codes. The use of the thread root area is conservative and agrees with typical practice.
NEDC 16-003 also presumes that the shear failure of steel is at roughly 75% of the UTS value.
Ultimate shear strength can vary between 58% and 80% of the UTS for steels. The higher values are typically quoted to estimate forces to punch through steel sheet metal. Testing of similar structural steel bolting documented in Reference 7 show that a value of 62% is consistent with the data from several sources. Therefore, the UTS could be about 20% higher than the value used in NEDC 16-003.
The hardness testing data performed as part ofNEDC 16-003 may also be used to estimate the UTS. UTS values derived from hardness testing are about 15% higher than the bolt UTS values used in NEDC 16-003. Hardness testing correlations are well-established and widely used throughout industry. These correlations arc defined in ASTM E 140 (Reference 11). Based on the hardness testing results and the uncertainties described above, MPR concludes that the bolt UTS values used io NEDC 16-003 are likely about 15% to 20% lower than the actual UTS val Illes of the lot tested. Considering the data available, the UTS may be closer to 77 ksi.
The bolts tested and used as the basis for the failure criteria may or may not have been from the same lot as the installed bolts in the turbine building structure. The bolts tested were stamped
'307 A', whereas bolts seen in photographs of the building do not appear to have the marking.
The marking requirement came into practice in 1987, so it is likely that the bolts tested are newer tha!ll those used in the building. Broadly, manufacturing methods have improved over time and the composition of materials have become more controlled over time. Older heats would likely have had their target material prope11ies set higher to ensure they satisfy the minimum specification given the presumed greater variability of the earlier process. The fact bolts of this grade are not heat treated would likely reduce the level of variability. Earlier bolts used in the building would therefore likely be stronger than bolts taken from later stock.
7 MPR also reviewed the yield strength elected for use in NEDC 16-003. The bolt yield strength was assumed to be 33% higher than the minimum specified yield strength ASTM for A36 steel (A307 has no specified yield strength)
(36.3 ksi x 1.33 = 48.28 ksi) citing NUREG/CR-213 7 (Reference 9). M PR assesses this to be conservative and appropriate for use in this evaluation.
0315-0084-RPT-001, Revision 0 Page 22 of 53
Some literature is available of the variability of steel properties. Per Reference 14, in 2003 data was collected for over 3,000 tests on Grade 50 and Grade SOW plates with varying thickness from six different No11h American mills and showed a coefficient of variance of up 4% for tensile strength. Coefficient of variance is the ratio of the standard deviation to the mean. So we can say that 95% of samples lay within +/-8% of the mean. It is possible that a similar difference exists between the bolts tested and those installed in the building. However, Reference 7, Section 4.2 showed that for A325 bolts (similar but higher grade bolt than the A307 bolts), the tested tensile strength was an average of 18% higher than the specified minimum, with a standard deviation of 4.5%. Using this information, a value of 77 ksi would be in the upper 2.5%
for A307 bolts, which have a minimum specified ultimate strength of 60 ksi. Consequently, it is expected that there is not additional variance expected between the lot tested and the bolts installed in the bujJding.
However, a more accurate failure limit for the installed bolting may be gained by either performing hardness on building bolts in place, or by removing a sample of bolts for tensile/shear testing.
Overall the errors introduced by the assumptions and methods in calculation EDC 16-003 to establish the bolt UTS could potentially introduce approximately 20% bias into the conclusions.
Specifically, the overal l margins beyond failure are small (if the stress concentration factor is not included) so any increases in bolt UTS could impact the conclu ions. The e uncertainties are considered when making our overall assessment of whether the bolts will fail in Section 3.2.2.
4.2.
Issue 2 - Failure Sequence and Load Sharing 4.2.1. Description Comment Issue 2 questions the failure sequence and associated load sharing between the connections in the NEDC 16-003 (Reference 2) analysis. Each girt analyzed in NEDC 16-003 is attached at either end to the building columns by an angle iron clip. The clip is bolted to the girl with two 3/4-inch bolls, and the clip is attached to the column via two 1/4-inch fillet welds. An example configuration is shown in Figure 4-1. When the internal building HELB pressure load is applied to the panels and tbe girt, both the bolted connection and the welded connection will react this load. The flexibility of both connections must be considered to accurately determine the load in both connections. It is requested that a technical justification for the loading application and failure sequence used in NEDC 16-003 be provided to ensure proper load sharing was considered.
As documented in Appendix A, the comment is as follows:
"Calculation NEDC 16-003 uses 11vn-co11serva1ivefailure mechanisms because it assumes that the inner and outer girt bolts, the near and far fillet welds, and the sag rod will fail in a sequential manner rather than being loaded in a shared load co11dirio11.
Assuming a shared load condition/or these componenrs would be more realistic and would significantly increase the amount of pressure required to ensure the bolts and rhe welds would.fail.
0315-0084-RPT-001, Revision 0 Page 23 of 53
Address the following:
Provide the technical justification as to why it is appropriate to model the loading of the bolts, welds, and sag rods in a sequential manner rather than modeling 1he load as being simultaneously distributed across all components."
"far weld" Design weld 4.2.2. Evaluation Girt Figure 4-1.
Example of Girt Connection to Column "near weld" Design weld (backside)
Connection Bolts As indicated in the comment, proper representation of the joints must be used to accurately predict the load on each joint and the potential for failure. Based on a review of the NEDC 16-003 analysis, the following relevant modeling approach was used:
The girt and angle clip are modeled with three dimensional (30) clements near the connection points in the models which predict failure of the bolts and welds. This is considered to be best and an appropriate representation of those components.
The connection between the angle iron clip and the girt is modeled with a frictionless contact during the initial loading case. Frictionless contact allows the angle iron to slide relative to the girt and to separate from the girt. This behavior would be expected if there was no significant friction or preload or once preload is overcome.
The bolts are represented in the models as either 3D elements or beam elements.
Modeling the bolts includes the flexibility of the bolts.
o Note that the cases where the bolts are modeled with 3D elements, the bolts fill the hole in the clip. This over estimates the stiffness of the bolts, but still allows the load to transfer in reasonable manner.
The 3D bolts are bonded to the clip and bonded to the girt at the cylindrical faces so that the load transferred from the girt to the clip is through the bolts, with the exception that 0315-0084-RPT-001, Revision 0 Page 24 of 53
some compressive loads will be carried through the frictionless contact between the girt and clip mentioned above.
o Using bonded contact on the cylindrical faces is an approximation of the connection, and will result in a stiffer connection than one that considers the load to be transferred the faces under the bolt head and nut.
The model which uses beam clements for the bolts, uses similar connections as the 3D elements.
The welds are not explicitly included in the models. Instead a line support is included in the models to represent the connection between the clip and the column. This representation of the welds will result in a stiffer model than if a 30 representation of the welds were included.
The building column is not included in the model. Therefore, its flexibility is not represented. The absence of the column is not expected to be significant since the column flange is relatively stiff, and the load path is unaffected.
Based on the above, the load path from the girt to the column is well represented. Therefore, the loads on the bolted joint and the loads on the welds calculated in EDC 16-003 are expected to be accurate, and the "load sharing" is expected to be accurate between the joints. Further, the failure sequences identified in NEDC 16-003 Section 4.5 are explaining the ways in which the girt connections are expected to fail by identifying the next weakest link in the load path after the inilial failure. These failure sequences are based on the re ult of the models discussed above (i.c*., the expected failure sequences arc not dictating the modeling approach or impacting the load path or load sharing in the models).
Failure mode one is explaining that the most highly stressed weld is the outer weld.
Consequently, the outer weld will fail at a lower pressure than the inner weld. After, the outer weld fails, the inner weld will be more highly stressed because it must carry the full load at the clip to column connection, and will fail shortly afterwards. A subsequent beam model is used to confirm that a failure of the welds will result in buckling, and therefore, a total loss of the girt.
Failure mode two is explaining a similarly sequence for the bolts, assuming that the bolts fail first. NEDC I 6-003 is stating that in the shared load condition, both failure modes are predicted, and it is therefore not necessary to demonsh*ate which occurs first. As such, the failure sequences in NEDC 16-003 Section 4.5 are considered to be appropriate. Overall, the analyses presented in NEDC 16-003 provide area onable representation of the load haring and the potential failure sequence for the girt connections when the turbine building panels are subjected to a 0.5 psig internal pressure. Thus, MPR does not agree that, "assuming a shared load condition for these co111pone11ts would be more realistic and would significantly increase the amount o_f"pressure required to ensure the bolts and the welds would/ail."
0315-0084-RPT-001, Revision 0 Page 25 of 53
4.3.
Issue 3 - Effect of Sag Rods 4.3.1. Description Issue No. 3 questions the impact that the sag rods may have on the stiffness of the blowout panel support structures. The sag rods are long, 5/8-inch diameter rods that connect each rnw of girts to the one above and below it. There are three sag rods between each row on a single girt (see Figure 4-2). The sag rods connect each girt along the entire wall to each other. Each sag rod has a single nut just below its lower girt and just above its upper girt to position the sag rods and support deadweight of the girt and the adjoining panels. At the top of the building, the sag rods on the uppermost girt are connected to a structural wide-flange beam. The claim in the calculation is that the sag rods do not have much flexural stiffness due to their long and slender nature and therefore, the twisting of the girt will not be affected by the presence of the sag rods.
Additional evidence is requested to prove that Assumption 3 in Section 2.2 of NEDC 16-003, which states the effect of the sag rods is insignificant, is val id.
As documented in Appendix A, the comment is as follows:
"Calculation NEDC 16-003 does not appropriately account for resistance from sag rods installed between channel girts. These sag rods would restrain girt channel twisting and should be factored into the analysis.
Address the.following:
Calculation NEDC 16-003, Section 2.2, Assumption 3, states that the "sag rods supporting the channel girts are flexible in bending because of the relatively large span (between girts) and small cross section, resulting inflexibility and low bending resistance. There.fore, the sag rods do 1101 pr<Jl'ide significant resiswnce to twisting of the girt channel. The relative stiffness of the blowout panels is likewise small compared to the stiffness of the girt channel and will not significantly affect the twisting of the girt channel."
Provide the technical ir!formation (dimensions, material properties, etc.) of the sag rods used to justify Calculation NEDC 16-003, Section 2.2, Assumption 3."
0315-0084-RPT-001, Revision 0 Page 26 of 53
Columns Sag Rods Figure 4-2.
South Wall Section Showing Sag Rods 4.3.2. Evaluation Discussion and Approach The internal pressure HELB load imparts a shear load on the girt. Because the shear load is eccentric to the girt section's shear center, a torque is induced about the girt's length causing it to twist. Twisting of the gi11 increases loading on the girt connections and promotes failure of the structw-e. The ability of the sag rod to resist girt twisting is controlled by the sag rod stiffness, the girt stiffness, and the connection between the girts and the sag rods. If the sag rods can react some of the pressure load applied to the girts by the blowout panels, the twisting of the girt will be reduced, and therefore, the loads on the girt connections (bolting and welds) will be reduced.
To understand the impact that the sag rods may have on the loads at the critical bolting and welds the following approach is used:
The connection between the girt and sag rods is evaluated A scoping finite element model of a single girt, similar to NEDC 16-003, is developed A model with several girts is included with:
o sag rods connecting each girt at the appropriate locations o
the constraints between the sag rods and girts consistent with the ability of the connection type o
the upper set of sag rods connected to the structural beam at the top of the wall Each model is used in a static structmal analysis where the same:
o loading conditions are applied to each girt, including deadweight (girts, panels, and sag rods) and the equivalent force of0.5 psi pressure load o
material properties are used, consideiing plasticity, o
model stiffness is adjusted with deflected shape (i.e., large deflection effects are included), and 0315-0084-RPT-001, Revision 0 Page 27 of 53
o the same end boundary conditions and model connections are applied:
Fixed supports at each weld Symmetry condition at the midspan of the girt Frictionless contact between the girt and angle iron piece Bonded contact between the bolts and the adjoining parts Base line loads at the critical welds and bolting arc extracted from the single girt model, and assessment loads at the critical bolting and welds are extracted from the model with multiple girts.
The difference in loads between the models is used to assess the impact.
Results The geometry for the single girt model is shown in Figure 4-3 and the geometry for 1he multiple girt model is shown in Figure 4-4.
Per Reference 4, the sag rod are connected to the girt by a single nut below the girt.
Consequently, the sag rods can only react loads in tension and the joint will not react any moment loads. The multiple girt model considered this and only axial tension loads were credited.
The deflection of a single girt without the effect of sag rods is as shown in Figure 4-5.
Figure 4-5 shows that as the girt twists, it also sags. lf the sag rods were only between girts (i.e.,
they did not connect to a relatively stiff structure at the top of the building) then each of the girts would sag in a similar fashion and the sag rods would likely have no significant effect.
However, the sag rods are attached at the top of the building to a relatively stiff support beam.
Consequently, the sag rods provide a resistance to stretching in the vertical direction and some of the twisting of the girt is restricted. A comparison of Figure 4-5 to Figure 4-6, which shows the dcncction of a girt within the multi-girt model, illustrates this effect.
The result is that the sag rods resist some of the applied loads and the loads in the welds and bolting are reduced by the presence of the sag rods. The load in each weld is reduced on the order of I 0%, and the bolting loads are reduced on the order of 30%. This result is consistent over the height of the wall (i.e., the top girt and bottom girl sec a similar impact). Based on these results, the effect of the sag rods are not insignificant. However, it is expected that the magnitude of this effect would change with a more rigorous modeling of the bolted connection, as in Section 3.2.2. In any case, this effect must be considered with other effects to determine if the structures will fail prior to the turbine building exceeding 0.5 p ig.
Note that these results are not expected to vary significantly for girts with diffcreot vertical spacing since the rods connections do not carry moments.
0315-0084-RPT-001, Revision 0 Page 28 of 53
Symmetry Plane Figure 4-3.
Single Girt Geometry for Sag Rod Evaluation 0315-0084-RPT-001, Revision O Page 29 of 53
Upper Structural Beam Sag Rods Symmetry Plane Figure 4-4.
Multiple Girt Geometry for Sag Rod Evaluation 0315-0084-RPT-001, Revision O Page 30 of 53
Edge is below/
undeformed y
frame Figure 4-5.
Single Girt Displaced Shape for Sag Rod Evaluation Undeformed
'I Edge is at approximately same height as undeformed frame Figure 4-6.
Multiple Girt Displaced Shape for Sag Rod Evaluation 0315-0084-RPT-001, Revision O Page 31 of 53
4.4.
Issue 4 - Friction Effect on Bolted Joint 4.4.1. Description Issue 4 questions how the joint girt is evaluated in NEDC 16-003 (Reference 2) analysis.
Specifically, the issue addresses how friction between the girt to clip fasteners is included and questions which design code the NEDC 16-003 is u ing for this joint.
As documented in Appendix A, the comment is as follows:
Calculation NEDC 16-003 does not use the correct method/or joint design in accordance with A/SC standards. The calculation uses a method that accounts/or joint friction for as-built girt joint that is not conservative and does not meet A/SC requirements.
In addition, address the following:
Provide the relevant construction standard used to assen1ble the turbine building column to girt joint Please specify the type of bolted joint design of the as-built girt joint (i.e. shear/bearing, slip-critical, or direct tension).
Provide the torque specifications, if any, for the girt bolting.
Provide the technical justification for NEDC 16-003, Section 2.2, Assumption 9, "Friction between steel swfaces is credited to tram.fer load between Jhe angle and channel and between the angle and column flanges following weld failure. A range of friction/or carbon steels is provided in various text (see f 16} as an example). A dry mild steel on mild steel 1*aL11e of 0. 78 is listed under static conditions, 111ith a sliding value around 0.42. A value o/0.35 has been commonly used as a lower bound friction value/or most steel on steel surface conditions.
Since selecting a low value is considered consen*ative, a value of 0.25 is utilized herein."
4.4.2. Evaluation Constmction codes are concerned with ensuring that structures resist design loadings and have the requisite margin to ensure failure does not occur under the design loading conditions. Failure is desired in the case of the turbine building blowout panels underlying structural supports, so design codes arc not specifically applicable. An assessment of the physics of the problem, rather than code compliance is required to determine the suitability of the design in its function as a pressure relief. NEDC 16-003 does not address preload or the friction that could develop between the girt and the clip if the bolted connection were preloaded. Crediting friction results in load sharing between the bolts and friction in the joints. Load sharing reduces the bolt loads and is non-conservative to neglect when evaluating failure. Section 3.2.2 provides scoping analyses that assess whether the joint will slip or not, and determines the resulting bolt loads.
Two cases arc performed - one case with high friction and preload and the other case with low friction and preload. The scoping analysis results show that the joint will slip under the 0.5 psig loading even when a large friction factor and large preload are considered. These analyses also 0315-0084-RPT-001, Revision 0 Page 32 of 53
confirm that the forces and moments applied to the limiting bolt are beyond its failure capacity based on actual material properties. See Section 3.2.2 for additional discussion.
4.5.
Issue 5 - Required of Blowout Area 4.5.1. Description Issue 5 questions whether or not the area of the blowout panels predicted to fail below 0.5 psig is sufficient to maintain the turbine building pressure below 0.5 psig. The design basis requirement for the blowout panels is that they blowout at 0.5 psig. Consistently, the design basis analyses consider the blowout panels to fail when the general room pressure reaches 0.5 psig.
Additionally, the design basis analyses assume that a certain area of the blowout panels will fail.
This area is not stated in Reference 17, which calculates the turbine building response to a HELB event. Consequently, lhere is no comparison for lhe area predicted to fail at 0.5 psig in NEDC 16-003, and NEDC 16-003 docs not make a conclusion on whether or not the predicted blowout area is acceptable. It is requested that an evaluation of the required area versus the predicted failure area be performed.
As documented in Appendix A, the comment is as follows:
Calculation NEDC 16-003 does nor demonsrrare rhar The 5362 square foor blowout area is adequate to preve11t pressure retention beyond 0.5 psig in the Turbine building following a High Energy line Break (HELB) event. A prel'ious analysis pe,formed by a contractor using GOTHIC modeling found that a blowout area of 30,000 square foot still resulted in pressure retention.
In addition, address the following:
Calculation NEDC 16-003, Section 5.0 states, "Using a HELB pressure of 0.45 psi, it is predicted.failure of the panel support system at this pressure will result in a loss (blowout) of total panel area of 5,362 ft2." Provide the engineering analysis or calculation that demonstrates that a total panel area of 5,362Jt2 is enough to maintain peak pressure in the turbine building below 0.5 psig.
4.5.2. Evaluation Discussion and Approach During a HELB event, the turbine building will pressurize locally around the break and the general pressure in the building will rise. When the general pressure in the building reaches 0.5 psig, the blowout panels must fail, per the requirement. To maintain the pressure at or below 0.5 psig, the area that blows out must be capable of passing the same flow at a 0.5 psi pressure differential as the flow entering the building from the HELB. Therefore, the following approach is used to assess adequacy of the predicted blowout area:
The mass and energy flows for the HELB event of concern are identified.
The volumetric flow of steam entering the building from the break as it expands to 0.5 psig is determined.
0315-0084-RPT-001, Revision 0 Page 33 of 53
The area required to pass the same volumetric flow rate at 0.5 psid is calculated using band methods considering the following:
o The maximum density of steam or air is considered, which increases the resistance and the required flow area o
The opening created by the blowout panels is considered to be a sharp edged orifice, which increases the resistance and the required flow area o
The maximum flow rate from the corresponding transient in Reference 17 is used.
The analysis considers the 30" line break in the turbine building operating floor.
Results The calculation presented in Appendix B determines that an area equal to approximately 2,300 ft2 is required to maintain a pressure of 0.5 psig for the 30" line break event analyzed in Reference 17 for the peak flow rate. This area is approximately half of the area predicted to blow out by NEDC 16-003. Therefore, it is expected that the blowout area i ufficient to maintain the pressure al 0.5 p ig or below. See Appendix B for additional calculation details.
Note that the comment also states that previous GOTHIC modeling found that a blowout area of 30,000 square fool still resulted in pressure retention. MPR does not have access lo this model, so cannot make any credible claim to its accuracy. However, MPR notes that GOTHIC utilizes a one dimensional node-connector modeling approach. Using GOTHIC results from a model intended to evaluate one phenomena lo provide a basis for something else that the model was not setup to evaluate may lead to overly conservative conclusions.
4.6.
Issue 6 - Effect of Columns on Blowout Area 4.6.1. Description Issue 6 questions whether or not the 5362 fl2 of blowout panel area that is predicted to fail in NEDC 16-003 took into account area that would be blocked by the girt support columns. The girts are supported at each end by an angle iron clip welded to the building columns. The blowout panels are attached to the girts and are continuous behind the support columns (i.e., they do not stop at the columns). Therefore, the flow area created by the blowout panels may be partially blocked by the support columns, and the actual flow area available may be less than predicted by NEDC I 6-003, if this was not taken into account. It is requested that the area calculation be reviewed.
As documented in Appendix A, the comment is a follow :
"The assumed turbine building siding blowoul area in NEDC 16-003 (5362 square foot) is based on column celllerli11e which would be 11on-co11servatil'e because colu11mj7a11ges 111011/d protrude beyond column centerline and reduce the actual blowow area. This would increase the pressure retention in the turbine building potelllially beyond 0.5 psig.
In addition, address the following:
0315-0084-RPT-001, Revision 0 Page 34 of 53
Provide engineering drawings that show the actual dimensions of the blowout area credited i11 NEDC 16-003."
4.6.2. Evaluation Discussion and Approach NEDC 16-003 analyzes three girt configurations that are considered to be the weakest. These are girt types 133A, 1338, and l 33C. Spans supported by other girt types are not expected to fail because they have more robust connections. The calculation then determines whether or not the girls identified to be weakest will fail due to the 0.5 psig pressure load and at what ve1tical spacing failure occurs (affects magnitude of pressure load over length). NEDC 16-003 then determines which sections of the girts would blow out based on these results. To address this issue, tJ1e following approach is used:
The calculation of blowout area in NEDC 16-003 is reviewed.
The sections stated to fail are compared to the design drawings in Reference 4.
An adjusted available area is calculated, as appropriate.
Results The areas predicted to fail in NEDC-16-003 are shown in Figure 4-7 and Figure 4-8 for the No:rth and South Walls, respectively. These region are stated to fail because they meet the criteria of having average pan above and below each girt of 6'-7" (79 inches) or greater. For this discussion, the areas are divided in Regions A, B, and C, as shown in Figure 4-7 and Figure 4-8.
Based on the calculations presented in NEDC 16-003 Section 4.7 and Reference 4 the following is observed:
Region A - Region A is calculated considering the width of Column C, but does not account for the width of Column Din the blowout area.
o Note that the area for upper portion of Region A where girt types 133A arc used is conservatively calculated.
Regions 8 and C - Regions 8 and Care not calculated considering the width of Columns D, E, or F.
The calculation below estimates that the total area blocked by the column over the height considered in NEDC 16-003 is 139 ft2 (or 2.6% of the total blowout area). Therefore, the remaining area available for t1ow from the total blowout panel area is 5223 ft2* This area remains significantly greater than the 2,300 ft2 t1ow area required to maintain the turbine building operating room pressure below 0.5 psig (see Section 4.5.2). Therefore, the width of the columns does not significantly impact the available flow area, and MPR concludes that this issue is not a technical concern.
0315-0084-RPT-001, Revision 0 Page 35 of 53
\\\\'ou -1 mm+..!.in - o."'l.t.t\\
16 HRA :- 99-lft - 932ft - 6.25tn
- 61 r9-ft HRc._ 7ft 2 + 6ft + IOm
- 20.833-ft Width of Columns D E and F. 18 WF 70 and 85 beam References 12 and 4 Height of Region A Height of Region B Height of Region C Each column at the boundary of a region only blocks an area equal to half the width of the column times the region height The columns in the midspan of a region block an area equal to the total column width times the region height The effective number of columns for each region are.
Number of Columns in Region A Number of Columns in Region B Number of Columns in Region C The total area blocked by the columns within the predicted blowout area is
~
~lock :* \\\\"ou{1'coH*HR.~... ~"com*HRB - ~colC HRd * !39.-l*ft*
The resulting flow area available from the predicted blowout area is 0315-0084-RPT-001, Revision 0 Page 36 of 53
- M
.ll!-..J_..J-1 __
I
, ~ cw ***':'f*,
.mt ' J_'. __
Jlll_L 153c.
4 rrr,,
.. Me- _..u.__.
l&!c.
I
- I
- 10_~ I II~
1,;,.
-f~-~-+-
.11!<-
I J"6_..,.!.O l1l<:
tm s.-. *...
.,:.. _....,.. 2 """'*
1*
d -,.. c,.
,1t1E
"*r- \\_ ~"
w
... ~
'1'c 1)3'
!~..arr--
- I Jt'it*A
- l
~-..,
~ l*Jc.
!I ~
- 1)
.JI.<.
,,lj c.
- 1
- I
',t.. tt"tll!'
,*~
~
mc.*r ('
..;J;
't **JHc.
- 1. *
- it*i,a o:-
me.
/..
I
.J *1 1~,P 1350
't-~
./'
IH~
w I :t
/Ha
- ' :~.,
/IJE
- I
,l:i
,ue
- l*i ~
/ /t "3P
~p*,
--l~
l I
-:r~f'.:*
Ii*-*
.~.:
UJ
-,:TT,.. ~se- "-
r.. -.u *,*.
-r
'I
'"I.
~
I.
V 21..
~
I DIR Si,\\
I
~
,i Figure 4-7.
North Wall Predicted Blowout Area 0315-0084-RPT-001, Revision 0 Page 37 of 53
J
~
-I..
~
,,,t JJIC f I
- !>IO /1Jc
/He i. / IJC
/ J3c.
i3JM,,,c.
Figure 4-8.
South Wall Predicted Blowout Area
- 4. 7.
Issue 7 - Load Application and Effect of Full Wall 4.7.1. Description I -
~ -.,_J Issue 7 questions whether the load applied in the NEDC 16-003 analysis appropriately considers the effects of adjacent girts (above, below, and to either side) on the specific girt being analyzed.
The turbine building wall is composed of several columns, girts, and blowout panels. The blowout panels create a continuous wall that spans the wall's width and height. The* blowout panels are secured to each other at their edges and to the girts where blowout panels join along horizontal lines. Because of tJ1is construction, the internal building pressure will apply to all blowout panels concurrently, and all girts must support tbc total load from all blowout panels. It is r,equcst.ed to confinn the analysis appropriately applied loads considering this simultaneously loading of the full wall.
0315-0084-RPT-001, Revision 0 Page 38 of 53
As documented in Appendix A, the comment is as follows:
"Calculation N£DC 16-003 includes non-conservative assumptions involving how rheforces against the turbine building siding are distributed during a high energy line break.
Specifically, the distributed force only assumes a single girt channel whereas the actual construction of the t11rbine b11i/ding blowo111 panel includes 11111/tiple girt channels.
Co11seque111/y, the resultant force may 1101 be enough to ensure bolt and weld failures at pressures less than 0.5 psig.
In addition, address the following:
Clarijj, whether the res11lta111 force is distrib111ed across only a single girt channel connection or across multiple girt channel connections supporting each turbine building blowout panel.
Provide a copy of the most curre111 revision ofCalc11lation NEDC 16-003.
4.7.2. Evaluation Discussion and Approach Since the pressure load is applied to the entire wall simultaneously, all girts must share the total load. The simplest estimation of the load each girt would be lo divide the total load by the number of girts. However, lhe spacing between girts varies at different locations. Consequently, some girts will have more load than others. Gi11s with larger spacing will have greater loads than girts with smaller spacing. It is expected that the load between two adjacent girts will be split evenly between those girts. Therefore, for a single girt the load is expected to be equal to 0.5 psi multiplied by the tributary area (i.e., the average of the area between the girt and the one above and between the girt and the one below it). This is the approach used in NEDC 16-003 to calculate the loads on an individual girt. To evaluate this theory, the following approach is used:
A simplified finite element model of a wall made up of blowout panels is developed Elastic material properties are used for this general load distribution estimate The wall is considered to be fixed by the girts along their length A 0.5 psi load is applied to the wall Loads are extracted at select locations The extracted loads are compared to loads determined by a hand calculation using tbe calculation method discussed above.
Results The model developed for this evaluation is shown in Figure 4-9, along with the boundary conditions and applied load. The vertical spacing of girts is varied in select rows to ensure generally applicability. The locations (A through E) assessed are shown in Figure 4-10. The vertical spacing above and below each girt is listed in Table 4-1. Figure 4-1 1 shows the displaced shape of the wall, which is consistent of the displaced shape of the turbine building wall prior to blowout of the panels. The only difference expected is that the girts wi II bow out 0315-0084-RPT-001, Revision 0 Page 39 of 53
and there will also be some bowing of the columns. However, these effects will not significantly impact the load distribution prior to blowout of the panels. Table 4-2 presents a comparison of the hand calculated loads and the loads from the finite clement model. The results show the difference in loads is 0.1 % or less. Therefore, the hand calculation method used in NEDC 16-003 is acceptable for calculating load on a single girt, and MPR concludes that this issue is not credible.
Table 4-1.
Location Information for Wall Load Distribution Evaluation Location Spacing Above Spacing Below Horizontal Lengthl1)
(ft)
(ft)
Position (ft)
A 6.25 7
Wall Edge 24 B
6.25 7
Mid Wall 24 C
5.75 5.75 Wall Edge 24 D
5.75 5.75 Mid Wall 24 E
5 6.25 Wall Edge 24 F
5 6.25 Mid Wall 24 Note:
- 1. Length is distance between vertical column centers.
Table 4-2.
Location Information for Wall Load Distribution Evaluation Average Area!1l Calculated<2)
FEA Load Location Load (ft2)
(lbf)
(lbf)
A 159 11448 11448 B
159 11448 11462 C
138 9936 9941.9 D
138 9936 9930.4 E
135 9720 9720.8 F
135 9720 9719.5 Note:
- 1. Calculated as average of areas above and below girt using data in Table4-1.
- 2. Calculated as area multiplied by 0.5 psi (i.e., 72 psf) 0315-0084-RPT-001, Revision 0 Error 0.00%
-0.12%
-0.06%
0.06%
-0.01%
0.01%
Page 40 of 53
FIXtd Support 2 FIXtd Support 3 Foxed Support4 F1Xtd Support 5 FIXtd Support 6 FIXtd Support 7 Foxed Support IE] Displacement PrtJSure: 0.5 pso 0.00***-====3=0*00***-====60::J.oocft) 15.00 45.00 Figure 4-9.
Model for Wall Load Distribution Evaluation Figure 4-10. Locations Considered for Wall Load Distribution Evaluation 0315-0084-RPT-001, Revision 0 Page 41 of 53
Figure 4-11. Displaced Shape for Wall Load Distribution Eval 0315-0084-RPT-001, Revision 0 Page 42 of 53
5.0 References I.
Cooper Nuclear Station Updated Final Safety Analysis Report, Amendment 25.
- 2.
NPPD Calculation 16-003, Revision. 0, "Structural Evaluation of the Turbine Building Blowout Panels Steel Support," (LP I Calculation A 15046-C-00 I, Revision I).
- 3.
Electronic Mail from T. Barker (Entergy) to C. Swanner (MPR), "Re: Calculation to Review," dated February 20, 2020 (provided in Appendix A).
- 4.
Turbine Building Design Drawings:
- a.
CNS DIR 454015485, Capitol Steel and Iron Contract 9 ISO, sheet 133,
- b.
CNS DIR 453015407, Capitol Steel and Iron Contract 9150, sheet 139,
- c.
CNS DIR 454015468, Capitol Steel and Iron Contract 9 I 50, sheet EI 0 I,
- d.
CNS DIR 454015471, Capitol Steel and Iron Contract 9150, sheet E L06,
- c.
CNS DIR 454015470, Capitol Steel and lron Contract 9150, sheet EL07,
- f.
CNS DIR 454015353, Capitol Steel and Iron Contract 9150, sheet 158,
- g.
CNS DIR 454015347, Capitol Steel and Iron Contract 9150, sheet 156,
- h.
CNS DIR 454015354, Capitol Steel and Iron Contract 9150, sheet 157, and
- i.
CNS DIR 454015352, Capitol Steel and lron Contract 9150, sheet 159.
- 5.
ASME Boiler and Pressw*e Vessel Code, Section VJll, Division 2, "Alternative Rules for Construction of Pressure Vessels," 2011 Edition.
- 6.
PVP2009-77l22, Snow, R. & Morton, D. (I L); Pleins, E. (WEC); and Keating, R.
(MPR), "Strain-Based Acceptance Criteria for Energy-Limited Events," published in the Proceedings of the 2009 ASME Pressure Vessels and Piping Division Conference, Prague, Czech Republic, July 26-30, 2009 (https://inldigitall ibrary.inl.gov/sitcs./sti/sti/4310585.pdQ
- 7.
"Guide to Design Criteria for Bolted and Riveted Joints," Second Edition, Research Council on Structural Connections, published by AISC, 200 I.
- 8.
A. S. o. Materials, ASM Handbook, Volume I.
- 9.
NUREG/CR-2137 "Realistic Seismic Design Margins of Pumps, Valves, and Piping,"
June 198 I.
l 0.
ASME/ ANSJ B 1.1-2003, "Unified Inch Screw Threads (UN and UNR Thread Form)."
- 11.
ASTM E 140-12, "Standard Hardness Conversion Tables for Metals Relationship Among Brinell Hardness, Vickers Hardness, Rockwell Hardness, Superficial Hardness, Knoop Hardness, and Scleroscope Hardness."
- 12.
ATSC, Manual of Steel Construction, Sixth Edition, American Institute of Steel Construction, 1967.
- 13.
E. Oberg, Machinery's Handbook 27th Edition, ew York: Industrial Press Inc, 2004.
0315-0084-RPT-001, Revision 0 Page 43 of 53
- 14.
. Suwan L. Manual and K. Frank PMFSEL R port o. 03-1, "Stati tical Analy i of Strnctural Plate Mechanical propertie," Phil M. Fergur on Structural Engineering Laboratory, Austin, TX, January 2003.
- 15.
ASA Reference Publication 1228, "Fa tener De ign Manual," 1990.
- 16.
A I/ I C 360-10 pecification for tructural teel Building, 20 I 0.
- 17.
DS Report 0 1-0840-1115 "Cooper uclear Station Environmental Effect Due to Pipe Rupture" Rev. 0.
I.
CRA Technical Paper o. 410, 'Flow of Fluid through Valve, Fitting, and Pipe,'
2010.
0315-0084-RPT-001, Revision 0 Page 44 of 53
A Issues Identified C S personnel provided MPR a list of the issu s identifi d r lated to ED 16-003 in an el ctronic mail (R fi r nc 3). Th I ctronic mail i attached.
0315-0084-RPT-001, Revision 0 Page 45 of 53
Swanner, Craig From: Barker, Troy S.<tsbarke@nppd.com>
Sent: Thursday, February 20, 2020 12:32 PM To: Bergquist, David; Swanner, Craig
Subject:
RE: Calculation to review Attachments: Issues list.pdf David/Craig, Attached are the list of issues that we have been provided anonymously. Please use these to develop the proposal for the review. As we talked yesterday, I would like the "Focused Review" of the response to the issues.
- Thanks, Troy From: Bergquist, David <dbergquist@mpr.com>
Sent: Monday, February 17, 2020 5:11 PM To: Barker, Troy S. <tsbarke@nppd.com>; Swanner, Craig <cswanner@mpr.com>
Subject:
RE: Calculation to review This email is from dbergquist@mpr.com. Do you know them and are you expecting this? - Look again!
Use the "Report Phishing" button if you think this is a phishing email.
Phishing is the #1 threat to NPPD. You are our best defense 11 Stay Vigilant!
Troy:
We have had some time to review the calculation. Our initial take is that the calculation seems reasonable, but there are a number of items we think would be worth looking into in more detail including: (1) the assumption that static load application bounds dynamic load application conditions, (2) the failure criterion selected, (3) the appropriateness of the model boundary conditions, (4) configurations selected to model, etc.
We would like to have a quick discussion with you regarding two main options for moving forward. The options include:
Full Independent Review - MPR would perform a complete independent calculation review, we would fully recreate the existing model to determine the impact of changes in assumptions on the conclusions.
Focused Review-The more efficient option would be for you to provide us the technical concerns raised regarding the calculation, and we focused our review on the merit of those concerns. In this case, MPR would still perform a high level review of the whole calculation, but we would focus any analytical effort on the specific concerns.
In either case, we would document our efforts in either a memo or short report including our review findings and conclusions.
Please let me know your thoughts on these two options and when you are available to discuss further.
Best Regards, David and Craig From: Bergquist, David Sent: Friday, February 14, 2020 5:25 PM To: 'Barker, Troy S.' <tsbarke@nppd.com>; Swanner, Craig <cswanner@mpr.com>
Subject:
RE: Calculation to review 0315-0084-RPT-001, Revision 0 Page 46 of 53
Thank you Troy. We've been reviewing and discussing today, and we expect to send you some preliminary options for what we could do on Monday.
Have a good weekend.
David From: Barker, Troy S. [mailto:tsbarke@nppd.com)
Sent: Thursday, February 13, 2020 6:05 PM To: Bergquist, David <dbergquist@mpr.com>; Swanner, Craig <cswanner@mpr.com>
Subject:
Calculation to review David/Craig, Per our telecon today, I have attached the calculation that is the subject of the allegation for your use to prepare the request for proposal and subsequent review. We can talk in more detail next week.
- Regards, Troy Barker Engineering Programs & Components Manager N PPD-Cooper Nuclear Station 402-825-5027 Contents of lssucs List.PDF:
Issue #I The girt bolt ultimate strength and shear strength assumed in Calculation EDC 16-003 is non-conservative and may not be representative of tJ1c bolts used in the blowout panel construction.
Spcci fically, the calculation improperly uses an average ultimate strength instead of maximum tested ultimate strength. Additionally, the average tensile strength is derived from testing of warehouse stock bolls which are likely from di ITerent heat and lots than of the bolts used in blowout panel construction. The average ultimate strength may not bound the ultimate crength of bolts used in construction.
Address the following:
Provide a discussion and evidence of any validation activities (walkdowns. testing, etc.)
performed to ensure that the currently installed girt bolts arc of the same grade (ASTM Grade A307) and/or have the same material [Properties a the bolts that were te led for Calculation NEDC 16-003.
Issue #2 Calculation NEDC 16-003 uses non-conservative failure mechanisms because it assumes that the inner and outer girt bolls, the near and far fi llet welds, and the sag rod will fail in a sequential manner rather than being loaded in a shared load condition. Assuming a shared load condition for these components would be more realistic and would significantly increase the amount of pressure required to ensure the bolts and the welds would fail.
0315-0084-RPT-001, Revision 0 Page 47 of 53
Address the following:
Provide the technical justification as to why it is appropriate to model the loading of the bolts, welds, and sag rods in a sequential manner rather lhan modeling the load as being simultaneously distributed across all components.
Issue #3 Calculation NEDC 16-003 docs not appropriately account for resi&tancc from sag rods i11Stallcd between channel girts. These sag rods would restrain girt channel twisting and should be factored into the analysis.
Address the following:
Calculation NEDC 16-003, Section 2.2, Assumption 3, states that the "sag rods supporting the channel girts are nexible in bending because of the relatively large span (between gi,1s) and small cross section, resulting in flexibility and low bending resistance. Therefore, the sag rods do not provide significant resistance to twisting of the girt channel. The relative stiffness of.the blowout panels is likewise small compared to the stiffness of the girt channel and will not significantly affect the twisting of the girt channel."
Provide the technical information (dimensions, material properties, etc.) of the sag rods used to justify Calculation NEDC 16-003, Section 2.2, Assumption 3.
Issue #4 Calculation N EDC 16-003 does not use the correct method for joint design in accordance with A I SC standards. The calculation uses a method that accounts for joint friction for as-built gi11 joint that is not conservative and docs not meet AISC requirements.
In addition, address the following:
Provide lhe relevant construction standard used to assemble the turbine building column to girt joint Please specify the type of bolted joint design of the as-built girt joint (i.c.
shear/bearing, slip-critical, or direct tension).
Provide lhe torque specifications, if any, for the girt bolting.
Provide the technical justification for NEDC 16-003, Section 2.2, Assumption 9, "Friction between steel surfaces is credited to transfer load between the angle and channel and between the angle and column flanges following weld failure. A range of friction for carbon steels is provided in various text (see [16) as an example). A dry mild steel on mild steel value of0.78 is listed under static conditions, with a sliding value around 0.42. A value of0.35 has been commonly used as a lower bound friction value for most steel on steel surface conditions.
Since selecting a low value is considered conservative, a value of0.25 is utilized herein."
0315-0084-RPT-001, Revision 0 Page 48 of 53
Issue #5 Calculation NEDC 16-003 docs not demonstrate that the 5362 square foot blowout area is adequate to prevent pressure retention beyond 0.5 psig in the Turbine building following a High Energy Line Break (H ELB) event. A previous analy is perfonned by a contractor u ing GOTHIC modeling found that a blowout area of 30,000 square foot still resulted in pressure retention.
In addition, address the following:
Calculation NEDC 16-003, Section 5.0 states, "Using a HELB pressure of0.45 psi, it is predicted failure of the panel support system at this pressure will result in a loss (blowout) of total panel area of 5,362 ft2." Provide the engineering analysi or calculation that demonstrates that a total panel area of 5,362 fu is enough to maintain peak pressure In the turbine building below 0.5 psig.
Issue #6 The assumed turbine building siding blowout area in EDC 16-003 (5362 square foot) is based on column centerline which would be non-conservative because column flanges would protrude beyond column centerline and reduce the actual blowout area. Thi would increase the pressure retention in the turbine building potentially beyond 0.5 psig.
In addition, address the following:
Provide engineering drawings that show the actual dimensions of the blowout area credited in NEDC 16-003.
Issue #7 Calculation NEDC 16-003 includes non-conservative assumptions involving how the forces against the turbine building siding are distributed during a high energy line break. Specifically, the distributed force only assumes a single girt channel whereas the actual construction of the turbine building blowout panel includes multiple girt channels. Consequently, the resultant force may not be enough to ensw*e bolt and weld failures at pressures less than 0.5 psig.
In addition, address the following:
Clarify whether the resultant force is distributed across only a single girt channel connection or across multiple girt channel connections supporting each turbine building blowout panel.
Provide a copy of the most current revision of Calculation EDC 16-003.
0315-0084-RPT-001, Revision 0 Page 49 of 53
B Calculation of Required Blow Out Flow Area This appendix estimates lhe blow out now area required to maintain the turbine building operating room pressure at or below 0.5 psig, consistent with the conditions discussed in Section 4.5.2 B. 1 Transient Conditions B.1.1 State Information There are two maior now cond1t1ons 1dent1fied for the 30" hne break event which occur at two different time periods in the transient This calculation considers both sets of conditions mass now rate enthalpy quahty lb BTU Qt = I m1 :m 9470-h t.= 1189j -
s lb Ref 6 pp 14 and 15 lb BTU m2 :m 31360-hi:,..594-Q2.- 0.07 s
lb 0315-0084-RPT-001, Revision 0 Page 50 of 53
There are two ma1or flow cond1t1ons identified for the 30" hne break event which occur at two different time periods in the transient This calculation considers both sets of conditions mass flow rate lb m1 '= 9470-s lb m1._ 31560-s enthalpy BTU h1.= 1189.5 lb Bru h2 :- 594 -
lb quahty Ref 17 pp 14 and 15 The pressure temperature and entropy associated with each cond11ion 1s determined as follows I P
) BTU H5t(P.Q) ** StmPQHj-Q.O -
\\ pst lb enthalpy as a function of pressure and quality finding pressure to use remaining winsteam functions pguess :* 500ps1 Gtnn P1 :- F111d(pguess)
- 1.018 x 103 psi T1 -- stmP/ ~.o)*r-553.827-~F SI.* StmPQs( ~.Q *. o)-1381 psi H5t(pguess. Q2) = h2 P2 :* Find(pguess)
- 1.043 x 103 psi T 2 :- stmPJ P2_
- o) 0r. 549. ~6-' *r
' l ps, s2 ** StmPQS( P2_.Q2.o)-0.79i psi 0315-0084-RPT-001, Revision 0 Steam pressure for condition 1 Temperature for cond1t1on 1 entropy for condition 1 Steam pressure for condition 2 Temperature for condrtion 2 entropy for condrt1on 2 Page 51 of 53
B.2 Volumetric Flow Rate Pr :* 0.5psi + latn'I
{
Pr l ft3 ft3 v1 :* StmPS\\ -
.s1.OJ -
- 19.26 -
pst lb lb B.3 Orifice Loss Coefficient T urbme building max pressure specrfic volume of steam after 1sentrop1c expansion to room cornditions at O 5psig for condition 1 specific volume of steam after 1sentrop1c expansion to room corndrllons at O 5ps1g for cond11ion 2 Volumetric flow rate mto building for condrtron 1 Volumetnc flow rate mto building for corndition 2 The opening created by the blowout panels 1s treated as a sharp edge onfice Reference 18 Page A-21 provides guidance for calculating the loss coefficient for an orifice f3
- 0.0001 Ratio of downstream to upstream areas conservatively taken to be very small Flow Coefficent. C. 1s taken to be O 6. which 1s approximately the min value conservative Flow Coefficient Cd as a function of p Loss coefficient based on smaller flow area as a function of p Loss coefficient for orrfice Three different fluids may be moving through! the hole created by the blowout 0315-0084-RPT-001, Revision 0 Page 52 of 53
B.4 Required Flow Area Three different fluids may be mOVtng through! the hole created by the blowout Density of air at 68 °F Ref 18. p A-9 Density of steam tn cond1t1on 1 density of steam in cond1t1on 2 lb kg Pair.* o.om -
"' 1.20s -
f\\3 m3 I
kg P1 :* -
- 0.832-:
,*1 m'
I kg P2 :* -
.. 1.SJS-3 m
Note that a lower air temperature would increase the air density but lower temperatures are not expected nor would they change the results of this calculation The maximum density will be conservatively used for calculaltng the required flow area kg PO.ow:- max(Pair*P1,P2)
- 1.848 - 3 m
The maximum volumetric flow rate 1s 3
\\'flow :* max{V 1. v2)
- 7.746 x 103 ~
s The specified maximum pressure differential is dP ** 0Jpsi The required flow area to pass the maximum flow is Required flow area 0315-0084-RPT-001, Revision 0 Page 53 of 53
LS2020018 Page l of l Enclo ure 5 LPI Ref. A16254-LR-001, Rev. 0, "Responses to RC Questions for NPPD Calculation
(48 page )
LPI, Inc. Consulting Engineers Adl*1w11*d A 11,1li1s,'\\ {1 f'1l11t*~.. )*" Sn,*,n*
/",1i/11r,*,;.,\\lat,*, 1,1/, rmf,,.,/11111
,\\fo1td,*.-.fr11t ltl',* l.n...,:i;"'c1,11g July 1, 2016 LPI Ref. A16254-LR-001, Rev. 0 Mr. Alan Able Nebraska Public Power District Cooper Nuclear Station PO Box 98 Brownville, NE 68321-0098
SUBJECT:
Responses to NRC Questions for NPPO Calculation No.16-003 (LPI Calculation A15406-C-001)- Cooper Nuclear Station NPPD Contract Order No. 4200002638 Amendment 2
Dear Mr. Alan:
At the request of NPPD, LPI, Inc. (LPI) has prepared responses to NRC questions related to NPPD Calculation No.16-003, Revision 0, (LPI Calculation No.
A15406, Revision 1), titled "Structural Evaluation of the Turbine Building Blowout Panels Steel Supports".
The NRC questions and the proposed responses are included as an attachment to this letter. The provided responses include clarifications, simple hand calculations, and finite element analysis. All responses have been checked and verified in accordance with LPI Quality Assurance program for nuclear safety related work.
This letter report is prepared in accordance with the requirements of NPPD contract number 4200002638 Amendment 2. If you should have any questions, or if we can provide additional assistance, please do not hesitate to contact us.
Lucius Pillon - established 1885 36 Mam Stroot
- Amesbury. MA 01913
- 978 517.3100 New York 80s1011 Richland Sydney wwwtpmycom l:.11.,111111~ fl,t* 111/t*gri/11 of /1~/11.11,; mftml 111/ra~tmcf11ri* Ji,r /011wm111*', wvrltl.
Mr. Alan Able LPI Ref. A 16254-LR-001, Rev. 0 July 1, 2016 Page 2 of 4 J,~((/,,1, B. Elaidi, Principal 0
7/01/16 Engineer Rev.
Date Prepared by Respectfully Submitted, LPI, Inc.
vlt!Jl/c_JI vlt/Jl/c~('
A.Chock, A. Chock, Principal Principal Engineer Engineer
µ;pr Andrew Smyth(1)
Principal Engineer Checked by Designed Verified by KW P. Bruck Principal Approved by
( 1) For review of finite element model and results In the response to Question 6.
Attachment A ( 44 pages)
1 2
3 4
5 6
7 8
9 10 Mr. Alan Able LPI Ref. A16254-LR-001, Rev. 0 July 1, 2016 Page 3 of 4 Document Software Record1 (include separate sheet for each software package used)
Computer Software Used ANSYS Release 14 (CodeNersion)
Software Supplier ANSYS, Inc.
Software Upd1te Review u Error notices; describe 181 Other; descnbe: Review of V&V report (see #8 below)
Nuclear Safety Related UNO 2 If YES, complete the following.
Software 181 YES2 Computer type*
Dell Precision 17500-4 desktop Computer SIN:
HC74VR1 Computer 0 /S:
Win 7 Pro 64 V & V (include as ref)*
(see #8 below)
Base for Identify the bases that support use of this ANSYS is a general purpose computer code Application software for the application herein; may be that Is widely used in calculating linear and separately discussed elsewhere in this nonlinear responses of compl.e)( configurations document (Indicate section) and/or may be in nuclear power plants. II is therefore suited for addressed in the V&V (identify reference the plastic analysis performed herein.
number)*
Input U Input listing and/or summary attached*
Llstlng/Summary3 181 Not attached; identify File/Oise ID: Input files are supplied separately on an electronic media aaaa.lnp 6/22/2016 5:20 PM Output L!5J Output results attached* Selected output results are included Data/ldentifler(1 )1 D Not attached; Identify File/Disc ID:
'e.g., run date/lime; use for reference, as app,oprlate, within body of calaJlation Comments V&V per LPI Report No. V&V-ANSYS-14, Rev. 0, "Verification and Validation of ANSYS Software Program."
Keywords*
SHELL 181, TB, TBDATA
- For use In describing software features used In this calculation: use common terms based on software user manual and/or help file&.
Project Manager Name: I B. Elaidi 1 If computer software is used on project, complete fonm with required information.
Update the LPI ~2m11ut~r §ofiwar~ !,!~e List per LPI Procedure 13.1 requirements.
Form: LPl-3. 1-Rev*B*F/g*S-B
Mr. Alan Able LPI Ref. A 16254-LR-001, Rev. 0 July 1, 2016 P
4 f4 age 0
~~
DESIGN VERIFICATION CHECKLIST u*
... \\ Document No(s)1: I A 16254-LR-001 I
Rev.:
Review Method: Ix I Document ReV!Ow I
IAtternate Calculation I
lrest Criteria 1
Were the Inputs correcily selecied and Incorporated Into design?
Ne assumptions necessary to perform the design activity adequately described and reasonable? Where necessa,y, are the 2
assumptions Identified f0< subsequent re-vertllcatlons when the detaUed design acthl!Ues are completed? If applicable, has on as built verification been performed and recooaled?
3 Are the appropriate quality and quality assurance requirements specified?
Are the applicable codes, standards and regulatory requirements Including Issue and addenda property ldentlfled and are their requirements for design met?
5 Have applicable construction and opereting experience been considered, Including operation procedures?
6 Haw the design Interface requirements been 11llsficd?
7 Was an appropriate design method used?
8 Is the output reasonable compared to inputs?
9 Are the specffted parts, equipment. and processes suitable for the required application?
10 Are the specified materials compatible v.ith ellCh other and the design cnwonmonlal condlllons to which tho materllll will be exposed?
11 Have adequate maintenance features and requirements bean specJ'led?
12 Are acccsslblllty and other design provisions adequate f0< perfomlanco of noodcd maintenance and repair?
13 Has adequato accenlbUlty been provided to per1orm the ln-s8fVlce lnspedton eipected to be required during the plant life?
1 Has the design properly considered radiation expos1Ke to the public end plant personnel?
15 Are the acceptance a'lterla Incorporated in the design documents sufficient to allow vertficaUon that design requirements have been satisfactorily accomplished?
16 Have adequate pre-operational and subsequent periodic test requirements been appropnately specified?
17 Are adequate handling, storage, cleanng and shipping requirements specified?
18 II software was used, have the computer type and operating system been properly Identified? Is use of the software, hardware and O/S approprtate for the concM1on1, components evaluated? Has a V&V bocn perlormed?
19 Are requirements for identifical.ion, record preperallon review, approval, retention, etc, adequately specified?
20 Has an Internal design review bean performed for applicable design projects? Have comments fr0<n the Internal 0eslgn Review been appropriately considered/addressed?
( 1) lncludo any drawings d&Vt/optd from roviowed documents, or Inc/IX# separate checklst sheet for drawings
{2) Design Verifier sha.U /nit/al Indicating review and marlc NIA where not applicable 0
ov2
~
~
8C,
{ll,
. fJ(,,,
- EJC, I):_,
Im.-
MA
~/A Aii,.,
A -1r AJ A1 le lat A) ~
,J* \\
~
~
l)~
DV Completed By: I Printed Name A. Chock I s~,Jqtf/( /ev I
Dal~ I /4t,U, Page I 1 I ofl 1 !Total Pages Include DV Checklist and : dmment Re solution sheets In page count v
Form LPl-3. 1 *Rev-8-Ftg-5-6
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A REVIEW COMMENTS RELATED TO ENTERGY CALCULATION NO 16-003 REVISION O (Including ANSYS Analyses)
In response to a request from Region IV, Division of Reactor Safety, (DRS) the Office of Nuclear Regulatory Research (RES) staff completed review of Entergy's Calculation No.16-003 Revision 0 titled, "Structural Evaluation of the Turbine Building Blowout Panels Steel Supports". Based on the review of this document, RES staff were unable to definitively establish whether the Turbine Building siding and it structural components will fail under a pressure level of 0.5 psi or less due to a High Energy Line Break (HELB). Accordingly, the following questions and clarifications were developed in order to address RES staff comments/questions. Pending the resolution of these issues, RES staff anticipate establishing the technical adequacy of the subject design basis calculation.
CALCULATION NUMBER 16-003, REVISION 0 Section 2.0 Input and Assumptions 2.1 Input Clarification Item 2. h (Pages 11 and 12)
This section states, in part, that the load due to the HELB pressure is transferred from the girt to the column by the angle via biaxial bending for the connection detail at Column line C. For the connection detail at Column lines D, E and F, load is transferred from the girt to the Column via both torsion and bending by the angle.
Question 1: Please provide the basis and justification for the assertion that; the bolts in both connection details experience bending forces and not just shear and tensile forces?
Response 1: Refer to Figure 1-1 (of Cale.16-003, RO) for connection at Col. Line D, E, F, and Figure 2-2 (of Cale.16-003, RO) for connection at Col. Line C. The bolts connect the angle with the channel web and are considered to be tight (see Figure 1-1 (of Cale.16-003, RO) as an example. Typical AISC practice for such a connection would be "snug tight". The angle leg thickness is 3/8" and the channel web thickness is 0.24".
The HELB pressure loading is transferred from the siding to the girt via siding fasteners attached to the girt flange. This load is thus eccentric with respect to the girt shear A1 of 44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A center which results in twisting of the girt. Figure 1-1 (below) shows the resulting deformation of the channel at the connection.
The bending in the bolts is a combination of two mechanisms. One bending mechanism results from the sliding of the channel from under the angle due to the lateral HELB load (see Figure 1-2, below). This sliding results in a guided cantilever mode of response in the bolt, with linear moment distribution over the bolt length. Shear forces are also developed in the bolts due to this mechanism. The mechanism is predicated on the presence of "slack", i.e., the diameter of the bolt is slightly less than the diameter of the hole.
The other bending mechanism results from the twisting moment acting on the channel at the connection (see Figures 1-1 and 1-3, below). This mechanism involves constant moment over the bolt length. The total bending in the bolts is the combination of those two bending modes.
Figure 1-1 Dt!!PJ,.i\\.COltfl' DHX *.J. U "
undeformed chennel undeformed angle Rotation of the Girt at the Support due to Torsion AN Ml U Ult
- 14. 46;.Sf A2 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A Figure 1-2 Guided Cantilever Model Bending of the Bolt, from sliding Figure 1-3 Constant Moment Bending of the Bolt, from twisting A3of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A Item 5. (Page 12)
This section states, in part, that the bolts used in the channel girt connection to the Turbine Building columns are specified as ASTM A307 Grade A. Based on the hardness testing for a group of 7 bolts, withdrawn from stock at Cooper Nuclear Station (CNS,) the hardness values (Rockwell I8 (HRB)) were determined and converted into approximate tensile strength values.
Question 2: Given that the derived and the actual failure capability of the bolts is predicated on the test data for a random sample of 7 bolts, please indicate whether any bolts, installed in the channel girt connection to the Turbine Building Columns, were removed and examined to confirm the material characteristics of the fasteners includino bolt shank length. Additionally, please provide the definitive basis and justification for the assumption that the sample fasteners, withdrawn from stock, are consistent with the actual material characteristics of the installed bolts.
Response 2: No bolts were removed from the connections of the Turbine Building siding girts in support of this assessment. The existing bolts are considered to be ASTM A307 as stated in the original plant design documents (Reference 19 of calculation 16-003, RO).
The LPI receipt inspection of the sample new bolts obtained from the plant (included in page C6 of calculation 16-003, RO) identified the material of the sample bolts as A307 based on stamping on the bolt head. The A307 bolt material is the typical low grade commercial bolting material for this size bolting and is similar to A36. Based on review of some of the earlier versions of the ASTM A307 specification, the minimum required strength of the A307 bolt specification did not change since the construction of the plant.
Therefore, the newer stock bolts can be used to define the tensile strength of the bolts Installed in the field.
A sampling of field bolts for the connections on the north side are shown in Figures 2-1 (bolt heads), Figure 2-2 (bolt projection) and Figure 2-3 (Plant Steel assembly drawing)
- - below. Bolt heads with no specification markings (Figure 2-1 shows mark "TB" only),
are typically "Grade 2* alloy with strength characteristics consistent with A307 Grade A.
The Capital Steel specification drawing (Figure 2-3) identifies the bolts as 3/4" M.B. The "MB" call out is considered as "Machine Bolt". The bolts are identified as 1-3/4" long (Figure 2-3). For the "plys* of 1/4" (girt web) and 3/8" (angle leg), the resulting grip is 5/8". Per ASTM A307 specification, thread form is defined by ASME 818.2.1. The nominal thread length is equal to 2 x the basic thread diameter plus 1/4". Thus thread length is 1.75", and the bolts are considered fully threaded. The projection of the bolt A4 of 44
LPI Letter Report A 16254-LR--001, Rev. 0 Attachment A beyond the grip of approximately 1-1/8" (Figure 2-2), is consistent with the bolt length specified in Figure 2-3 (I.e. 5/8" grip+ 1-1/8" projection= 1.75"). Bolts in the south wall connections have same projection above the nut as shown in Figure 2-4 and therefore they are fully threaded as well.
AS of 44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A Figure 2-1 Field Condition, Bolt Head Markings of Sample of Bolts in Evaluated Connections -
North Wall A6 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A Figure 2-2 Bolt Assembly Projection - North Wall
-l -
~
A7 of 44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A CAPITOL nHL 41411 IAON 00,
.. t.AKOM.\\Off, NOl.llHN Ult ol fltl, ltltt or.,.,, I A8 of 44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A Figure 2.. 4 Boll Assembly Projection South Wall A9 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A 2.2 Assumptions Item 7. (Page 14)
This assumption states, in part, that "...for added conservatism with respect to applied torsion, the dead weight and associated torsional loading are applied to the girt".
Question 3: Please provide the basis and justification for not applying both forces (twisting moments due to the pressure and twisting moment due to dead load to the girt}, accounting for the proper direction of the moments?
Response 3: Assumption #7 (Section 2.2 of calculation 16-003, RO} states that the torsional effects of the deadweight were applied In the direction that counters the torsion due to the HELB pressure. This was the case in Models 1, 2, and 3. Consideration of deadweight is considered conservative, in that it reduces the twisting from the HELB pressure applied through the fastener.
The analysis of Model 4 did not include, the deadweight effect. Therefore, the analyzed model configurations (1, 2, 3, and 4) effectively bound all possible configurations of the deadweight distribution of the panels, (either included or not included} including eccentricity of the applied deadweight load.
By review of the building siding detail drawings (see for instance Figures 3-1 and 3-2, below (Ref. Plant Drawing 49054 Sheet 12T}, it is clear whether the panel dead weight is transferred to the base or supported from the girts. It is for this reason that models 1, 2 and 3 incorporated the siding dead weight and model 4 does not include the panel dead weight. The bolt and weld stress results from all these models therefore, represent bounding values.
Note: Additional Ref. Plant Drawing 49054 Sheet 12T A10 of 44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A
-r
.,,,.,0
_ r_r. ~,0-810 __.:.. -
""'1-:-
c.C-9 ____... FF-~~~----+-__:
w c.c.-Js Figure 3-1:
Siding Attachment Configuration at Base of Turbine Building 903'-6" Floor (Ref Plant drawing 49054 Sheet 12T)
A11 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A Figure 3-2:
Photo of Exterior Siding at Base of Wall A12 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A Item 9. (Page 14)
This assumption states in part that, "Friction between the steel surfaces is credited to transfer load between the angle and the channel.... following weld failure."
Question 4: RES staff acknowledges that there may be some frictional force between the angle and the column flange following weld failure. However, this frictional force appears to be insignificant, given that the sag rods support a significant portion of the dead load of the girt. Accordingly, please provide the technical basis and justification for this assumption.
Response 4: The consideration of friction here was to ensure the failure pressure was not under-predicted. The friction considered is invoked to keep the girt channel level following weld failure. i.e. following weld failure, the leg of the angle remains in contact with the flange of the building column, from a bearing load associated with reaction of the HELB pressure. To resist the angle sliding down the face of the building column under self-weight (and thus inducing weak axis bending in the channel), friction is credited between the angle leg and the building column flange, see Figure 4-1, below.
With the channel level, this provides stability for the channel and as stated above, eliminates potential weak axis bending due to deadweight on the outer segments of the girt acting as an overhang, beyond the adjacent sag rod (see depiction in Figure 4-1, below). Eliminating this weak axis bending and ensuring the girt remains level, improves the resistance of the girt to HELB pressure, and therefore provides for a conservative estimate of failure pressure.
A13 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A WELDS Figure 4-1 WHEN WELDS FAIL, HELB PRESSURE WILL HOLD THE ANGLE AGAINST THE COLUMN FLANGE.
FRICTION DUE TO CONTACT BETWEEN THE ANGLE AND FLANGE SURFACE RESISTS DOWNWARD MOVEMENT OF GIRT END.
Configuration of Loads at the Angle to Building Column, following weld failure.
A14 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A 3.0 Methodology and Failure Criteria 3.1 Analysis Methodology Item 6. {Page 21)
Model 4 is used to evaluate the girts between column lines C and D. Potential failure modes that were evaluated by this analysis include item 6, shear and or bending failure of the bolts. Specifically, in this analysis the two bolts are subjected to single shear mechanism and bending due to twisting of the girt.
Question 5: Please provide the basis and justification for the assertion that the bolts experience bending forces (moments) due to the twisting of the girt, given that there is no significant gap between the two surfaces?
Response 5: See the response to question 1 for description of the bending modes of the bolts.
A15 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A Item 8. (Page 22)
As discussed above, Model 4 is used to evaluate the girts between column lines C and D. Potential failure modes that were evaluated by this analysis include item 8, fastening of the panels to the girt. As noted in this section, because of the complexity of the analysis the evaluation did not consider the twisting of the girt channel which would cause additional forces on the siding attachment screws and panel sheet metal components.
Question 6: RES staff noted that, the effect of the siding on the rotational capacity of the girt should be taken Into account. Specifically, the siding may provide little resistance to lateral deflection of the girt. However, for any torsional rotation to occur, the screws attaching the siding to the girt have to shear off or the siding has to either bend, buckle or stretch, without the screws failing. Please provide the basis and justification for not considering the resistance to twisting of the girt provided by* the siding, the stresses in the connection angle (and subsequently the welds and bolts), that are higher than they would have been, otherwise.
Response 6: The evaluation documented in calculation 16-003, RO considered the rotational stiffness of the siding to be low and to not have a significant effect on the rotation of the girt (see Section 2.2, item# 3 of calculation 16-003, RO).
The siding panels inside lining that is fastened to the girts is 18 gage which has a maximum thickness of 0.046" (i.e. 1.17 mm) per the vendor data sheets (included in sheet A 10 of calculation 16-003, RO and the siding thickness is listed in the table of properties for L 10 panels). Per Plant drawing No. 49054 Sht 2T, the panel-to-girt screws are stainless steel# 14 x 3/4" Type B with neoprene washer. Specification of this screw is included in page A-6 of the Inland-Ryerson test report (included in page B20 of CNS calculation NEDC 13-028). The material specified therein is #410 alloy.
The bolt head is hex washer with 9/16" diameter.
The main mode of response of the panels under the HELB pressure load is that of bending between the girts. This mode is supplemented with local deformation of the panel at the screw heads due to twisting of the girts, from eccentric loading of the siding attachment screw in the girt flange, relative to the shear center of the girt channel. This local mode of response involves high local strains in the siding panel and results in low panel stiffness by comparison to the girt stiffness.
This is explicitly demonstrated by a finite element model of a panel segment shown in Figure 6-1 below. The model includes shell elements for the panel and the screw head A16 of 44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A Is simulated by a rigid disc with a diameter equal to the screw flanged head diameter (9/16"). The panel model Is subjected to 0.5 psi pressure and a nominal twisting moment via a force couple on the rigid disc as shown in Figure 6-2 below. The rotation of the rigid disc is obtained and used to calculate the panel stiffness associated with rotation of the screw as follows:
Diameter of screw flanged head = 9/16" Nominal applied force couple, F = 50 lbs This Is a nominal value and is less than the actual HELB torque transmitted through the screw to the channel. The HELB torque per screw Is the tributary screw HELB force times its vertical eccentricity with respect to the shear center of the channel (see page B4 of Cale.16-003 for guidance). Thus:
Vertical eccentricity of the screw with respect to the channel shear center
= half width of the C1 0 flange+ shear center from the back of the C10
= 2.6251n/2 + 0.796in = 2.1in HELB torque per screw = 0.5 psi x 7ft x 12in/ft x 6in x 2.1 in = 529 lbs-in per screw. This Is significantly higher than the torque applied in the present local model of 50 lbs x 9/16in = 28.1 lbs-in per screw. Use of actual screw torque increases the plasticity in the panel and reduces the panel stiffness. Thus, the use of this low screw torque in the present model is conservative.
The torque couple is applied on the diameter of the screw head.
Applied moment per screw = 50 lbs x 9/16" Resulting deflection at applied force couple, U = 0.13646" (from FEA model - see Figure 6-2 below)
A17 of 44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A girt
--~---7--.
F U
7L ---i applied twist force angle couple HELB pressure panel screw head Panel rotational stiffness per screw= 50 lbs x 9/16" / tan-1(0.13646" x 2 I (9/16") )
= 0.062 kip-in / rads The torsional stiffness of the girt is calculated for half the span using shear modulus of steel and torsional modulus of the girt C1 O section as follows:
Kc10 = G J / L = 11.5 x 103 ksi x 0.21 in4 / 12 x 12 in= 16.8 kip-in / rads This rotational stiffness of the girt is compared to a summation of rotational stiffness of the panels to demonstrate the relative stiffening effect provided by the panels. Since the twisting angle of the girt is maximum at mid span and decreases to a small value at the supports, the potential for the panels to interfere with girt twisting is greatest at mid span and panels at and near the columns have much lesser to no impact. Thus, over a girl's 12 ft. half span, consider the stiffness for 7' length of panels to be effective (two screws per 1 ft. of girt). Thus, the stiffness ratio is calculated as follows:
A18 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A Stiffness Ratio= 7 ft x 2 screws/ft x 0.062 kip-in/rad (per screw)/ 16.8 kip-in/rad (for girt)
Stiffness Ratio = 0.05 x 100% = 5%
Thus, the stiffness associated with local deformation of the panel at the screw head is considered to be less than 5% of the torsional stiffness of the girt. As such, the panels are not considered to provide appreciable resistance to twisting of the channel. This is consistent with Assumption #3 of Section 2.2 in calculation No.16-003, RO.
Note: Additional
Reference:
Plant Drawing No. 49054 Sht 2T A19 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A Figure 6-1 NODAL SOLUTIOII StlP*l SUB *ZZ Tinl*l UY I\\SYS*O l)JI)( *3. 89899 S1III *-3. 89899 6'11X *. 136457 NODAL I OLUTIOII STIP*l SUB *U TIHJ*l UY C.\\VGI P.SYS*O 1)1!)( *3. 89897 81111 *-3 tHU SID< *. 136457
-i. HIH
-.oot
.,.,so, 6"
girt CL J\\N JIJII lZ ZOU 17: 10, SS J\\N JUN ZZ Z016 17: lZ:33 symm BC 1H4J1 finite Element Model of Panel Segment Corresponding to One Screw Showing Contours of Out-of-Plane Displacement at 0.5 Pressure and Nominal Twisting Moment
- Lower View Shows Enlarged Model at Screw head.
A20 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A Figure 6-2 IIODlk IOWTIOII l tlf*l sua *U U ll* l ur Q1'-a DIDI -,
11111
- l Zoom in View of the Panel at Screw Head Showing Applied Force Couple and Resulting Out-of-Plane Displacement A21 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A RES staff also noted that the welds between the girt angle and the column flange experience bending about the two orthogonal axes and torsion about the third. Please provide the basis and context for determining that this connection is considered "more rugged."
Response: The RES staff question is related to the statement made that the girt connection at column line C (see Cale.16-003, RO Figure 2-2) is considered more rugged by comparison to the connections at column lines D, E, and F (see Cale 16-003, RO Figure 1-1 ). Design of the connection at column line C is shown in Figures 2-1 and 2-2 of the calculation. A discussion of the design and load path for this connection is in item 2.h. In Section 2.1 of the calculation. The main aspects of this connection that differentiate it from the other connections are:
- a. The connection angle reacts the HELB pressure in bending without significant twisting (i.e. the angle is perpendicular to the girt axis).
- b. The weld connecting the angle to the building column is larger in length.
- c. The girt extends eastward beyond the bolted connection. If there is a positive connection with the girt on the east side of the building (see Cale.16-003, RO Figure 2-1 ), additional load path and load sharing with the angle increases the connection strength.
All these three aspects contribute to higher strength and stiffness in this connection compared to girt connections at column lines D, E, and F.
A22 of44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A 3.2 Bolt Failure Testing (Page 22)
As noted above, in the RES Staffs Question concerning Item 5, the shear testing of the supplied bolts was performed by LPI Inc. Consulting Engineers, to demonstrate the actual shear failure stress for the bolts and to derive the ultimate tensile strength of the bolt material. Fully threaded bolts were tested in double shear configuration in accordance with National Aerospace Standard NASM 1312-13. However, no objective evidence was provided to establish that the installed bolts are fully threaded or if the bolts have a solid shank. Specifically, if the bolts have a solid shank with the threads excluded from the shear plane, the effective cross sectional area and ultimate strength of the bolts would be different. Additionally, RES staff noted that bolts with solid shanks were assumed in the previous calculation submitted by the licensee (Entergy Calculation No: NEDC 13-028 Rev. 1, page E4-10).
Question 6a: Please provide justification for using fully threaded bolts in the calculation.
Response: ASME B18.2.1, referenced in ASTM A307, provides standards for hex bolt dimensions. Per Table 2 therein, the nominal threaded length for 3/4" hex bolts is 1.75" for bolts 6" long and shorter (2 x basic thread diameter plus 1/4M). Per field measurements (see Response to Question 2), the bolts projects 1-1/8" above top surface of the angle (i.e. above the "grip"). Adding the plate thicknesses, total bolt length is 1.7" (1-1/8" + 0.24" + 3/8"), i.e. nominally 1.75" as specified in Figure 2-3 of Response to Question 2). Thus, the bolts are threaded along their complete length.
The bolts in the south side connections are 1/4" longer. Resulting threads remain in the shear plane and at the nut attachment point.
A23 of 44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A 3.3 Failure Criteria (Page 23)
As noted in the analysis, fillet welds contain a complex state of stresses that involve tension, shear, compression, and bearing. As such the analysis used a conservative limit for the fillet weld failure to take place that was postulated to occur between 75% Fu (shear failure) and Fu (tensile failure). However, RES staff noted that a value of 70 ksi was previously used for the tensile strength of the weld on page E4-11 of Entergy's Calculation No: NEDC 13-028 Rev. 1.
Question 7: Please identify the value used for the tensile strength of the fillet weld including the basis and technical justification for this value.
Response 7: The welds in the connections in column lines D, E, and F are two vertical fillet welds (see Cale.16-003, RO Figures 1-1 and 3-1 ). Failure of this weld can occur along the base metal fusion surface parallel with the siding (fusion with the building column), along the base metal fusion surface normal to the siding (fusion with the connection angle), or along any plane within the weld.
Under the HELB pressure, the state of stress along any of the possible failure :surfaces stated above is complex. Depending on the plane of failure, the failure can be dominated by shear stress or tensile stress. Therefore, it is stated that failure of the weld would occur in the range of 0.75 Fu to Fu where Fu is the postulated ultimate strength of the weld. The 0.75 Fu limit is considered applicable to shear stress failure and the Fu limit is applicable to failure in tension.
Per assumption #5 in Section 2.2 of the calculation, the weld material is assumed to be as strong as the base-metal. This is considered to be a reasonable assumption, consistent with typical engineering practice. Based on base-metal mechanical properties obtained from original construction Certified Material Test Reports (CMTRs) and listed in item #4 in Section 2.1 of Calculation 16-003, RO. A value of 78 ksi for tensile strength of the weld material is considered a reasonable value.
The tensile strength value of 70 ksi referred to by the staff above is the design value used for electrode E70 weld rods. This value may not be conservative for failure prediction because it is a minimum design value and it may under represent the actual material strength. The value of 78 ksi is considered conservative relative to the design value, and more appropriate for predicting failure of the weld.
A24 of 44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A 4.0 Analysis 4.2 Nonlinear Load Step Analysis (Pages 29 thru 32)
As indicated in Figure 4-5 of the analysis, the initial stress in the weld of approximately 25,000 psi, is attributed to deadweight loading. The analysis also indicates that, the high strain intensity is localized at the bolt hole and is dominated by bearing action against the bolts. The analysis further states that the bolt head and nut were not modeled.
Question 8: Please describe how the stress level of 25,000 psi was derived, and why the tensile stress in the bolt is dominated by bending and not pure tension. Additionally, please describe how the tensile stress/strain value in the bolt was derived, if the bolt head and nut were not explicitly modeled.
Response 8: The question concerns bolt stress and the statement that precedes the question, includes weld stress. This response therefore addresses both aspects.
Weld Stress:
The weld is modeled in the finite element model by restraining the displacements in the three orthogonal directions, Ux, Uy, and Uz, at the nodes along the weld lines of the back side of the angle leg as shown in Figure 8-1, below. The resulting force reactions Fx, Fy, and Fz are the weld forces. In the ANSYS coordinate system, x is along the girt axis, y is up, and z is normal to the siding. The weld stress is calculated at nodal points as follows:
J Fx2 + Fy2 + Fz 2 T
= -,---------,--,---,---c---
wcld
- 0. 707 x tw x nodal tributary length Where the weld size, tw, is 0.25". The nodal tributary length is obtained from the finite element mesh as the distance between the two mid side points relative to the two adjacent nodes, identified as distance "Lw" in Figure 8-1 herein. Weld stresses in Figure 4-5 of the calculation are for the second mesh node from the bottom of the angle for which Lw is equal to 0.07109"/2. This is conservative, because it ignores the higher weld stresses at the bottom of the weld leg. In cases where Fz (weld force normal to face of the angle) is compressive, Fz was excluded from the stress equation above.
This is also considered to be conservative and it was performed in this manner because failure of the weld under compression is difficult to predict. Including this compressive term would result in a lower failure pressure. Typical Mdesign* practice is to include A25 of 44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A compressive stresses in weld joint design. In Figure 4-5 of the calculation, Fz is compression for the near weld and is tension for the far weld.
As an example, the weld stress calculation for the dead weight load is presented.
- a. Using ANSYS Model 1, the Fx, Fy, and Fz reactions at node 29763 (second node from bottom at near weld). The reactions are obtained at ANSYS time=2.
This corresponds to the deadweight load prior to application of the HELB pressure. The resulting forces are:
Fx = -162.313 lbs Fy = 17.1539 lbs Fz = 51.1783 lbs (this is compression and therefore will be ignored in calculating the total weld force)
- b. Calculate total weld force: Fw = (Fx2 + Fy2)0 5 = 163.2169 lbs
fw = Fw / (0.707 x 0.25" x 0.071093"/2) = 25,978 psi This is the initial value of the green and orange curves in Figure 4-5 of Calculation No.16-003.
Bolt Stress:
The bending stress in the bolt is the result of two modes of response. One mode results from sliding of the channel from under the angle creating guided cantilever bending in the bolt with linear moment distribution through the length of the bolt. The second mode involves the twisting of the girt which causes rotation of the channel cross section with respect to the angle, resulting in uniform bending through the length of the bolt (see also response to Question 1 ).
The bolt is modeled with beam elements and is connected to the top surface of the angle and bottom surface of the channel web by radial rigid beams that simulate the nut and bolt head as shown in Figure 8-2, below. The use of the radial rigid beams is a
'Simplified method to simulate the head and nut without the need to explicitly model the head and nut with solid elements and thereby increase the number of nodes in the model. Bolt bending stress is obtained using the maximum bending moment along the bolt length divided by the bolt (bending) section modulus. The bending modulus is calculated based on the tensile stress area of the bolt. Since the maximum bending stress occurs in the threads, a conservative stress concentration factor of 2.5 is used based on review of stress concentration values reported in Peterson (Ref. 6 of Calculation 16-003, RO) based on testing of threaded fasteners. Refer to response to Question 15 for additional discussion of this stress concentration factor.
A26 of 44
LPI letter Report A16254-LR-001, Rev. 0 Attachment A Figure 8-1
[!.MllTS u
~
~
Weld Modeling in the Finite Element Analysis (x axis parallel with girt, y axis is up, z axis normal to siding)
I\\N J illi 13 2016 14: 08: 46 Lw A27 of 44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A Figure 8-2 Bolt Elements Attachment to Angle (Similar Connection to Channel at the other end of the bolt). The bolt is modeled using a beam element, the nut and bolt head are simulated with radial rigid links around the bolt hole.
A28 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A 4.3 Post Weld Failure Analysis (Page 38)
As indicated in the analysis, the deadweight reaction on the girt corresponds to 687.6 lb.
RES staff noted that, in order to simulate bending and lateral torsional buckling of the girt. Model 3 excluded the welds and the angle connection and accounted for friction between the column flange and the girt angle. However, RES staff noted that full surface contact for frictional forces is not likely, at the time the girt starts to rotate (torsion), because as indicated in Figures 4-3 and 4-4, the top of the vertical leg of the angle (around the near weld) pulls away from the column flange. RES staff also noted that the rotation of the girt at midspan appears to disregard the presence of the wall sidings and the screw connection of the sidings to the flange of the channel girt which does not appear to be a conservative assumption.
Question 9: Please describe how the deadweight reaction on the girt of 687.6 lbs.
translates to a dead weight stress in the weld of more than 25,000 psi as shown in Figure 4-5 of the analysis.
Response 9: The initial weld stress is due to dead weight forces and twisting moment associated with the panel weight. Because of the twisting moment, the stresses are not uniform along the weld lines. The calculated stress is the peak stress that occurs at the lower end of the welds as indicated by the stresses in the back of the angle shown in Figure 4-4 of the calculation. It is postulated that weld failure is initiated at the lower end of the weld and progresses up. See response to Question 8 and Questions on Figure 4-5 for explanation of the weld stress derivation.
Note: Consideration of weld stresses without consideration of the applied deadweight loads was addressed in Section 4.4 of the calculation (i.e. Model 4).
1<2.9 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A 4.4 Girt Between Column Lines C & 0 (Pages 44 and 45)
RES staff review noted that, the welds at this connection realize shearing forces as well as bending and torsional moments about two orthogonal axes due to the eccentricity of the bolts with respect to the welds.
Question 10: Please provide the basis and justification for why the welds at column line Dare failing at a higher pressure than those at the other end of the girt (column line C).
Response 10: The analysis does not provide failure load for the weld at column line C.
It is postulated that the weld at column line C is stronger than the weld at column lines D, E, and F. This is because column line Chas more weld length than column line D, E, F and also because the girt extends beyond the bolted connection as shown in Figure 2-1 (of calculation 16-003, RO) which causes a portion of the girt reaction to bypass the angle weld and follow a load path through the extended girt to the corner weld with the girt running along the east side of the building.
Comparison of weld stresses in Figure 4-5 of the calculation for the weld at column lines D, E, and F and in Figure 4-16 of the calculation for the weld at column line D for a girt between column lines C and D shows reduction in weld stresses In the latter case. This is because two different models were used for the two cases as explained below:
Figure 4-5 of the calculation is based on modeling the girt with line elements using ANSYS BEAM189 with HELB pressure applied as element line pressure (force/linear distance) as shown in Figure 10-1 below. Due to BEAM189 limitation in modeling exact cross section, the flanges were represented with uniform thickness and the fillets at the corners were not represented. The model included twisting effect from the deadweight of the panels.
Figure 4-16 of the calculation is based on modeling the girt with solid elements and matching exact shape of the channel section, including the tapered flanges and fillets at the inside corners. Deadweight was not considered and HELB pressure was applied as nodal forces at mid width of the flange as shown in Figure 10-2 below.
The above modeling differences explain the difference in the weld stress in the two figures. The two figures are considered to provide conservative bounds of the failure HELB press11re A30 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A lUll"lNTI HELB pressure girt mid-span symmetry BC girt beam elements
-1* i )e
-1.u..-n
_., 11 u.*u1
>
- n.t
- 0.H.Ul
,.uon It.UH a,,o,o U
Figure 10-1 Girt Beam Element Simulation and HELB Pressure Application Figure 10-2 Girt Solid Element Simulation and HELB Pressure Application A31 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A As noted by RES staff, the connection angle is already in contact with the inside face of the column flange prior to failure of the welds.
Question 11: Please provide the basis and justification for why failure of the welds would disconnect the girt at column line D and cause the girt to "slide over the edge of the column flange under High Energy Line Break (HELB} pressure"? Specifically, have both bolts failed at this point in time?
Response 11: Two separate scenarios of failure are postulated for the girt between column lines C and D. One is initiated with bolt failure at column line D that causes the girt to lose its support and become a cantilever supported at column line C. The other scenario is initiated by weld failure at column line C which causes the release of the rotational restraint that is effected by the two weld lines.
In the scenario of bolt failure, it is shown in the calculation that the resulting cantilever moment at column line C exceeds the plastic moment capacity of the girt. The two bolt connection capacity at column line C will also be exceeded. This would lead to complete release of the girt.
In the scenario of weld failure at column line D, the channel will be free to rotate and slide at column line D under the HELB pressure. This causes increase in bending deflection of the channel. The failure mechanism becomes similar to that demonstrated in Section 4.3 for the girt between column lines D & F, though the connection at column line C is not postulated to fail at this point. This similarity is because of the relatively large length of the girt span. With this large span, the condition at column line C does not significantly influence the girt rotation and sliding at column line D. As shown in Section 4.3 for the girt between column lines D, E, and F, as the HELB pressure increases the girt threshold for lateral torsional buckling is reached (i.e., the failure mode of the channel subsequent to weld failure is a lateral torsional buckling mode combined with high bending strain in the channel}. This instability increases the mid span bending stresses in the channel and leads to the development of a mid-span plastic hinge, at which point the load resistance of the channel is reduced.
A32 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A Additionally, as noted by RES staff, if and when the connection at column line D fails prior to that at Column line C, the girt now becomes an overhang beam. The typical girt at column line C is bolted to an angle at column C and also to the girt at the same elevation running in the north-south direction along the east end of the structure. The failure of the near and far welds described in Section 4.2 does not appear to be consistent with the description of the weld failure in Section 4.4, in that, the weld configurations at Column Line D are the same as those describe in Section 4.2.
_Question 12: Please provide the basis and justification for the apparent inconsistency in the weld configurations at Column Line D described in Section 4.4 and those described in Section 4.2.
Response 12: The weld stresses reported in Sections 4.2 and Section 4.4 are for the same weld configuration but are produced by two different models, deemed to bound the loading conditions. Thus differences are not considered to be inconsistencies. See response to Question 10 for detailed description of the two models and load application.
The analysis in Section 4.4 did not account for the deadweight including the twisting effect which opposes the twisting from the HELB pressure. The effect of deadweight and associated twisting from deadweight are considered in the Section 4.2 analysis.
A33 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A As noted during RES staff review, for a plastic hinge to be formed at Column Line C due to failure of the bolts at Column Line D, it is assumed that the bending moment can be developed in the cross-section of the channel at the bolt line. Specifically, the bolts have to be able to stay intact to develop that moment. The shearing force in the two bolts that are 3 inches apart resulting from the "cantilever moment" of 1594 kip-in appear to be significantly larger than the bolts shearing capacity and the bolts should fail prior to any plastic hinge formation in the girt. Accordingly, if the behavior of the girt after failure of the connection at column line D is similar to an overhang beam, as suggested in the comments above, then this moment appears to require re-calculation.
Question 13: Please provide a detailed description of the basis and justification for the conclusions described in Section 4.4 regarding the failure of the turbine building siding.
Response 13: LPI concurs with the staff that the shear capacity of the two-bolt arrangement at column line C has lower bending capacity compared to the plastic moment capacity of the channel section about its strong axis. However, failure of the bolts at column line C may not immediately cause complete failure of the girt connection, since the girt extends beyond the bolts as shown in Figure 2-1 of calculation 16-003, RO and will provide additional strength if there is a positive connection with the girt along the east wall. The plastic hinge analysis in the calculation is therefore made irrespective of bolt failure. The plastic hinge moment is conservatively compared to the moment resulting from HELB at the location of the bolts rather than the larger moment resulting at the east end of the girt.
A34 of 44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A 4.5 Failure Mechanism
- 1. Failure of the weld (Page 47)
As indicated in the analysis, the panel fasteners have to either shear off or the siding has to bend, buckle or stretch for there to be sufficient channel rotation to cause bending about the weak axis. However, as noted during RES staff review, previous testing of the siding panels, performed by the vendor, indicated that the panel fasteners did not fail at pressures less than 0.5 psi.
Question 14: Please specify how much rotation can occur in the channel, such that failure is through bending about the weak axis?
Response 14: Figure 4-11 in calculation 16-003, RO illustrates the total bending stress in the girt under HELB pressure for various rotation angles using hand calculations.
Note that the analysis in this figure is based on use of the plastic section moduli of the channel and the assumption that bending is decoupled from twisting (i.e. warping normal stresses are conservatively ignored). As shown, the total stress reaches the ultimate stress of the channel material at a rotation angle of 20 degrees.
The failure of the girt is due to a combination of high bending strains and buckling into a lateral torsional mode where the compression flange bends about its weak axis. This buckling mode is the result of the excessive span of the girt and its low weak axis bending which results in low capacity in lateral torsional buckling.
It is to be noted that the previous vendor testing of the siding panels did not involve representation of the girt design at CNS and therefore did not simulate the C10 channel response.
A35 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A
- 2. Failure of the bolts (Page 47)
As stated on Page 30, of the analysis, bolt forces were obtained from Model No. 2 with the bolts modeled as beam elements. This resulted in obtaining shear and tensile forces in the bolt. The tensile forces were attributed mainly to bending in the bolts.
However, as noted during RES staff review, it is more likely that the bolts undergo shearing and pure tensile forces due to the twisting of the girt. Additionally, RES staff noted that, the capacity of the bolt is based on the threads included in the shear plane and stress concentration factors applied to the bolt because of the presence of bending moments.
RES staff review also noted that, Enter1gy's previous calculation (No. NEDC 13-028 Rev. 1) assumed that the bolt was not fully threaded and the threads were excluded from the shear plane. Specifically, the bolts were applied snug tight thus ensuring that there was full contact between one of the legs of the angle and web of the channel. For bending to occur in the bolt, the two surfaces must have separated under the loading, significant enough to cause bending in the bolt.
Question 15: Please provide a detailed description of the analysis of the bolting forces including thread inclusion in the shear plane and stress concentration factors applied to the bolt.
Response 15: Please refer to response to Question 6a for description of the bolt threaded length where it was demonstrated that the bolts are fully threaded in the shear plane and at the nut attachment point. Refer also to response to Question 8 for description of the bending mechanism in the bolts. The bolt is modeled with several beam elements and the maximum moment is obtained and used to calculate the bending stress using the tensile area of the threaded plane.
Peterson ("Stress Concentration Factors", 1974 edition - Ref. 6 of calculation 16-003, RO) provides a study of stress intensification in threaded joints. The study is primarily based on review of results of load testing of bolts and measurements of threaded section stresses using photoelasticity. The study presents data from several sources and various stress concentration values are reported based on stresses calculated using the nominal diameter and also based on stresses calculated at the root diameter.
The stress intensification value of 2.5 used herein is a lower bound of the values reported in the study and is therefore considered acceptable to predict bolt failure.
A36 of 44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A Additional question (with two parts)
Figure 4-5 shows the change in the maximum weld stress for the near and far welds. For the near weld, Figure 4-5 also shows the maximum stress calculated excluding the bearing stress as well as the maximum stress calculated excluding the bearing stress and after failure of the far weld.
Question 1 on Figure 4-5 Describe the calculation of the weld stress excluding the bearing stress prior to the calculated failure of the far weld. In particular, clarify if this recalculation involved identifying the bearing stress from the results of the finite element analysis and then excluding it from the maximum stress shown In the curve labeled near weld. If so, please describe how the bearing stress was identified in the finite element results (from which model of the four models), calculated and then excluded from the maximum stress shown by the purple curve. If possible, provide an example to show how the calculation of the weld stress and bearing stress is performed to obtain the final values on the curves. Clarify also if another finite element analysis or model were involved.
Response to Question 1 on Figure 4-5:
First, a clarification of the graph in Figure 4-5 is presented as follows:
- a. Solid blue curve is for the near weld stress obtained at ANSYS node number 29763 using Model 1 (ANSYS input file Span North3).
- b. Solid magenta curve is for the far weld stress obtained at ANSYS node number 32767 using Model 1 (ANSYS input file Span North3).
- c. Solid green curve is the same as item "a" but the Fz force component (indicating compression between the angle and column) is excluded in calculating the weld stress and hence the weld stresses are lower.
- d. Solid orange curve Is for near weld stress obtained at ANSYS node number 29763 using Model 3 (ANSYS input file Span North4) which simulates failure of the far weld at pressure value of 0.25 psi. Fz force component is excluded as it indicates compression or bearing stress in the weld.
The weld is modeled by Ux, Uy, and Uz restraints of the nodes in the back of the angle along the weld lines (refer to Figure 8-1 herein). The resulting weld forces Fx, Fy, and Fz are obtained at each node along the weld lines. In the ANSYS model, x is along the girt, y is up, and z is normal to the siding into the building. The peak force values along the weld lines occur at the lower ends of the welds. The analysis of weld stresses is performed in ANSYS /POST26 and Excel. ANSYS /POST26 is used to obtain the forces Fx, Fy, and Fz at the designated nodes. Excel is used to combine the forces divide by the weld area to obtain stresses.
A37 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A The weld stress analysis conservatively skips the bottom node where the weld forces are highest and uses the second node from the bottom to define peak weld stress. The weld stress, as explained in the response to Question 8, is calculated as the SRSS combination of the three forces divided by the weld throat size and the tributary length associated with the node (I.e. distance between midpoint on both sides of the node as shown in Figure 8-1 herein).
The direction of the force component that is normal to the siding, Fz, is used to determine if the weld is in compression or tension in that direction. A state of compression Implies that the angle is bearing against the flange of the column where the load path for Fz is presumed to be by direct bearing of steel members and thereby bypasses the weld. This occurs at the near weld. The far weld is on the other hand under tension where the angle is pulling away from the column. Thus, for the near weld, where Fz indicates compression, this force component is not considered in the evaluation of the weld stress. If Fz is included, weld failure would be unconservatively predicted at lower pressure. The sign of Fz from the finite element analysis is consistent with this explanation (being negative at the far weld, for tension, and positive at the near weld, for compression or bearing).
Question 2 on Figure 4-5 After the calculated failure of the far weld, the stress in the near weld excluding bearing stress increases quickly after 0.25 psi. Describe the calculation of the bearing stress after the far weld falls, Including why the bearing stress decreases and how it is excluded from the total stress. Clarify if another finite element analysis was done, that is, clarify if the results of a single finite element analysis, that is, a single nonlinear finite element analysis using the same model and increasing pressure load intensity were used to derive all curves for the near weld stress, or the same finite element model but with changed boundary conditions or loading conditions? Again, if possible, provide an example to show how the hand calculation of the weld stress, bearing stress and total stress is performed by using the data from FEA.
Response to Question 2 on Figure 4-5.
The following ANSYS commands were used to retrieve the forces in the welds:
/post26 rforce,2,29763,f,x rforce,3,29763,f,y rforce,4,29763,f,z rforce,5,32767,f,x rforce,6,32767,f,y rforce, 7,32767,f,z A38 of 44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A The arrays resulting from these commands Include the history of the forces at the designated weld nodes. Arrays 2, 3, and 4 are for a peak stress point in the near weld and arrays 5, 6, and 7 are for a peak stress point at the far weld. The results for arrays 2, 3, and 4 are shown below for load steps from 2 to 3 which correspond to the application of the HELB pressure:
Load Step Fx Fy Fz 2
-162.313 17.1539 51.1783 2.0904
- 23.8268 9.15582 315.722 2.1904 41.9122 45.5778 354.099 2.2904 47.3186 64.7878 383.587 2.3904 41.3685 16.108 437.92 2.4904 29.0532
-67.2244 503.151 2.5904 5.34804
-162.438 564.27 2.6904
-44.4952
-259.342 614.466 2.7904
-128.059
-3 77.186 656.353 2.8904
-257.378
-535.507 679.82 2.9904
-403.651
-702.821 692.716 3
-414.938
-718.863 693.783 As can be seen, the sign of Fz is positive indicating that the reaction at the back of the angle is acting towards the inside of the building and therefore is indicative of compression or bearing.
Therefore, it is ignored in the calculation of the weld stress.
These arrays are then imported into Excel for calculation of the stresses. The load step column is adjusted to initialize the starting point to zero (this done by subtracting 2 from all values) so that the array indicates the fraction of the maximum HELB pressure load of 0.5 psi. The following illustrates the weld stress calculation at load step 3 with and without Fz:
Weld stress (near weld) with Fz
= ((-414.938)2 + (-718.863)2 + (693.783)2)05 I (0.707 x 0.25 x 0.071093/2)
= 172,182 psi (last value of the solid blue curve)
Weld stress (near weld) without Fz = ((-414.938)2 + (-718.863)2)0-5
/ (0.707 X 0.25 X 0.071093/2)
= 132,109 psi (last value of the solid green curve)
The above demonstrates that if Fz is not excluded in the stress calculation, the weld stress will be higher and failure HELB pressure will be underestimated.
As explained in Question 1 of Figure 4-5 response, Model 1 is used to obtain the weld stresses assuming no failure (solid blue, magenta, and green curves). Model 3 is used A39 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A to simulate response when the far weld fails (solid orange curve). Far weld failure occurs at 0.25 psi (0.5 x 0.5 psi) based on results of Model 1 (solid magenta curve).
Thus, in the analysis of Model 3, the far weld is removed from the model at 0.25 psi pressure. This causes sharp increase in the stresses of the near weld (solid orange curve) because all reactions are now supported by one weld rather than two welds. The solid orange line is based on excluding Fz from the weld stresses.
A40 of 44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A AN SYS ANALYSES
- 1. Do any of the computer models, used in the ANSYS analyses, account for the restraint against twisting of the girt provided by the siding?
Response to ANSYS Analysis Question 1:
No. It is assumed that the siding being 18 gage and connected with two screws per panel is flexible and does not possess enough stiffness locally in the vicinity of the fasteners to have significant influence on girt twisting. See response to Question 6 for additional explanation.
- 2. The pressure loading applied to the girt via the siding is applied as a uniform surface load that is always perpendicular to the flange of the girt. Is this a true representation of the pressure loading transmitted to the girt? Pressure on the siding is applied to the girt as point loads through the screws attaching the siding to the girt. Is the uniform loading a true representation of the loading on the girt?
Response to ANSYS Analysis Question 2:
The models used for the girt between column lines D, E, and F use ANSYS beam elements with the HELB pressure applied as side pressure (see Figure 10-1 herein).
The model used for the girt between column lines C and D, the girt is modeled with solid elements and the HELB pressure is applied as nodal forces on a line at the center of the flange (see Figure 10-2 herein).
The siding screws are 2 per panel and the panel is 12" wide. So, there is a screw every 6" along the girt. Over an approximate girt span of 24 ft, and total of 48 fasteners, this distribution of the screws is considered to be closely spaced.
Therefore, the pressure load transferred to the girt can be adequately represented by uniformly distributed load.
- 3. On page 20 of the calculation, it states that the deadweight is fully applied to the girts in models 1 and 2 and not in 4. Results for models 1 and 2 are supposed to form the bounding cases. Are the stresses in both the welds and the bolts also the bounding cases?
Response to ANSYS Analysis Question 3:
We consider models 1, 2, and 4 as bounding for bolt and weld stresses based on the difference in modeling and load application.
The models included in the calculations are explained in the table below.
A41 of 44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A ANSYS Input Model Description File Name Model Span North3 Girt modeled with beam 1
elements**, bolt modeled with solid elements, two weld lines are present, dead weiQht included.
Model Span North qb Similar to Model 1 but 2
with bolts modeled with beam element to facilitate extraction of bolt moments, shear and axial forces.
Model Span North 4 Similar to Model 1 but 3
with far weld removed at 0.25 psi pressure.
Span North 5 Girt modeled with solid elements at mid span and beam elements for the remainder. Girt slides at support.
Deadweight excluded.
Model Span North t Full girt length modeled 4
with one bolt with solid elements, deadweight excluded.
Purpose Used to obtain weld stresses Used to obtain bolt stresses Used to obtain far weld stresses considering failure of the near weld Girt stability analysis post weld failure Used to obtain weld and bolt stresses
- Girt at connection and angle are modeled with solid elements.
Girt Results Location Between Section Column 4.2 Lines D, E, and F Between Section Column 4.2 Lines D, E, and F Between Section Column 4.2 Lines D, E, and F Between Section Column 4.3 Lines D, E, and F Between Section Column 4.4 Lines C and D
- 4. Would it be possible to include the siding in the models and apply the pressure to the siding?
Response to ANSYS Analysis Question 4:
It would be possible to expand the model and include the siding. Some items for consideration would be:
A42 of44
LPI Letter Report A16254-LR-001, Rev. 0 Attachment A The tributary length of the siding for a single girt would be included {for 7ft girt spacing, a 7ft width of the panel should be included in the model). Shell elements can be used for the siding and pressure would be applied directly on the shell surface. The screws would be simulated to establish the interface with the girts.
Detailed mesh would be required in the panels and girt at the screw connections
{total of 24 connections for the half symmetry girt model). Additional mesh refinement in the panels would be required to establish the mesh transition away from the screws. In addition to the plastic properties in the girt, plasticity of the panel material would need to be included since the panel material at the screw connection would buckle and exceed yield. The resulting model size, load step size, and execution time would be significantly larger than in the current model.
It was concluded the complexity of such a model could be simplified as per the current models utilized and considered to bound the field conditions, and provide equivalent or conservative results to that of such a described complex model.
- 5. Assumptions 2 and 7 {on page 13 of the calculation) seem to contradict each other.
Is one used to predict the upper bound capacity and the other the lower bound?
Response to ANSYS Analysis Question 5:
Assumption 2 is related to the location on the girt flange where the HELB pressure load is applied. This load should be applied at the location of the panel screws.
Since the shear center of the girt channel is above the inverted channel, the higher the pressure load application point ( on the girt flange towards the web) the less the eccentricity of this load and the less the torsion associated with this load. Therefore, the pressure load was assumed at half the width of the flange. Based on field observations (see calculation 16-003, RO Figures 2-3, 2-4) the screws appear to be mostly below the midpoint of the flange which would result in higher torsional loads, than considered in the model. This is considered as a conservative load application point.
Assumption 7 is related to the deadweight of the panels and the associated eccentricity of that load with respect to the shear center of the channel. The resulting torsion from the dead weig1ht of the panels act in the direction opposite to the twisting moment of the HELB pressure. Thus Assumption #2 and #7 are not considered to contradict, they are both considered in the modelling approach.
A43 of 44
LPI Letter Report A 16254-LR-001, Rev. 0 Attachment A The resulting state of stresses in the weld and bolts is complex and therefore the approach elected was to perform the girt analysis for two girt models with and without the deadweight effect, as explained earlier.
A44 of 44
LS2020018 Page l of l Enclo ure 6 PPD Engineering E aluation (EE)13-041, Revision 3, "Turbine Building Blowout Panel / Metal WaU ystem" (39 page )
~
NUCLEAR QUALITY RELATED 3-EN-DC-115 I REV. 15C6
~ Entergy MANAGEMENT MANUAL INFOffMATIONAL USE Engineering Change Process ATTACHMENT 9.1A ENGINEERING CHA.NGE COVER SHEET SHEET 1 OF 2 jec# I (EE)13-041 I Rev. No. 3 I
Page 1 of 21 I
Created from:
ECR 11156311 EC Type:
EVAL SubType.
Engineering Evaluation Proj. Pri:
None Facility:
CNS EWO#:
None System:
Turbine Building/Structure Outage: (YIN)
NO WIO Required:
NO Alt Ref.:
None Temp?:
Keywords I
Setpoint? (YIN)
N IO&M Mod?I N Software?
N Cyber? N Quality y
Proprietary N
(YIN)
(YIN)
(YIN)
Related? (YIN)
Info? (YIN)
I Safoguards? l NI EO?
I N
I I
I I I I I I I (YIN}
(Y/N)
Title:
Turbine Building Blowout Panels I Metal Wall System 11 Summary I
The north and south Turbi ne Building panel walls have been evaluated to blowout at 0.5 psld, as stated In FSAR Amendment 25. The resulting blowout panel wall debris will not affect the surrounding.structures and equipment.
0 Vendor EC NIA Vendor company name/Contract #/Data Resp. Engr. I Taylor Sutton I Date: I if~
lb
~!,~.1/4-Preparer I
REVIEW I
0 Technical Reviewer Date:
~ Design Verifier r-'ltH&t,.> N,tvittbtr Mu/11 l,o, fl_,.,",t..f/,_
Date: 2-24 * /6
~ EQRT Review l<O'~"" ti\\ f ~R.AQ.<\\ \\ \\)_
~
, m
./.
~ -
Date: '"3' --c1-,f, 0 Reviewers (See Attachment 9.21) (P)13 APPROVAL IE Supervisor t:'\\\\..,..l.. JH,lt-QJ-,e~
Date:
,-~:i-I?
I~'<:.
"J, j <,- / t.
0 OED Manager Date:
0 GMPO Date:
Expiration Date = (Four (4) years after approval data Date:
ATTACHMENT9.1A SHEET20F2 Copy Distribution:
[X) Master Data Group
[X] Originator NUCLEAR MANAGEMENT MANUAL
[X] Maintenance Rule Group
[X] Licensing
[X] Work Planning QUALITY REl ATED I FORMATlO Al. USE
[X] Simulator Services
[X] System Engineer
[X] Training DERC 3-EN-DC-115 REV. 15C6 ENGINEERING CHANGE COVER SHEET IXI wee (if Field Work Required)
I I SRAB (Required if TS change or NRC approval is needed)
EC # (EE)13-041 I Revision 3 I
Engineering Evaluation I Page3of 21
Title:
Turbine Building Blowout Panels / Metal Wall System Table of Contents Section Description Page
- 1.
Description......................................................................................... 4
- 2.
Documents........................................................................................ 6
- 3.
Evaluation/Design Summary....................................................... 8
- 4.
Impact on Current Operational Basis........................................ 20
- 5.
Engineering Requirements Necessary to Implement the Change....... 20
- 6.
Engineered Materials List and Vendor Technical Information........... 20
- 7.
Special Process Requirements.......................................................... 21
- 8.
Tests & lnspections............................................................................. 21
- 9.
Attachments.....
.21
EC # (EE} 13-041 I Revision 3 I
Engineering Evaluation I
Page 4 of 21
Title:
Turbine Building Blowout Panels / Metal Wall System
- 1. Description Girt Spacing 7'
1.1. Structure, System, or Component Description Turbine Building The Turbine Building is designated as a Principle Class II Structure [Ref.
2.3.31).
The Turbine Building houses the turbine-generator and associated auxiliaries i11cluding the Condensing, Feedwater, and Water Treatment Systems. Space Is provided in this building for other auxiliary power plant equipment. The Water Treatment Area, Machine Shop, Exhaust Fan Room and Heating Boiler Room are located adjacent to the Turbine Building and are referred to as the Turbine Building appendages [Ref.
2.3.32).
The Turbine Building is a reinforced concrete structure up to the operating lh;\\
floor (EL. 932'-6"). Concrete shield walls surround the turbine-generator
~
and structural steel framing rises above the operating floor. The building is enclosed with insulated metal siding and roofing. The interior walls are reinforced concrete with concrete block enclosing smaller areas. The Turbine Pedestal is a massive reinforced concrete structure supported by the same foundation mat as the building [Ref. 2.3.32].
Girt Bay 23'11"
"\\
Column I
I I
I I
I I
I I
I I
I I
I t
I I
I I
t I
I I
I I
I I
I t
I I
I I
I I
I I
I I
I I
I t
t t
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
t I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I J
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I t
I I
I I
I Figure 1: Typical (Simplified) Wall Section View I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I A I
I I
I I
I I
I I
I I
I :y I
I I
I
[
I I
I I
I I
I I
I I
I I
I I
I I
I I
I k'L
.Ill.
/
/
V V
C onnection Angle Panel
/
V Structural Girt
EC *# (EE)13-041 I Revision 3 I
Engineering Evaluation I
Page 5 of 21
Title:
Turbine Building Blowout Panels / Metal Wall System Metal Wall Panel System Interior Pressure t
I t
I I
I I
I t
I Structural Girt I
I Connection Angle Column
/i i i i i i i i i Figure 2: Typical (Simplified) Girt Top View Figure 3: Typical 5w~ <.lr:t Metal Wall Panel Cross Section Exterior Panel Interior Panel Structural Girt Figures 1 and 2 depict the typical layout of the north and south Turbine Building walls between column lines D, E and F. Figure 3 depicts the typical assemblage of a metal wall panel on the north and south Turbine Building walls.
1.2 Reason for Change/Evaluation EE 13-041 Revision 2 During a High Energy Line Break (HELB), such as a Main Steam Lioe Break, the Turbine Building will pressurize. This event will lead to over pressurization of the Control Room, and other environmentally sensitive I areas (DG Rooms), if the pressure is not relieved. The peak pressure that the panel wall system (metal siding) can achieve is stated in Final Safety Analysis Report (FSAR) Amendment 25. This Engineering Evaluation looks at the north and south walls of the Turbine Building to determine the failure mechanism that will control the release of the panel wall system when 0.5 psld pressure Is present in the Turbine Building.
EE 13-041 Revision 3 The Nuclear Regulatory Commission (NRC) identified a Green Non-Cited Violation (NCV) of 10 CFR Part 50, Appendix B, Criterion Ill, "Design Control", on NEDC 13-028, Revision 1 [Ref. 2.3.42]. In response to this NCV, a consulting engineering vendor was hired to perform a finite element analysis (FEA) of the Turbine Building blowout panels. The analysis is captured in NEDC 16-003, Revision 0. This EE revision
EC# (EE)13-041 I Revision 3 I
Engineering Evaluation I Page6of 21 TIiie: Turbine Building Blowout Panels / Metal Wall System incorporates the results from NEDC 16-003, which replaces the previous analysis (NEDC 13-028) as the design basis analysis.
The generation of these calculations was dnven by the requirement of maintaining design basis documentation that supports FSAR Amendment
- 25. The ongmal Turbine Building siding blowout analysis could not be located; new analysis was performed to substantiate the 0.5 psi blowout pressure.
1.3 Design Oblective to Resolve Problem This Engineering Evaluati'on is an assessment of the weak link and subsequent internal pressure that can be maintained as well as the impact of the effects of blowout on the surrounding structures and equipment.
FSAR Amendment 25 states a value of 0.5 psid for the pressure at which the building siding fails. This evaluation conservatively confirms that panel wall systems on the north and south Turbine Building walls will fail at 0.5 psid.
- 2. Documents 2.1 Affected Documents List There-are-AOdoouments--affeGted by this Engineering Change,, This EC implements NEDC 16-003. See Attachment 9.2 for the list of identified affected documents.
2.2 Affected Equipment List There are no SSCs or equipment affected by this Engineering Change.
2.3 Reference Documents 2.3.1 2.3.2 2.3.3 2.3.4 2.3.5 2.3.6 2.3.7 2.3.8 2.3.9 2.3.10 NEDC 13-028, Rev.2; "Ultimate Internal Pressure of Turbine Building Blowout Panels and Metal Wall System" FSAR Amendment 25 DWG. 4088, Sht.1, Rev.OD, Ver.AA DWG. 4089, Sht.2, Rev.OD, Ver.AA DWG. 9150, Sht.E106, Rev.OD, Ver.AA DWG. 9150, Sht.E107, Rev.OD, Ver.AA DWG. 9150, Sht.10, Rev.OD, Ver.AA DWG. 9150, Sht.133, Rev.OD, Ver.AA DWG. 9150, Sht.156, Rev.00, Ver.AA DWG. 9150, Sht.157, Rev.DO, Ver.AA
EC# (EE)13-041 I Revision 3 I
Engineering Evaluation
Title:
Turbine Building Blowout Panels / Metal Wall System 2.3.11 DWG. 9150, Sht.158, Rev.OD, Ver.AA 2.3.12 DWG. 9150, Sht.159, Rev.00, Ver.AA 2.3.13 DWG. 49054, Sht.2N, Rev.00, Ver.AA 2.3.14 DWG. 49054, Sht.2R, Rev.00, Ver.AA 2.3.15 DWG. 49054, Sht.2T, Rev.OD, Ver.AA 2.3.16 DWG. 49054, Sht.2W, Rev.00, Ver.AA 2.3.17 DWG. 49054, Sht.6T, Rev.00, Ver.AA 2.3.18 DWG. 49054, Sht.7T, Rev.OD, Ver.AA 2.3.19 DWG. 49054, Sht.8T, Rev.00, Ver.AA 2.3.20 DWG. 49054, Sht.9T, Rev.OD, Ver.AA 2.3.21 DWG. 49054, Sht.9W, Rev.DO, Ver.AA 2.3.22 DWG. 49054, Sht.10T, Rev.00, Ver.AA 2.3.23 DWG. 49054, Sht.10W, Rev.00, Ver.AA 2.3.24 DWG. 49054, Sht.11T, Rev.00, Ver.AA 2.3.25 DWG. 49054, Sht.12T, Rev.00, Ver.AA 2.3.26 USAR Vl-4.3, Rev.loep.xxvii4 2.3.27 USAR Vl-4.4, Rev.loep.xxvii4 2.3.28 USAR Xl-2.4, Rev.loep.xxvil4 2.3.29 USAR Xl-9.3, Rev.loep.xxvil4 2.3.30 USAR Xll-2.1.2.3, Rev.loep.xxvii4 2.3.31 USAR Xll-2.1.3.1, Rev.loep.xxvii4 2.3.32 USAR Xll-2.2.2, Rev.loep.xxvil4 2.3.33 USAR Xll-2.2.3, Rev.loep.xxvli4 2.3.34 USAR Xll-2.2.6, Rev.loep.xxv114 2.3.35 USAR Xll-2.3.3.1, Rev.loep.xxvii4 2.3.36 USAR Xll-2.3.3.2, Rev.loep.xxvii4 2.3.37 USAR Xll-2.3.5.1.3, Rev.loep.xxvli4 2.3.38 USAR XIV-6.5.3.1.i, Rev.loep.xxvll4 2.3:-39 T eGhnical ~pacification 3.5.2 2.3.40 Bums and Roe Calculation Book 58 2.3.41 CR-CNS-2015-05705 2.3.42 CR-CNS-2015-06544 I
Page7of 21 2.3.43 NEDC 16-003, Rev.O; "Structural Evaluation of the Turbine Building Blowout Panels Steel Supports" 2.3.44 Burns & Roe Inc., "Computer Analysis of CNS Multi-Compartment Pressure History Main Steam, Feedwater, and Extraction Line Breaks In the Turbine Building", Dated 10-23-73, Media ( 1}
08317-1718 or (2) 64158-1094 Other applicable calculation references are contained in NEDC 13-028, Revision 1, and NEDC 16-003, Revision 0.
EC # (EE)13-041 I Revision 3 I
Engineering Evaluation
Title:
Turbine Building Blowout Panels / Metal Wall System
- 3. Evaluation/Design Summary 3.1 Evaluation Resolution I Page 8 of 21 To remedy the missing design basis document that analyzed the Turbine Building siding blowout pressure, CNS performed a simplified hand calculation. NEDC 13-028 contains that analysis, and it concludes that failure of the panel system is achievable. However, the analysis that was performed was disputed by the NRC, which eventually resulted in CNS receiving a NCV. Therefore, a finite element analysis was performed to validate FSAR Amendment 25. NEDC 16-003 contains that analysis, and it also concludes that failure of the panel system Is achievable. NEDC 16-003 is the design basis calculation of the Turbine Building blowout panels.
- o.
Figure 4: Location of Blowout Panels t fAAll (0 "'~"st -cM1i1, M1 ta.mt llOG
!tSCO ~
SlOIRAC(fA
-~~riJILu.
~
( ""°"
Calculation NEDC 1-3-028 16-003 concludes that the wall system will fail in the middle of the north and the south walls, as noted in Figure 4. (The east and west walls of the Turbine Building have more robust connections to the girts and will require a higher internal pressure for the connecting elements to fail.) The calculation al-68 validates that the walls will fail at the FSAR Amendment 25 pressure value of 0.5 psi.
EC # (EE)13-041 I Revision 3 I
Engineering Evaluation I
Page 9 of 21
Title:
Turbine Building Blowout Panels / Metal Wall System NEDC 13-028 Summary*
The internal pressure acting on the panels puts the girts in bending. The connection angles have smaller plastic moment capacities compared to the girt channel. After the connection angles deform plastically, plastic hinges form that induce catenary behavior in the connection angles. A catenary is formed by lack of flexural strength such as is present in a string or suspension bridge. This is reasonable as deflection will continue to increase with no increase in load. The catenary action leads to a much larger tensile force that must be resisted by the bolted connections between the girt channel and the connection angles. In the end, the bolts fail due to excessive loading which results in the blowout (release) of the structural girt with the panels attached.
The loss of a connection angle (by way of bolt failure) causes a redistribution of the loads on the whole girt bay. Loads supported Initially by the broken connection are distributed to the connections above and below, which then fail under increased load. The bays that will experience panel release by the HELB pressure (0.5 psi) on the north and south walls are between column lines D, E, and F. As a result of the panels blowing out, a pressure relief area is created in the Turbine Building walls. The calculation contains a simple evaluation of the total blowout area, which was determined to be 6,576 tt2.
As stated previously, a NCV was received on this calculation. Therefore, NEDC 16-003 was perfomied to replace the analysis used in this calculation. NEDC 13-028, however, will not be superseded but will be revised to non-design basis status. The calculation will be retained for the purpose of maintaining a history of the design control analysis of the Turbine Building blowout panels.
NEDC 16-003 Summary:
The purpose of this analysis is to predict the failure of the structural supporting system of the siding (panels) with respect to the stated value of 0.5 psi in the FSAR Amendment 25. For failure to initiate In the support system, the load path in the girt channel beam and its corresponding end connections were investigated. The analyses of the girt channel, connecting angles, bolts and welds were performed by means of using the ANSYS computer modeling software, which performs finite element analysis.
The FEA was performed by application of the pressure loading In a *static manner. Material properties of the structural elements were included in the model in a non-linear fashion, crediting the materials yield point, and the material behavior following the onset of yielding. In the analysis, the
EC # (EE)13-041 I Revision 3 I
Engineering Evaluation I Page 10 of 21
Title:
Turbine Building Blowout Panels / Metal Wall System load was applied in a load step manner, where the load was increased with each subsequent load step. This approach was used to develop the response curve of the girt and the angle connections, considering the non-linear response of the material as the connection was subjected to the applied loads.
Hand calculations were performed to determine deadweight and the HELB pressure load acting on the girt and angle connections. Both the, lateral HELB pressure and vertical deadweight are eccentric with respect to the girt channel section. These eccentric loads result in torsional moments on the girt; these loads were included as inputs to the ANSYS analysis.
(Appendix B contains the hand calculations.)
In addition, bolt shear testing and structural steel Certificate of Material Test Reports (CMTRs) were used as inputs in the determination of the ultimate strength of the bolts and supporting structural steel members, respectively, in the ANSYS models. For a failure analysis, the use of actual material strength values versus minimum specification values is necessary to better predict failure. (Appendix A contains the CMTRs.)
Two failure mechanisms were determined to control the blowout panels release on the north and south walls between column llnes D, E and F: (1) bolt failure is to will occur in the shear plane between the girt channels and the connection angles, or (2) weld failure will occur between the connection angles and columns which will lead to lateral torsional buckling of the girt channel. This failure mechanism controls-the release-On the elGweut-f}aAelS:- Both of these failure mechanism can initiate at about the same HELB pressure load level; however, only one mechanism will occur, which will lead to failure of the supporting structural system and the blowout of the panels.
The mechanism for bolt failure involves failure of the outer bolt at about 0.25 psi due to bending stresses that result from rotation of the channel.
The outer bolt failure is rapidly followed by failure of the inner boll, which is subjected to a rise in bending stresses after the failure of the outer bolt.
After failure of both bolts occur, the supporting steel girt channel becomes unsupported and will blowout under the lateral pressure load.
The mechanism for weld failure begins with failure of the fillet weld farthest from the girt at about 0.25 psi. The other fillet weld, the near weld, will fall at about 0.38 psi. The response of the girt following the failure of both welds involves significant twisting of the channel and lateral torsional buckling at 0.425 psi. With the failure of the welds and the rotation of the girt, a plastic hinge Is formed In the mid-span of the girt. In this state, the bending strains at mid-span increase significantly and reach failure limits,
EC# (EE)13-041 I Revision 3 I
Engineering Evaluation I Page 11 of 21
Title:
Turbine Building Blowout Panels / Metal Wall System causing the complete collapse of the g,rt until the ends push passed the column flanges.
The failure mechanism for girts between column Imes C and D on the north wall is similar the mechanisms discussed above where failure initiates at the connection at column D by bolt failure and/or weld failure.
The connection at column C is considered rugged. Post connection failure is similar to the failure mechanisms discussed except for the girt reacts to the pressure load as a cantilever and fails in bending at column C.
The pressure values for the bolt failure mechanism and the weld failure mechanism were determined by analyzing a 7 ft girt spacing. To account for smaller girt spacings in the north and south Turbine Building walls, a limiting spacing of 6 ft - 7 in was calculated with a corresponding blowout pressure of 0.45 psi. The smaller limiting spacing allows for more blowout area to be created.
The Turbine Building bays that were investigated in NEDC 16-003 are between column lines D, E, and Fon the north and south walls, and between C and O on the north wall. The resulting blowout area that is achieved by the release of the panels in these bays was calculated to be 5,362 ft'. This blowout area value, in comparison to the value obtained In NEDC 13-028, is considered to be the controlling blowout area. A more in depth analysis was performed in NEDC 16-003 for the determination of the controlltng failure mechanisms in comparison to the techniques used in NEDC 13-028.
Commonalities between both Ca/culat,ons:
Localized dynamic loading was not considered In NE-OG-1-3-028.
Localized pressures from a HELB event could break the panel fasteners, which would result in releasing the panels. If this were to occur, it would establish a vent path sooner than what is calculated. From the point of view of the HELB event, ignoring dynamic effects is conservative.
After the first connection falls, there will be a redistribution of load, which in turn, will causes other more bolts to shear girt connecting elements to fail. This affect is the result of increased tributary area on the neighboring girts. Sag rods tie the girts to each other from the lowest girt to the celling.
These rods are supported from above in the eave strut. Dependent on how tightly fastened and where they are located in the girt web, the sag rods may provide dead load and lateral support of the panels and girts.
Both calculations discuss the sag rods; however, they were analyzed differently in each calculation.
The interior and exterior panels do not fail within this evaluation. The panels were subjected to a vacuum test (retrieved from microfilm and is
EC # (EE)13-041 I Revision 3 I
Engineering Evaluation I Page 12 of 21
Title:
Turbine Building Blowout Panels / Metal Wall System attached to oalculaoon NEDC 13-028) that demonstrated the load carrying capacity of the IW21 A Aluminum wall panel. The failure mechanism that was ultimately realized in that test was sheet pullover of the exterior panel at an applied pressure of 0.5417 psi (78.0 psf). The liner (interior) panels, as tested in the test setup, do not fail and remained undamaged. The underlying failure of the panels In this evaluation and the associated calculations is the failure of the supporting structural system.
The Turbine Building~ concrete walls and metal roof were not considered.
Using engineering judgement, they are not the weakest components of the building envelope. An alternate blowout location, such as the Office Building or Turbine Building ground level exit, does not offer a large enough vent path and so are conservatively Ignored.
Panel Blowout Impact Concerns.
Howo¥or, It is foreseeable that unusual conditions may cause panels to separate from the main mass of the wall. The resulting pieces would be small and would fly anywhere within a wide arc of the breach. It is also foreseeable that larger sections of the wall could break off. These larger pieces would not travel far and may land close to the north or south walls.
In addition to wall debris, small, lightweight, loose items within the Turbine Building may also be swept up in the resulting airstream and become airborne.
As stated in USAR Xll-2.3.3.2.1, all Class I structures are designed for tornado loading. Also, USAR Xll-2.3.3.2.2 states Class I structures are designed for tornado generated missiles. Any missile generated by release of the blowout panels will have less kinetic energy than a 35' utility pole with a velocity of 200 miles per hour due to the significantly smaller mass and slower velocity.
Class I structures include the Service Water Pump Room, the Control Building, the Diesel Generator Building, the Controlled Corridor, and tho Drywell and Suppression Chamber. Not included is the refueling floor which is enclosed in a metal wall similar to the Turbine Building. This part of the structure is not located near the blowout, so it is protected from damage by separation.
Diesel fuel oil storage tanks are classified as Class I equipment as stated in USAR Xll-2.1.2.3. The diesel fuel oil storage tanks are protected from tornado missiles by a double manhole and a soil cover. They are not affected by airborne debris.
USAR Section Xll-2.2.6 states that the ERP was not designed for tornado loading. It is not required for safe shutdown. It's failure does not cause an accident.
EC # (EE)13-041 I Revision 3 I
Engineering Evaluation I Page 13 of 21
Title:
Turbine Building Blowout Panels/ Metal Wall System While the Emergency Station Service Transformer (ESST) and Start-up Station Service Transformer (SSST) are not located directly adjacent to the walls that blow out, it is possible that panel fragments or other airborne missiles could damage the transformers or short the buses. Per USAR XIV-6.5.3.1.i, during a main steam line break accident, the plant assumes a Loss Of Offsite Power (LOOP). The Emergency Station Service Transformer (ESST) and Start-up Station Service Transformer (SSST) are both assumed down at this point. Therefore, they are not required.
The Condensate Storage Tanks (CST) provide station system makeup, receive system reject flow, and provide condensate for any continuous service needs and intermittent batch type services. While CST-A i&U6e8 in Tech Spec3.6.2 and is referenced as ECCS suction In USAR sections Xl-9.3, Vl-4.3, and Vl-4.4, both Condensate Storage Tanks A and B are not designed to withstand tornado missiles. +he Tuch Spec and--USAR sections are-under revision to remove-this requirement due lo the lack of seismic qualification of the tanks. CST's are not required after a Main Steam Line Break.
Summary/Conclusions:
The 0.5 psid value stated in FSAR Amendment 25 was shown to be calculated achievable by a simplified analysis (NEDC 13-028) and by a FEA (NEDC 16-003). This value is shown to be cooseJVative bounding when compared with the shearing force to the pressure levels that Is requir,ed can cause feF the controlling failure mechanisms in the system to occur.
The oort-A aAG-60Uth Turbine Building bays that were evaluated for panel blowout are between column lines D, E, and Fon the north and south walls, and between column lines C and D on the north wall. These panels will experience blowout oofere when a HELB pressure of 0.5 psid is aGAieved when-the--boll-oonnestions-fail present in the upper Turbine Building, which will cause the pressure inside the structure is to be relieved to the atmosphere. The controlling blowout area, from NEDC 16-003, is 5,362 ft'.
When the wall panels detach form the Turbine Building, they do not impact structures or equipment important to nuclear safety.
3.2 Licensing and Design Bases Discussion FSAR Amendment 25 states that the Turbine Building siding will blow out at 0.5 psid thereby venting the released steam and steam/water mixture to atmosphere pressure causing complete pressure relief for the confined space (Turbine Building).
EC# (EE)13-041 I Revision 3 I
Engineering Evaluation I Page 14 of 21
Title:
Turbine Building Blowout Panels / Metal Wall System A review of historical records identified a Burns & Roe document, titled "Multi-Compartment Containment Pressure History: Main Steam, Feedwater, and Extraction Line Breaks in Turbine Building" [Ref. 2.3.44).
The document consists of computer software printout files, speciflcally the files document runs that were performed subsequent to the FSAR Amendment 25 submittal. The files appear to use the same computer code that was used to obtain the submittal data, and contain a record of a Turbine Building run that validates the FSAR Amendment 25 pressure value (0.5 psi) for siding/panel blowout.
A single flow path was used in the computer analysis, representing flow from the Turbine Building to the atmosphere. A vent opening area of 5,000 fl? was used to represent the vent path from the Turbine Building upper level to the atmosphere upon blowout of the sidin~ panels.
Therefore, this evaluation must result in at least 5,000 ft of blowout panel area to validate the results of the Burns & Roe analysis and FSAR Amendment 25. NEDC 16-003 does indeed result in moro than 5,000 fl2 of blowout area with 5,362 ft2.
The blowout panel area calculated in NEOC 16-003 is considered to be the controlling blowout panel area value. This conclusion is based on the in depth analysis performed,n the FEA to determine the controlling failure mechanisms that result in the supporting structural members ultimate failure, and subsequent panel release. NEDC 13-028 also contains a blowout panel area calculation; however, the calculated value (6,576 ft2) was based on hand calculations that employed a simpler approach for solving the blowout pressure. Therefore, the blowout panel area contained in NEDC 13-028 Is considered to be inferior, and not the controlling value.
An evaluation of the Turbine Building concrete structure was performed to confirm that it is capable of remaining structurally intact without gross structural failure following a postulated SSE. This allows crediting the dose consequence mitigation assumptions related to leakage holdup and the resulting iodine plateout within the Main Condenser following a postulated Loss-of-Coolant Accident [Ref. 2.3.31). The blowout panels do not impact r the ability of the Turbine Building to resist seismic loads; therefore the iodine plateout assumptions are not impacted.
The Turbine Building is a Class II structure [Ref. 2.3.31] that is designed to withstand a minimum wind velocity of 100 mph [Ref. 2.3.35]. Class I structures are designed to withstand tornado loads, such as tornado missiles [Ref. 2.3.36); the Turbine Building is not designed to withstand tornado loads.
EC# (EE)13-041 I Revision 3 I
Engineering Evaluation I Page 15 of 21
Title:
Turbine Building Blowout Panels / Metal Wall System The Turbine structure is designed to withstand missile effects for normal operation, I.e., the inner and outer casings are designed to resist disk missile impact at more than 120% of rated speed [Ref. 2.3.28].
Burns and Roe Calculation Book 58 contains the structural design of columns, roofing, girts and other wall panel components in Turbine Building. This book was reviewed in the preparation of NEDC 13-028.
The design pressure for wind loading on the Turbine Building siding, in Book 58, is 40 psf (0.28 psi). This pressure Is significantly lower than the blowout pressure (0 28 psi < 0.5 psi) that will cause failure of the supporting structural system (girts). Therefore, a wind load at the design magnitude (100 mph) will not cause a blowout failure of the Turbine Building siding panel support system.
Samples of key Turbine Building substructures (e.g., walls, floor slabs, and columns) were evaluated for Increased seismic loading resulting from a postulated SSE. The horizontal seismic acceleration input to the operating floor of the Turbine Building at elevation 932 ft-6 in. due to the Turbine Building response was assumed to be 0.3g based on a comparison with Class I structures (Reactor Building and Control Building). The evaluations show that the increase in design loadings from the original seismic Class II criteria to the postulated SSE condition do not result in stresses that exceed the allowable limits applicable to the SSE load case. Therefore, it is concluded that there is sufficient margin in the original design to ensure that the concrete portion of the Turbine Building structure will remain intact during and following an SSE. These results are based primarily on the fact that allowable stresses are increased for the SSE load case and, consequently, the increase in seismic loading Is offset by the increase in allowable stresses [Ref. 2.3.32].
The Turbine Building was specifically evaluated for the potential to impact adjacent Class I structures. It was determined that the only zone of the Turbine Building that can adversely affect the integrity of an adjoining Class I structure Is the area in close proximity to the Diesel Generator Building. This particular area has been designed to Class I standards [Ref.
2.3.37].
The Control Building houses instrumentation and switches required for station operation. Also located in this building is a computer room, station batteries, and components of the Reactor Protection System [Ref. 2.3.33].
This evaluation is in support of protecting the Control Building. These other areas are evaluated separately.
The applicable drawings of the Turbine Building panel walls were utilized in NEDC 13-028 and NEDC 16-003, and are listed as references to this I~
document. When possible, the drawings were field verified to validate F
EC# (EE)13-041 I Revision 3 I
Engineering Evaluation I Page 16 of 21 Tille: Turbine Building Blowout Panels / Metal Wall System materials and dimensions of the metal wall system. In such cases, the drawing details did match the field conditions. The drawings that could not be easily verified were assumed to be accurate.
The use of several structural design codes, such as the AISC Manual of Steel Construction 6th Edition, were used, when applicable, to determine the ultimate strength of the metal wall system girt channels, connection angles, bolts and welds. Design codes are typically used to evaluate the acceptability of structural members, and not the point of ultimate f aiture.
This evaluation NEDC 13-028 uses engineering principles when the standard engineering practices (design codes) lack specific design criteria for evaluating material failure. CMTRs and material testing data were used in NEDC 16-003 for inputs to the ultimate strength of the structural members and bolts 1n the analysis 3.3 Design Input Considerations CNS Safety Analysis Report FSAR Amendment 25: 0.5 psid Turbine Building siding blowout pressure Design Codes AISC Manual of Steel Construction, 6th Ed.
AISC Manual of Steel Construction, 13th Ed.
AISI North American Specification for the Design of Cold-Formed Steel Structural Members, 2007 Ed.
Standards ASTM A307: Carbon Steel Bolts and Studs, 60,000 PSI Tensile Strength, 1978 Ed.
Physical/Measured Components
~tr;,uctural Girt Channels: C10x15.3 (A36)
Connection Angles: L5x3-1/2x3/8, L4x3x3/8 (A36)
Connecting Bolts: 3/4" machine bolts (A307, Gr.A)
Angle Welds: 1/4", 5/16" (FE10)
Performance Requirement Turbine Building siding will blowout at the FSAR defined blowout pressure thereby venting released steam and moisture to the atmosphere, completely relieving the confined space.
EC# (EE)13-041 I Revision 3 l
Engineering Evaluation I Page 17 of 21
Title:
Turbine Building Blowout Panels / Metal Wall System Other Documents All other documents are referenced in Section 2.3 or referenced as inputs to the analyses in NEDC 13-028 and NEDC 16-003.
3.4 Discussion of Impact Screen Results See Attached 9.1 for the Impact Screening Summary. A Mechanical Design Engineering review is necessary for Revision 3.
3.5 Relationship with Other Engineering Changes This EC is not associated with any other Engineering Change document.
However, the following discussion has been added for clarification on the different calculations associated with Turbine Building blowout panels and HELB.
NEDC 12-012, "Turbine Generator Building Siding Blowout Pressure, other than EQ Purposes," is superseded by NEDC 13-028, "Ultimate Internal Pressure of Turbine Building Blowout Panels and Metal Wall System." However, NEDC 12-012 supports NEDC 11-075, "Turbine Building High Energy Line Break," which supports a closed CAP item. For this purpose, these two calculations have not been deleted. Calculation NEDC 03-005, "Turbine Generator Building Siding Blowout Pressure,"
which also looks at the blowout of the Turbine Building walls, has been left in place. It provides conservative results to NEDC 02-006, "HELB EQ -
RB Gothic Model." Calculation NEDC 11-122, "Design Check for Door Mark H-307 for an Overpressure of 0.75 psi," evaluates door H-307 for an overpressure of 0. 75 psid. Because the blowout pressure has been determined to be lower than 0.75 psid, there are no pending changes.
3.6 Operating Experience Search Results 3.6.1 The search criterion of "Turbine Building Blowout" was used in the INPO ICES database; there were 21 results using that criterion. Only two reports appeared to be applicable to this evaluation.
3.6.1.1 Report #164109 - Safety Analysis Overpressure Protection Not Provided for the Turbine Building Event Summary: On March 3, 1997, with Vermont Yankee operating at 100 percent power, it was discovered that the design for Turbine Building overpressure protection, as described in the final safety analysis report (FSAR), was never installed. The Vermont Yankee FSAR description of Turbine Building overpressure protection
EC# (EE)13-041 I Revision 3 I
Engineering Evaluation I Page 18 of 21
Title:
Turbine Building Blowout Panels / Metal Wall System Includes a 1,000 square foot blowout panel that would actuate at an internal pressure of 0.25 psi. No venting areas of this size have been found in the Turbine Building, and investigation to date has been unable to identify any other venting methods that may have been assumed, in lieu of the blowout panels, when the plant was constructed. The analysis for high energy line breaks of the main steam and feedwater piping in the Turbine Building credits the blowout panel. Once the blowout panels actuate, a steam cloud would be released In a matter seconds. Absent the blowout panel, the potential higher pressure raised concern for the structural capability of certain block walls and ductwork In operation.
The preliminary FSAR states that steam would travel up through the Turbine Building, partially condensing on the walls and equipment. The increase in pressure in the Turbine Building would result in portions of the siding and roof of the building being tom from the building structural members releasing steam to the environment. It Is not clear from the current FSAR description and that of the preliminary FSAR as to whether the anticipated change to blowout panels was to be made as a result of a decision that an engineered feature needed to be installed or simply a case of interpretation.
Due to the age of this event, it is not possible to determine the root cause for the failure to install the device described in the FSAR. The difference in the terminology between the plant's preliminary safety analysis report (PSAR) and the original final safety analysis report alludes to an apparent cause. The PSAR identified that building overpressure was to be prevented by failure of Turbine Building siding. In contrast, the original FSAR described a blowout panel, indicating that a formally engineered relief path was intended. It appears that a design change was planned which would have Installed a specific relief device in the Turbine Building. However, because the investigation was hampered by the lack of documentation and unavailability of Individuals involved in the early licensing process, it could not be positively determined if the change was simply a change in wording or the result of actual change of intent.
Corrective actions include modification of the vital switchgear room heating, ventilation and air conditioning system to prevent an HELB in the Turbine Building from
EC # (EE)13-041 I Revision 3 I
Engineering Evaluation I Page 19 of 21
Title:
Turbine Building Blowout Panels I Metal Wall System introduction steam into the switchgear rooms and developing operability determinations to support bases for continued safe operation of the plant.
Applicability: The original Vermont Yankee PSAR indicated that the Turbine Building siding would fail and thereby create a relief path for internal pressure to escape to the atmosphere.
3.6.1.2 Report #156158 - Turbine and Reactor Building Blowout Panels Determined to be Outside Plant Design Basis Event Summary: At 1000 hours0.0116 days <br />0.278 hours <br />0.00165 weeks <br />3.805e-4 months <br /> on October 25, 1993, with the reactor at 100 percent power, it was determined that the relief (blowout) panels in the Nine Mlle Point Unit 1 (NMP1 ) Turbine and Reactor Buildings would not blow out at the design pressure of 45 psf because the bolt fasteners for the panels were larger than designed and had a higher ultimate strength than designed. The initial engineering evaluation of this condition erroneously determined that the Turbine and Reactor Building panels would blow out at 60 and 53 psf, respectively, to relieve internal building pressure prior to structural failure of the buildings.
On March 27, 1995, during refueling outage 13 (RFO13),
it was determined that there was an error in the design assumptions for load distribution. Revised calculations confirmed that the relief panels would not blow out until the internal building pressure exceeded the minimum documented building structural design of 80 psf. The cause of this event was inadequate quality control measures during initial construction which resulted in oversize bolts being installed in the relief panels.
Prior to restart from RF013, the panel attachment design was revised, the size of the bolts was verified, and every other bolt was removed from the relief panels to reduce their blowout point to a value below the documented building structural capability. Additionally, appropriate Engineering personnel were counseled regarding verification of design assumptions.
The blowout panel design uses panels of corrugated siding approximately 2 feet high by 20 feet long. The horizontal ends of these panels are firmly fastened to steel angle bars which are then connected to the building
EC # (EE)13-041 I Revision 3 I
Engineering Evaluation I Page 20 of 21
Title:
Turbine Building Blowout Panels / Metal Wall System
- 3. 7 Margin Analysis structural steel by shear bolts. The top and bottom of the individual panels are not structurally fastened to adjacent panels, and, because of the corrugation, the panels are relatively flexible in the vertical direction. For these reasons, the load resulting from an internal pressure is transferred horizontally only, however, the engineer evaluated both a horizontal and vertical transfer of load in the design calculation. As a result, the incorrect calculation conclusions were used to determine the blowout panels would separate from the building structures at internal pressures of 60 psf (Turbine Building) and 53 psf (Reactor Building) by tearing of the panel metal at the top and bottom of the Blowout Panels.
Applicability: The Nine Mile Point Turbine and Reactor Building Blowout Panel analysis determined the limiting failure mechanism of the system to be the bolt fasteners.
The analysis preformed in NEDC 13-028 16-003 determined the metal wall systems on the north and south Turbine Building walls will fail ( create a vent path to the atmosphere) before 0.5 psid pressure is present reached in the Turbine Building. However, the purpose of the calculation, and this EC, Is to reconstitute validate the FSAR Amendment 25 pressure value of 0.5 psid and not create a new value; therefore, margin is available in the calculation to confidently Glaim confirm that the applicable panels will fall at 0.5 psid.
- 4. Impact on Current Operational Basis This Engineering Change does not adversely impact the current operational basis of the plant, but on the contrary, by validating the FSAR Amendment 25 blowout pressure the current operational basis is Fe+RforGeG verified.
- 5. Engineering Requirements Necessary to Implement the Change There are no engineering requirements associated with this Engineering Change.
- 6. Engineered Materials List and Vendor Technical Information 6.1 Engineered Materials None.
EC# (EE)13-041 I Revision 3 I
Engineering Evaluation I Page 21 of 21
Title:
Turbine Building Blowout Panels / Metal Wall System 6.2 Vendor Technical Information The applicable vendor as-built drawings are listed in Section 2.3. Refer to NEDC 16-003 for information supplied by the vendor (LPI, Inc.).
- 7. Special Process Requirements No special process requirements are associated with this Engineering Change.
- 8. Test & Inspections No new tests or inspections are associated with this Engineering Change.
- 9. Attachments 9.1 Impact Screening Summary 9.2 Affected Document Identification List 9.3 10 CFR 54.37(b) Review Determination 9.4 PAD (Revision 2) 9.5 Additional Reviewers Form
-:---Entergy NUCLEAR MANAGEMENT MANUAL EC I (EE) 13*041 Re'ilslon 3 Allildlment 9 1 QuAUTY RELATED INFORMATIONAL Use Engineerina Chanae Process p
f 7 aoc 1 o 3-EN-DC-115 I REV. 15C4 ATTACHMENT 9.3 IMPACT SCREENING
SUMMARY
SHEET 1 OF 7 INSTRUCTIONS:
Identify the potentially impacted and the unaffected areas by checking the appropriate Yes or NQ. block on the Impact Screening Criteria.
If a direct determination cannot be made based on the impact screening questions, the detailed questions in Attachment 9.4 may be used to identify potential impacts.
The completion of detailed questions in Attachment 9.4 Is recommended but not required.
Contact potentially impacted groups if assistance is required.
Engineering Change No.: (EC) EE 13-041 Rev. No.: 3 Prepared by:
Taylor Sutton Date:
0E~l!zN ENGINEERING DISCIPLINE~
Poton"al lm~act Clvll / Structural Design Englnoertng l8l YES ONO
. Does the proposed act1vijy Involve any civil / structural Onciuding seismic) design changes or activities? OR Does the proposed activity Involve any piping engineering desion chanqes or activities?
Paintings and Coatings Engineering YES l8J NO
. Doos the proposed act1v1ty lnvolvo the use of a coating or a component coated with e coatino not previously evaluated under the Paintinos and Coatinos ornorem?
Electrical Design Engineering YES 181 NO
. Does the proposed acllv1ty Involve any station or swltchyard etectncal design, large oower transformers. or setbnos chanaes or activities? (SOER 10-1)
Instrumentation and Controls Design Engineering YES 181 NO Ooos the orooosed activity involve any l&C desion or sett"nas chanoes or acbvities?
Mechanical Design Engineering 181 YES O NO
. Does the proposed activity add/remove/replace lnsulaUon, aluminum or other metallic'non-metalllc sources or debris in the reactor/containment building or involve any mechanlcal deslqn chanQes or actiV1Uas?
PSA Engineering YES 181 NO Does Iha p<0posed activity impact or Involve changes to plant evaluations or probablhslic salety assessments?
Does the proposed activity Impact or involve changes to the Emergency Operabng Procedures, Abnormal Operating Procedures, or Severe Accident Procedures or does it add, remove or modify SSCs Included In a Maintenance Rule function?
Nuclear Analysis 0 YES 181 NO Does the proposed activity impact or Involve changes to plant evaluations, T echnlcal Soecifications, Technical Reouirements Manual or reouire a lull 50.59 Evaluation?
Cyber 8111:urily A11aly11is YES 181 NO
. Does Iha proposed activity Impact or involve changes to Cyber Security Controls, or require cyber security assessment. evaluation as per 10CFR73.54 and Procedure 11.CYBER-SECURITY?
~ Entergy NUCLEAR MANAGEMENT MANUAL EC I (EE)13-041 RevtSlOO 3 Alt ctunent 9 1 a
0\\JAI.ITY RELA TEO INFORMATIONAL USE Engineering ChanQe Process P
2 I 7 aao o 3-EN-DC-115 I REV. 15C4 A TTACHMENT 9.3 IMPACT SCREENING
SUMMARY
SHEET 2 OF 7 DESIGN ENGINEERING DISCIPLINES !CONTINUED)
Potontlal lmeact FERC / NERC Impact YES f81NO
. Does the proposed ac1iv1ly impact or involve change to FERC / NERC or Regional Enllly compliance documents?
8.5.b and other "Beyond 0&1lgn Baal*" considerations YES 181NO
. Does the proposed activity impact or involve changes to strategies related to B.5.b or other Issues considered 'beyond design basis' including Severe Accident Procedures and Guidelines?
DESIGN ENGINEERING PROGRAMS PQtontlal lrn11aot ASME Section Ill Specifications YES 181 NO
. Does the proposed activity add, delete, or modify 1nformahon required by ASME Section Ill to bo contained In a design soocificat1on?
Cable and Raceway Program YES 181 NO
. Does the proposed activrty Involve any changes to cable trays, raceways or the associated documentation?
Electronlc Databases (EDB)
YES 181 NO
. Does the orooosed act1v1tv Involve anv chanaes to electronic databases?
ea Program (10CFR50.49, NUREG 0588, Rog Guida 1.89)
YES 181 NO
. Doos the orooosod actlvitv Involve any changes to tho ea Program?
Hydrogen Control Program (10 CFR 50.44) (If appllcable)
YES 181NO
. Does the oronnsed activity Impact equipment or matenals related to hvdroaen control?
Human Factors Program YES 181 NO
. Does the proposed acti111ty involve control panel design Including layout and labeling
- visual displays, operator aids, auditory signals or enVJronment In the control room, cablo spreading room, and other control panel lo<:atlons?
Margin Management YES
~NO Does the proposed activity Impact or In-.olve any change to design, licensing or ooerations maroins?
Reg. Guida 1.97 / PAM (Post Accident Monitoring)
YES 181 NO
. Does the proposed activity Involve Reg. Gulde 1.97 (Post Accident Monitoring)
Indications?
MAINI~Nat:l~E PQtentl!!I lm11a!il Electrlcal Maintenance YES 181 NO Does the proposod activity require an Electrical Maintenance review to Identify effoctod PfOcedures. reaulred actions and reoulred tralnina?
l&C Maintenance YES 181 NO
. Does the proposed activity require en l&C Meintenence reviow lo ldontiry offcctod procedures. reauired acbons and rA<>ulred tralnma?
EC f (EE)13-041 Revis,on 3 Atlaehmenl 9 1 3 f 7 Pano o NUCLEAR 0uALITY RELATED 3-EN-DC-115 1 REV. 15C4
-=:---Entergy MANAGEMENT MANUAL INFORMATIONAL U SE Eni:iineerina Chance Process ATTACHMENT 9.3 IMPACT SCREENING
SUMMARY
SHEET 3 OF 7 MAINTENANCE (CONTINUED!
Pot2nt111l lm[!RC!
Mechanical Maintenance YES
@ NO
. Does the proposed activity require a Mechanical Maintenance review to identify affected procedures, roquired actions and required training?
NUCLEAR ENGINEERING Potential Impact Nuclear Fuel Design YES
@ NO
. Does the proposed activity impact or involve the design, perf0<mance or storage of nuclear fuel?
Reactivity Management Program YES
@NO
. Does the proposed activity impact or involve the reactor system reactor controls, reactor chemistry, related systems, poten~ al core and spent fuel damage, spent fuel, reactor coolant pressure boundary or reactor svstem nmcedures?
PRQ~E~S QB eBQ~BAM IMPA~I ~~BE!;~IN~ ~HE~KLl~I PQten!lal lml!iJS:!
Computer Support and Software YES
@NO
. Does the proposed activity impact 0< involve changes to plant computer software or firmware or Impact Software Quality Assurance ISQAl?
Chemistry and Environmental Impact YES
@NO
. Does the proposed activity impact or,nvolve any changes to plant chemistry requirements, operations or procedures, or any changes to the environment?
. Does the proposed activity impact or involves the srto mdiolog1cal ground water monltorinq prOQram?
Radiation Protection (RP) Program Impact YES
@ NO
. Does the proposed activity impact or involve anv chanoes to the RP nm,,ram?
Operations YES
@ NO Does the proposed activity Impact or Involve changes to Operations procedures, trainino or ooerator actions?
Planning, Scheduling and Outage (PS&O)
YES
@NO
. Does the proposed activity require a PS&O review to identify required design and Installation information?
MP&C (Inventory)
YES 181 NO Does the proposed activity Impact or involve any addition or removal of equipmont from the inventory?
. Does the proposed activity Impact or involve any Procurement of Quality material from non-Qualified suoollers?
Procurement Engineering YES
@ NO
. Does the activitv imoact or involve any orocurement activities?
Vendor Manual Control YES
@NO
. Doos the chanoo add, remove, or rnodifv a structure, svstom, or =onent?
~
-=-Entergy NUCLEAR MANAGEMENT MANUAL EC f (EC)13-041 Rellislon 3 Allachmenl 9. I QuAUTY RELATED INFORMATIONAL US!
Enaineerina Chanae Process 4 17 Page o 3-EN-DC-115 I REV. 15C4 A TTACHMENT 9.3 IMPACT SCREENING
SUMMARY
SHEET 4 0F7 PBQ~R8M~ 8~0 COMPONENT~
Pgten1Ial l!!lllilC!
ASME Appendix J (Primary Containment Leak Rate Testing) Program YES 181 NO
. Does the proposed activity Impact or involve any changes to pnmary containment leak rate testil'lQ?
ASME Containment In-service ln1pection (IWE / IWL) Program YES 181 NO
. Does the proposed activity Impact or Involve the containment pressure boundary or associated moisture barriers or a support fOI' the containment p<essure boundary?
This Includes 'software only' changes that do not physically change the hardware such as re-rating of press1Kes or temperatures to include changes to documented information on these itoms or additional documented lnfOffllatJOn for these Items.
. Does the proposed activity llm1t access to containment surfaces tor Inspection?
. Involve d sassembly of a bolted connection wh ch forms a portion of the conlalnment boundary?
ASME In-service Inspection (ISi) Program YES 181 NO
. Does the proposed acUv1ty add, delete, or modify an ASME SectJon XI pressure boundary item or a support for an ASME Sect10n XI pressure boundary Item? ThlS Includes 'software only" changes that do not physically change the hardware like such as re-rating of pressures or temperatures (ASME SOctlon XI items include ASMC: Class 1, 2, 3, or 831.1 treated as ISi Class 2 or 3 (T2 and T3) components, parts, or appurtenances such as pipe or pressure vessel walls, valve bodies and pump caslnQs.
ASME Section XI Repair / Replacement Program YES 181 NO
. Does the proposed acbv,ty involve any mechanical component within the ASME XI program I boundaries? Does tho proposed activity add, delete, or modify an ASME Section XI pressure boundary Item or a support for an ASME Sectton XI pressure boundary item? (ASME Section XI Items Include ASME Class 1, 2, 3, MC, and CC, as well as 831.1 treated as ISi Class 2 or 3 (T2 and T3) components, parts, or appurtenances such as pipe or pressure vessel walls, valve bodlos and pump casings).
ASME In-service THtlng Program YES 181 NO
. Does the proposed activity impact or involve any Item (safety related or non safety related) that may affect the performance or tasUng of a safety related pump or valve?
. Doos lhe proposed activity impact tho function or functional classiflCiltlon of any oump or valve as stated In lhe 1ST orooram documents?
Air Operato<I Valve (AOV) Program YES
@NO Does the proposed activity impact or involve the design, operat10n or testing of AOVs?
Burled Piping and Tanks (BP&T) Program YES
@NO
. Does the proposed activity Impact or Involve any changes to plpmg or cathodic orotectlon of Buried Piplno & Comoonents?
Boric Acid Corrosion (BAC) Program YES 181 NO Does the proposed activity impact or involve any increase In the llkel1hood of boric acid formalion or corrosion?
Check Valve Program YES
@NO Does the proposed activity Impact or Involve the design, operaUon or testing of check valves?
~
-=--=-Entergy NUCLEAR MANAGEMENT MANUAL EC # (EE)13-041 Revision 3 Altachmcnl 9.1 Ou.t.1.IIY RELATED l~FORMA TION.t.L USE EnQineerinQ ChanQo Process Paoe 5 of7 3-EN-DC-115I REV. 15C4 ATTACHMENT 9.3 IMPACT SCREENING
SUMMARY
SHEET 50F 7 PROGRAMS ANO COMPONENTS (CONTINUED!
Potential lm(;!act Cobalt Reduction YES r8J NO
- Does the proposed activity involve the polential to create conditions that could Impact personnel dose such as the mtroducllon (or reintroduction) of any cobalt contaIrnng materials (such as SteMe or high coba t stainless steel) Into the primary system or processes such as electro-oohs hi no or advanced surface treatments?
Control Room Habltablllty YES r8J NO
. Does the proposed activity impact or involve changes that affect the temperature 0<
radlolOQical onvironmontal conditions In tho Main Conlrol Room?
Electrical Circuit Broakor, Relay and Electrical Equipment Test.Ing YES r8J NO
. Does the proposed actlvity Impact or Involve changes to the functional testing or circuit breaker relav or electrical eou1oment?
Flow Accelerated Corrosion (FAC) Program YES r8J NO
. Does the proposed activity involve any changes (e.g.: configurahon, veloe1ty, flow rato, pressure, temperature, material, weld location, etc.) to plpjng systems included In the FAC Proqram.?
Heat Exchanger Program YES 181 NO
. Does the proposed activity impact or involve the design, operation or testing of a comoonent In tho heat exchanoer oroaram?
Predictive Maintenance Program YES 181 NO
. Does the proposed activity impact or involve any aspect or the storage, use and testing of lubricants?
. Does the proposed activity Involve thermography?
. Docs the orooosed activity involve vibration monitoring?
Microbiological Induced Corrosion (MIC) Program Impact YES r8J NO Docs lho proposed activity Involve piping containing untreated or stagnant water or ooen to the atmosohere?
Motor Operated Valve (MOVJ Program YES r8J NO
. Does the proposod activity impact or invo,ve tho dosign, operation or testing of MOVs?
Plant Thermal Performance Program YES
~NO
. Does the orocosed activity lmo;ict or involve Diani thennal oerformance?
Preventive Maintenance Program YES
~NO Does the proposed activity Impact or Involve periodic testing or performance or SSCs? or Does the proposed activity add, modify or delete any Environmental Qualification (EQ) maintenance requirement or replacement frequency to ensure the comoonent(s) maintains Its quahficallon status?
PT Curves YES r8J NO
. Does the proposed actiVtty add, delete, or modify the basis (as contained in each sites reactor vessel surveillance material testing program) ru1 lhtt P1tlllsu1t1 /
Temoerature Llmlt Curves /fiuence. ores sure, temoorature, reactOI' materials l?
-:-:-:-Entergy NUCLEAR MANAGEMENT MANUAL EC I (EE)13-041 R81/lsion 3 At 9
tachment 1
Ck.lAUTY AEI.ATEO INFORMATIONAL US!!
Eni:alneerlnA Chan9e Process P
6 f7 aoe o 3-EN-DC-115I REV. 15C4 A TTACHMENT 9.3 IMPACT SCREENING
SUMMARY
5HEET6 OF 7 PROGRAMS AND COMPONENTS (CONTINUED}
Potentiol lm11ac1 Relief Valve Program YES
£81 NO
. Does lho proposed actIvIty impact or involve tho design, operat10n end testing of relief valves, saretv valves. vacuum breaker valves or ruoture disc?
RPV Internals Program (BWR VIP)
YES tislNO
. Does the proposed activity impact or in110lve any change to the reactor internals?
This Includes "software only' changos thal do not physically change tho hardWare such as re-rating of pressures or temperatures.
. Does the proposed activity Impact or involve or related documentation to include neutron fluence or neutron fluence calculations?
. Does tho proposed activity Impact or involve or changes 10 coro now characteristlCS or core flow characlerlshcs?
Welding Program YES
£81 NO
. Does the proposed activity Impact or involve a special process, such as welding, brazino or solderina?
Safety Program YES
@NO
. Does the prooosed impact or activttv involve personnel or induslnal safely?
System Englneorlng DYES
£81 NO
. Does the proposed activity Impact or invo,ve any changes to system configuration, function or performance, etc., for Maintenance Rule or other system?
. Does the proposed activity impact or involve any changes lo system procedures, malnlenanco or oporalion, etc.?
Training Program YES tislNO
. Does the proposed actlV1ty Involve ex1Sllng traln,ng requirements or create the need for new trainlno?
Simulator Impact YES
£81 NO
. Does the proposed aclivilv lmoact or Involve chanqcs to the Slmulalor?
£81 NO
. WIii the proposed activity be performed as 'Work at Risk" thus requiring notJllcatlon of NEIL representative at conceptual design phase?
. Does the proposed activity result in a change to any eXJSting or install a new fire protection water supply system (tanks, pumps, underground fire main, hydrants), nre suppress10n system (water, gaseous), fire alarm or deted!On system?
. Does tho proposed activity result in a change to any ex1sbng fire barrier, fire barrier penelralion seal, or install new fire barners or fire barrier penetration seals?
Docs the proposed activity result In e change to any ln*sltu combusbblo loading or add new in-situ combustible materiflts (cable insulation, thermal Insulation, oil, lubricants, plastics, etc.)?
. Does the proposed activity result in a change to any admnislrative control or tho Fire Protection Program (FPP) (e.g., hot work, transient combustible, impairmenls, fire brigade composition or training)?
. Does the proposed activity result in a change to any system or component credited in the Nuclear Safety capability Assessment (NSCA)? Reforonce NEDC 11-019.
. Does the proposed activity result In a change to the Fire Protecllon Program (FPP) as described In Procedure 0.23? Also, refer to the Fire Safety Analysis calculation for the specific flre areas affected by the change.
NUCLEAR MANAGEMENT MANUAL EC# (EE) 13--041 Revi5lon 3 At1achmen1 9 I QuAIJTY R ELATED INFORMATIO Al use Pa e 7 or7 3-EN-DC-115 REV.15C4 A TTACHMENT 9.3 IMPACT SCREENING
SUMMARY
5HEET7 OF 7 PROGRAMS AND COMPONENTS (CONTINUED}
Potential lml!act Fire PRA YES 181 NO
. Does the proposed act lty resuh in a change that would affect the PRA fire model?
Change to a pump, motor, fan. MOV.
Change to any cable{s) or cable routing Change to equipment location.
Change to In-situ combusUbles (cable m ulatK>n, lubricants. plastics)
Change to or addll on or an igmoon source (transformer, motor. electncal cabinet. junctJon box or switchgear).
NEIL Design Review YES 181 NO Is this design change a new structure?
. Does addition. renovation or alteration change the occupancy dassificalion of any part of a NEIL insured structure?
. Does the design change involve the addition of a new Ell required fire protection system?
. Does the change significantly add to, remove or a ter an e>Cisting NEIL required fire detection or fire protection system? For example, this does not include relocation of less than 10% or the fire detectors or suppression heads/nozzles for a single system while still maintaining code compliance?
. Does the change create an addition to an existing NEIL insured structure?
. Does the change involve replacement of the roof decklng and/or covering?
. Does the design change affect an interior finish such that it would not meet the requirements of the NEIL Property Los Control standard (Pl.CS)?
. Does the design change add to. renovate or alter e cooling tower f':11 or supports?
. Does the design change add 011 filled components over SO gallons oil capacity?
. Does the change add to. renovate or alter od co lectlon systems, fire barrier or flre orotect1on svstems for 011 filled components?
NUCLEAR Q\\JAUTY REl.A "reD
~Entergy MANAGEMENT MANUAL lt<FORMATIONAL USE EC I (EE)13-041 Revision 3 Attachmenl 92 3-EN-OC-115 I REV.15C5 PAGE 1 OF 169 Enaineerina Chanae Process ATTACHMENT 9.17 SHEET1 OF 6 EC Number: (EE)13-041 Rev.3 Work Order Number(s)4*7: None DOCUMENT TYPE2 TRAINING CALC CALC DOCUMENT NUMBER N/A 16-003 13-028 SHEET NUMBER NIA NIA NIA See Attachment 9.17 Sheet 2 for Form Footnotes.
AFFECTED oocuMEHT loENT1F1cATION e**10 DOCUMENT CHANGE NUMBER3 Rev.0 Rev.2 PENDING CHANGES None None Page 1 of 1 Page 1 of 1 A/S Code1
[X) DPD
[ ] EHP - CR Docs
( ) EHP - RTS
( ) PRTS6 DATE DOCUMENT A/S4*5
NUCLEAR QUALITY REJ.ATED 3-EN-DC-115 I REV.15C5
-===-Entergy MANAGEMENT MANUAL INFORMATIONAL USE Englneerina Chanae Process ATTACHMENT9.13 10 CFR 54.37(b) REVIEW D ETERMINATION Sheet 1 of 1 NOTE IRefer to Attachment 9.14 for guidance on tlhis attachment This attachment determines if a review for 10 CFR 54.37(b) reporting is required. This Determination is required for Nuclear Change, Commercial Change, Engineering Evaluation, Safety Classification Evaluation, and Fire Protection Evaluation ECs
- 10 CFR 54.37(b) Review Determination Yes No
- 1, Does the change Involve changing the classification of non-essential SSCs to essential?
181 2, Does the change result In existing non-essential SSC containing liquid or steam now being located In the same space as essential SSCs, where before it was not?
181 3, Does the change result in existing non-essential SSCs providing structural support to essential SSCs where before they did not?
181 4, Does the change Involve existing SSCs being newly Identified as required to protect essential SSCs during design basis events such as HELB, tornados, flooding, and seismic events?
181 5, Does the change Involve existing SSCs being newly Identified as required to demonstrate compliance with any of the following NRC regulations?
181 Fire Protection (FP, 10 CFR 50.48)
Station Blackout (SBO, 10 CFR 50.63)
Anticipated Transient without Scram (ATWS, 10 CFR 50. 62)
Pressurized Thermal Shock (PTS, 10 CFR 50.61)
Environmental Qualification (EQ, 10 CFR 50.49)
- 6. Does the change involve the addition of a system safety function requiring components 181 not previously identified as being required for accomplishment of a system safety function?
7, Does the change create or Identify a new failure mode such that failure of existing non-essential SSCs could prevent the satisfactory accomplishment of a safety function?
181 8, Does the change remove or relocate a seismic boundary anchor such that existing non-181 essential components are now Included in the seismic analysis of essential components when previously they were not?
If any of Questions 1-8 are answered yes, contact the site renewed license program coordinator.
EC#(EE)13-041, Rev.3.3 Page 1 of 1
EC# (EE)13-041 Rev 3 - Attachment 9 4
~ Entergy NUCLEAR QIJAI.ITY RELATED 0-EN-Ll-100 I REV. 12C1 MANAGEMENT MANUAL INFORMATIONAL USE PAGE 1 OF 8 I
Process Appllcability Determination ATTACHMENT 9.1 PROCESS APPLICABILITY DETERMINATION FORM I.
OVERVIEW PAO Tracking#: ~11~3~7 __
PAD Rev. #: =2 __
Proposed Activity / Document: EE13-041 (and NEDC 13-028 16-0031 Change/Rev.#: _3 _
Description of Proposed Activity: Turbine Building Blowout Panels/ Metal Wall System II.
DOCUMENT REVIEW Provide the requested information for each Item below.
- 1.
For documents available electronlcally:
- a.
List search engine or documents searched, and keywords used: ESAR, Tech Specs.
"Siding", "HELB", "BLDG", "Panel", "Girt". "Turbine Building*, "Blowout', Finite"
~
- b.
List relevant sections of controlled electronlc documents reviewed:
CNS USAR Section 11-3.2.2, Wind".
CNS USAR Section Xll-2.1.3.1, "Principal Class II Structures*.
CNS USAR Section Xll-2.2.2, "Turbine Building".
CNS USAR Section Xll-2.3.3.1, Wind Loads".
CNS USAR Section Xll-2.3.3.2.2, "Tornado Generat6d Missiles" CNS USAR Section IV-12, "High Energy Line Break (HELB) Study" CNS USAR Section XIV-6.2. 7.2, "fission Products Released From the Tl/rblne Bu/ldl11g" CNS USAR Section X-4.6.3.3, "RPV Head Drop Analysis" CNS USAR Appendix C-2.3. "Governing Codes*
CNS USAR Appendix C-2.5. 7.1.6, "Method of Analysis" CNS USAR Appendix C-2.5.8, 'Ma,n Steam Tunnel" CNS USAR Appendix C-4.42, "References for Appendix c*
- 2.
Documents reviewed manually (hardcopy):
NEDC 03-005, Rev. 0 thru 4 NEDC 12-012, Rev. 1 NEDC 11-075, Rev. 1
- 3.
For those documents that are not reviewed either electronically or manually, use the specific questions provided In Sections Ill and IV of Attachment 9.2 of 0-EN-Ll-100 as needed. Document below the extent to which the Attachment 9.2 questions were used.
t:liA
EC # (EE)13-041 Rev 3 - Attachment 9 4 t!AE NUCLEAR QuA!.ITY RRATED 0-EN-Ll-100 I REV. 12C1
~ ntergy MANAGEMENT MANUAL INFOIIMATIOflo\\l. Use PAGE 2 OF 8 Process Applicability Determination A TTACHMENT 9.1 P ROCESS APPLICABILITY DETERMINATION FORM Ill.
PROCESS REVIEW Does the proposed activity affect, invalidate, or render incorrect, OR have tho potential to affect, Invalidate, or render Incorrect, Information contained In any of the following processes? Associated regulations and procedures are identified with each process below.
PROCESS (Regulations / Procedur&s)
YES NO REVIEW RES UL TS Radwaate / Procon Control Program (PCP)
(0.PCP.1 or contact the Radiation Protecuon Dept.)
0 181 Radiation Protoctlon / ALARA 0
181 (10 CFR 20 / 9.EN-RP-110 or contact the Radiation Protection Dept.)
lnsarvice Inspection Program (10 CFR 50.55a / EN-OC--120. -351) 0 181 lnaorvlca THllng Program (10 CFR 50.55a / 3-EN-OC-332)
D 181 Maintenance Rule Program (10 CFR 50.65 / EN-OC-203, -204, -205, -206,
-207)
D t8l Containment Leakage Rate Testing (Appendix J) Program (10 CFR 50 Appendix J / 3.40)
D t8l IF any box Is checked "Yes," THEN contact the appropriate department to ensure that the proposed change is acceptable and document the results In the REVIEW RESULTS column.
IV.
LICENSING BASIS DOCUMENT REVIEW Does the proposed activity affect, Invalidate, or render incorrect, OR have the potential to affect, invalidate, or render incorrect, Information contained In any of the following Licensing Basis Document(s)? Associated regulations and procedures are identified with each Licensing Basis Document below.
LICENSING BASIS DOCUMENTS YES NO REVIEW RESULTS OR SECTIONS (Regulations/ Procedures)
AFFECTED OR L8OCR #
Quality Assurance Program for Operation Policy Document (QAPD)
(10 CFR 50.54(8) / 0.29.4)
D l8I fire Protection Program (FPP) [lncludee the Fire Huard* Anelyala (FHA), Appendix R SSAR, 0.23) 0 181 OL Condition, 10 CFR 60.48 / 0.29.4)
Emergency Plan (10 CFR 50 54(q) I 0.29.4)
D 181 Safeguards Plan and Cyber Security Plan 0
110 CFR 50.54(p) IO 29.4 or contact the site Security / IT Dept I t8l Operating License (OL) / Technical Sp1clflc1tlon1 (TS)
(10 CFR 50.90/0.29.1, 0-EN-Ll-103) o*
181 TrS Bases (10 CFR 50.59 / O-EN-Ll-100 /O-EN-Ll-101, 0.29 1) 0 181 Technical Requirements Manual (TRM) (lncludlng TRM Baaes)
( 10 CFR 60.59 / 0-EN*Ll*100 / 0-EN-Ll-101, 0.29.1) 0 181 Coro Operating Limits Report (COLR), (TS Administrative Conlrols, 10.32, D
181 O-EN*Ll-100, 0-EN-Ll-101 )
Off-site Dose Assassmont Manual (ODAM)
ITS Admlnlsiratlve Controls or 10 CFR 50.59 / 0.29 1 or O-EN*Ll-100 / 0-EN*
D t8l
EC # (EE)13-041 Rev.3 - Attachment 9.4 NUCLEAR QuALITY RELA TEO O-EN-Ll-100 I REV. 12C1
~ Entergy MANAGEMENT MANUAL INFORMATIONAL Use PAGE 3 OF 8 Process Applicability Determination ATTACHMENT 9.1 PROCESS APPLICABILITY DETERMINATION FORM LICENSING BASIS DOCUMENTS YES NO REVIEW RESULTS OR SECTIONS (Regulations / Procedures)
Ll-101)
Updated Safety Analysis Report (USAR) 0 (10 CFR 50.71(e)/ 0.29 2, 0-EN-U-100, 0-EN-U- 01)
Storage Cask Certificate of Compliance (10 CFR 72.244 / 0.29 9) o*
Cask Safety Analysis Report (CSAR) o*
(10 CFR 72.70 or 72.24810.29.9, 0-EN-ll-100, O-EN-Ll-112) 10 CFR 72.212 Evaluation Report (212 Repor1)
D (10 CFR 72.48 1 O-EN-ll-100, 0-EN-Ll-1 2, 0.29.1)
NRC Orders (10 CFR 50.90 / 0-EN-ll-103 or as directed by lhe Order) o*
NRC Commltmonts and Obllgallona (0.42. I) o*
Site Specific CFR Exemption o*
(10 CFR 50.12, 10 CFR 55 11, 10 CFR 55.13, 10 CFR 72. 7)
- STOP. Conlact lhe site Ucoosmg Department for further guidance.
!E any box is checked "Yes," THEN ensure that any required regulatory reviews are performed in accordance with the referenced procedures. Prepare an LBDCR per procedure 0.29.1 if a LBD Is to be changed, and document any affected sections or the LBOCR #. Briefly discuss how the LBD is affected in Section VII.A.
EC # (EE) 1 ~
1 Rev.3 - Attachment 9 4 NUCLEAR QvALITY RELATED 0-EN-Ll-100 I REV. 12C1
~ Entergy MANAGEMENT MANUAL INFORMATIONAL US PAGE 4 OF 8 Process Applicability Determination ATTACHMENT 9.1 PROCESS APPLICABILITY DETERMINATION FORM V.
10 CFR 50.59 / 10 CFR 72.48 APPLICABILITY Can the proposed activity be dispositioned by one of the following criteria? Check the appropriate box (if any).
D An approved, valid 50.59/72.48 Evaluation covering associated aspects of the proposed activity already exists. Reference 50.59/72.48 Evaluation# N/A (if applicable) or attach documentation. Verify the previous 50.59/72.48 Evaluation remains valid.
D The NRC has approved the proposed activity or portions thereof or an approved license amendment addresses the proposed activity. Reference the approval document:
NIA D
The proposed activity is controlled by one or more specific regulations.
Examples of specific regulations are:
Quality Assurance Program (10 CFR 50 Appendix 8)
Safeguards Plan (50.54(p))
Emergency Plan (50.54(q))
Fire Protection (operating license condition)
See NEI 96-07 Section 4.1 for additional guidance on specific regulations.
Reference the controlling specific regulation(s):
!.Ethe entire proposed activity can be dispositioned by the criteria in Section V, THEN proceed to Section VII and provide basis for conclusion in Section VII.A.
Otherwise, continue to Section VI to perform a 50.59 and/or 72.48 Screening, or perform a 50.59 and/or 72.48 Evaluation In accordance with 0-EN-Ll-101 and/or 0-EN-Ll-112.
EC # (EE)13-041 Rev 3 - Attachment 9 4 NUCLEAR QuALIIY RELATED 0-E N-Ll-100 I REV. 12C1
~ Entergy MANAGEMENT MANUAL INFORMATIONAL USE PAGE 5OF8 Process Applicability Determination ATTACHMENT 9.1 PROCESS APPLICABILITY DETERMINATION FORM VI.
50.59 / 72.48 SCREENING REVIEW VI.A 50.59/72.4B SCREENING (Check the appropriate boxes.)
~ 10 CFR 50.59 Screening criteria are met. (10 CFR 50.59(c)(1)1 The proposed activity meets all of the following criteria regarding design function:
Does not adversely affect the design function of an SSC as described In the USAR; AND Does not adversely affect a method of performing or controlling a design function of an SSC as described In the USAR; AND Does not adversely affect a method of evaluation that demonstrates Intended design function(*) of an SSC wlll be accomplished as described in the USAR; AND Does not Involve a test or experiment not described In the USAR.
0 The proposed activity does not Involve structures, systems, or components controlled by 10 CFR 50.59.
D 10 CFR 72.48 Screening criteria are met. (10 CFR 72.48(c)(1 )I (Applicable to sites with an ISFSI)
The proposed activity meets all of the following criteria regarding design function:
Does not adversely affect the design function of an SSC as described in the CSAR; AW2 Does not adversely affect a method of performing or controlling a design function of an SSC as described In the CSAR; AND Does not adversely affect a method of evaluation that demonstrates Intended design function(s) of an SSC will be accomplished as described In the CSAR; ANQ Does not Involve a test or experiment not described In the CSAR.
(81 The proposed activity does not Involve structures, systems, or components controlled by 10 CFR 72.48.
.!E either of the 50.59 or 72.48 Screening criteria are met, THEN complete VI.B below as appropriate and proceed to Section VII.
!E the proposed activity does not meet the applicable criteria, THEN perform a 50.59 or 72.48 Evaluation in accordance with 0-EN-Ll-101 or O-EN-Ll-112, as appropriate, attach a copy of the Evaluation to this form, and proceed to Sectiion VII.
IF the activity does not Involve systems, structures, or components controlled by 10 CFR 50.59 or by 10 CFR 72.4B, THEN a 50.59 or 72.4B Screening is not required, as appropriate, and proceed to Section VII.
EC# (EE)13-041 Rev.3 - Attachment 9 4
~
NUCLEAR QuAl.llY REL.A TI:O 0-EN-Ll-100 I REV. 12C1
-~ Entergy MANAGEMENT MANUAL
'"'fORMA TIONAL Use PAGE 6 OF 8 Process Applicability Determination ATTACHMENT 9.1 PROCESS APPLICABILITY DETERMINATION FORM VI.B BASIS Provide a clear, concise basis for determining the proposed activity may be screened out such that a third-party reviewer can reach the same conclusions. Refer to NEI 96-07 Section 4.2 for guidance.
Provide supporting documentation or references as appropriate.
Pu_!Pose (!f EC# (EEJ-13-941_. Revision 3 The purpose of this activity is lo validate the Turbine Bu1kilng blowout pressure va1ue of O 5 psi as staled In rsAR Amendment 25. The original blowout panel am:ilyl>ls could not be located 1n CNS Rer.ords Therelore,.a finite olement 11nalysis wns usod 1n NEDC 16 003, Revist-0n 0, to evaluate tho vahcf1ty of tho hcnnsing bac;,s valuit, os stated 1n tho FS/\\R Tho fintto olemonl 1Jnalys1s was used to validate tho rn~ull of tho original mis~ ng Jnalys1s, Jnd wJs approved as a design oasis calculation or record. NEDC 16-003 does not replace tho original analysis that 1s missing.
Does not adversely affect the design function of an SSC as described In the USAR; AND The siding (such as: panels. girts, sub-girts, sag rods and connections) is considered as an architectural component and Is not credited as a structural SSC for the design of the Turbine Generator Building (TGB) superstructure (Note: the siding only transfers load to the TGB Superstructure SCC). In the event or a I l[LB, the Turbine Building s1d1ng 1s designed to blowout to reheve pressure before 1t w1ll lmpacVfa1I critical HELB boundary doors thnl protect design funcbons of SSCs 1n lhe Control Room. 0G Bu11d1ng, and other locations from a harsh onv1ronmenl (steJm) that would occur The HELB Event has the potential to generate debris and/or missiles, but will bo minimized by the sag rod and attachment to the roor truss and girders. In addition, !he-Reactor Au1l<11nq all Class I Structures aro designed against the impact or Tornado Generated Missiles 6vent for m1~les tmf13G1
[Ref. CNS USAR Section Xll-2.3.3.2 2), wh,ch envelope this siding systom missile type.
Also to note. Certificates of Material resting Reports (CMTRs) nnd double shear test data was used for the upper bound (fmlure) analysis of tho siding support system Tho envolop,ng yield and tensile strength values, from the CMTRs and double shear tostmg, wero used to belier predict tho fmlure mochanlsms Tho governing tleslgn code for structural steel, AJSC 6th edition. does not provide guidance on failure analysis. but Is s,tent on the topic.
Tho code Is focused on the design or structures. to prevent failure by comparing des.gn loads to spocifiod material m1n1mum yield strengths. Using mm,mum spoclfled malonal strengths 1n failure analysis would result in failure mochontsms occurring before the actual or upper bound,trongth of the material 1s reached, causing the analysis results to be nonconscrvallvc (failure occuring sooner). Therelore the use of CMTRs and double shear test data do not invalidate tho AISC code but provide occ.urato inputs lo tho lailure analysis to volidalo lho dosign !unction of the blowout panels.
This acli111ty does not change the design function ol the Turbine Building s1d1ng; it vahdates the design function.
Does not adversely affect a method of performing or controlling a design function of an SSC as described in the USAR; MfQ There are no changes lo any procedure, or any other document, as a result of this assessment. EE 13-041 Rev.2
~
3, and/or Calculation NCOC 13-026, Rev O N[OC 16-003, determines the maximum Internal pressuro at which
~
the TGB siding can be maintained. Therefore, there are no adverse effects on YSAR a method of performing or controlling a design function of an SSC as described in the USAR.
Does not adversely affect a method of evaluation that demonstrates Intended design functlon(a) of an SSC will be accomplished as described in the USAR; AMQ.
This assessment of the TGB sidlng(s) capability to resist an Internal pressure does not Impact either the design bases for the structure or safety analyses for the HELB. The HELB analyses' bounding peak and temperature curves will are not be altered and/or changed by this assessment. However, this evaluation does demonstrate the codA.ipproac:hamUor mitthodology 11nd values used to arrive at the FSAR Amendment 25 [Ref. CNS USAR Section IV-12, "High Energy Line Break (HELBJ Study", 110) bounding value for the TGB limiting Internal pressure capacity as being conservative.
Nl:OC 16-003 documents the use or ANSYS Vorsion 14.0, which Is a finite element analysis (FEA) program ANSYS was used to model and simulate the behavior of the g,rt channel, connecting angles, bolts and welds of tho panel siding support system ot various levels of tho HELB pmssuro valuo (0 5 psi) Tho nnlto elomont analysis methodology validated the result of the methodology that was used In the original construction calculation. which was performed al or before the time of 1n1llal plant start-up as part of CNS's onglnal design and hcenslng requirements, as conservative
EC # (EE)13-041 Rev.3
- Attachment 9 4 NUCLEAR QUALITY RELA TEO REV. 12C1 0-EN-Ll-100 I
~ Entergy MANAGEMENT MANUAL INFORMATIONAL Use PAGE 7 OF 8 Process Applicability Determination ATTACHMENT 9.1 PROCESS APPLICABILITY DETERMINATION FORM NCI 96-07, Rov1s1on 1, definition 3 10 prov,dos lhe rollowng gu1danco; "Examples or melhods of evalualton are prc5ented below Changes to 6uch methods of ev,1luatlon require ovalus11ton under 10 CFR !>0.59{c)(2J(v111) only for evaluaI1ons used e,lher in UFSAR sa'oly analyses or In eslabli~hing tho design bases, ano only 1f the melhods are descnbed, outlined or summarized m the UFSAR.
Mothudology changos lhul mo subjot:1 lo 1 O CFR 50 59 ndude ch.1nges to elemonls of exIs1tn9 molhods descnbed In the UFSAR and to changes that,nvol~c rcplaLcmcnl ol exIstI~ molhods or ev,iluallon with alternative moIhodolug1ei,
- lhoraforo, changes to MOEs ruquIre evalual1on under 10 CFR 50.59(c)(2)(v111)
ONLY for evaluations used eilher in UFSAR safely dnalyses ot establishing desl!}n bases, AND ONL V IF lhe mothods aro descr bod, outhnod or summanzed In tho UFSAR No desc.rlpt,on or the actual MOE that was utilized was lound initial plant licensing suhm,ltals Tho Turbine Bu11d,ng panel siding blowoul pressure valuo Is stated In FSAR Amendment 25 All associated supporting documents that were submitted to the NRC wero rev,ewcd. The blowout pressure value Is not slated In any other locution w1th,n the,nitial Safety Analysis Report FSAR Amendment 20 has a slatemont that claims "the Turbine Building siding may blow out* but does not provide II p111ssuro value or any justification f SAR Amendmenl 75 As prov1souly slated, this amendmont contdins the onl~ releml'lC'.u to the 0.5 p~l blowout pressure value The method used lo develop that v,11\\Je is not discussed 111 any length therein current USAR cJecnpllon No discussion Within lh1t hody of 1h11 USAR stains the Turh,no Building blowout panel prossuro value. nor the MOE used to calculate the value.
Based on the above, the use of ANSYS vahdatcs the result of a calculation Iha! has never been 'described, outlined or summari1od m lho USAR
- Therefore,,t Is not subJoct lo further review under 10 CFR 50.59(c)(2)(vlll)
Does not Involve a teat or experiment not described In the USAR.
'2.
No USAR-<186Cribed tests or experiments were performed on SSCs ara involved in the evaluation of the TGB
~
siding system Internal pressure capacity. Therefore, the ac1Jv1ty does not Involve a test or experiment not described m the USAR Summary:
Engineering Evaluation 13-041 and subsequent supporting analyses (NEDC 13-020 16-003) provide Is an 1&
assessment of the architectural siding's structure, system and/or components (SSC) to determine the weak link and subsequently the maximum Internal pressure that can be maintained as a result of a High Energy Line Break Event. This assessment does not change the Design Basis of the TGB and therefore has no impact on the design requirements for this Class II structure as defined In tho USAR. However, this evaluation or progressive collapse of the TGB architectural siding system, consisting or the girt and connections, documents the code methodology and Input values to derlvo an internal pressure value or 0.5 psid as reported in tho FSAR Amendment 25. In addition, the analysis also demomtrales identifies this system's inlemal pre68Ufe limiting r comllllon as 0.3 paid-0couuln<J at run yi&kl or angle connec:llona relallve-to thl&ev8'll 8C6A8Fio woakost 2
component that wtll fail and subsequently 1rnliate the release or panels rrom the north and south TGB walls when Iha l'SAR defined pressure Is present In tho TGB. Theroforo, the result or the screening criteria questions Indicates there Is no further evaluation required under 10 CFR 50.59
EC # (EE)13-041 Rev.3
- Attachment 9 4 NUCLEAR 0UALITY REI.A lED 0-EN-Ll-100 I REV. 12C1
~ bntergy MANAGEMENT MANUAL INfORMATlOIIAL USE PAGE 8 OF 8 Process Applicability Determination ATTACHMENT 9.1 PROCESS APPLICABILITY DETERMINATION FORM VII.
REGULATORY REVIEW
SUMMARY
VII.A GENERAL REVIEW COMMENTS (Provide pertinent review details and basis for conclusions if not addressed elsewhere In form.)
None VU.B CONCLUSIONS
- 1.
Is a change to an LBD being initiated?
!f "Yes," THEN enter the appropriate change control process and include this form with the change package.
- 2.
Is a 10 CFR 50.59 Evaluation required?
!.E "Yes," THEN complete a 50.59 Evaluation in accordance with 0-EN*LI*
101 and attach a copy to the change activity.
- 3.
Is a 10 CFR 72.48 Evaluation required?
!.E "Yes," THEN complete a 72.48 Evaluation In accordance with 0-EN-LI-112 and attach a copy to the change activity.
- 4.
Are any other regulatory reviews required?
!f "Yes," THEN specify the review(s) required, below, complete the review, and attach a copy of the review result(s) to this PAO.
Revlew(s) required prior to PAD completion: NJA VIII.
SIGNATURES 1
Preparer:
TQD-901P, R, or926 Reviewer:
TQD-901R or 926 Name (print) / Signatu Process Appllcablllty Excluslon Procedure Owner:
Name (print) / Signature / Department/ Date Yes 181 No Yes 181 No D Yes
~ No Yes
~ No Upon completion, forward this PAD form to the appropriate organization for record storage. lftha PAD form Is part of a process that requires transmittal of documentation, Including PAD forms, for record storage, then the PAD form need not be forwarded separately.
1 Signatures may be obtained via manual methods (e.g., Ink signature) or telecommunlcatlon.
~
~ : Entergy NUCLEAR MANAGEMENT MANUAL OUALITYREI.ATEO J-EN-DC-115 1 REV. 15C5 ATTACHMENT 9.21 SHEET 1 OF 1 EC Number: EE 13-041 Rev.3 INFORM~ TIONAL Use Engineering Chan9!' Process A DDITIONAL REVIEWER F ORM @/
Page 1 of 1 EWO Number: None
~~---------- -------
REVIEWER DEPT.
SI DATE PRINT NAME OED-Civil 22'/./h M.,-flhe..; MeAAber-OED-Mech
(?-7~
l/17//1,
~M/lv ~. St,.,.,,J:.Do::,
EC # (EE)13-041, Rev.3.5 Page 1 of 1
NLS2020018 Attachment Page I of 17 ttachment Independent valuation Re ult, Allegation RIV-2019-A-0065 Investigation Report
NLS2020018 Attachment Page 2 of 17 Prepared for:
I DEPENDENT EV ALU TIO RESULTS ALLEGATION RIV-2019-A-0065 uclear Regulatory Commis ion Prepared by:
Troy Barker Director of Engineering ebra ka Public Power District - Cooper uclear Station PO Box 98 Brownville, NE 68321 Report Date:
May 6, 2020
NLS2020018 Attachment Page 3 of 17 1
TABLE OF Co TE TS 1.1 Introduction and Background........................................................................................... 4 1.2 Summary of Conclu ion................................................................................................. 5 1.3 Evaluator Independence and Qualification..................................................................... 5 1.4 Inve tigation Process and cope....................................................................................... 5 l.4.1 Investigation m thodology........................................................................................ 5 1.5 Summary of re ult........................................................................................................... 6 1.5. l Overall A sessment................................................................................................... 6 1.5.2 Asses ment of Specific ls ues:................................................................................. 8 1.6 Allegation I ue with Re ponses.................................................................................... 9 1.6.1 Issue# l..................................................................................................................... 9 1.6.2 I ue#2................................................................................................................... 10 1.6.3 I ue#3................................................................................................................... 10 1.6.4 I ue#4................................................................................................................... 12 1.6.5 I sue #5................................................................................................................... 14 1.6.6 ls ue #6................................................................................................................... 14 1.6.7 I ue #7................................................................................................................... 15 1.7 Conclu ions.................................................................................................................... 16 1.8 Reference...................................................................................................................... 16
NLS2020018 Attachment Page 4 of 17 1.1 Introduction and Background The U.S. Nuclear Regulatory Commission (NRC) transmitted to the Nebraska Public Power District (NPPD), a Request For Information dated January 16, 2020, focused on an allegation received about the Calculation NEDC J 6-003 [Enclosure 11 being non-conservative and may not adequately predict the failure of turbine building blowout panels used to mitigate a postulated high energy line break (HELB). More specifically, the RFI stated that the NRC had received seven areas of concern to be addressed. These areas of concern (Issues) are outlined in Section 1.5 below. For each issue, there is an initial response from the calculation preparer, LPI, Inc (LPD, along with the results of the independent review by MPR Associates, Inc (MPR).
The purpose of calculation NEDC 16-003, Revision 0, "Structural Evaluation of the Turbine Building Blowout Panels Steel Supports" determines the Turbine Building pressure from a High Energy Line Break (HELB), which would cause failure of the siding strucLUral support system, resulting in failure (blowout) of the structure's siding.
This calculation is implemented by Engineering Evaluation (EE)13-041, Revisfon 3, "Turbine Building Blowout Panels / Metal Wall System" [Enclosure 6]. The calculation's conclusions under its curTent form as well as prior versions have been the subject of controversy and disagreement by various parties during the past several years.
In 201 1, CR-CNS-2011-0050 I was generated to document an error within N EDC 03-005, "Turbine Generator Building Siding Blowout Pressure." Revision 3 to this calculation was later the subject of an NRC allegation, as covered under RTV-2013-A-0059. This calculation was revised several times and was ultimately superseded by NEDC 12-012, "Turbine Generator Building Siding Blowout Pressure, Other EQ Purposes". The quality of this calculation was later questioned as documented in CR-CNS-2012-03586, which led to the generation of a new calculation NEDC 13-028, Revision 0, "Ultimate Internal Pressure of Turbine Building Blowout Panels and MetaJ Wall System".
In 2015, issues were identified with Revision I of calculation NEDC 13-028. As a result of these issues, several condition repo11S were generated (CR-CNS-20 I 5-02718, CR-CNS-2015-02759, which culminated into on-Cited Violation (NCY 2015003-01) from the NRC as documented in CR-CNS-2015-05705 and CR-CNS-2015-06544. At the time, the NCY documented that the,calculation methodology and assumptions employed were not adequate and could not conclude that it ensured siding failure as required.
In response to this NCY, NEDC I 6-1003 was developed as a corrective action to CR-CNS-2015-06544 and documented as the design ba is calculation of record for the Turbine Building blowout panel capability, replacing NEDC IJ-028. As a part of this developmenl, LPI was chosen as the vendor to develop the calculation. The decision to use LPl was based on their industry-recognized experience in the nuclear indusb*y for their analytical capabilities and sttuctural knowledge. C S and Entergy both had prior experience using LPl, with satisfactory performance. In this case, the calculation was approved by Mr. Paul Bruck, LPI Principal Engineer who is widely known across lhe nuclear industry as a subject matter expen in structural analyses. This assessment of Mr.
NLS2020018 Attachment Page 5 of 17 Bruck was provided during an interview during this investigation by an Entergy Chief Engineer, who had no prior involvement wi1h the CNS calcula1ion in question.
Following the initial development of calculation NEDC 16-003, it was the subject of an NRC inspection, as documented under NRC lnspcction Report 2016003 dated November I 0, 2016. The results of this inspection found no additional issues related to calculation NEDC 16-003.
1.2 Summary of Conclusions 1.3 Overall, based on an independent review and the opinion of NPPD, the conclusions of calculation NEDC 16-003 are valid and meet the design and licensing basis at Cooper Nuclear Station. While the calculation i valid, it is noted that there are several non-conservatisms within the calculation. Specifically, MPR found non-conservatisms related to the context of the subjects or the caJculation documented in Issues I, 3, and 4.
However, for each of these issues, MPR conducted scoping evaluations that demonstrated that they were not of significance to negate the overall conclusions of the calculation and that the turbine building blowout panels will perform its safety function as designed.
Evaluator Independence and Qualifications The NPPD evaluator responsible for investigating and preparing the response to the RFI was Troy Barker, the Director of Engineering and most recently the Programs and Components Engineering Manager at Cooper Nuclear Station for - 8 years. Mr. Barker has been employed in the nuclear industry for over 20 years, during which time he has worked for Nebraska Public Power District at Cooper Nuclear Station. During this period, he has also worked in Quality Assurance and performed duties a an external and internal ANSI Certified Lead Auditor before serving a Quality Assurance Manager.
While in QuaJity Assurance, he led several internal and external supplier audits and surveillances and oversaw several others.
The Evaluator was independent of the subject maner of this RFJ, as he was not involved in associated events or decision-making. While he was a member of the CNS Engineering department during the same timeframe as the development or N EDC 16-003, he was in the Programs and Component Engineering group, a separate division within the Engineering department. He was also not a part of any associated decision-making, analysis, or other activities that led to the completion or the subject calculation.
1.4 Investigation Process and scope 1.4.J Investigation methodology
- a. The originator or NEDC 16-003 (LPl) was initially sought for feedback relative to the calculation allegations described in RIV-20 I 9-A-0065. LPI was provided with the concerns raised in the RA and given an opportunity to respond accordingly. This response, documented under LPI Ref. A 16254-LR-002, Rev. 0, in its ent1rety is contained in Enclosure 2. Excerpts from this response are contained within the body
NLS2020018 Attachment Page 6 of 17 of this report in the respective sections. For fonnatting purposes, text taken from this enclosure has been italicized.
- b. Separate from the LPI response, MPR Associates, Inc. (MPR) was contracted to complete an independent review under NPPD Ta k Authorization #4700002182. To accomplish this review, MPR was provided the allegation described in RIV-20 I 9-A-0065 and NEDC 16-003. MPR was not provided any correspondence from other entities prior to their review. This ensured the independence of their activities was maintained throughout. The results of this review are contained in its entirety in. Excerpts of the MPR review are contained within the body of this report in the respective sections. For fon11atting purposes, text taken from this enclosure has been italicized.
- c. Regarding this independent review, the following individuals were contacted ancl/or interviewed to seek their input or involvement in calculation NEDC 16-003:
- Vince Bacanscas - Chief Engineer, Entergy
- Paul Bruck, Principal Engineer, LPl
- Craig Swanner, Engineer, MPR
- David Bergquist, Engineer, MPR
- Alan Able, Design Engineering Supervisor, CNS
- Taylor Sutton, Engineer, CNS During an interview with Alan Able, who approved NEDC Calculation 16-003, he stated that his role at the time was the De ign Engineering Supervisor over the "Analysi "group. Hi primary focu a hi role a supervisor was to ensure that the correct proces had been followed during the development, review and approval of the calculation a well a ensuring that the indi vidual involved in the calculation were quali fied. Additionally, he tated that he had no concerns with the outcome and conclusion of the calculation nor had any concerns using LPI as the vendor for the calculation.
As the CNS Preparer of NEDC Calculation 16-003, Taylor Sutton was interviewed regarding this role. He provided information regarding the development of the calculation. He had no concerns with the conclusions of the calculation, nor did he have any concerns related to the quality or performance of LPL 1.5 Summary of results 1.5.1 Overall Assessment Based on LPI's response and the independent review by MPR, NPPD has concluded that the conclusions of calculation NEDC 16-003 are valid and each issue documented in RJV-20 J 9-A-0065 is not substantiated.
NLS2020018 Attachment Page 7 of 17 The following excerpt has been taken from the MPR report 03 15-0084-RPT-00 I, Rev. 0
[Enclosure 3J:
"MPR performed cm i11depe11de11t review of NEDC 16-003 focusing on the technical approach and 0 11erall validity of the report's co11clusions. As part of our review, MPR developed a number of scoping el'afuatio11s using ANSYS to test key assumptions and behaviors of the underlying steel support structure for the turbine building blowout panels. Note that MPR only considered the same subset of blowout panels as those determined in NEDC 16-003 to be 111ost susceptible to failure (Girt Types 133A, 1338, and /33C).
It should be noted that standard design codes are i11te11ded to build i11 margin to failure in a design and therefore, the results are not generally intended for calculating accurate actual failure loads. This was recognized in NEDC 16-003, and MPR considers the overall NEDC 16-003 methodology to be appropriate for estimating failure limits. In general, assumptions and analytical techniques are consistent with the physics of the problem and current state-of-the-art. Strucwral engineering code guidance, which would build margin into the load or stress limits, is appropriately not followed. MPR notes that the 0.5 psig HELB internal pressure load applied to the struclllre results in stresses for many of the connections that are well beyond the limits of design codes.
The key components in the structural load path are the girt-to-clip bolts and the fillet welds that connect the a11gfe clip to Jhe turbine building column. Based 011 our review of the structural behavior, if either one of these co1111ections Jails, then the 1111derlying support structure for a blowout panel will fail. To assess each component, MPR pe1formed a series of three dimensional finite eleme11t models of the bolted cmd welded connections including plasticity with acwal material properties. MPR's assessment of each of these connections is provided below:
Girt to Clip Bolts - Analyses were performed with (a) high.friction and high prefoad and (b) low friction and low preload. /11 both cases, MPR's model demonstrated that the joint slips through the clearance between the bolts and the bolt holes such that the bolts begin carrying load directly through shear. The slipping is caused by prying action and a high shear loadi11g. The model predicts that the combination of shear and tensile loadi11g with the moments that result from the prying action of the clip on the girt significantly exceeds the failure criteria. Therefore, MPR concludes that the girt to clip bolts are likely to Jail upon a 0.5 psig pressure loading. This is consistent with the conclusions of NEDC 16-003.
Clip to Colllm11 Welds - This model considered that the bolts remained intact and transferred all the load through their co1111ectio11 witho111 slipping. Results demo11strated high plastic strains in the weld well beyond the material's.failure strain. Other effects includirrg sag rods, potential.for Ol'er-sized welds, bolt slippage were evaluated individually. The strains are s11fficie11tly high in the base case that these other effects are 1101 expected to impact the conclusion that the welds will fail. Therefore, MPR co11c/11des that the welds would likely fail due to
NLS2020018 Attachment Page 8 of 17 the 0.5 psig pressure loading. This is consistent with the conclusions of NEDC 16-003.
MPR concludes that ca least one component in the load path of the turbine building blowout panels underlying structural steel will fail for the Girt Types 133A, 1338, and
/ 33C. Therefore, MPR agrees with the overall conclusions of NEDC 16-003 and concludes that the specified blowo111 panels will fail if subjected to a pressur,e loading of 0.5 psig. However, it is noted that MPR estimated the required blowout area to maintain a 0.5 psig pressure is approximately half of rite area predicted to blowout in NEDC 16-003. Therefore, MPR concludes that there is e,*en more margin than credited in NEDC 16-003.for uncertainty in the calculation inputs and assumptions. See Section 3.0
[Enclosure 3]for additional details on MP R's overall assessment of NEDC 16-003."
1.5.2 Assessment of Specific Issues:
In addition to the overall summary, MPR assessed each of the issues against the overall purpose of the calculation NEDC 16-003. These are detailed below.
A summa,y of MPR's assessment of each of the individual issues related to NEDC 16-003 is provided in Table 2-1. MPRjinds that Issues 2, 5, 6, and 7 are either not technical concerns or are already adequately addressed in NEDC 16-003. MPR notes that Issue 5 (required blowout area) was,wr addressed in NEDC 16-003.
To address the issue, MPR estimated the required area to maintain the internal pressure from the limiting HELB break at or below 0.5 psig robe approximately 2300 ji2. NEDC 16-003 predicts that turbine panels with 011 area equal to about 5400 Jt2 will Jail and blowout. Thus, there is comiderable margin between the predicted blowout area and the required blowout area. This margin provides added confidence that the potellfial input uncertainties (e.g., material limits) and 11011-conserl'atisms in NEDC 16-003 will not impact the conclusion that the turbine building blowo11t panels will satisfy their design basis function after a HELB event.
Issues I, 3, and 4 identify 11011-conse,vatisms in t/Je NEDC 16-003 calculation approach.
011 aggregate, MPR believes that these 11011-conse11*atis111s are balanced by other factors 1101 credited in N EDC I 6-003 ( e.g., dynamic effects of the fast pressure load application are neglected) and the quantified margin beyondft1i/11re. F11rther, MPR evaluated each of these effects and it is our opinion that these 11011-co11se1'1'C1tis111s will not impact our overall conclusions identified in Section 2.1 (i.e., at least one portion of the girt connection load path will Jail 1111der the 0.5 psig inremal pressure loading from a HELB event).
See Section 4.0 [Enclosure 3]for additional details on MPR's individual assessments of each iss11e.
NLS2020018 Attachment Page 9 of 17 1.6 Allegation Issues with Responses 1.6.J Issue #1 Assertion:
The girt bolt ultimate strength and shear su*ength assumed in Calculation NEDC 16-003 is non-conservative and may not be representative of the bolts used in the blowout panel con truction. Specifically, the calculation improperly use an average ultimate strength instead of maximum tested ultimate strength. Additionally, the average tensile strength is derived from testing of warehouse stock bolts which are likely from different heat and lots than of the bolts used in blowout panel construction. The average ultimate strength may not bound the ultimate strength of bolt u ed in construction.
Address the following:
Provide a discussion and evidence of any validation activities (walkdowns, testing, etc.) performed to ensure that the currently installed girt bolts arc of the same grade (ASTM Grade A307) and/or have the same material properties as the bolts that were tested for Calculation NEDC 16-003.
NPPD Response to Issue #I:
As documented in LPI Ref. A 16254-LR-002 [Enclosure 21, it is demonstrated that the bolt grade and material that was used during construction met the A307 grade material as seen through waJkdowns and material testing. Based on this information, the bolt properties used in NEDC 16-003 were appropriately utilized. Thi information was further supported by the independent review completed by MPR, whose overall conclusion within MPR Report 0315-0084-RPT-OOI, Rev. 0 [Enclosure 3), states "NEDC 16-003 assumptions regarding bolt strength may potentially be non-consen,ative. While MPR 's opinion is that the assumptions made in NEDC 16-003 to determine the bolt ultimate tensile strength (UTS) are reasonable, there is evidence that supports a higher bolt ultimate tens;/e strength by as much as 20%. While, there is uncertainty in the equivalence of the material properties of the bolts tested and those installed in the building, the magnitude of this uncertainty would be bounded by the 20%
above, which would put the bolts in the upper range of possible bolt strengths for A307.
These uncertainties were considered when making our overall assessment that the bolts will fail."
The assessment by MPR regarding bolt strength that may potentially be non-conservative listed above was addressed by the development of scoping evaluations by MPR, which concluded that they arc consistent with the methodologies of NEDC 16-003.
Therefore, taking into consideration the comments by LPI along with the results of the independent assessment conducted by MPR documented in MPR Report 0315-0084-
NLS2020018 Attachment Page 10 of 17 RPT-001, Rev. 0 !Enclosure 3), PPD concludes that the concern outlined in Issue #I is not substantiated.
1.6.2 Issue #2 Assertion:
Calculation NEDC 16-003 uses non-conservative failure mechanisms because it assumes that the inner and outer girt bolls, the near and far fillet welds, and the sag rod will fail in a sequential manner rather than being loaded in a shared load condition. Assuming a shared load condition for these components would be more realistic and would significantly increase the amount of pressure required to ensure the bolts and the welds would fail.
Address the following:
Provide the technical justification as to why it is appropriate to model the loading of the bolls, welds, and sag rods in a sequen1ial manner rather than modeling the load as being simultaneously distributed across all components.
NPPD Response to Issue #2:
As documented in LPl Ref. Al6254-LR-002 lEnclosure 21, "The evaluation pe1fonned in the Ref. 1 [Enclosure 11 calculation considered load sharing - and was in-fact "simultaneously distributed across all components". Once a sub-component (i.e. bolt, weld, girt) reaches the load corresponding to its failure, it is removed from the analysis and the remaining components then carry the load."
MPR conclusion states, "MPR concludes that the overall failure sequence and the load sharing documented in NEDC 16-003 between the girt-to-clip connection bolts and the clip to column welds is based on the actual physics of the problem and is appropriate for use in the evaluation. See conclusions for Issues #3 and #4 which specifically address the sag rods and bolr friction."
Therefore, taking into consideration the comments by LPI along with the results of the independent assessment conducted by MPR documented in MPR Report 0315-0084-RPT-OO I, Rev. 0 fEnclosure 3), PPD concludes chat the concern outlined in Issue #2 is not substantiated.
1.6.3 Issue #3 Assertion:
Calculation NEDC 16-003 does not appropriately account for resi tance from sag rods installed between channel giJts. These sag rods would restrain girt channel twisting and should be factored into the analysis.
NLS2020018 Attachment Page 1l of 17 Address the following:
Calculation NEDC 16-003, Section 2.2, Assumption 3, states that the "sag rods supporting the channel girts are flexible in bending because of the relatively large span (between girts) and small cross section, resulting in flexibility and low bending resistance. Therefore, the sag rods do not provide significant resistance to twisting of the girt channel. The relative stiffness of the blowout panels is likewise small compared to the stiffness of the girt channel and will not significantly affect the twisting of the girt channel.
Provide the technical information (dimensions, material properties, etc.) of the ag rod used Lo justify Calculation NEDC 16-003, Section 2.2, Assumption 3.
NPPD Response to [ssue #3:
As documented in LPI Ref. A 16254-LR-002 [Enclosure 21, "The calculation considers rhe sag rods robe flexible in bending, based 011 rheir lengrh, lack of resistance to bending, and end connecrion of rhe sag rod passing rhrough a hole in rhe connected girt with gravity attachment only. As a result, they will not influence the twisting of the girt channel. (See Ref 1 LEnclosure I], Section 2.2, Assumption No. 3).
The sag rods are identified on plant Drawing No. 4089 (Ref 6) [Enclosure 4], as "5/8" dia., Typ. ". Sag rod length would be co11sisre111 wilh girt spacing. Ref 8 calls/or structural steel for rhe Turbine Building to be ASTM A36 material, Figure 3. Sag rods pass through holes in the girts and are considered as pin-ended (i.e., the sag rod-to-girt connection is not rigid enough in bending to provide significant rotational restraint)."
MPR states in section 4.3.2: "The result is thar rhe sag rods resist some of the applied loads and the loads in the welds and bolting are reduced by the presence of the sag rods.
The load in each weld is reduced 011 rhe order of I 0%, and the bolting loads are reduced on the order of 30%. This result is consistent over the height of the wall (i.e., the top girt and bottom girt see a similar impact). Based on these results, the effect of the sag rods are not insignificant. However, it is,expected that the magnitude of this effect would change with a more rigorous modeling of the bolted connection, as in Secrion 3.2.2. In any case, this effect must be considered with other effects to determine if the structures will fail prior to the turbine building exceeding 0.5 psig."
MPR conclusion states, "NEDC 16-003 neglects the effect of the sag rods. The sag rods would reduce girr twisring and loading on the girr connecrions, so this assumption is therefore non-conservative. Scoping analyses pe,formed by MPR show rhat this reduces the loading on the girt connecrions by as much as 10% for each weld and 30% for the bolts. The margins beyond failure i11 NEDC /6-003 would be reduced by this amoullf, but would still result in a failure. This effect is considered in MP R's overall assessment." Sec section 1.4.2 of this report above.
Therefore, taking into consideration the comments by LPI along with the results of the independenl assessmenl conduc1ed by MPR documented in MPR Report 0315-0084-
NLS2020018 Attachment Page 12of 17 RPT-00 I, Rev. 0 I Enclosure 3), PPD concludes that the concern outlined in Issue #3 is not substantiated.
1.6.4 Issue #4 Assertion:
Calculation NEDC 16-003 does not use the correct method for joint design in accordance with AlSC tandards. The calculation use a method that accounts for joint friction for as-built girtjoim that is not conservative and does not meet ATSC requirements.
ln additjon, address the following:
Provide the relevant construction standard used to assemble the turbine building column to girt joint. Please pecify the type of bolted joint design of the as-built girt joint (i.e. shear/bearing, lip-critical, or direct ten ion).
Provide the torque specilicat ions, if any, for the girt bolting.
Provide the technical justification for NEDC 16-003, Section 2.2, Assumption 9, "Friction between steel surfaces is credited to transfer load between the angle and channel and between the angle and column flanges following weld failure. A range of friction for carbon steels is provided in various text (see l 16J as an example). A dry mild steel on mild steel value of 0.78 is listed under static conditions, with a sliding value around 0.42. A value of 0.35 has been commonly used as a lower bound friction value for most steel on steel surface conditions.
Since selecting a low value is considered conservative, a value of 0.25 is utilized herein."
NPPD Response to Issue #4:
As documented in LPJ Ref. A I 6254-LR-002 [Enclosure 2], "The calcula1ion is pe1fonned to ascertain 1he point of failure of the bolted joint. A/SC standards are associated with design. Applying A/SC s1andards to predict failure when A/SC equations are not met will result in a lower prediction ofjoi111 strength, and thus lower fa Nu re pressure.
In applying friction at the girl joint, a similar question was raised and responded 10 in Ref 3 [Enclosure 5], Question 4. In swm11a0*, consideration of some degree of friction here is conservative, to maximize the resistance of the girt lo HELB pressure. Use of fric1io11 in this part of 1he evaluation is considered conservative. Fricfion in 1his parf of 1he calculation is associa1ed with "posi1ioni11g" the girt, once bolt failure has occurred, not the strengfh of the bolted joint evaluation."
LPI response to bullet I: "The column to girt joinf bolled connection is a bearing type connection. This is consistent with A/SC requiremenfs for an A3O7 bolted joint."
NLS2020018 Attachment Page 13 of 17 LPI response to bullet 2: "The bolts are unfinished A307, in a bearing connection. LP/ is 1101 aware of any torque specification for the girt bolts. Our assumption would be they
\\Vere i11sta/led using good bolting field practices. The assessment pe,formed in Ref I
[Enclosure I] does not rely 011 a minimum value of torque being prese11t i11 the bolt."
LPI response to bullel 3: "The purpose of considering friction is to hold or position" the girt in a horizontal position, thus 111axi111izi11g its resistance to bending u11der HELB pressure. The result of this approach is a higher pressure to Jail the girt component. If friction \\Vere to be lower/lost, rotation of the girt would occur, resulting in weak axis bending of the girt, and thus lower resista11ce to the HELB pressure.
As 1Vas demonstrated in Section 4.3, frictio11 at the lower value used of 0.25 is sufJkient to maintain the deadweight reactioll' of the girt at the column. If the friction were higher this conclusion 1Vould not change, the answer would not change the failure pressure derived for the girr. ff.friction was sig11ijicantly lower or zero, the girt would rotate and fail at a lower pressure tha11 currently predicted."
MPR conclusion states, "NEDC 16-003 does not address preload or the friction that could develop between the girt and the clip if the bolted connection were preloaded.
Crediting friction results in load sharing bet1Vee11 the bolts and frictio11 in the joints. Load sharing reduces the bolt loads and is 11on-conservati1*e to neglect when evaluating failure. Scoping analyses pe,formed by M PR demonstrate that for the largest friction coejJ,cient a11d the bolts preloaded to two-thirds yield stre11gth, the girt to clip joint is expected to slip through the clearances such that the bolts will be loaded in shear through contact with the clip and girt. These analyses also confirm that the forces and moments applied to the limiting bolr are beyond its failure capacity based on actual material properries. See Section 3.2.2 for additionc,I discussion."
Regarding the aforementioned non-conservatism related to load sharing, MPR states, "Load sharing reduces the bolt loads and is 11on-consen1ative to neglect when evaluating failure. Section 3.2.2 provides scoping analyses that assess whether thejoinr will slip or not, and detennines the resulting bolt loads. Two cases are pe,formed - one case with high friction and pre load and the other case with low friction and pre load. The scoping analysis results show that the joint will slip under the 0.5 psig loading even when a large frictio11 factor and large preload are considered. These analyses also corifim, that the forces and moments applied to the limiting bolt are beyond its failure capacity based on actual material properties. See Section 3.2.2 for additio11al discussion."
Therefore, taking into consideration the comments by LPI along with the results of the independent assessment conducted by MPR documented in MPR Report 03 15-0084-RPT-00 I, Rev. 0 [Enclosure 3), PPD concludes that the concern outlined in Issue #4 is not substantiated.
NLS2020018 Attachment Page 14 of 17 1.6.5 Issue #5 Assertion:
Calculation NEDC I 6-003 does not demonstrate that the 5362 square foot blowout area is adequate to prevent pressure retention beyond 0.5 psig in the Turbine building following a High Energy Line Break (HELB) event. A previous analysis performed by a contractor using GOTHIC modeling found that a blowout area of 30,000 square foot sti II resulted in pressure retention.
In addition, address the following:
Calculation NEDC 16-003, Section 5.0 states, "Using a HELB pressure of0.45 psi, it is predicted failure of the panel support system at this pressure will result in a loss (blowout) of total panel area of 5,362 ft2." Provide the engineering analysi or calculation that demon trates that a total panel area of 5,362 ft2 is enough to maintain peak pressure in the turbine building below 0.5 psig.
NPPD Response to Issue #5:
As documented in LPl Ref. A 16254-LR-002 [Enclosure 2], the purpose of NEDC 16-003
[Enclosure I] calculation was to demonstrate the blowout siding panels would fail at or below 0.5 psi. CalcuJating the required area was not in the scope of the analysis.
Additionally, as MPR concluded: "NEDC /6-003 does not address this issue. Scopi11g calculations pe,formed by MPR estimate that an open area of only about 2300 Jt2 is required to maintain the turbine building general pressure to 0.5 psig or below during a HELB event. This is less than half of the predicted blowout area from NEDC 16-003 (5400 jt2)_"
Therefore, talcing into consideration the comments by LPI along the results of the independent assessment conducted by MPR documented in MPR Report 0315-0084-RPT-001, Rev. 0 [Enclosure 31, NPPD concludes that the concern outlined in lssue #5 is not substantiated.
1.6.6 Issue #6 Assertion:
The assumed turbine building siding blowout area in NEDC 16-003 (5362 square foot) is based on column centerline which would be non-conservative because column tlanges would protrude beyond column centerline and reduce the actual blowout area. This would increase the pressure retention in the turbine building potentially beyond 0.5 psig.
Jn addition, address the following:
Provide engineering drawings that show the actual dimensions of the blowout area credited in NEDC 16-003.
NLS2020018 Attachment Page 15 of 17 See Enclosure 4 for drawings of Turbine Building blowout panels.
NPPD Response to Issue #6:
As documented in LPl Ref. A 16254-LR-002 [Enclosure 2], "The siding pa11els provide continuous coverage of the Turbine Building. The girt detail is such that it projects beyond the flange of 1he column... As such, siding "behind" column flanges are exposed to Turbine Building HELB pressures, and the pressurized area considered in the Ref I
[Enclosure l] calculation is appropriate. "
MPR conclusion states, "Evaluations pe,formed by MPR demonstrate that the width of the columns does 1101 significantly impact the available flow area. This result combined with the conclusions to Issue #5 led MPR to conclude that available flow area is not a technical concern."
Therefore, taking into consideration the comments by LPI along with the results of the independent assessment conducted by MPR documented in MPR Report 03 15-0084-RPT-00 I, Rev. 0 [Enclosure 3), NPPD concludes that the concern outlined in Issue #6 is not substantiated.
1.6.7 Issue #7 Assertion:
Calculation NEDC 16-003 includes non-conservative assumptions involving how the forces against the turbine building siding are distributed during a high energy line break.
Specifically, the dist1ibuted force only a sume a single girt channel whereas the actual construction of the turbine building blowout panel includes multiple girt channels.
Consequently, the resultant force may not be enough to ensure boll and weld failures al pressures less than 0.5 psig.
In addition, address the following:
Clarify whether the resultant force is distributed across only a single girt channel connection or across multiple girt channel connections supporting each turbine building blowout panel.
Provide a copy of the most current revision of Calculation NEDC 16-003. [See Enclosure I)
NPPD Response to Issue #7:
As documented in LPJ Ref. A 16254-LR-002 [Enclosure 21, "The HELB pressure is applied over the entire area of the siding panels. The analysis of the girt channel in NEDC 16-003 considers the panel pressure tributary to a single girt. Other girts of similar span and spacing will have similar loading and show similar behavior. Use ofa tributary area is appropriate."
NLS2020018 Attachment Page 16 of 17 MPR conclusion states, "Scoping evaluations performed by MPR of the entire wall showed a negligible difference in the load sharing between girts when compared to the simplified single girt models. Therefore, MPR concludes thar the Tributary merhod used 10 apply rhe pressure load in NEDC 16-003 is acceptable for calculating loads 011 a single girt".
Therefore, taking into consideration the comments by LPI along with the results of the independent assessment conducted by MPR documented in MPR Report 0315-0084-RPT-001, Rev. 0 lEnclo ure 3], PPD concludes that the concern outlined in Issue #7 is not substantiated.
- 1. 7 Conclusions Based on the LPJ response and the independent review by MPR, NPPD has determined that the conclusions of calculation NEDC 16-003 are valid and each issue documented in RlY-2019-A-0065 are not substantiated. While some non-conservatisms were identified during the independent review, the overall conclu ion were not adver ely impacted by these items. Specifically, MPR found non-conservatisms related to the context of the subjects of the calculation documented in Issues I, 3, and 4. However, for each of these issues, MPR conducted scoping evaluations that demonstrated that they were not of significance to negate the overall conclusions of the calculation and that the turbine building blowout panels will perform its safety function as designed.
1..8 References
- a. NRC RIY-2019-A-0065
- b. NPPD Calculation NEDC 16-003, "Structural Evaluation of the Turbine Building Blowout Panel Steel Supports"
- c. 4530l5347_DRW_OOO_AA Capitol 9150 Sh 156
- d. 4530 I 5352_0RW _OOO_AA Capitol 9150 Sh 159
- e. 453015353_DRW_OOO_AA Capitol 9150 Sh 158
- f. 4530 l5354_DRW _OOO_AA Capitol 9150 Sh 157
- g. 453015407_DRW_OOO_AACapitol9l50Sh 139
- h. 4530 I 5485_DRW _OOO_AA Capitol 9150 Sh 133
- i. 454015468_DRW_OOO_AA Capitol 9150 Sh EIOI
- j. 4540l5470_DRW _OOO_AA Capilol 9150 Sh El07
- k. 45401547 I_DRW _OOO_AA Capitol 9150 Sh E106 I.
EDS Nuclear Report O 1-0840- 11 15, Rev 0
- m. FSAR Amendment-25
- n. INRYCO Vacuum Load Test Job No 49054
- o. NEDC 13-028 Rev. I Approved
- p. LPI Ref. A 16254-LR-002 Rev. 0, "Responses to NPPD Supplied Questions for Calculation No.16-003 (LPI Cale. A 15406-C-OO I) - Cooper Nuclear Station"
- q. NPPD Task Authorization #4700002 182
NLS2020018 Attachment Page 17 of 17
- r. MPR Report 0315-0084-RPT-001, Rev. 0, "Independent Review of Underlying Structural Steel for Turbine Building Blowout Panel "
- s. LPI Ref. A16254-LR-001, Rev. 0 "Responses to RC Que tions for NPPD Calculation o.16-003 (LPI Calculation Al5406-C-001)- Cooper Nuclear Station'
H Nebraska Public Power District NLS2020018 May 12, 2020 ATTN: Jesse M. Rollins Senior Allegation Coordinator
'Alw...,., clurt 14fl,'fl w11 nw.l,u*
U.S. Nuclear Regulatory Commission 1600 East Lamar Boulevard Arlington, Texas 76011-451 1 Signed Cover Letter
Subject:
Reference:
Response to Request for lnfonnation -Allegation RIV 2019-A-0065 Letter from Jeremy Groom, U.S. Nuclear Regulatory Commission to John Dent, Jr., Nebraska Public Power District, "Request for Information,"
dated January 16, 2020
Dear Sir:
ln response to the referenced letter, Nebraska Public Power District (NPPD) has completed the requested evaluation of certain activities at Cooper Nuclear Station. Results of NPPD's independent evaluation arc provided in the attachment to this letter.
In accordance with your requests, this correspondence is not docketed and is provided only to you. Internal distribution of this report at NPPD has been limited to a "need to know" basis.
This letter contains no new commitments.
Please call Rhonda Aue, Employee Concerns Coordinator at 402-825-5020 should you have any further questions regarding the matter.
Sincerely, w
~
~
Vice President and Chief Nuclear Officer Ira COOPER NUCLEAR STATION 72676 648A Ave I P O Box 98 1 Brownvtlle, NE 68321 http:Jtww.v nppd.com
NLS2020018 Page 2 of 2
Attachment:
Independent Evaluation Results, Allegation RIV-20 I 9-A-0065
Enclosures:
- 1. NPPD Calculation NEDC 16-003, Rev. 0, "Structural Evaluation of the Turbine Building Blowout Panels Steel Supports"
- 2. LPI Ref. Al6254-LR-002, Rev. 0, "Responses to NPPD Supplied Questions for Calculation No.16-003 (LPI Cale. Al5406-C-O0l) - Cooper Nuclear Station"
- 3. MPR Report 0315-0084-RPT-001, Rev. 0, "Independent Review of Underlying Structural Steel for Turbine Building Blowout Panels"
- 4. NPPD Drawings
- 5. LPI Ref. Al6254-LR-001, Rev. 0, "Responses to NRC Questions for NPPD Calculation No.16-003 (LPI Calculation A 15406-C-0O 1) - Cooper uclear Station"
- 6. NPPD Engineering Evaluation (EE)13-041, Revision 3, "Turbine Building Blowout Panels / Metal Wall System"
NLS2020018 Enclo ure 2 Page I of I LPI Report Ref. A16254-LR-002, Rev. 0, "Re pon e to PPD Supplied Questions for Calculation o.16-003 (LPI Cale. A15406-C-001) - Cooper ~uclear Station
(11 pages)
Mr. Troy Barker February 14, 2020 LPI Ref. A16254-LR-002, Rev. 0 Nebraska Public Power District Cooper Nuclear Station PO Box 98 Brownville, NE 68321-0098
SUBJECT:
Dear Mr. Barker:
Responses to NPPD Supplied Questions for Calculation No.16-003 (LPI Cale. A 15406-C-001) - Cooper Nuclear Station NPPD Contract Order No. 4200002638 Amendment 2 At the request of NPPD, LPI, Inc. (LPI) has prepared responses to NPPD supplied questions related to NPPD Calculation No.16-003, Revision 0, (LPI Calculation No. A15406, Revision 1), titled "Structural Evaluation of the Turbine Building Blowout Panels Steel Supports".
The questions and our responses are included as an attachment to this letter. The provided responses include clarifications. All responses have been reviewed in accordance with LPI Quality Assurance program for nuclear safety related work.
This letter report is prepared in accordance with the requirements of NPPD contract number 4200002638 Amendment 2. If you should have any questions, or if we can provide additional assistance, please do not hesitate to contact us.
0 2/14/20 Rev.
Date Respectfully Submitted, LPI, Inc.
~/,,. J,
.Atldll?Jf B. Elaidi, A.Chock, Pr. Engineer Pr. Engineer Prepared by Checked by
.At/d!/c,_,I
?.;#
A.Chock, P. Bruck Pr. Engineer Principal Design Verified by Approved by*
'The approver of this document attests that all project examinations, Inspections, tests and analysis (as applicable) have been conducled usln a roved LPI Procedures and are In conformance to the contracV urchase order.
Attachment A: Responses Luaus P,t.liin E.sl 1B85 * --tp,nycom Nt;w Yor~
- Mo$Siiehuaeus
- w~s,,mgtor,
- ca,,1om11
- flotlda
- A11wn1,,
L11s11m,.~ ll1,* 111ti*.~ritv of t,\\lav', cnt,rnl 111fr11,tri,ct11rc fi,r timwrrou*~ mJrld
1 2
3 4
5 6
7 8
9 10 11 12 13 14 15 16 17 18 19 20 21 (1)
(2)
Mr. Troy Barker LPI Ref. A16254-LR-002, Rev. 0 DESIGN VERIFICATION CHECKLIST IJ1~
D.
Document No(s)1: I A 16254-LR-002
~
Review Method:
I x I Document Review I
!Alternate Calculation Cri1teria Were the inouls correcllv selected and lncomnraled into desian?
Rev.:
!Test Are assumptions necessary to perfonn the design activity adequalely described and reasonable? Where necessary, are the assumptions identified ror subsequent re-verifications when lhe detailed design activities are completed? II applicable, has an as built verification been oerformed and reconctled?
Are the aoorooriate aualitv and aualilv assurance reauirements S""cified?
Are the applicable codes, standards and regulatory requirements including issue and addenda properly Identified and a re their reaulrements for desian met?
Have applicable construction and operatir,q experience been considered. indudillQ operabon procedures?
Have the desiqn Interface re<1uirements been s~ttsfied?
Was an appropriate design method used?
Is the outoul reasonable comoared lo inouts?
Are the soeclfied parts, eaulpmenl, and orocesses suitable for the reauired aoalicalion?
Are the specified materials compatible with each other and the design environmental condll10ns to which the material will be exposed?
Have adeauate maintenance teatures and reaulrements been specified?
Are accesslbllltv and other desl<1n orovlslons adeauate for performance or needed maintenance and repair?
Has adequate accessibility been provided to perform the in-service insoection exoected to be required durinQ the plant life?
Has the deslQn properly considered radiation exoasure to lhe public and plant personnel?
Are the acceptance criteria Incorporated in the des,gn documents sufficient to allow verification that design requirements have been satisfactorilv accomo!ished?
Have adeouate ore-ooerational and subseauent oeriodrc tosl reauirements been aooroPnatelV soeclfied?
Are adeouate hancmno, storaae, cteanlno and sh1pplno reauirements soec11ied?
If software was used, have the computer type and operating system been property identified? Is use of the software, hardware and O/S aooroorlate for tha conditions, comoonents evaluated? Has a V&V been oerformed?
Are reaulrements for identification, record oreoaratron revrew. anrvoval. retention etc., adeauatelv soecified?
Has an Internal design review been performed for applicable design projects? Have comments from the Internal Design Review been appropnalely considered/addressed?
Are the necessary design Inputs for Interfacing organizations speofied in the design documents or m supporting procedures or lnstrucUons?
Include any drawings developed from reviewed documents, or include separate checkhsr sheet for drawings Design Verifier shall /n,tial indicating review and mark NIA where not applicable DV Completed By: I A. Chock I Signature: p1.:1//4,,.1 02/20/2020 Page I 1 lot I 1 !Total Pages (include DV Checklist and Comment Resolution sheets in page count) 0 ov2 AC AC AC AC NA AC AC AC NA NA NA NA NA NA NA NA NA NA AC NA NA Form: LPt-3.1-Rev-10-Flg-5-5 Page 2 of 2
LPI Ref. A16254-LR-002, Rev. 0 Attachment A LPI, Inc. Responses to Request from NPPD (Ref. 5), regarding the Reference 1 NEDC 16-003 document.
References:
"Structural Evaluation of the Turbine Building Blowout Panels Steel Support'"
(Completed for NPPD under agreement No. 4200002638). Dated 2/4/2016
- 2. LPI Letter A15406-L-007, "Transmittal of Owner Comment Responses for LPI Calculation A 15406-C-00 1, Rev. 1.
- 3. LPI Letter Report A 16254-LR-001, Rev. 0 "Responses to NRC Questions for NPPD Calculation No.16-003 (LPI Calculation A15406-C-001)-Cooper Nuclear Station (Completed for NPPD under agreement No. 4200002638). Dated 7/1 /2016
- 4. NRC Letter to NPPD, dated 11/10/2016, "Cooper Nuclear Station - NRC Integrated Inspection Report 05000298/2016003", page 26, para.2a, bullet 1, "Review of Calculation NEDC-16-003 associated with turbine building blowout panels", Adams Accession No. ML16315A141.
- 6. NPPD Drawing 4089, Rev. 6 "Structural Turbine Generator Building Elevation - Sheet No. 2"
- 7. NPPD Drawing 4094, Rev. 1 "Structural Miscellaneous Steel Details"
- 8. Contract No. E-69-16, "Structural Steel for Turbine Generator and Reactor Building and Intake Structure," Rev. 11, 7/11/69.
- 9. AISC Manual of Steel Construction, 61h Edition. 1971.
Discussion:
The subject calculation, NEDC 16-003 (LPI Calculation No. A15406-C-001, Rev. 1), [1] was prepared for NPPD to address failure of blowout panels on the Turbine Building (TB), under High Energy Line Break (HELB) pressurization. The calculation also resolved issues identified in a previous calculation (NEDC-13-028) by an earlier NRC inspection.
Calculation NEDC 16-003 was extensively reviewed by NPPD [2], with all technical issues resolved to the satisfaction of NPPD. In 2016, the calculation was further reviewed during a Plant NRG inspection and subsequent review by NRC's NRR branch, with resulting RAl's addressed in Reference 3. Based on a technical review by RIDE branch and an independent finite element analysis, the staff confirmed that the subject calculation "adequately established the range of failure of the connecting welds at high energy line break pressures below 0.5 pounds per square inch." [4]
Page A1 of 9
LPI Ref. A 16254-LR-002, Rev. 0 Attachment A LPl's responses to additional questions submitted by NPPD in the February 6, 2020 email [5]
are provided below in red.
Issue # 1 The girt bolt ultimate strength and shear strength assumed in Calculation N EDC 16-003 is non-conservative and may not be representative of the bolts used in the blowout panel construction. Speci ti cal ly, the calculation improperly uses an average ultimate strenglh instead of maximum tested ultimate strength. Additionally, the average tensile strength is derived from testing of warehouse stock bolts which arc like ly from different heat and lots than of the bolts used in blowout panel construction. The average ultimate strength may not bound the ultimate strength of bolts used in construction.
Address the following:
Provide a discussion and evidence of any validation activities (walkdowns. testing, etc.) performed to ensure that the currently installed g irt bolts are of the same grade (ASTM Grade A307) and/or have the same material properties as the bolts that were tested for Calculation EDC J 6-003.
Response to Issue # 1:
Within Ref. 3, NRC Question 2 documents response to a somewhat similar question.
Further information can be found in response to that question in Reference 3.
Notes on drawing 4094 [7], - see Figure 1 below, indicate requirements for bolts, rivets, welds are as described in the specification. The specification for the Turbine Building (8) specifies A307 bolting-see Figure 2 below.
In summary, plant design documentation for the bolts specifies A307 material (see Ref. 8, herein).
GE~&l?AL STEEL NOTES
. ). ~H)IJl2EMffl DE-Scet~EO IN C:
sot r.s, 12/V~TS ANO NctOS SHAU. l}Jf-AS
.,-ff-elf ICIT/01/S Figure 1 - Note 1 of Drawing 4094 (7)
Page A2 of 9
LPI Ref. A16254-LR-002, Rev. 0 Attachment A 2,3,4 Connections "One sided or other type;' o*l ~cdentric" "c"on~ect ions will not be permitt ed unless shown in detail and approved by the Engineers on the shop drawings,
Surfaces of joints for riveted, welded, or high-strength-bolted connections shall comply with the clean-liness requirements of all joint surfaces as specified ln Section "3b, Bolted Parts of the Specification for Structural Joints using ASTM A-325 Bolts.
Unless othenrise indicated on the dTawings, or in paragraph 2.3.4.7 of this specification, bolted connections for purlins, girts and secondary framing connections when approved on the shop drawings shall be made with ASTM A-307 unfinished bolts. All other bolted connections Ah.. 11 be made u11inll hillh stren2th bolts."
~=***---~v..... ~.. 0... - ~- :.s.
w...... v....n.u.. ~ Wl.CO t oe aeta11s shown on the dnwings.
2.2. 6 Paint
.Q,nfinished Bl
~-.;.;::.:::~~~o~t~~
~ ~~,1 conform to Federal Spec Unfinished b requirements of olts, nuts and 2
- 3 W~~ut1;:lQ7._" Ann, w, as hers shall vv, CflAN,,,-,,.-.1:& conform
<.::>t=
7,.!:,
to the Figure 2 - Excerpts from Contract No. E-69-15 Steel Requirements for Turbine Building [8]
NPPD elected not to remove bolting from the Turbine Building but provided a sample of A307 bolts from plant stock to LPI in support of the Ref. 1 calculation. ASTM bolt specification strength requirements have not changed from the time of construction of the plant to current day specification
[3]. As a result, it is reasonable to conclude the supplied bolts to A307 requirements would have similar properties to A307 bolting supplied at the time of construction Field walkdowns were performed to identify bolt head markings, bolt thread projection and confirmation of assembly configuration. Bolt head markings for the field observed bolts were assessed to ensure strength was not greater than A307.
For the tested bolts, an average of the tests resulted in an approximate tensile strength (Fu) value of 66 ksi. This value is 10% greater than the code specified 60 ksi tensile strength value, with the highest tested value of 70 ksi reported. Given the large number of bolts within the Turbine Building being evaluated in this connection detail, it is reasonable to conclude that they would have a distribution of strength properties. Assuming all bolts have strength equivalent to the highest strength bolt tested Page A3 of 9
LPI Ref. A 16254-LR-002, Rev. 0 Attachment A would be unreasonable.
As such, the average value utilized was considered as reasonably representative of the bolt strength for the bolts of this type in the Turbine Building.
If the highest tested value of 70 ksi had been selected, it can be seen as outlined in Section 4.2 of Ref.
1, the resulting calculated bolt thread stress of 85 ksi under 0.25 psi HELB pressure remains well above the highest tested value of 70 ksi.
Failure of the bolts as outlined within the calculation are demonstrated (i.e. calculated stress of 85 ksi at 9.25 psi) versus tensile strength values of 66 ksi (used) or 70 ksi (maximum tested value).
Bolts within Column Line D were evaluated in Figure 4-15. It can be seen that the resulting "HELB Pressure Load Fraction" would adjust from 0.8 to 0.82, using a bolt stress failure value of 70 ksi vs the 66 ksi used. Thus, using this pressure load fraction of 0.82, would result in a pressure of 0.82 x 0.5 psi
= 0.41 psi. This remains below the 0.5 psi value summarized in the Section 5.0 Conclusion. Failure of the bolts are demonstrated at tensile strength values of 66 ksi or 70 ksi.
As such conclusion of bolt failure using the highest tested bolt strength of 70 ksi vs the average value of 66 ksi used in Ref. 1, would not significantly change the failure pressure, and do not change the conclusions reached.
Issue #2 Calculation NEDC 16-003 uses non-conservative failure mechanisms because it assumes that the inner and ouler girt bolts, the near and far fillet welds, and the sag rod will fail in a sequential manner rather than being loaded in a shared load condition. Assuming a shared load condition for these components would be more realistic and would significantly increase the amount of pressure required to ensure the bolts and the welds would fail.
Address the following:
Provide the technical justification as to why it is appropriate to model the loading of the bolts, welds, and sag rods in a sequential manner rather than modeling the load as being simultaneously distributed across all components.
Response to Issue # 2:
The evaluation performed in the Ref. 1 calculation considered load sharing -
and was in-fact "simultaneously distributed across all components". Once a sub-component (i.e. bolt, weld, girt) reaches the load corresponding to its failure, it is removed from the analysis and the remaining components then carry the load.
Page A4 of 9
LPI Ref. A 16254-LR-002, Rev. 0 Attachment A lssuc #3 Calculation N EDC 16-003 does not appropriately account for resistance from sag rods installed between channel girts. These sag rods would restrain girt channel twisting and should be factored into the ana lysis.
Address the following:
Calculation NEDC 16-003, Section 2.2, Assumption 3, states that the "sag rods supporting the cha1mel girts are ilexible in bending because of the relatively large span (between girts) and small cross section, resulting in flexibility and low bending resistance. Therefore, the sag rods do not provide sig nificant resistance to twisting of the girt channel. The relative stiffness o f the blowout panels is likewise small compared to the stiffness of the girt channel and wi II not significantly affect the twisting of the girt channe l."
Provide the technical information (dimensions, material properties, etc.) of the sag rods used to justify Calculation EDC 16-003, Section 2.2, Assumption 3.
Response to Issue # 3:
The calculation considers the sag rods to be flexible in bending, based on their length, lack of resistance to bending, and end connection of the sag rod passing through a hole in the connected girt with gravity attachment only. As a result, they will not influence the twisting of the girt channel. (See Ref. 1, Section 2.2, Assumption No. 3).
The sag rods are identified on plant Drawing No. 4089 (Ref. 6) as "5/8" dia., Typ.". Sag rod length would be consistent with girt spacing. Ref. 8 calls for structural steel for the Turbine Building to be ASTM A36 material, Figure 3. Sag rods pass through holes in the girts and are considered as pin-ended (i.e., the sag rod-to-girt connection is not rigid enough in bending to provide significant rotational restraint).
2.2 Materials 2.2.1 Structural Steel All structural steel shapes, plates and angles 11 11n1Pct<l nthPl"w1~P !'l.nP~ifi Pti. Rhall confor m to ASTM A36 Where i ndi cated on the drawings, columns for the I ntake Structure shall conform to ASTM A-441." AOO. I C, HAIVGE P 2. '-t C:. c:
- c:
1.,.1.*ru1c nc...1...1. o Figure 3 - Structural Steel Specification, Excerpt from [8]
Page A5 of 9
LPI Ref. A 16254-LR-002, Rev. 0 Attachment A lssue #4 Calculation NEDC 16-003 does not use the coJTect method for joint design in accordance with A ISC standards. The calculation uses a method that accounts for joint friction for as-built girt joint that is not conservative and does not meet AlSC requirements.
Response to Issue # 4:
The calculation is performed to ascertain the point of failure of the bolted joint. AISC standards are associated with design. Applying AISC standards to predict failure when AISC equations are not met will result in a lower prediction of joint strength, and thus lower failure pressure.
In applying friction at the girt joint, a similar question was raised and responded to In Ref. 3, Question
- 4. In summary, consideration of some degree of friction here is conservative, to maximize the resistance of the girt to HELB pressure. Use of friction in this part of the evaluation is considered conservative. Friction in this part of the calculation is associated with "positioning" the girt, once bolt failure has occurred, not the strength of the bolted joint evaluation.
In addition, address the following:
Provide the relevant construction standard used to assemble the turbine building column to gi1t joint Please specify the type of bolted joint design of the as-built girt joint (i.e. shear/bearing, slip-critical, or direct tension).
Response to Issue # 4 - bullet #1 :
The column to girt joint bolted connection is a bearing type connection. This is consistent with AISC requirements for an A307 bolted joint, [9).
Provide the torque specifications, if any, for the girt bolting.
Response to Issue# 4 - bullet #2:
The bolts are unfinished A307, in a bearing connection. LPI is not aware of any torque specification for the girt bolts. Our assumption would be they were installed using good bolting field practices. The assessment performed in Ref. 1 does not rely on a minimum value of torque being present in the bolt.
Provide the technical justification for N EDC 16-003, Section 2.2, Assumption 9, "Friction between steel urfaces is credited to transfer load between the angle and channel and between the angle and column flanges following weld failure. A range of friction for carbon steels is provided in various text (sec [ 16) as an example). A dry mild steel on mild steel value of 0.78 is listed under static conditions, with a sliding value around 0.42. A value of0.35 has been commonly used as a lower bound friction Page A6 of 9
LPI Ref. A 16254-LR-002, Rev. 0 Attachment A value for most steel on steel surface conditions. Since selecting a low value is considered conservative, a value of 0.25 is utilized herein."
Response to Issue# 4 - bullet #3:
The purpose of considering friction is to "hold or position" the girt in a horizontal position, thus maximizing its resistance to bending under HELB pressure. The result of this approach is a higher pressure to fail the girt component. If friction were to be lower/lost, rotation of the girt would occur, resulting in weak axis bending of the girt, and thus lower resistance to the HELB pressure.
As was demonstrated in Section 4.3, friction at the lower value used of 0.25 is sufficient to maintain the deadweight reaction of the girt at the column. If the friction were higher this conclusion would not change, the answer would not change the failure pressure derived for the girt. If friction was significantly lower or zero, the girt would rotate and fail at a lower pressure than currently predicted.
Issue #5 Calculation NEDC 16-003 does not demonstrate that the 5362 square foot blowout area is adequate to prevent pressure retention beyond 0.5 psig in the Turbine building following a High Energy Line Break (HELB) event. A previous analysis performed by a contractor using GOTHIC modeling found that a blowout area of30,000 square foot still resulted in pressure retention.
[n addjtion, address the following:
Calculation NEDC 16-003, Section 5.0 states, "Using a HELB pressure of 0.45 psi, it is predicted failure of the panel support system at this pressure will result in a loss blowout) of total panel area of 5,362 fl2." PrO\\ idc the engineering analysis or calculation that demonstrates that a total panel area of 5,362 ft2 is enough to maintain peak pressure in the turbme building below 0.5 psig. (f-NPPO Action)
Response to Issue # 5 is an NPPD action.
The purpose of the Ref. 1 calculation was to demonstrate the blowout siding panels would fail at or below 0.5 psi.
LPI was not contracted to derive pressure build-up in the Turbine Building as a result of HELB and subsequent panel area blow-out or review the validity of any such existing calculations.
Page A7 of 9
LPI Ref. A 16254-LR-002, Rev. 0 Attachment A Issue #6 The assumed turbine building siding blowout area in EDC 16-003 (5362 square foot) is based on column centerline which would be non-conservative because column flanges would protrude beyond column centerline and reduce the actual blowout area. This would increase the pressure retention in the turbine building potentially beyond 0.5 psig.
In addition, address the following:
Provide engineering drawings that show the actual dimensions of the blowout area credited in NEDC 16-003. (~ NPPD Action)
Response to Issue# 6:
The siding panels provide continuous coverage of the Turbine Building. The girt detail is such that it projects beyond the flange of the column, see a drawing excerpt in Figure 4 and walkdown image, Figure 5 (below). As such, siding "behind" column flanges are exposed to Turbine Building HELB pressures, and the pressurized area considered in the Ref. 1 calculation is appropriate.
lRU5'f>
\\.OWIER C.. QIU)
Gl"T Page AB of 9
LPI Ref. A 16254-LR-002, Rev. 0 Attachment A SIDING Figure 5: Girt Projecting Outside TB Column Flanges Issue #7 Calculation NEDC 16-003 includes non-conservative assumptions involving how the forces against the turbine building siding are distributed during a high energy line break. Specifically, the distributed force only assumes a single girt channel whereas the actual construction of the turbine building blowout panel includes multiple girt channels. Consequently, the resultant force may not be enough to ensure bolt and weld failures at pressures less than 0.5 psig.
Response to Issue# 7:
The H ELB pressure is applied over the entire area of the siding panels. The analysis of the girt channel In NEOC 16-003 considers the panel pressure tributary to a single girt. Other girts of similar span and spacing will have similar loading and show similar behavior. Use of a tributary area is appropriate.
In addition, address the following:
Clarify whether the resultant force is distributed across only a single girt channel connection or across multiple girt channel connections supporting each turbine building blowout panel.
Response to Issue# 7 - first bullet:
See Response# 7 above.
Provide a COJ>Y of the most current revision of Calculation l:DC 16-003. ( ~ NPPD Action)
Page A9 of 9