ML20195D513
| ML20195D513 | |
| Person / Time | |
|---|---|
| Site: | Portsmouth Gaseous Diffusion Plant, Paducah Gaseous Diffusion Plant |
| Issue date: | 11/10/1998 |
| From: | Toelle S UNITED STATES ENRICHMENT CORP. (USEC) |
| To: | Paperiello C NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM), NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS) |
| References | |
| GDP-98-0241, GDP-98-241, TAC-L32043, TAC-L32044, NUDOCS 9811180083 | |
| Download: ML20195D513 (30) | |
Text
-
l USEC
. A Global Energy Company November 10,1998 GDP 98-0241 j
Dr. Carl J. Paperiello
~ Director, Oflice of Nuclear Material Safety and Safeguards Attention: Document Control Desk 4
U.S. Nuclear Regulatory Commission Washington, D.C. 20555-0001 Paducah Gaseous Diffusion Plant (PGDP)
Portsmouth Gaseous Diffusion Plant (PORTS)
Docket Nos. 70-7001 & 70-7002
)
Response to NRC Request for AdditionalInformation - Safety Analysis Report Update (TAC NOs. L32043 and L32044)
Dear Dr. Paperiello:
j By letter dated July 14,1998 (see the reference), the U.S. Nuclear Regulatory Commission (NRC) forwarded to the United States Enrichment Corporation (USEC) various questions on the certificate amendment requests to Update the Application Safety Analysis Reports (SARUP) for the Paducah, Kentucky and Portsmouth, Ohio gaseous diffusion plants.
USEC's responses to the NRC questions / comments contained in the July 14,1998 letter are provided in Enclosure 1. provides a status of the response to each of the N~RC questions / comments on the SARUP submittals.
~
If you have any questions on USEC's responses, please contact Steve Routh at (301) 564-3251.
There are no new commitments contained in this submittal.
I Sincerely,
/j-S. A.
Ii Steven A. Toelle Nuclear Regulatory Assurance & Policy Manager 9811100083 981110
~
PDR ADOCK 07007001 l
PDR 6903 Rocidedge Drive, Bethesda, MD 20817-1818 Telephone 301-564-3200 Fax 301-564.3201 http://www.usec.com Offices in Livermore, CA Paducah, KY Portsmouth, OH Washington, DC
x Dr. Carl J. Paperiello November 10,1998 i
GDP 98-0241, Page 2
Reference:
Letter from Charles Cox (NRC) to Mr. James H. Miller (USEC), "Paducah and Portsmouth Certificate Amendment Requests - Update of the Application Safety Analysis Reports (TAC NOS. L32044 and L32043)," dated July 14,1998.
j
Enclosures:
1.
Responses to Questions 1 - 15 from the July 14,1998 NRC Request for AdditionalInformation - Safety Analysis Report Update (TAC NOs. L32043 and L32044) 2.
Status of Responses to NRC Questions / Comments on SARUP cc:
Mr. Robert C. Pierson, NRC HQ l
NRC Region III Office NRC Resident Inspector - PGDP NRC Resident Inspector - PORTS Mr. Randall M. DeVault, DOE 4
y.
GDP 98-0241 l
Responses to Questions 1 - 15 from the July 14,1998 NRC Request for Additional Information Safety Analysis Report Update (TAC No. L32043 and L32044)
(26 pages total) f-L l
I i:
L F
4 s
W 9
~
i 6
SARUP Q&R - PGDP/ PORTS November 10,1998 Q1 (NRC 7/14/98 Letter)
NRC Information Notice 96-29. " Requirements in 10 CFR Part 21 for Reporting and Evaluating Software Errors," describes deficiencies in the structural analysis program GTSTRUDL that the GTSTRUDL vendor has issued Part 21 notifications. GTSTRUDL was used in the analysis of structures in the SARUP. Did any of the analyses invoke any subroutines or functions that contain errors? If so, what effect do the errors have on the analyses?
Response.
None of the GTSTRUDL errors referenced in NRC Information Notice 96-29 have an impact on the SARUP structural analyses. The verification process described in the following paragraphs adequately 1
evaluated the salient features of the GTSTRUDL program used for analyses of both the Paducah and Portsmouth structures.
The SARUP structural evaluations were originally performed by Lockheed Martin Energy Systems (LMES)in support of DOE. A verification process was developed by LMES to corroborate the accuracy of the GTSTRUDL computer code used in the evaluation of structures for the GDP SARUP. This process covered: a) commercially available structural analysis software from external vendor sources, e.g.,
GTSTRUDL, b) commercially available spreadsheets used for calculations, e.g., Microsoft Excel and c) miscellaneous in-house software programs, typically pre-and post-processors. The process permitted software h
. verification by any of three methods: 1) running sample problems for which theoretical solutions are known,
- 12) procurement of vendor supplied software QA verification, and 3) evaluation of relevant, specific software features on a job-by-job basis.
- Method I was the primary basis of GTSTRUDL verification. A suite of twenty-seven problems was evaluated and compared to theoretical solutions. These problems include such applications as plane truss and plane frame static and dynamic problems, 3-D frame and truss analyses, an eigenvalue solution, response spectrum modal analysis, time history analysis, support settlement, and AISC code checking. The adequacy
. of the solutions was documented and approved by responsible management personnel.
Additionally, " sanity" checks were incorporated into'a majority of the SARUP evaluations to further provide assurance of the accuracy of the GTSTRUDL computer code. A typical check was to confirm that the summation of the participation factor times the mode shape (r*&) for a particular node for all significant frequencies of an eigenvalue analysis is equal to approximately 1.0. This provided confidence that sufficient dynamic mass had been included in the analysis.
SARUP Revision:
No revision required.
i
~
1 b
I i
lo
(
- ~
~ -,,.-
-. - ~.. - - - _ ~ ~
- SARUP Q&R'- PGDP/ PORTS November 10,1998,
- Q2 (NRC 7/14/98 Letter)
Describe the horizontal force resisting systems in each of the' structures amlyzed and how these systems were modeled, s
' Response:
Structures evaluated as part of the SARUP were one of three structural configuration types:
(1)
. Predominantly structural steel moment or braced frames with reinforced concrete slabs and shallow spread footings, (2)
Reinforced' concrete shear wall or reinforced concrete frame structures supported upon shallow spread footings, and l
(3)
- A combination of(1) and (2).
Type I structures are predominant at both Paducah and Portsmouth and are represented by the process buildings, e.g., Paducah Buildings C-333 and C-331, Portsmouth Building X-330. Type 2 structures are laboratory buildings or control buildings, e.g., Paducah Building C-710, Portsmouth Buildirgs X-710 and X-
-.300. Paducah Building C-315 represents a type 3 structure.
' Figures 1,-2,' and 3 provide a generalized load path description of the three types of structural configurations.
u.
f9 The GTSTRUDL computer code was utilized to model the structures evaluated as part of the SARUP.
- In general, for Type 1 ~ structural steel frame structures, the columns and beams were represented by six degree-of-freedom (DOF) beam elements with cross bracing represented by simple truss elements. In some analyses, roof trusses were considered equivalent to be2m elements in lieu of detailed modeling.' Infill walls (masonry walls between columns), if present, were modeled as ' simple truss elements to represent equivalent struts.' Simple 2-D or 3-D lumped mass models were generated for the Type 2 shear wall structures and the calculated global seismic demands distributed on the basis of stiffness for the determination of detailed stresses.
Reinforced concrete floors were considered as rigid diaphragms for both type models. The Type 3 model for Paducah Building C-315 was a 3-D assemblage of beam elements representing structural steel columns and reinforced concrete piers, truss elements representing infill walls, and finite elements representing floor slabs and reinforced concrete shear walls.
For the geometrically nonlinear evaluations of Paducah Buil6ings C-333/C-337 and C-331/C-335, a sequential, phased approach was undertaken to determine the state of stress within the structures. The previously described 3-D GTSTRUDL model, with base uplift incorporated, was utilized to calculate the load Ldeflection chattenristics of the model when subjected to variations of lateral loads. The load deflection relationship was usad.to create a simplified, but equivalent, ABAQUS 2-D computer model simulating the geometric non-linearity of the'more complex 3-D GTSTRUDL models. Selected resuhs derived from a transient time history analysis calculated using the simplified models were backsubstituted into the 3-D GTSTRUDL model to determine detailed member stresses. Non-linear evaluations using the ABAQUS
. general-purpose computer code were performed only for Paducah Buildings C-333, C-337, C-331, and C-335.
1of5
.:i l
+
SARUP Q&R - PGDP/ PORTS November 10,1998 SARUP Revision:
No revision required.
1 1
l 2 of 5 j
i
Figiere 1.
g Paducah and Portsmouth Gaseous Diffusion Plants
- =
C Lateral Load Path for Type I Structural Steel Frame m-Typical for Process Buildings and Shop
?.
o Oa2 M czzanine Floo r
-9 G round Floor Slab o
C ran e s Roof E x te rio r S id..in g.
(IF A n nlic a ble)
I c!
P u rlin s G irts R oof B eam s U ru sse s C ell Floo r (If A p plicable)
E v
V Beams C olu m n s' V
C ro ss B racin g U nderlyin g S oil i..... - c. i. -..
6....., c. i....
r; 9
k b
2
- -- i
Figure 2 Paducah and Portsmouth Gaseous Dimision Plants j
Lateral Load Path for Type 2 Reinforced Concrete Structure
]
Typical for Laboratory and Control Structures en 3-oa B
Roof' Diaphgram Intermediate Floor (s)
Ground Floor Slab' x
E!
u 2,
V
) Reinf'orced Concrete Shear Walls or Columns 5
V Underlying Soil
- i. - ca.
um - ca m...
Footings / Grade Beams y
3 k
3 b
=
r L
L---.-.---. -
. Figure 3 -
y Paducah and Portsmouth Gaseous Dimision Plants
_g.
u Lateral 1.oad Path for Type 3 Structure Typical for Paducah Building C-315 -
- o 3o s
"O b
C ran es Roof Exterior Sidin g Interm ediate Floor G round Floor Slab g'
d l
l P u rlin s G irts Roof B eam s
/T r u s s e s V
> C olum ns d Be s
v V
C ross B racing /R ein fo rced C on crete S h ea r W alls U nderlying Soil
- i. iu... c. i....
r......,c.....
I n
B k
sL
=
f r
l
- SARUP Q&R - PGDP/ PORTS.
November 10,1998
' Q3 (NRC'7/14/98 Letter)
Design drawings, design specifications and walkdown packages were used to model existing conditions l
' of' structures. Design drawings and specifications were used to define geometry and material properties. Describe the basis for concluding that the as built structures were modeled accutately.
The response should include a discussion of the availability, completeness, and accuracy of existing drawings reviewed; how the accuracy of the drawings was determined; the extent and completeness of walkdowns performed (include a typical walkdown package); and information obtained from design specifications.
Response
' At the start of the DOb efforts to prepare the PGDP and PORTS SAR Upgrades (Documents KY/EM-174 and POEF-LMES-89, respectively), a detailed structural walkdown plan was developed. This plan 5'
outlined the process to be followed, and the appropriate structural information to be collected, for the purpose i
of providing inputs into the assessment of the adequacy of existing Natural Phenomena Analysis (NPH) evaluations (i.e., the 1982 EDAC analyses) and inputs into the updated NPH evaluations of Paducah and
~ Portsmouth structures. This plan ensured consistency in the data collection process.
1 This approach was implemented in order to develop confidence that each facility surveyed was constructed in accordance with its design drawings, that any major configuration changes to the principle structure were identified, and that the location of major equipment loading was captured.
1 The building structural walkdown provided the following:
1.
Documented overall description of the general condition of a building.
2.
Documented statistical evidence that members are consistent with the design drawings.
3.
Redlines of the structural drawings to indicate major changes. Major changes, in this context, were those changes that could conceivably modify the lateral load carrying system'and included strength discontinuities, changes in member sizes and connections, undocumented or en'arged openings in walls or floor, etc.
4.
Documented the magnitude of loads on the roof and floors, and provided a load map of the building.
5.
Documented the configuration of masonry walls carrying lateral loads.
6.
Provided a complete list of material properties and reference drawings.
The walkdown plan was peer reviewed by Dr. R.P. Kennedy of RPK Structural Mechanics Consulting Inc. Dr. Kennedy concurred with the walkdown plan.
L
- C, 1 of 2 L
.. _ ~. _ _ _. _. _ _ _ -. _. _ _
SARUP Q&R - PGDP/ PORTS -
November 10,1998
- Structural walkdowns were performed for the following buildings:.
Paducah Portsmouth C-333 X-330
- C-333-A X-342A C-331 X 326
{
C-335 X-333 C-337 X-343 C-337-A
-X-344A C-360 X-300 C-310 -
X-745 C-310-A X-705 C-315 X 710 C-300 Process Tie Lines C-409 C-400 Process Tie Lines In general, the walkdowns confirmed the following:
-1.
Structural drawings were available and, in most cases, were accurate. Identified differences'
' were generally conservative; the majority of differences were in column sizes which were generally larger than specified on the drawings, e.g., the W10x49 column specified on the drawing for column lines C and 16 intersection of Paducah Process Building C-331 was actually confirmed to be a W10X54.
' 2.
Design specifications were difficult to obtain and, in many cases, were not available. If specifications were unavailable, the code values at the time of construction were specified.
3.
Major structural changes were rare.
The walkdown calculation packages and the walkdown plan are available for NRC review at USEC Headquarters or at the Paducah and Portsmouth Gaseous Diffusion Plant sites.
SARUP Revision:
No revision required.
2 of 2 m
w wn-e I
y i
.. __ _.-. _.,._. _ _. ~ -
SARUP Q&R - PGDP/ PORTS November 10,1998 Q4 (NRC 7/14/98 Letter)
What values were used for "Fp" in the structures analyzed? Ilow were these values determined?
Response
i The values of F tabulated in Table 2.4 of DOE-STD-1020 assume reasonably uniform inelastic behavior and good detailing practice conforming to current code requirements. The detailing for structures built in the 1950s, while adequate for the codes in effect at the time of their design, does not necessarily meet current code requirements for lateralload carrying elements. Therefore, for Paducah ami Portsmouth, a value.
of Fp=1.0 was typically used for the evaluation of all structural steel connections (DOE-STD-1020 requirement) and for most structures. A limited number of building evaluations, on a case-by-case basis, used
)
an F value greater than 1.0. For example, Paducah Building C-710, a reinforced concrete structure with adequate detailing and anticipated uniform nonlinear behavior, used the 1.5 value from DOE-STD-1020.
Further, the seismic upgrade calculations of Paducah Buildings C-331 and C-335 document the determination of an effective F of 1.2 to 1.4 based upon the approximate, simplified procedures in Section C.4.4.3 of DOE-STD-1020. The basis for the values of Fp tabulated in DOE-STD-1020 is provided in UCRL-CR-111478, " Basis for Seismic Provisions of DOE-STD-1020."
SARUP Revision:
No revision required.
1
)
)
i f
l i
1
- ~ _ -.
.h SARUP_Q&R - PGDP/ PORTS November 10,1998 7
Q5 (NHC 7/14/98 Letter)
..The DOE-STD-1020 procedure requires that the strength and detailing of the entire load path be checked. Describe the assessment method, the extent of the assessments, and the acceptance criteria employed to assure that good detailing exists. Were any joints found where the detailing was inadequate? If so, how was this handled?
1
Response
4
- First, the walkdown plan controlling d' ata collection for NPH evaluations mandated that connections be included in the statistical sample to ensure the adequacy of design drawings. Further, a 100% walkthrough
- of the facilities was performed to identify changes in the structural configuration from the design drawings.
j strength discontinuities, changes in member sizes and connections, undocumented or enlarged openings in j
walls or floors, etc. This requirement provided a high level of confidence that structural steel connections were designed to complement the strength of the structural steel members and no obvious " weak links" j
existed, e.g., a 4-bolt connection with only 3-bolts installed.- Subsequent analyses were intended to confirm suspected " weak links," if any.
Second, the SARUP evaluations were designed to identify and evaluate the structural element 1
controlling the " seismic capacity" of each building. Each structural element comprising the lateral load carrying system of a building was screened to identify the component with the lowest " seismic capacity "
" Seismic capacity" as used in the SARUP and here is defined in the response to Question 13 (NRC 7/14/98
' ! Letter). ' This component, whether a connection, cross brace, "J-bolt", etc, was labeled as the " weak link" of the lateralload resisting system. The calculated capacity of structural connections and structural members '
was based upon the criteria of AISC for structural steel and ACI for reinforced concrete.
Joints were often identified by analysis to fall into the " weak link" category. One of two options was I
l then exercised: 1) accept the "seimic capacity" of the joint as the controlling " seismic capacity" for the building or 2) perform additional evaluatior.s to permit load redistribution and a potentially higher " seismic-capacity." For example, the " seismic capacity" of Paducah Building C-400 (0.05g) is controlled by inadequate "J bolt" capacity at the connection of a structural steel column to its reinforced concrete footing. No additional b
load redistribution load studies were performed to substantiate a higher " seismic capacity." Furthermore, j
L "J bolt" anchorage and beam-to-column moment connections for Paducah Building C-333 were identified by L
analysis as " weak links." In a subsequent geometrically non-linear evaluation of this building using the
' ABAQUS and GTSTRUDL computer codes, structural steel column bases with compromised "J-bolts" were allowed to uplift and beam to column moment connections, when demand exceeded capacity, were pinned to I
permit load redistribution. This methodology, which allowed load redistribution within the global system, demonstrated a higher " seismic capacity" for Building C 333 rather than one controlled principally by the "J bolt." -
SARUP Revision:
No revision required.
w i
e I
1 SARUP Q&R - PGDP/ PORTS November 10,1998 Q6 (NRC 7/14/98 Letter)
Describe the peer review performed on the analyses. What are the qualifications of the peer review panel?
Response
Dr. R. P. Kennedy, Structural Mechanics Consulting, Inc, performed the peer review of the natural phenomena evaluations. Dr. Kennedy is internationally recognized as one of the leading experts in natural phenomena engineering, particularly earthquake engineering. lie has had key involvement in the development of the earthquake engineering requirements for the Department of Energy, the Nuclear Regulatory Commission, other federal agencies, plus the overall engineering community.
Dr. Kennedy's review focused on the walkdown plan, the natural phenomena evaluation criteria, arxl the application of these documents in the evaluations of the main process buildings. In addition, Dr. Kennedy performed a peer review of the proposed modifications to Paducah Buildings C-331 and C-335.
In addition to Dr. Kennedy's peer review, EQE, International, Inc. performed a peer review of the soil-structure interaction evaluations. Dr. James J. Johnson, Stephen A. Short, and Richard L. Tiong from EQE performed this review. All of these engineers are recognized as experts in earthquake engineering, and Dr. Johnston is recognized as one of the leaders in developing methods for performing soil-structure interaction analyses. Dr. Kennedy also reviewed the soil-strecture interaction analyses and concurred with the analyses.
SARUP Reviskm:
No revision required.
SARUP Q&R - PGDP/ PORTS November 10,1998 Q7 (NRC 7/14/98 Letter)
Describe how in-structure spectra were generated, considering that the structures are in the non-linear range. Were seismic anchor movements also included in the analysis of SSCs?
Resiase:
i In-structure response spectra were generated using linear elastic GTSTRUDL models for all buildings er. cept Paducah Buildings C-333/C-337 and C-331/C-335. Section 2.4.1 of DOE-STD-1020-94 (page 2-19) j states: " Floor response spectra should be developed accounting for the expected response level of the supporting structure even though inelastic behavior is permitted in the design of the structure (see Section 2.3.3)." Spectra generation was consistent with the requirements of DOE-STD-1020-94 and ASCE 7.
1 For Paducah Buildings C-333/C-337 and C-331/C-335, the ABAQUS computer code was utilized to calculate time histories of a simplified 2-D model constructed to be dynamically equivalent to the more complex 3-D GTSTRUDL computer model. Bot'n the GTSTRUDL and the ABAQUS computer models incorporated the geometrically non-linear base uplift of these buildings. The calculated ABAQUS time histories were used to generate the in-structure spectra. Spectra generated are consistent with the requirements j
of DOE-STD-1020-94. The following paragraphs describe the use of these spectra in the evaluation of 1
equipment and components.
Building expansion joints exist between the structural units comprising the large process buildings at Paducah and Portsmouth. Thus, piping and equipment crossing expansion joints between the structural units are supported by multiple structures and subject to relative movements or seismic anchor movements (SAM).
Eaducah Paducah Buildings C-333 and C-337 respond linearly up to peak ground accelerations in the range of 0.10g to 0.12g:
- y then respond somewhat nonlinearly up to 0.15g, the peak ground acceleration (PGA) of the site-specific 250-year return period Evaluation Basis Earthquake (EBE) currently identified in the SARUP submittal. Building responses for the 0.15g EBE, when determined from linear elastic structural analysis, over-predict the building accelerations and under-predict the building displacements. Typically, SSCs located in the process buildings were coNrvatively evaluated using the peak of the acceleration response spectra.
SSCs, particularly piping expansion jo nts (bellows) and piping spanning building expansion joints in process Buildings C-333 and C-337, were evaluated for displacements (seismic anchor movements) which accounted for the geometrically non-linear response. 4 these buildings.
Buildings C-331 and C-335 are being structurally upgraded to withstand the 0.15g EBE. The evaluations of SSCs in these two buildings were based on buildmg responses, i.e., floor accelerations and building displacements, for the conceptual design of the proposed modifications. [The modifications to Buildings C-331 and C-335 have also been shown to satisfy the proposed updated seismic hazard (0.165g EBE.)]
1 of 2
l.
s e
SARUP Q&R - PGDP/ PORTS November 10,1998 L
. Portsmouth
%e peak ground acceleration of the 250-year return period EBE for the Portsmouth site is 0.05g atxi the buildings respond in the elastic range. Seismic anchor movements were considered in the evaluation of
. SSCs where relative movements were present (i.e., across process building expansionjoints).
SARUP Revision:
. No revision required.
(-
l 4
t 1
J
-l l
1 i
i 2 of 2
l e l
l SARUP Q&R - PGDP/ PORTS November 10,1998 l
l Q8 (NRC 7/14/98 Letter)
State and justify all damping coefficients that were used in the analyses.
Response
Damping values for the SARUP structural evaluations were taken directly from DOE-STD-1020 (formerly UCRL-15910); the damping value used for concrete and structural steel structures was 10 percent.
The values contained within DOE-STD-1020 are consistent with those contained within NUREG/CR-0098,
" Development of Criteria for Seismic Review of Selected Nuclear Power Plants." A structural damping level i
of 7 percent was used for the generation of in-structure acceleration response spectra.
l-SARUP Revision:
)
l No revision required.
l l
i i
i i
I i
j l'
I l
i.
1 4
r
l SARUP Q&R - PGDP/ PORTS November 10,1998 l
Q9 (NRC 7/14/98 Letter) a.
How are member capacities, described in the SAR as code ultimate or yield values, defined for the different types of structural elements and materials? How were column and lateral buckling considered? For each building, how many members, by type, entered the non-linear
~
range when the structure is subjected to a 0.15g earthquake? Are there any instances where member total inelastic demand (Dn) exceeds seismic capacity (Cc)? If so, describe these instances and how they are handled in the analyses. Are all concrete structures that exceed a ductility of one under-reinforced?
' b.
Page 4.3-29 in the Paducah and Portsmouth SARs states " member capacities were evaluated from code ultimate or yield values." Page 4.3-138 of the Paducah SAR states " capacity is defined as no building collapse." From the description in the Paducah SAR, it appears that a
after it was determined whether member capacities were exceeded, it was determined whether building collapse was predicted to occur. Define " building cc!! apse." Describe the methodology and acceptance criteria for determining whether building collapse was predicted to occur. Were there any instances where building collapse (or partial collapse) was predicted y
to occur? If Paducah was handled differently than Portsmouth with respect to member and l
building capacities, please describe.
l
Response
i
- a.
For both Paducah and Portsmouth, the capacity of concrete members was based upon the Ultimate Strength Design (USD) requirements of the American Concrete Institute (ACI). The capacity of
~
structural steel members was based upon a factor of 1.7 times the Allowable Strength Design requirements for bending and axial loading and 1.4 times the Allowable Strength Design requirements for shear of the American Institute of Steel Construction (AISC). These are essentially yield limits and are consistent with the requirements of DOE-STD-1020-94.
For both Paducah and Portsmouth, structural steel column and beam lateral buckling was determined l.
by solving the applicable equations of the AISC criteria as programmed within the GTSTRUDL computer code. The actual unbraced length (determined from appropriate structural drawings) of any structural steel element was used in the AISC equations and defaults or conservative values for C,and C were used. Concrete columns and beams were evaluated considering ACI requirements; these 5
evaluations were either by hand calculations or by use of spreadsheets.
L Most of the buildings at Paducah have a seismic capacity less than 0.15g (i.e., the criteria of DOE-
- STD-1020 is not explicitly satisfied). However, no precise tabulation has been made of the exact number of members in any building not meeting the acceptance criteria of DOE-STD 1020 for a j
0.15g earthquake level. The seismic capacity of all evaluated buildings is cominonly tied to an identificci " weak link (s)." For example, the seismic capacity for Paducah Building C-400 identified in SARUP Table 3.15-10 is 0.05g and is defined by the slip of "I" type anchor bolts at braced column locations and could be considered a point capacity.
All Portsmouth buildings identified in SARUP Table 3.8-6 satisfy the acceptance criteria of DOE-STD-1020, i.e., the demami imposed on all structural members is less than their corresponding j
capacity.
1 of 4
. _ ~ - - - -.
';4 SARUP Q&R - PGDP/ PORTS November 10,1998 Although F, was typically taken as 1.0, there were limited evaluations where the value of F, used for certain elements of the lateral load carrying structural system was taken as greater than 1.0 (e.g., the seismic retrofit analysis for Paducah Buildings C-331 and C-335. Paducah Building C-710 and Portsmouth Building X-710). In these instances, the F, value used is supported by discrete calculations (e.g., Building C-331) or the structural detailing supports the use of DOE-STD-1020-94 values. The approach in the use of F, values was to divide the calculated seismi demand by F, and compare to the member yield based capacity. This is consistent with the requirements of DOE-STD '
1020-94.
Paducah and Portsmouth structures with a predominately reinforced concrete lateral load resisting system (e.g., Paducah Builidngs C-300 and C-710 and Portsmouth Buildings X-300 and X-710) satisfied the acceptance criteria of DOE-STD 1020 when evaluated for the 250-year return period seismic event. Paducah Building C-710 and Portsmouth Building X-710 used a value of F, greater than 1.0. The reinforced concrete colunms of Portsmouth Building X-710, a moment frame, satisfied the minimum and maximum reinforcement limits of ACI. The reinforced concrete shear walls of Paducah Building C-710 satisfied the minimum reinforcing steel requirement of the ACI. Paducah Building C-300 and and Portsmouth Building X-300 are reinforced concrete structures with cylindrical walls (18" thick) and domed roofs (18" thick). Similarly, the vertical reinforcement in the cylindrical walls, the primary lateral k)ad carrying element, satisfies the minimum steel requirements of the ACl.
[
b.
The individual buildings at Paducah and Portsmouth were seismically evaluated using the criteria in UCRL-15910 and DOE-STD-1020-94 and the 250-year return period earthquake ground motion. The i
seismic evaluations were based on the evaluation criteria for natural phenomena performance category (PC) 3 buildings. The performance goals associated with the PC 3 building criteria are designed to j
ensure occupant safety, continued operation, and hazard confinement. A more qualitative description of these performance goals from DOE-STD-1020-94 is as follows:
1.
no structural collapse occurs; failure of contents is not severe enough to cause injury or death or to prevent evacuation, 2.
concrete walls are cracked, but small enough to maintain pressure differential with normal ilVAC: largest cracks are no greater than 1/8 inches, j
i 3.
metal liner will remain leak tight, 4.
components remain anchored and functional, 5.
possible vir. role damage will occur, but permanent distortion will not be immediately apparent to the naked eye.
L Based on these performance goals, the buildings were evaluated to determine if their capacities were l
greater than the demands imposed by the evaluation basis earthquake, and if not, the building L
components and their capacities were identified. The member capacities of the building elements were
. calculated using ACI(USD) for concrete elements and AISC (1.7 x ASD for bending / axial and 1.4 x ASD for shear) for structural steel elements. Capacity calculations are consister.t with DOE-STD-1020-94 recommendations. In order to satisfy the performance goals for a PC 3 building, the l
behavior of the building is primarily limited to the linear response range.
4 l
2 of 4 i
jL
l 4
SARUP Q&R - PGDP/ PORTS November 10,1998 The seismic evaluations of the buildings at Portsmouth determined that the buildings have capacities greater than the demanis imposed by the evaluation basis earthquake (0.05g peak ground acceleration) using the criteria in DOE-STD-1020-94. The seismic behavior of the Portsmouth buildings during the evaluation basis earthquake is essentially linear.
The seismic evaluations of the buildings at Paducah determined that most of the buildings do not have a seismic capacity greater than the demand imposed by the evaluation basis earthquake (0.15g peak ground acceleration), i.e., some of the buildings linear capacities are exceeded. The identified weak link in most buildings was the foundation "j-bolts" at braced column lines while the identified weak link in Process Buildings C-331 and C-335 was the rocker supports between building units. Based on
- this, the performance goals for the Paducah buildings were reassessed considering the remaining life of the facilities risk evaluations, and subjective cost-benefit judgement.. The basis for this reassessment is contained in Justficationfor the Acceptability of the Seismic Risk with a 250 Year
- Return Level Earthquake Event, prepared by J. C. Carter for the DOE Regulatory Oversight Office, Oak Ridge Operations February 1996. Although no specific remaining life of the Paducah facility has been identified, the remaining life of the facilities is significantly less than that of a new facility.
In addition, the risk from a release during an earthquake is small due to the free zone out to one mile and the relatively low population density, concentrated in areas away from predominant wind directions, between one and two miles. The risk of unacceptable consequences is low compared to normal voluntary risks and previously proposed reactor safety goals. Considering the remaining plant life and the low risks, the subjective cost-benefit judgements concluded that the risks would not be significantly reduced if the facilities were upgraded to meet the performance goals of DOE-STD-1020-94.
Considering all of the above, it was determined that the appropriate performance goals were to ensure that the buildings do not collapse, the siding stays in place, and permanent distortions of the buildings are such that the ability of important to safety piping and equipment to maintain confinement is not impaired. To assist in evaluating these performance goals, non-linear analyses considering anchor bolt uplift (weak link) were performed for the process buildings. These evaluations, which considered the non-linear behavior of the anchor bolts, increased the overall building capacities and identified the next weak link in the buildings. The next weak link was typically the cross bracing and/or the beam-column comwcuons leading to increased building lateral displacements and eventually to building a
collapse, i.e., as the building lateral displacements increase, the overturning forces from the building vertical loads increase (P-A phenomenon), leading to building collapse.
L in addition to the non-linear analyses of the process buildings, considerations were given to the buildings' fragility curves, risk of building failures, acceptable consequences, etc. The seismic criteria in DOE-STD-1020-94, in general, are equivalent to defining the building capacity as the capacity at which there is high confidence (95%) of a low probability (5%) of failure (HCLPF). The median capacity of buildings similar to those at Paducah is typically about 2.5 times the HCLPF capacity.
t 3 of 4 L
~ -,. _.. _ _ _.
_ ~ _, _.
l
. SARUP Q&R - PGDP/ PORTS November 10,1998 I~
Based on the above, it wasjudged that the Paducah plant buildings (with one exception) satisfy the less stringent performance goals. As defined above, these performance goals, (i.e., no building collapse, siding stays in place, and permanent distortions of the building), are such that that the ability of important to safety piping and equipment to maintain confinement, when subjected to the evaluation 1
basis earthquake, is not impaired. The one exception was the C-331 and C-335 process buildings.
Modifications to the C-331 and C-335 process buildings have been designed, and are being installed; these modifications will increase the seismic capacity to meet performance goals for the evaluation basis earthquake.
There were no instances where building' collapse (or partial collapse) was predicted to occur for the evaluation basis earthquake (except as noted above for the C-331 and C-335 process buildings).-
.SARUP Revision:
]
..s.
No revision required.
1 q,
l
)
.i
-l
)
I l'
i l-4 of 4
...=
SARUP Q&R "GDP/ PORTS November 10,1998 Q10 (NRC 7/14/98 Letter)
How were story drifts calculated? Did the calculation account for inelastic behavior? What acceptance criteria was used for story drifts? DOE-STD-1020 states " lateral seismic forces are not reduced by Fp when computing story drifts." Was this followed?
Response
For the majority of the Paducah and Portsmouth structures evaluated as part of the SARUP, a linear elastic analysis was used to calculate story drifts. For these cases, the value of F, was set equal to 1.0 and the guidelines provided in DOE-STD-1020 were used to judge the acceptance of the story drift. For the case where a geometric non-linear analysis (Paducah Buildings C-333 and C-337) was used, the story drift and resulting P-delta effects were calculated using the computer program ABAQUS. Conservative controlling loads from this analysis were used as input into a geometrically non-linear 3-D GTSTRUDL model. The calculated deflections were compared to the DOE-STD-1020 criteria and found to be acceptable. For the modification analysis of Paducah Buildings C-331 and C-335, the calculated loads from the geometrically non-linear model were used as input into a geometrically non-linear GTSTRUDL 3-D model. The resulting deflections were compared to DOE-STD-1020 criteria and found to be acceptable.
SARUP Revision:
')
No revision required.
l f
I-i l
i-
..___.m
._._-.m--..
l l
I-
- SARUP Q&R - PGDP/ PORTS November 10,1998 l
l Q11 (NRC 7/14/98 Letter) a.
Are any masonry structures part of the load resisting system to resist seismic forces? If so, describe the location of the wall, the size, material properti:s, and composition of the wall and the method used to assess its capacity, b.
' For all masonry walls, whether part of the primary load resisting system or not, list all Q or AQ equipment and their support systems that are either supported by the wall, or closer than the height of the wall. For each case that may exist, state whether the masotuy wall is j
l reinforced or unreinforced. Assuming that all such equipment is concurrently disabled by a i
common-mode seismic event, describe the effects upon public and worker safety, considering i
events that will be occurring during an earthquake and subsequent events.
l
Response
Masonry walls are part of the lateral load resisting system of the following buildings, X-705 at a.
Portsmouth, X-344A/X342A at Portsmouth, and C-720 at Paducah. These walls are constructed of concrete masonry units (6" and 12" for Building C-720, 8" and 12" for Building X-344A/X342A, and 8" for Building X-705) and are infilled between building columns. The walls were modeled as equivalent struts in the 3-D GTSTRUDL seismic models. Their capacity was conservatively based l
upon the results of the Hollow Clay Tile Testing Program at the Department of Energy Y-12 Plant, Oak Ridge, TN. Concrete masonry unit (CMU) walls found in other structures are not aligned
[
sufficiently to serve as lateral load carrying elements.
b, The equipment selected for evaluation in the SARUP for potential adverse seismic interac a
l consisted of equipment that provided the UF containment or pressure boundary, Cases where me 6
equipment was either supported by a masonry wall or located within a distance equal to or less than j
the height of the wall were considered in the walkdown inspection and subsequent evaluation. The cases identified are as follows.
Paducah Table Q11. SSCs Potentially Adversely Affected by Possible Failure of Masonry Walls.
Equipment Equipment Location Type Reinforced or Structural or Masonry Unreinforced Nonstructural Autoclaves #1, 2, 3, 4, C-360 8-in. CMU Reinforced Structural Relief Piping Systems Wall (Note 1) l Autoclaves #1, 2, 3, 4,-
C-360 8-in. CMU Reinforced Structural Condensate Piping Wall (Note 1)
Systems i
i-.
I of 2 l
l l
l f'
SARUP Q&R - PGDP/ PORTS November 10,1998 l
Table Q11. SSCs Potentially Adversely Affected by Possible Failure of Masonry Walls.
(continued)
Masonry Walls Equipment Equipment Location Type Reinforced or Structural or Masonry Unreinforced Nonstructural Autoclaves # 1,2, 3,4, C-360 8-in. CMU Reinforced Structural Sample and Evacuation Wall (Note 1)
Piping System Autoclaves #1,2,3,4 C-360 8-in. CMU Reinforced Structural Steam Supply Piping Wall (Note 1)
Systems Note 1: The 8-in. CMU wall was determined to be structurally adequate to resist the EBE.
Q or AQ equipment whose failure would release insignificant quantities of UF was excluded from evaluation for interaction effects; the exclusion criterion was based upon the amount of UF. released. No other Q or AQ equipment that could potentially be damaged by interaction was determined by the SARUP accident analysis to be required to perform its safety function during or after an EBE.
Portsmouth There was no equipment identified for evaluation in the SARUP calculations for potential adverse interactions that were either supported by a masonry wall or located within a distance equal to the height of the adjacent masonry wall, i
SARUP Revision:
No revision required.
l l-2 of 2 l
m SARUP Q&R - PGDP/ PORTS November 10,1998 i
l
' Q12 (NRC 7/14/98 Letter)
' Is any operator action relied on during or after a seismic event? If so, describe these actions and how it is assured that the actions will occur in a timely manner as assumed. Also, describe how it is assured i
that the desired result of the operator action will occur considering any seismic vulnerabilities that may exist in SSCs relied on to accomplish the desired result.
Response
)
. No operator action is relied upon to :nitigate the consequences of a seismic event. As analyzed in SARUP Section 4.3.2.5.3, the seismic source term is self-limiting in 10 minutes.
SARUP Revision:
No revision required, i
'I i
L-i I
I k
i 1
I 1
-......~
SARUP Q&R - PGDP/ PORTS November 10,1998 Q13 (NRC 7/14/98 Letter)
In Tables 3.15-4 through 3.15-11 of the Paducah SAR and 3.8-4 through 3.8-6 of the Portsmouth SAR, describe what the term seismic capacity' (shown in terms of "g" value) represents.
Response
Paducah For buildings, the term " seismic capacity" in SARUP Table 3.15-10 represents the threshold zero period acceleration (ZPA) of a response spectrum curve (identical in shape to the Evaluation Basis Earthquake spectnau) that produces the building responses (seismic combined with dead and live loads) that just meet without exceedi0g the acceptance criteria given in DOE-STD-1020-94. The basic intent of the deterministic
)
seismic evaluation criteria defined in DOE-STD-1020 is to achieve less than a 10% probability of unacceptable performance for a structure, system, or component subjected to an evaluation basis earthquake (EBE)
J multiplied by a facto of 1.5. For a PC-3 structure, system, or component, this is equivalent to the HCLPF j
capacity.
For equipment, the term " seismic capacity" in SARUP Tables 3.15-4 through 3.15-9 and 3.15-11 represents the threshold zero period acceleration (ZPA) of a response spectrum curve (identical in shape to the Evaluation Basis Earthquake spectrum) that produces responses (i.e., loads, stresses, deflections, accelerations, etc.) of equipment located in the building that just meet, without exceeding, the acceptance criteria given in DOE-STD-1020-94, when combined with dead and live loads.
Portsmouth The term " seismic capacity" has the same meaning for buildings in SARUP Table 3.8-6 and equipment in SARUP Tables 3.8-4 and 3.8-5 as for Paducah buildings and equipment.
SARUP Revision:
j No revision required.
.)
1
. N-SARUP Q&R - PGDP/ PORTS November 10,1998 i
1-
. Q14 (NRC 7/14/98 Letter) l DOE-STD-1020 states "to determine the. response of SSCs which use F > 1, note that for fundamental periods lower than the petiod at which the maximum spectral amplification occurs, the j
maximum spectral acceleration should be used. For higher modes, the actual spectral accelerations should be used." Was this followed in the analyses?
Response
j L
DOE-STD-1020 further states that "the actual spectrum may be used for all modes if there is a high l
confklence in the frequency evaluation and F is taken to be unity". Fp was set to 1.0 for a majority of the 1
structures at both Paducah and Portsmouth. For those structures at Paducah where Fp was taken as greater than 1.0 (e.g., Buildings C-331 and C-335 non-linear evaluation), the fundamental f:wquency fell within the
. displacement region (Iow frequency) of the response spectrum curve. Therefore, the stated requirement that was included within DOE-STD-1020 to compensate for softening of a structure from inelastic behavior with' resulting higher input spectral accelerations (i.e., as the fundamental frequency " moves" toward the peak of the spectrum) is not germane to structures with extremely low frequency response.
i SARUP Revision:
- No revision required.
~
c l
i l
+
l l
I c
i i
t i
l, l.
w M
w tregse "r
w ad
.---- we
>.e'>
w s-ry
--w-=rr--
-3 L
SARUP Q&R - PGDP/ PORTS --
November 10,1998 Q15 (NRC 7/14/98 Letter) n Were soil-structure interaction effects included in the analyses? If not provide your justification.
I
Response
r Studies were performed to address the inertial interaction effects of Paducah and Portsmouth structures
' with their supporting soil foundation. In addition, traveling seismic wave effects on seismic leads for these
, structures were evaluated. Further, the studies were peer reviewed by EQE, International who concurred with, and expanded somewhat, upon the results of the studies. The conclusions and recommendations of the EQE peer review were incorporated into the NPH evaluations of Paducah and Portsmouth structures.
SARUP Revision:
l No revision required.
i t.-
e' 1
E I
i GDP 98-0241 STATUS OF RESPONSES TO Page1 of2 NRC QUESTIONS / COMMENTS ON SARUP 1.
2/5/98 NRC Ouestions Letter from Robert C. Pierson (NRC) to Mr. James H. Miller (USEC), "Paducah Ccitificate Amendment Request - Update of the Application Safety Analysis Report-(TAC NO. L32043)," dated February 5,1998.
Submitted 2/27/98:
Ql,Q2,Q3,Q4 Working:
None 2.
2/25/98 NRC Ouestions Letter from Charles Cox (NRC) to Mr. James H. Miller (USEC), "Paducah and Portsmouth Certificate Amendment Requests-Update of the Application Safety Analysis Reports (TAC Nos. L32043 & L32044)," dated February 25,1998.
Submitted 3/27/98:
Q4,Q8,Q10,Q11,Q12,Q13,Q14,Q15,Q19 Submitted 4/21/98:
Ql,Q2,Q3,Q5,Q6,Q7,Q9,Q16,Q18 Submitted 5/1/98:
Q17 Working:
None 3.
6/1/98 NRC Ouestions Letter from Charles Cox (NRC) to Mr. James H. Miller (USEC), "Paducah and Portsmouth Certificate Amendment Requests-Update of the Application Safety Analysis Reports (TAC Nos. L32044 & L32043)," dated June 1,1998.
l Submitted 7/20/98:
Ch 2: Q1 Ch 3: Q4(a, f), Q5, Q6(a, g), Q7(a, d), Q8, Q9, Ql 1(a, b), Q 12, Q16(a, b), Q17(a, b), Q18(a, b), Q22(c), Q24(a, g)
Ch 4: Q2,Q5 Revised 10/5/98:
Ch 3: Q8 Working:
Ch3: Q1-Q3, Q4(b-e), Q6(b-f, h), Q7(b, c), Q10, Q11(c-f), Q13, Q14, Q15, Q16(c-f), Q17(c, d), Q18(c, d), Q19-Q21, Q22(a, b, d-f), Q23, Q24(b-f), Q25-Q35, Q37 i
Ch 4: Ql, Q3, Q4, Q6-21 l
i o
I
e l
GDP 98-0241 STATUS OF RESPONSES TO Page 2 of 2 NRC QUESTIONS / COMMENTS ON SARUP 4;
7/9/98 NRC Ouestions Letter from Charles Cox (NRC) to Mr. James H. Miller (USEC), "Paducah and Portsmouth Certificate Amendment Requests-Update of the Application Safety Analysis Reports (TAC Nos. L32044 & L32043)," dated July 9,1998.
Working:
Ch1: Q1 Ch 3: Q1,Q2 Ch 4: Ql,Q2 TSR: Q1 - Q130 5.
7/14/98 NRC Ouestions Letter from Charles Cox (NRC) to Mr. James H. Miller (USEC), "Paducah and Portsmouth Certificate Amendment Requests-Update of the Application Safety Analysis Reports (TAC Nos. L32044 & L32043)," dated July 14,1998.
Submitted 11/10/98: Q1 - Q15 Working:
None