...

• Transactions oflhe \Slh International Conference on
. ~ . Structural Mechanics in Reactor Technology (SMiRT ·15)
• ~ . Seoul, Korea, August 15-20, 1999
Comparison of Russian PNAE, ASME BPVC and Japan JEAG 4601
Regulations for Seismic Design of NPP's Piping Systems
1. Stevenson!), V. Kostarev
2
>, A. Bercovsky2
l
, A, Schukin
2
) and R. Masopuse)
1) Stevenson and Associates, Europe, USA
2) CKTI- Vibroseism Co. Ltd., Russia
3) Stevenson and Associates, Office in Czech Republic, Czech Republic
ABSTRACT
K14/2
This paper presents a summary of current earthquake design criteria used in Russian
(former Soviet) Regulations for equipment and piping systems of nuclear power plants (NPP)
in light of those used in United States and Japan. The detailed comparative seismic analysis
of PWR (WWER) Primary Coolant Loop System (PCLS) according to Russian PNAE Code
and ASME BPV Code with some comments regarding to Japan Code JEAG - 4601 was un-
dertaken for better understanding of the differences and coincidences of seismic design crite-
ria and requirements. The selection of these three guides for the study has very simple expla-
nation: according to ASME BVPC, JEAG and PNAE the huge majority of existing NPPs has
been designed.
1. INTRODUCTION
The international cooperation in safety and seismic upgrading of existing and new design
nuclear power plants as well as increasing coordination role of International Atomic Energy
Agency (IAEA) leads to necessity of more clear understanding of the criteria and standards
used in different countries in earthquake protection design.
On the other hand since the breakup of the Soviet Union there has been considerable concern
from World community and international organizations regarding the safety and, particularly,
seismic resistance of Soviet (Russian) designed reactors and NPPs in whole. Sometimes this
concern was substantiated, sometimes not.
Formerly, there were many attempts to compare different national nuclear safety standards.
Among them have to be mentioned one of the first research made in the 1979 and in early
80's, [1,2]. These and other efforts deal primarily with the contents of the Guides and Stan-
dards only, [3, 4]. However the most effective and beneficial way, as it seems, is to undertake
the direct compare analysis of the same NPP representative system using the given criteria,
material properties, formulas and methodology introduced in national or international Codes
and Guides, [5].
The present paper contains two comparative calculations of the PCLS (Loop No.4) of the
former Soviet design NPP WWER-1000 MWt Unit. These analyses are based on the applica-
tion of two national codes: Russian PNAE and American ASME BPVC with some references
to Japan JEAG 4601, [6-25]. The comparison of seismic analysis and calculation methodolo-
gies has been performed only for piping systems classified according to PNAE as category I,
class 1 following the ASME and class As according the importance classification in JEAG
4601.
Only sustained (pressure and weight) and seismic loads were considered in presented analy-
sis. The seismic excitation was chosen as MRZ (Maximal Design Earthquake) according to
IX-237
PNAE, SSE (Safe Shutdown Earthquake) in tenus of A8ME BPVC and 82 (Extreme Design
Earthquake) as in JEAG 4601. All these levels of design earthquakes are roughly equivalent.
All analyses presented in this paper have been carried out by application of the mostly wide-
spread in design practice RSMAM method.
2. EQUIPMENT CLASSIFICATION
All NPP equipment and piping systems are divided in PNAE as well as in ASME BPVC
on the groups A, B, C and Safety Classes 1,2, 3. The basis for such classification is the im-
portance for nuclear safety of these systems. Taking into account these circumstances the cur-
rent consideration contains the seismic comparative analysis of WWER-IOOO peLS, that is
classified as Class 2 ofPNAE and as Class 1 of ASME BPVC and JEAG.
PNAE divides NPP equipment and pipes on the two seismic categories. In contrast with
PNAE, ASME BPVC contains only one seismic category I. This category includes all com-
ponents and equipment for which is designated to remain their functionality if a SSE occurs.
In the JEAG all major equipment is divided on four Categories (Classes) of Aseismic Impor-
tance Classification: As, A, Band C.
The following Table illustrates various classifications of the considered PCLS according to
different Codes.
Table 1 CODE CLASSIFICATION
Code Group Class Seismic Category
PNAE B 2 I
ASME A I I
JEAG - I As
3. PIPING COMPONENTS STRENGTH ANALYSIS
All Codes: PNAE (for all Classes) as well as ASME BPVC and JEAG (for Class 1) re-
quire performing of strength analysis by checking the primary stresses on the basis of maxi-
mum shear stress theory of failure. These primary stresses are divided to the general mem-
brane stresses, local membrane stresses and bending stresses.
Taking into account that according to ASME BPVC the strength analysis is performed only
on the basis of membrane stresses and general bending stresses, the comparative seismic
strength analysis in this report has been carried for the same PNAE (0'5)2 stress category.
Table 2 contains the dependencies for the nominal allowable stresses applied for pipe's ele-
ments. The JEAG allowable nominal stresses Sm roughly the same as in ASME.
Table 2 ALLOWABLE NOMINAL STRESSES
I
Code
I
Symbol
I
Allowable Nominal Stresses
I
PNAE [0']
for all steels - min
ASME 8
m
ferrous steels - min {ST /3; /3; SU1.5}
austenite steel - min {ST /3; 1.1Si /3; Sy ) 1)
In result the values of allowable nominal stresses are practically the same for all codes.
lX-238
The main difference between ASME BPVC and PNAE is identification in accordance with
ASME BPVC four Levels of Service Limits Loading for each component or support. These
Service Limits may be designated in the Design Specification and defined as different Levels
(Levels A, B, C and D). The seismic loads are considered in strength analysis only for Levels
Band D (Appendix A, SRP 3.9.3).
The JEAG Code prescribes totally even 8 Load Combinations in seismic analysis of Class 1
piping (four with SI and also four combinations with S2 earthquakes).
In the further consideration only the Level D Service Limits in tenns of ASME BVPC will be
applied for seismic analysis ofPCLS according to the NCA-2I42.4.
The PNAE does not postulate Levels of Service Limits, which allow some damage of equip-
ment and piping for given set of design loading. Anyway, the different combinations of
loading sets present in PNAE (NUE, NNUE, AS) as well as ASME BPVC (SL, LOCA,
DBPB, MS/FWPB). So, herein only influence of seismic loading (MRZ or SSE) will be con-
sidered. Table 3 contains the comparison of allowable stresses for pipes.
Table 3 ALLOWABLE STRESSES
Code Level Class Category Loading S,
PNAE - 2 I NUE+MRZ 1.8[cr]
ASME D I I SL+SSE 3.0 Sm
JEAG - I As (I-IV)+S2
3.0 Sm
The comparison of allowable stress values for different materials in accordance with ASME
BPVC and PNAE is presented in Table 4.
Table 4 ALLOWABLE STRESSES MPa ,
MATERIAL PNAE ASME(JEAG) ASME (JEAG)/PNAE
St. 20 234 390 1.67
15GS 270 450 1.67
08HI8NIOT 212 354 1.67
Table 4 shows that the level of allowable stresses calculated according to PNAE (Category I)
are essentially lower than corresponding values obtained from ASME BPVC.
The ASME BVPC and PNAE codes use different formulas for calculation stresses in straight
pipe and in bend elements.
Analysis shows that fonnulas for stress calculation according to PNAE and ASME BPVC
give practically the same result for the resultant moments in the range of service pressures
when allowable stresses are the same.
However, the ratio between resultant moment values becomes less than 1 when the differ-
ences between allowable stresses are taken into account. The difference between PNAE and
ASME BPVC allowable stresses level results about 70% increasing of allowable moment Mi
(ASME) in comparison with Mi (PNAE).
Analysis Model
The finite-element approximation of the pipelines and the attached equipment components
has been used to create the calculation model of the Loop 4 piping system, Figure I.
IX-239
4G12
4GS4
4Cl
4C5
YP>
/
(Pl1
" YP2
'l'PlQ
~ B
)
YPB
Y'4
Figure 1. Dynamic Analysis Model of the WWER-l 000 PCLS
Input Seismic Excitation
The input seismic excitation for seismic analysis ofPCLS has been chosen in tenus ofRe-
sponse Spectra given in ASME BPVC Appendix N, N-1211. The considered Spectra have
been modified according to damping values for applicable Codes.
The ratio between vertical and horizontal components of seismic excitation according 10
AS ME BPVC is equal to 2/3. The JEAG and PNAE specified the ratio between seismic verti-
cal and horizontal components as 0.5.
According to ASME BPVC the following scaling coefficients of Response Spectra were ap-
plied for comparative analysis: the horizontal direction -- 0.6, the vertical direction -- 004.
Thus, for considered Response Spectra the ZPA values for the PCLS floor level were ac-
cepted as ZPA" ~ 0.6g (the horizontal direction) and ZPA, ~ O.4g (the vertical direction).
ASME BPVC Seismic Analysis
The seismic calculations based on the ASME BPVC NB-3600 requirements have been per-
formed using the "dPIPE" computer program developed by CKTI-VIBROSEISM Co. Ltd.
The internal seismic loads for pipeline and equipment of peLS were calculated with use of
RSMAM, when the following assumption have been made: SRSS rule for summation of
mode shapes and spatial components of response; Cut-off frequency at 33 Hz; Missing mass
effect to be considered.
The PCLS pipeline stress values have been obtained from Eq. 9 NB-3650.
The following output results have been obtained from ASME BPVC analysis: natural fre-
quencies and modal properties; nodal dynamic displacements and accelerations; stress values
for the straight pipe, bend and tee elements; dynamic loads in piping and equipment supports
and nozzles; resulting static and dynamic internal element loads.
IX-240
PNAE Seismic Analysis
The seismic calculations according to PNAE G-7-002-86 requirements also have been per-
fanned using the same "dPIPE" computer program.
The methodology used for determination of PCLS seismic response is based on the ASME
BPVC recommendations.
The dynamic and static stresses of the PCLS pipe elements have been calculated in accor-
dance with PNAE, Appendix 5, and Chapter 2 requirements.
Comparison of Analysis Results
The following parameters were chosen for comparative seismic analysis: dynamic displace-
ments; dynamic loads for supports and nozzles; stress level in the weakest elements; values of
seismic margin capacity.
The response of pipeline is strongly depended on acceleration level taken from the input Re-
sponse Spectra. At the same time Response Spectra acceleration depend on damping values
accepted for the analysis. Thus, for identical seismic excitation the response of piping system
may be quite different when different damping values are implemented. For example, the ra-
tio between maximal dynamic displacements (node 4p2, X-direction) calculated according to
PNAE and ASME BPVC respectively is equal to:
D max (2%) / D max (5%) = 43 II320 = 1.35.
It should be noted that this ratio for given system and seismic excitation corresponds to the
ratio between spectral accelerations at frequency 2.5 Hz from PNAE and ASME curves:
A
2
.5(2%) / A
2
., (5%) = 4.25/3.13 = 1.35.
The ratio between dynamic load values for supports and nozzles depends only from intensity
of the given Response Spectra and as was mentioned above is equal to 1.35. Thus it may be
concluded that results according to JEAG also should be more conservative versus ASME
due to recommended damping values: 2,5% and 5% correspondingly.
Figure 2 shows the comparison of stress values for the weakest PCLS elements.
1000
902
870
900 806
782
800
l.--'--------1---
• 700
0.
"
600
u;
w
500
0)
0)
400
w
'"
300
e-
O)
200
100
0
590
~
556
SIr. Pipe(4H1-4H2) Bend (4H4-4HS) Tee (Node 3)
DPNAE DASME
Figure 2. Comparison of stress values for the weakest peLS elements
The higher stress value for PNAE in comparison with ASME BPVC is explained by higher
magnitude (about 36 %) of the Response Spectra in the resonance frequency domain.
The ratio between maximal calculated stress values and allowable stresses for the weakest
peLS elements are shown in Figure 3. For PNAE case these values are about 2 times higher
than corresponding values for ASME BPVC (2.65/1.45=1.83).
IX-241
3
2,65
2,56
2,5
2,37
2

fiJ 1,5
0
tJ)
1,45
1,1
1,03
0,5
° Str. Pipe Bend (4H4-4H5) Tee (Node 3)
DPNAE DASME
Figure 3. Ratio between maximal Sss and allowable Sa stress values for the weakest elements
ofPCLS
In recent years the Seismic Margin Assessment Methodology in western engineering practice
is widely used. This methodology is based on the analysis of probability of failure for safety-
related structures, systems and components. On the basis of Conservative Deterministic Fail-
ure Margin (CDFM) the values of High Confidence Low Probability Failure (HCLPF) seis-
mic capacity have to be estimate in terms of maximum earthquake level. Regarding consid-
ered pipeline system this value may be defined according to the following equation:
S -s
HCLPF(CDFMj= a ,/ X ZPA.
Sss -Ssl
In this expression the value of HCLPF (CDFM) defines the level of seismic excitation corre-
sponded to the low probability of pipeline failure. ZPA is the maximum intensity of seismic
excitation on the pipeline floor level (the value of ZPA is used here instead of ZPGA value in
traditional SMA consideration for comparative purposes only),
Figure 4 shows the HCLPF values of seismic capacity for weakest elements of PCLS.
0,6
0,54
0.6 0,6
0,58
0,6
0,5
z
Q 0,4

0,3
0,19
w

0,2
w
"
I
"
0.1
'"
°
J,. ••
0,22
0,17
I
Str. pipe Bend Tee (Node 3)
IDPNAE DASME GZPA I
Figure 4. HCLPF values of seismic capacity for selected elements of PCLS
lX-242
For both code cases the minimum HCLPF seismic capacity is lower than input ZPA level
equal to 0.6g, that meanS the seismic resistance of PCLS for considered analysis is insuffi-
cient. However, the seismic capacity of system analyzed by ASME BPVC is more than two
times higher in comparison with values obtained by PNAE (004110.17 = 2041). It should be
noted that the later versions of ASME BVPC, 1995 and 1998 editions have increased this gap
and gives essentially less conservatism in seismic analysis ofNPP piping.
6. CONCLUSIONS
I. The basic principals of ASME BPVC, Section III, Subsection NB and Russian PNAE as
well as some features of Japan JEAG 4601 Code for seismic analysis ofNPP piping sys-
tems have been considered. Generally all three Codes are practically identical in main
principals of seismic analysis of piping system and are in good agreement with IAEA re-
quirements.
2. The principal distinctions between ASME BPVC, JEAG and PNAE codes are the values
of allowable stresses and damping ratio. The requirements of ASME BVPC and JEAG
4601 are much closer in definition of damping values and allowable stresses.
3. The obtained results show that using ofPNAE requirements for RSMAM seismic analysis
involve essential conservatism for seismic qualification of Class 1 piping systems under
SSE (MRZ, S2) seismic excitation in comparison with ASME BPVC and JEAG 4601. The
limit value of piping seismic capacity calculated by ASME BPVC more than two times
higher than corresponding level for PNAE Code. That means that Class I piping system
seismic analysis perforroed by RSMAM and according to demands of PNAE Code satis-
fies ASME BPVC Class I Service Level D and JEAG 4601 requirements.
4. The quantity results and ratios obtained in this study can oscillate significantly depending
on material properties, service temperature, pressure and types of pipe elements. Thus, for
every certain case it is necessary to carry out individual analysis to make definite conclu-
sion about relative degree of conservatism of each code.
5. The using ofTH seismic analysis for Class I piping systems, Service Level D leads to es-
sential decreasing of PNAE conservatism against ASME BPVC Codes results due to
Codes recommended damping.
6. In cases of analysis Class 2 and 3 piping systems and/or analysis under other than Service
Limit D conditions approximately similar or even vice-a-versus effect can be obtained in
range of conservatism of PNAE and ASME BVPC both for RSMAM and THA methods.
7. REFERENCES
I. Stevenson, J.D., "Standards - Status and Development in the Nuclear Industry", Proceed-
ifigs of ASCE Specialty Conference on Design of Nuclear Plant Facilities, Boston, April
1979.
2. Stevenson, J.D. and Thomas, F.A., "Selected Review and Evaluation of U.S. Safety Re-
search Vis-a-Vis Foreign Safety Research for Nuclear Power Plants", NUREG/CR-3212,
US. Nuclear Regulatory Commission, March 1983.
3. Charles H. Hofmayer, Young 1. Park, James F. Costello, "Understanding Earthquake De-
sign Criteria Used in Japan", Proceedings of the Fifth Symposium Orlando, Florida, De-
cember 1994. North Carolina State University, Raleigh, NC27695-7908IUSNRC/EPRJ.
4. Minematsu A., "Seismic Design of Equipment and Piping Systems for NPPs in Japan",
International Symposium on Seismic Safety Relating to NPPs, Kobe, Japan, 1997.
IX-243
5. Kostarev V., Masopust R., Bercovsky A., Schukin A. "Former Soviet Regulations for
Seismic Design of NPPs and Comparison with Current International practice", Interna-
tional Symposium on Seismic Safety Relating to NPPs, Kobe, Japan, 1997.
6. "Technical Guidelines for Aseismic Design of Nuclear Power Plants", JEAG 4601-1987,
Japan Electric Association, 1987.
7. PNAE G-5-006-87. "Seismic Design ofNPPs", 1988.
8. PNAE G-7-002-86. "Strength Analysis of Equipment and Systems of Nuclear Facilities",
1989.
9. ASME BPVC, 1992 Edition, Section III, Division I, Subsection NB-Class I, Nuclear
Power Plant Components, ASME (1993).
10.ASME BPVC, 1992 Edition, Section III, Division I, Subsection NC-Class 2, Nuclear
Power Plant Components, ASME (1993).
II.ASME BPVC, 1992 Edition, Section III, Division I, Subsection ND-Class 3, Nuclear
Power Plant Components, ASME (1993).
12.ASME BPVC, 1992 Edition, Section 1Il, Division I, Subsection NF-Supports, Nuclear
Power Plant Components, ASME (1993).
13.ASME BPVC, 1992 Edition, Section III, Division I, Appendix N "Dynamic Analysis
Methods", ASME (1993).
14.ASME BPVC, 1992 Edition, Section III, Division I, Appendix F "Rules for Evaluation of
Service Loading with Level D Service Limits", ASME (1993).
15.RG 1.29, " Seismic Design Classification" (Rev. 3, 9/78).
16.RG 1.60, "Design Response Spectra for Seismic Design of Nuclear Power Plants" (Rev. I,
12173).
17.RG 1.61, "Damping Values for Seismic Design of Nuclear Power Plants" (Rev. I, 10/73).
18.RG 1.92, "Combining Modal Responses and Spatial Components in Seismic Response
Analysis"(Rev. 1,2/76).
19.5RP 3.2.1. "Seismic Classification" (Rev.l, 6/81),Report NUREG-0800. NRC, Washing-
ton, 1989.
20.SRP 3.2.2. "System Quality Group Classification" (Rev.l, 6/81),Report NUREG-0800.
NRC, Washington, 1989.
21.SRP 3.7.1. "Seismic Design Parameters" (Rev.2, 8/89),Report NUREG-0800. NRC,
Washington, 1989.
22.SRP 3.7.2. "Seismic System Analysis" (Rev.2, 8/89),Report NUREG-0800. NRC, Wash-
ington, 1989.
23.SRP 3.7.3. "Seismic Subsystem Analysis" (Rev.2, 8/89),Report NUREG-0800. NRC,
Washington, 1989.
24.SRP 3.9.3. "ASME Code Class 1,2, and 3 Components, Components Supports, and Core
Support Structures" (Rev. I , 6/81), Report NUREG-0800. NRC, Washington, 1989.
25.ASME BPVC, Case N-411-I."Alternative Damping Values for Response Spectra Analy-
sis of Class 1,2 and 3 Piping ". Section III, Division 1,20.02.1986.
IX-244

The JEAG allowable nominal stresses Sm roughly the same as in ASME. In the JEAG all major equipment is divided on four Categories (Classes) of Aseismic Importance Classification: As. In contrast with PNAE. SSE (Safe Shutdown Earthquake) in tenus of A8ME BPVC and 82 (Extreme Design Earthquake) as in JEAG 4601.R:'~. Band C.1Si /3. PNAE divides NPP equipment and pipes on the two seismic categories. EQUIPMENT CLASSIFICATION All NPP equipment and piping systems are divided in PNAE as well as in ASME BPVC on the groups A.j1.min {ST /3. Taking into account these circumstances the current consideration contains the seismic comparative analysis of WWER-IOOO peLS. Sy /1. local membrane stresses and bending stresses. 1. The following Table illustrates various classifications of the considered PCLS according to different Codes.2. All analyses presented in this paper have been carried out by application of the mostly widespread in design practice RSMAM method. B. These primary stresses are divided to the general membrane stresses.9S~ ) In result the values of allowable nominal stresses are practically the same for all codes.min {ST /3.5} austenite steel .. the comparative seismic strength analysis in this report has been carried for the same PNAE (0'5)2 stress category. This category includes all components and equipment for which is designated to remain their functionality if a SSE occurs. Table 2 contains the dependencies for the nominal allowable stresses applied for pipe's elements. 2. lX-238 .1~ /3. that is classified as Class 2 ofPNAE and as Class 1 of ASME BPVC and JEAG. Table 1 CODE CLASSIFICATION Group Code PNAE ASME JEAG 3. All these levels of design earthquakes are roughly equivalent. Taking into account that according to ASME BPVC the strength analysis is performed only on the basis of membrane stresses and general bending stresses. C and Safety Classes 1. ASME BPVC contains only one seismic category I. A.5. B A - Class 2 I I Seismic Category I I As PIPING COMPONENTS STRENGTH ANALYSIS All Codes: PNAE (for all Classes) as well as ASME BPVC and JEAG (for Class 1) re- quire performing of strength analysis by checking the primary stresses on the basis of maximum shear stress theory of failure. The basis for such classification is the im- portance for nuclear safety of these systems.:'/2. SU1.5} I I I I 1) ASME 8m ferrous steels .PNAE. 1.6.0. Table 2 ALLOWABLE NOMINAL STRESSES Allowable Nominal Stresses Symbol Code PNAE [0'] for all steels . 3.min {R.

MS/FWPB).8[cr] 3.The main difference between ASME BPVC and PNAE is identification in accordance with ASME BPVC four Levels of Service Limits Loading for each component or support. C and D). B. Analysis Model The finite-element approximation of the pipelines and the attached equipment components has been used to create the calculation model of the Loop 4 piping system. Analysis shows that fonnulas for stress calculation according to PNAE and ASME BPVC give practically the same result for the resultant moments in the range of service pressures when allowable stresses are the same. LOCA. Anyway. Table 3 contains the comparison of allowable stresses for pipes. AS) as well as ASME BPVC (SL. 1.0 Sm The comparison of allowable stress values for different materials in accordance with ASME BPVC and PNAE is presented in Table 4.4.3). the ratio between resultant moment values becomes less than 1 when the differences between allowable stresses are taken into account. The difference between PNAE and ASME BPVC allowable stresses level results about 70% increasing of allowable moment Mi (ASME) in comparison with Mi (PNAE).9. These Service Limits may be designated in the Design Specification and defined as different Levels (Levels A. NNUE.67 1.0 Sm 3. Table 4 ALLOWABLE STRESSES . the different combinations of loading sets present in PNAE (NUE. In the further consideration only the Level D Service Limits in tenns of ASME BVPC will be applied for seismic analysis ofPCLS according to the NCA-2I42. which allow some damage of equipment and piping for given set of design loading. The JEAG Code prescribes totally even 8 Load Combinations in seismic analysis of Class 1 piping (four with SI and also four combinations with S2 earthquakes). So. MPa ASME(JEAG) MATERIAL PNAE 390 St. SRP 3. The PNAE does not postulate Levels of Service Limits. However.67 1. Figure I. herein only influence of seismic loading (MRZ or SSE) will be considered. 20 234 15GS 450 270 354 08HI8NIOT 212 ASME (JEAG)/PNAE 1. DBPB.67 Table 4 shows that the level of allowable stresses calculated according to PNAE (Category I) are essentially lower than corresponding values obtained from ASME BPVC. The seismic loads are considered in strength analysis only for Levels Band D (Appendix A. IX-239 . The ASME BVPC and PNAE codes use different formulas for calculation stresses in straight pipe and in bend elements. Table 3 ALLOWABLE STRESSES Category Code Level Class I PNAE 2 I ASME D I JEAG I As - Loading NUE+MRZ SL+SSE (I-IV)+S2 S.

IX-240 . The JEAG and PNAE specified the ratio between seismic vertical and horizontal components as 0. The considered Spectra have been modified according to damping values for applicable Codes. The ratio between vertical and horizontal components of seismic excitation according 10 AS ME BPVC is equal to 2/3. N-1211. for considered Response Spectra the ZPA values for the PCLS floor level were accepted as ZPA" ~ 0. when the following assumption have been made: SRSS rule for summation of mode shapes and spatial components of response. Missing mass effect to be considered. the vertical direction -. Dynamic Analysis Model of the WWER-l 000 PCLS Input Seismic Excitation The input seismic excitation for seismic analysis ofPCLS has been chosen in tenus ofRe- sponse Spectra given in ASME BPVC Appendix N. bend and tee elements. Thus. 9 NB-3650. The internal seismic loads for pipeline and equipment of peLS were calculated with use of RSMAM.6g (the horizontal direction) and ZPA.4G12 4GS4 4Cl 4C5 YP> (Pl1 " 'l'PlQ / YP2 ~B Y'4 ) YPB Figure 1. ~ O. dynamic loads in piping and equipment supports and nozzles.0. Ltd. ASME BPVC Seismic Analysis The seismic calculations based on the ASME BPVC NB-3600 requirements have been performed using the "dPIPE" computer program developed by CKTI-VIBROSEISM Co. stress values for the straight pipe. The PCLS pipeline stress values have been obtained from Eq.004. nodal dynamic displacements and accelerations. resulting static and dynamic internal element loads.6. According to ASME BPVC the following scaling coefficients of Response Spectra were applied for comparative analysis: the horizontal direction -. Cut-off frequency at 33 Hz.5.4g (the vertical direction). The following output results have been obtained from ASME BPVC analysis: natural frequencies and modal properties.

Appendix 5. X-direction) calculated according to PNAE and ASME BPVC respectively is equal to: D max (2%) / D max (5%) = 43 II320 = 1. " 0) 0) 0. Pipe(4H1-4H2) l.PNAE Seismic Analysis The seismic calculations according to PNAE G-7-002-86 requirements also have been perfanned using the same "dPIPE" computer program. The methodology used for determination of PCLS seismic response is based on the ASME BPVC recommendations. (5%) = 4. Comparison of Analysis Results The following parameters were chosen for comparative seismic analysis: dynamic displacements. the ratio between maximal dynamic displacements (node 4p2. • w e'" w O) 900 800 700 600 500 400 300 200 100 0 870 ~ Bend (4H4-4HS) DPNAE DASME 556 SIr. values of seismic margin capacity.. At the same time Response Spectra acceleration depend on damping values accepted for the analysis.35. stress level in the weakest elements.83).5% and 5% correspondingly. for identical seismic excitation the response of piping system may be quite different when different damping values are implemented. The ratio between maximal calculated stress values and allowable stresses for the weakest peLS elements are shown in Figure 3. For PNAE case these values are about 2 times higher than corresponding values for ASME BPVC (2. The ratio between dynamic load values for supports and nozzles depends only from intensity of the given Response Spectra and as was mentioned above is equal to 1.35. IX-241 .13 = 1.5 Hz from PNAE and ASME curves: A2.35. Thus. and Chapter 2 requirements. Figure 2 shows the comparison of stress values for the weakest PCLS elements. For example. Comparison of stress values for the weakest peLS elements The higher stress value for PNAE in comparison with ASME BPVC is explained by higher magnitude (about 36 %) of the Response Spectra in the resonance frequency domain.25/3. Thus it may be concluded that results according to JEAG also should be more conservative versus ASME due to recommended damping values: 2.65/1.45=1. 1000 902 806 590 782 u. It should be noted that this ratio for given system and seismic excitation corresponds to the ratio between spectral accelerations at frequency 2.5(2%) / A 2. The dynamic and static stresses of the PCLS pipe elements have been calculated in accordance with PNAE.--'--------1--Tee (Node 3) Figure 2. dynamic loads for supports and nozzles. The response of pipeline is strongly depended on acceleration level taken from the input Response Spectra.

Regarding considered pipeline system this value may be defined according to the following equation: HCLPF(CDFMj= S a -s.6 0.3 0.56 fiJ 0 • 2 1. HCLPF values of seismic capacity for selected elements of PCLS (ZPA~O.1 ° I Str.17 0.5 0.5 1.6 0.19 0. On the basis of Conservative Deterministic Failure Margin (CDFM) the values of High Confidence Low Probability Failure (HCLPF) seismic capacity have to be estimate in terms of maximum earthquake level.22 0.6g) lX-242 .58 0.6 Q 0. 0. Figure 4 shows the HCLPF values of seismic capacity for weakest elements of PCLS. Ratio between maximal Sss and allowable Sa stress values for the weakest elements ofPCLS In recent years the Seismic Margin Assessment Methodology in western engineering practice is widely used.65 2.5 2. •• I Bend (4H4~4H5) Tee (Node 3) IDPNAE DASME GZPA I Figure 4. pipe (4H1~4H2) J.2 ~ ~ 0.37 2.3 2. ZPA is the maximum intensity of seismic excitation on the pipeline floor level (the value of ZPA is used here instead of ZPGA value in traditional SMA consideration for comparative purposes only).6 0.4 z " " '" w w 0.5 ° Str.. Pipe (4H1~4H2) Bend (4H4-4H5) DPNAE DASME Tee (Node 3) Figure 3. systems and components.54 ~ 0.1 tJ) 1.45 1. Sss In this expression the value of HCLPF (CDFM) defines the level of seismic excitation corresponded to the low probability of pipeline failure./ -Ssl X ZPA. This methodology is based on the analysis of probability of failure for safetyrelated structures.03 0.

5.6g. The quantity results and ratios obtained in this study can oscillate significantly depending on material properties. Service Level D leads to essential decreasing of PNAE conservatism against ASME BPVC Codes results due to Codes recommended damping. REFERENCES I. pressure and types of pipe elements. 1997. However. The using ofTH seismic analysis for Class I piping systems. Young 1. Generally all three Codes are practically identical in main principals of seismic analysis of piping system and are in good agreement with IAEA requirements. Park. Proceed- ifigs of ASCE Specialty Conference on Design of Nuclear Plant Facilities. The limit value of piping seismic capacity calculated by ASME BPVC more than two times higher than corresponding level for PNAE Code.S. Boston. April 1979. "Selected Review and Evaluation of U. F. The principal distinctions between ASME BPVC... S2) seismic excitation in comparison with ASME BPVC and JEAG 4601. 6. J. Raleigh.D. for every certain case it is necessary to carry out individual analysis to make definite conclusion about relative degree of conservatism of each code. J. Stevenson. Costello. "Seismic Design of Equipment and Piping Systems for NPPs in Japan". March 1983. 4. JEAG and PNAE codes are the values of allowable stresses and damping ratio. Subsection NB and Russian PNAE as 2. 7. Thus. In cases of analysis Class 2 and 3 piping systems and/or analysis under other than Service Limit D conditions approximately similar or even vice-a-versus effect can be obtained in range of conservatism of PNAE and ASME BVPC both for RSMAM and THA methods. NC27695-7908IUSNRC/EPRJ. Charles H. 6. Safety Research Vis-a-Vis Foreign Safety Research for Nuclear Power Plants". 4. International Symposium on Seismic Safety Relating to NPPs. Florida. That means that Class I piping system seismic analysis perforroed by RSMAM and according to demands of PNAE Code satisfies ASME BPVC Class I Service Level D and JEAG 4601 requirements. It should be noted that the later versions of ASME BVPC. that meanS the seismic resistance of PCLS for considered analysis is insufficient. 3.. US. The basic principals of ASME BPVC. North Carolina State University.Status and Development in the Nuclear Industry". and Thomas. The obtained results show that using ofPNAE requirements for RSMAM seismic analysis involve essential conservatism for seismic qualification of Class 1 piping systems under SSE (MRZ. "Understanding Earthquake Design Criteria Used in Japan". "Standards . Japan. Nuclear Regulatory Commission. well as some features of Japan JEAG 4601 Code for seismic analysis ofNPP piping systems have been considered. 1995 and 1998 editions have increased this gap and gives essentially less conservatism in seismic analysis ofNPP piping.17 = 2041). CONCLUSIONS I.A. The requirements of ASME BVPC and JEAG 4601 are much closer in definition of damping values and allowable stresses. December 1994. Stevenson. Kobe. Proceedings of the Fifth Symposium Orlando. 2. IX-243 . service temperature. Section III. Hofmayer. NUREG/CR-3212. 3.For both code cases the minimum HCLPF seismic capacity is lower than input ZPA level equal to 0. the seismic capacity of system analyzed by ASME BPVC is more than two times higher in comparison with values obtained by PNAE (004110. Minematsu A.D. James F.

PNAE G-5-006-87.7.2. 16. "Seismic System Analysis" (Rev. 23. Washington. 6. Section 1Il. 1987.RG 1.3.Report NUREG-0800.ASME BPVC. Appendix N "Dynamic Analysis Methods". NRC. 1989. and Core Support Structures" (Rev. I.SRP 3. 9/78). and 3 Components. Nuclear Power Plant Components. 1992 Edition.RG 1. 14. 19. 25. "Damping Values for Seismic Design of Nuclear Power Plants" (Rev.ASME BPVC. 1.5. "Technical Guidelines for Aseismic Design of Nuclear Power Plants". 8. 8/89).l.7.2.29.2. "Strength Analysis of Equipment and Systems of Nuclear Facilities". II. Section III.02. NRC. 1992 Edition. 6/81).l. 13. 17.RG 1. "ASME Code Class 1. "Seismic Classification" (Rev.RG 1. Section III. 15.SRP 3. Subsection NB-Class I..61.5RP 3. Division I. 1989. 1989. 1989. 10/73). 9.1. ASME (1993). 1992 Edition. 10. "System Quality Group Classification" (Rev. I . Division I.2 and 3 Piping ". 12173). "Former Soviet Regulations for Seismic Design of NPPs and Comparison with Current International practice". Nuclear Power Plant Components.SRP 3. Subsection NF-Supports.Report NUREG-0800.3.Report NUREG-0800.SRP 3. 6/81).60. 1989. 1997. "Seismic Design Parameters" (Rev. 1988. Washington. ASME (1993). Washington.92. Japan.9. ASME (1993). JEAG 4601-1987. " Seismic Design Classification" (Rev. ASME (1993). Division I.2/76). Section III. 22. 8/89). Masopust R. Division 1. 12. NRC. Bercovsky A.ASME BPVC. Kobe. Report NUREG-0800. "Design Response Spectra for Seismic Design of Nuclear Power Plants" (Rev. Section III. ASME BPVC. Japan Electric Association..2.ASME BPVC. 6/81). I. 8/89).2. "Seismic Subsystem Analysis" (Rev. Washington. 21.1986. Appendix F "Rules for Evaluation of Service Loading with Level D Service Limits".2. 7. Division I. 3. 1989.ASME BPVC. "Seismic Design ofNPPs". Washington. Division I.SRP 3. ASME (1993). Section III. NRC. NRC. Washington. Kostarev V. 20. PNAE G-7-002-86. Subsection ND-Class 3.20.. Components Supports. International Symposium on Seismic Safety Relating to NPPs.Report NUREG-0800.2. ASME (1993). 1992 Edition. Schukin A. Section III.1. Subsection NC-Class 2."Alternative Damping Values for Response Spectra Analysis of Class 1.Report NUREG-0800. NRC. 1992 Edition. 1992 Edition. 18.2. 1989. Nuclear Power Plant Components.ASME BPVC. 24. Division I. "Combining Modal Responses and Spatial Components in Seismic Response Analysis"(Rev. Case N-411-I. IX-244 .7. Nuclear Power Plant Components.