Professional Documents
Culture Documents
1601651
1601651
ABSTRACT NOMENCLATURE
This paper presents the seismic design and evaluation pipe cross-sectional area, net after the
guidelines for underground piping for the Department corrosion allowance
of Energy (DOE) High-Level-Waste (HLW) apparent wave velocity traversing the site
Facilities. The underground piping includes both pipe diameter
single and double containment steel pipes and depth to the top of the pipe
concrete pipes with steel lining, with particular stress intensity factor
emphasis on the double containment piping. internal pressure
peak ground acceleration of the seismic
The design and evaluation guidelines presented in this wave
paper follow the generally accepted beam-on-elastic- peak ground displacement
foundation analysis principle and the inertial response peak ground velocity
calculation method, respectively, for piping directly pipe radius
in contact with the soil or contained in a jacket. A Code allowable stress value
standard analysis procedure is described along with primary stress indices
the discussion of factors deemed to be significant for soil adhesion (units of stress)
the design of the underground piping. Young's modulus of the soil
Young's modulus of the pipe material
The following key considerations are addressed: 1. the secant moduIus of elasticity of the buried
the design feature and safcty requirements for the Pipe
inner (core) pipe and the outer pipe, 2. the effect of maximum friction f o r c e h i t length of pipe
soil strain and wave passage, 3. assimilation of the axial force from thermal expansion
necessary seismic and soil data, 4. inertial response axial force from both anchor movement and
calculation for the inner pipe, 5. determination of wave passage
support anchor movement loads, 6 . combination of the frequency (in Hz) at the peak spectral
design loads, and 7. code comparison. Specifications velocity
and justifications of the key parameters used, stress coefficient of soil earth pressure at rest
components to be calculated and the allowable stress modulus of subgrade reaction
and strain limits for code evaIuation are presented. moment of inertia of the pipe cross section
the dominant wave length of the passing
Finally, a numerical example is provided to illustrate seismic wave
the application of the guidciines. resultant moment from sustained loads
resultant moment (use only one-half the
moment range from earthquake) from
occasional loads
' This study was sponsored by the U. S . Department of Energy, Office of Environmental Restoration and
i
deadweight and earthquake inertia response during a seismic event. Because it is continuously
moment (use only one-half the moment supported by soil, the inertial response of the buried
range from earthquake), respectively outer piping is rarely significant. Instead, its stress
range of resultant moment from thermal and strain are more significantly controlled by the
expansion soil strain from ground motion. The beam-on-elastic-
resultant from scismic loads due to both foundation principle (Reference 1) is generally used
anchor movement as well as wave passage, to calculate the response of the outer pipe.
use half the range
MWP moment due to wave passage On the other hand, the inner piping (and also the
moment due to oscillatory S A M outer piping when it is not in direct contact with the
bearing capacity factor soil) is expected to behave in many ways similarly to
expansion stress above ground piping. The inertial response could
bending stress (one-half the range) become a significant factor in this case, and the finite
allowable stress at temperature element response calculation method can be
torsional Stress (one-half the range) employed.
stresses due to occasional loads and
sustained loads There are currently no industry codes or standards
material yield stress which provide necessary guidance for the analysis of
f, nominal pipe thickness the buried double containment piping where both
v, the apparent wave speed of the seismic piping inertial effects and the effects of soil strains
motion are present. Chapter 7 of Reference 2, fulfills such a
weight of pipe and its contents need.
wave Coeffcient, 1 for compression or
Rayleigh wave, 2 for shear wave This paper presents a summary discussion of the
the Poisson's ratio for soil methods, procedures, and criteria recommended in
axial strain Reference 2 for buried piping and conduits.
total strain of the pipe
backfill weight density
apparent angle of pipe wall friction (degrees) 2. CONTROLLING FACTORS FOR THE
average contact pressure bctween soil and ANALYSIS OF BURIED PIPING
the pipe due to soil weight and surcharge
loadings The inertial response analysis method adopted for the
above ground piping is not entirely suitable even for
the analysis of the inner piping when it behaves fairly
1. INTRODUCTION similar to the above ground piping, because of the
presence of the outer pipe. On the other hand, the
Underground piping for the DOE HLW facilities is beam-on-elastic-foundation analysis approach
normally encased (or double containment) piping traditionally employed for the close-formed static
system, i.e., consists of a "pipe in a pipe" analysis of the single-walled buried piping needs to
arrangement. The inner (or core) pipe is designed to be modified when a computer aided finite element
convey waste and is intended to remain leak tight for analysis is desired.
all design loads, including scismic loads. The outer
pipe is designed to provide additional assurance that Table 1 presents a comparison of some of the more
the waste carried by the core pipe will be contained, important factors affecting the design evaluations of
in the unlikely event of leakage from the core pipe. various piping arrangements for the buried piping.
The guidelines provided in Reference 2 and discussed
There are many possible design combinations of the in detail in this paper are intcnded to ensure that the
encased underground piping. For instance, the outer key factors affecting the design evaluation of the
pipe could be steel, ceramic tile, or concrete. The buried piping are properly addressed.
outer pipe could also be either in direct contact with
the supporting soil, or supported and housed in
underground concrete trenches. 3. REQUIRED SEISMIC AND SOIL DATA
Piping which is in direct contact with the soil is Appropriate seismic and soil data must be assembled
confined by soil and controlled by soil movements to allow for evaluation of the maximum response
DISCLAIMER
Soil properties of specific interest include: For long straight sections of piping, ground
movement induced forces, moments and stresses are
grain size distribution description found to be relatively insensitive to the specific value
plasticity characteristics of the stiffness parameter, except for the relatively
density and relative density unusual case of piping traversing areas of
equivalent SPT blow count information significantly different soil properties.
effective coefficient of subgrade reaction
liquefaction potential However, ground movement induced loads computed
equivalent cyclic properties obtained from at or near pipe bends and structural intersection
either site specific laboratory testing or points can be strongly dependent on the specific
gencric data for appropriate soil types value of the modulus. It is therefore important that
conservative estimates of this modulus be selected for
Enough data must be available to allow for evaluation any analysis. This is typically accomplished by
of the effect of seismically induced pore pressures on incorporating the effects of variability in this
ground movements and thc loads induced on the parameter on the computed stress resultants.
underground system from relative movements
between the ground and the piping. It should be noted that, for buried piping analyses,
conservative estimates of the subgrade modulus tend
A range of generic values may be used with sufficient to lie on the stiffer side of the range of potential
variability to account for undnowns. variability. This is different from the normal
assumptions of foundation analyses which typically
are made to underestimate the stiffness parameter so
1. DETERAIINATION OF KEY INPUT as to overestimate settlement values.
PARABlETERS FOR ANALYSIS
Due to greater uncertainties associated with the
Modulus of Suberade Reactions selection of the subgrade modulus, the buried piping
analysis performed should incorporate a wider
ypically, the buried piping system is analyzed as a variation in the modulus than is typically used in
series of beam elements on an elastic foundation, standard soil-structure interaction analyses so a s to
requiring the assumption of modulus of subgrade lead to bounding estimates of induced loading. Site
reaction for the soiWpipe system. In a computer specific measurements, when available, can be used
analysis, the modulus of subgrade reaction (lbs/in3) is to reduce the variability in parameter that must be
used to determine the soil spring constants at the pipe incorporated into the evaluations, since the
i
rccommendations for gcncric soils incorporate such In addition, if the plate load test is conducted, for
\ d c variations in thc approprintc cocfficicnts. example, using a 1 foot diamcter rigid plate, the dope
of the pressure-displacement curve must be suitably
The form of the reaction-displacement relations modified to account for the difference between the
rcquired to perform the analysis of buried piping is pipe and plate diameter; that is:
shown in Figure 1. In each case, the soil reaction to
the specified displacement can be approximated by k = 41 f(D>
means of a linearly varying relationship to the
maximum value defined by the soil strength as well where k,, is the coefficient of subgrade reaction for
as the piping configuration. The slope of the straight the 1 foot diameter plate and f@) is a function of the
line segment is defined as the modulus of subgrade pipe diameter as well as the soil type. For plate tests
reaction. For application to simple bcam models, the conducted on the surface of sandy soils, the
Winkler spring coefficient is then defined as the experimentally determined relationship is found to be:
product of the subgrade modulus and the pipe
diamcter.
For the case of piping placed near the surface in while for typical cohesive soils, the relationship is
uniform soils, with a soil cover of at least three
diameters, the effective subgrade modulus in the k = k,, W I (3)
horizontal direction, K,, can normally be expected to
be larger than the vertical modulus, k,,, since the Typical values of coefficient of subgrade reaction
resistance of the soil cover above the piping is less from plate load tests are listed in Tables 2 and 3.
effective for resisting displaccmcnt. Thercfore, if the
pipe is buried thrcc or more pipc diameters deep, the Analytic Estimates
horizontal stiffness is rccommendcd to be at least
equal to the vcrtical stiffness, unless special Several formulas have been used in the past to
conditions indicate such is not the case. estimate the modulus of subgrade reactions for the
horizontal lateral direction. One of these is Vesic's
Limiting Values of Lateral Load approximation (Reference 5 ) in the following form:
Estimates of in-situ stiffness can be obtained from Equation (6) is not a direct function of the pipe
simple field tests using plate loads, although such test rigidity as in the case of Equation (4) but it includes
results tend to undcrcstimate thc stiffncss of fully the depth of the pipe below grade (H) as a parameter.
confined soils. In addition, the results quoted from Therefore, the modulus of subgrade reactions
such tests are typically associated with shear strains calculated from Equations (4) and (6) may differ
in the range of 0.1% to I%, highcr than the strains markedly. Hence, care should be exercised when
associated with the initial slope of the load determining the appropriate equation to be used for a
displacement curve. The stiffness from these tests particular situation.
can be increased by a factor of about 2 to amve at
results appropriate or the lower strain levels Because piping is normally modelled as a concentric
associated with the vibratory problem. (line) model, the Spring constant of the soil per unit
Icngth of the pipc is calculatcd by multiplying the The following loads are normally considcrcd in the
modulus of subgrade reaction with the pipe diameter. buried piping analysis:
Considering the ground wave to vary sinusoidally as where g, is a wave coefficient dependent on wave
it traverses the site, the maximum ground strain that type and listed in Table, 4.
can be developed will occur over a quarter wave
length of the predominant seismic wave associated The maximum forces and moments developed in long
with the PGV. Thus the maximum strain that can be straight sections of pipe as computed above typically
transmitted to the buried pipe through shear is limited do not lead to the maximum stress conditions in the
to: pipe (Reference 19). Rather, the controllin_econdition
(EJmax 5 FmaxLwY(4EsctA) (8) usually occurs at pipe bends and elbows as a result of
differential motions developed from the wave passage
The maximum force pcr unit length ,F which can effect.
be transmitted through friction to the pipe is related
to the characteristics of both the pipe and the soil. Effect of PiDe Bends. Elbows. and Tee's
The dominant wave length of the passing seismic
wave L, can be estimated from the site response The effect of changes in direction, size and material
spectrum using the relation C/f, where C is the of the underground piping must be considered for a
apparent wave velocity, f,, is the frequency (in Hz)at complete evaluation of critical conditions. Using the
which the peak spcctral velocity occurs (typically of ground strain calculated based on Equation (7,the
I
forccs and moments dcvciopcd at the critical Equation 9, together with the displacement estimate
locations of thc underground piping can be for full mobilization leads to an estimate of the axial
detcrmined by following the usual procedures as stiffness as follows:
outlincd in References 11 through 20 for pipe
elcmcnts of different configurations.
For double-walled piping, one additional item which The Winkler spring coefficient in the axial direction,
needs to bc considcred in the beam-on-elastic- ,k has units of force per length of pipe over unit
foundation analysis is the increase in sccondary pipe displacement (such as psi) and is a direct input into
stiffncss due to the prcscnce of the primary pipe. the beam on elastic foundation analysis.
More rigid pipcs gcncrally lead to higher pipe strains
and stresses, particularly at the elbows, due to the Discretization Recommendations
applied soil strains. To be conservative, the potential
increase in the overall axial and bending stiffnesses When analytical solutions for beam on elastic
of the secondary pipe should be assessed. foundation models are used, stresses, displacements,
etc., are written directly in terms of the system
The amount of stress at the clbow is proportional to characteristic parameter defined by
the deformation (displacement and rotation) at the
elbow. To determine the deformation at the elbow,
it is necessary to know first if the longitudinal leg has
the length requircd to allow thc friction force to be The parameter p relates soil and beam stiffnesses and
fully developcd at both cnds of the pipe. Whether controls the magnitude of responses.
the longitudinal leg is long or short has a direct
bearing on the development of the close-formed In regions of high bending (near concentrated loads
solution. or at pipe bends), the pipe displacements generally
decrease relatively rapidly with distance from the
Likewise, for the transverse leg, the important item is discontinuity. The length of this zone is sometimes
whether the pipe run is longer or shorter than the called the "bearing span" and is estimated to be
influence length charactcnzed by Hetenyi's system
Characteristic factor j3 (Reference 11). L, = 3X/(4P) (14)
Each combination o l the long and short legs of an Within this distance from the discontinuity, spacing
elbow results in one set of closc-formed solutions. of the Winkler springs should be limited to about 1.5
Because there arc many possible combinations of the to 2 times the pipe diameter to obtain reasonable
long and short legs and piping configurations, care accuracy in the computed stresses. For zones away
should be exercised to ensure h a t all possible from these points of discontinuity, discretization of
combinations of the required analyses have been soil springs could be relaxed since changes in
considered when a close-formed solution approach is bending moment per element decrease.
used.
SUDDO
Anchor
~ ~ Movements (SAM')
For the finite element computer analysis using soil
springs, several additional factors, discussed as In addition to the stresses developed in the
follows, need to be considcrcd. underground piping system from the wave passage
effects, the stresses induced in the piping system from
Asia 1 S t i f fne ss Est ima tc s relative movement at anchor points and other building
penetrations or attachment points (SAM) must be
The effective asial stiffncss uscd in the clastic considered. It is usual in this situation to consider a
foundation model for the soil surrounding the pipe is static analysis of the piping system subjected to out-
based upon studies of pile behavior subjected to axial of-phase anchor movements placed on the system to
loads (Reference 20). For example, the displacement produce worst case conditions.
required to develop the full adhesion force on a pile
which is 12 inches in diameter is estimated to be 0.2 In analyzing the support anchor movement loads, the
to 0.5 inch. procedures of beam-on-elastic-foundation must again
be utilized, with the same degree of variability
The maximum frictional force, F,, as specified in required to ensure conservative estimates of stress
L
rcsultants. As recommended in Reference 7, the A. Stress Limits for Deadweight and Inertial
strcss rcsultants computed for anchor point movement -
Loads
can be combined with those computed for passage
effects using the SRSS method. The allowable stress and strain criteria specified
herein are intended to compliment the requirements
Permanent Differential Ground Movements of the ASME Codes and be consistent with the needs
of the DOE HLW facilities.
Both horizontal and vertical long term total and
differential ground displacements must be estimated The DOE HLW underground piping is typically
which account for gross settlement or consolidation designed in accordance with the ANSUASME B3 1.3
of foundation soils as well as potential maximum Code [Reference 251. Section 319.4.4 of the B31.3
movements of building attachment points. Stress Code specifies the equations and allowable stress
resultants induced by these movements can then be limits for flexibility stresses based on a combination
estimated by considering the buried piping to be of the torsional stress and the bending stress, as
supported on an elastic foundation. follows:
s, = (S,’ + 4 s y (16)
The stiffness coefficients selected for the analysis of
long term response should take into account any In contrast, the B3 I. 1 Code [Reference 261 requires
differences in soil behavior associated with the effects the combination of the axial stress from pressure plus
of long term loading as opposed to the effective the bending stress in the following form:
stiffness associated with the short durations associated
with wave passage effects. Again, sufficient
variability in soil stiffness properties must be
included in the analysis to assess worst case There are obvious differences between Equations (16)
conditions. and (17). One key factor being that Equation (16)
includes the axial stresses in the calculation.
A summary of approaches to evaluate the potential of
inducing permanent settlements from a variety of For the buried piping, the axial stress in the piping
sources is presented in Reference 30. Also, standard induced by the axial soil strain is a factor which
procedures are available (References 31 and 32) to should not be ignored.
estimate induced vertical strain levels from the
maximum induced cyclic shear strains developed in A comparison of the ASME Section I11 Subsection
the soil layer during the seismic loading. NC Code equations with the B31.1 Code would show
that both codes use similar equations and material
Alloivable Stress and Strain Limits for Inner Piping specifications.
Concrete Conduit Requirements For instance, the left hand side equation of the ASME
Section 111, Subsection NC Code, adopted for the
The concrete componcnt of any piping system should occasional loads can be written as follows:
meet the general requirements of the current version
of the ACI Code 349-85. Compressive strains
devcloped in the concrete should be limited to no
more than 0.3%. Tensile requircmcnts should be While Equations (16) and (17) may appear to be
controlled by the amount of rcinforcing steel added to substantially equal if the minor differences in the
the concrete as specificd in the ACI Code. If stress indices are ignored and the moments are
required to prevent any cracking from developing in properly accounted for, there is one important
the concrete, tensile strains should be limited to: difference in both Codes, i.e., B3 1.1 does not have a
complete range of stress criteria as required by
Section I11 Subsection NC, particularly when faulted
condition loads are considered.
*
where f Cis the strength and E, is the elastic modulus
of the concrete. Thus, to be consistent with the dcsign needs of the
buried piping, the stress equation can be written in
Steel Pining and Componcnts the nomenclature of the ANSUASME B3 1.1 Code but
using the allowable limits specified in the Subsection
NC Code for the faultcd condition loads as follows: as an alternative to the time consuming nonlinear
plastic analysis.
In addition to meeting the requirements for the If and when Equation (19) is not satisfied, a strain
inertial loads, ASME Section 111, Subsection NC also based criterion would seem to be appropriate to
requires that the support anchor displacements due to further determine the piping design acceptability.
earthquake effects be considered in Equation (10) of
the Code, if they are not included in Equation (9). As a substitution for the plastic analysis, a stress-
Equation (19), above, includes only the seismic strain correlation method [Reference 281 could be
inertial response loads. Equation (20), below, can be .used to convert the elastically calculated code stress
used to satisfy the intent of Equation (10) of the to the equivalent total strain.
Code.
(F,+F,,,)/A+i(MC+MDISP)/Z The following allowable strains can be used:
13S,, but 52S, (20)
where q S 1% for carbon steel
MD,P=[Mwa+%*M21 In (2 1) S 2% for stainless steel
If permanent settlement is expected following the These allowable values are consistent with the
seismic event, then the axial stress due to the recommendations of Reference 28.
resulting SAM shall be limited to the smaller of 3S,
and 2S, For carbon steel, the allowable strain is more
conservative than the 2% strain allowed by Code case
Equation (20) diffcrs from Equation (10) of the Code N-47 [Reference 291.
in scveral aspects. The most important difference is
that Equation (20) includes the asial force generated Finally, the axial strain of Equation (20) should be
from both thermal expansion and support anchor limited to:
movements. This is to recognize the potential . < 0.2/Wt,,)
E (23)
significance of the axial force in the inner pipe
induced by the axial movement of the outer pipe. to avoid compressive wrinkling.
C. Strain Limits Allowable Stress and Strain Limits for Outer Pipinq
If both Equations (19) and (20) cannot be met, the The outer piping is required to maintain its structural
piping design is still considcred acceptable if the integrity. For piping which is not in contact with the
maximum piping strain is limited to a reasonable soil, meeting Equations (19) and (20) or Equations
level where racheting fatigue is not a concern. (22) and (23) will suffice.
Recent tests (Reference 27) indicate that the failure
mode of piping components subjected to very large For piping which is in direct contact with the soil, the
seismic loading is not "collapse" under "sustained inertial stresses are not a significant contributing
primary loading". Rather, failure of piping factor to the total stress. Therefore, Equation (1 9) is
components is due to fatigue racheting. Fatigue not appropriate. Instead, Equation (20) or (22) and
racheting will not occur to any significant level if the (23) should be used for the combined effect of
total accumulated strain for seismic loading is within thermal and soil constrained seismic loads.
5% (Reference 27).
Finally, when the strain limit is used for the outer
A set of new design critcria has bcen proposed piping for seismic anchor movemcnts due to ground
rccently [Reference 271 to increase the membrane settlement, the total strain should be limited to 4%, in
plus bending stress for Equation (19) from 3 s h to 6 addition of meeting the requirement of Equation (23).
Sh for reversing dynamic loads in class 2 and 3
piping to be consistent with the 5% strain level, and
a
As shown in Figure 2, the buried piping consists of . analyzing the inner pipe independent of the
a inner pipe which is used to convey hazardous outer pipe by treating the spacers as supports
fluids. The outer pipe is intended to contain the to ground and applying the 5% damped
hazardous fluid in the event of leakage of the inner ground response spectrum, or
pipe.
modelling the inner and outer pipes in one
The inner pipe is supported off the outer pipe through single model.
cross bars (which provide vertical downward
restraint) and spacers (for vertical and lateral The approach taken herein is to use the 5% damped
restraints). ground response spectnun of Reg. Guide 1.60
(Reference 33).
The outer pipe is supported from piers on concrete
pedestals. The piers are spaced to support the Two seismic anchor movements are considered in this
combined weight of the inner and outer pipe, prior to example. The first is the f 2" differential oscillatory
backfill. with negligible deadweight stress and with movement between the tanks and the adjacent
no sag in the pipe. supports. The second is the gradual settlement of the
soil which results in differential movement between
The piping model consists of both inner and outer tanks and the adjacent outer pipe supports. The
pipes. The inner pipe model is linked to the outer absolute settlement is 6" down at the tanks and 4"
pipe model with constraints to reflect the design down along the buried pipe. The differential
configurations and dimensions of the spacers in all settlement is therefore 2" down at the tanks relative
three directions. to the buried pipe.
Both inner and outer pipe anchors are connected to The soil springs are spaced in two zones as shown in
the soil in a11 6 degrees-of-freedom. The outer pipe Fig. 3. Zone 1 is determined based on L, calculated
is supported vertically down and laterally to the from Equation (14) which is approximately 10 ft.
ground, in addition to the restraining action provided Zone 1 is within L, from each side of elbows or tees.
by the soil. Soil springs are spaced at 1.5 times to 3 times the
pipe diameter.
Soil stiffnesscs along all thrce directions have been
determined to be 400 lb/in/in. Zone 2 is determined from the pipe slippage length
required to accommodate the soil strain which is
The PGV is 15 in/sec. Therefore, the maximum soil calculated to be approximately 100 ft. Therefore,
strain calculated from the horizontal shear wave Zone 2 includes the length beyond L, within each
velocity (e.g., 2000 fps) and the vertical compressive straight section. The soil spacing is approximately 20
wave velocity (e.g., 2800 fps) is 0.05%. times the pipe diameter.
The thcorctical maximum axial strain in the outer Table 8 shows the maximum stresses calculated for
pipc due to soil-pipe axial force resulting from wave all load cases. As expected, the highest stress for the
passage (from Equation (8)), is 0.1%. The maximum inner pipe is the thermal expansion stress. As for the
strain in the pipe can not exceed the maximum strain outer pipe, the oscillatory SAM has the highest stress.
in the soil. Thus, the 0.03% maximum soil strain However, the settlement due to S A M and the wave
governs. This value is then used to determine the passage effects are also significant.
equivalent temperature increase for the outer pipe and
applied uniformly to the outer pipe to simulate the Table 9 compares the combined stresses and
seismic wave strain imposed by the soil. maximum strains with the allowable limits established
in this paper. The results indicate that the calculated
In addition to the wave passage effect, the inertial stresses and strains are well within the limits
provided. Thercforc, thc pipe design is acceptable. critcria development and are sincerely acknowledged.
7. CONCLUSIONS 9. REFERENCES
The design and evaluation guidelines presented in this 1. Hetenyi. M., Beams on Elastic Foundation,
paper follow the generally accepted beam-on-elastic- University of Michigan, Ann Arbor, 1967.
foundation analysis principle and the inertial response
spectrum calculation method, respectfully, for piping 2. US Department of Energy, BNL 52361,
directly in contact and not in contact with the soil. Seismic Design and Evaluation Guidelines
A standard analysis procedure is described along with for the Department of Energy High-Level
the discussion of significant factors which are deemed Waste Storage Tanks and Appurtenances,
to be essential for the acceptable design of the Chapter 7, Buried Piping and Conduits,
underground piping. Draft, Jan. 3, 1995.
The following key considerations have been 3. Regulatory Guide 1.132, "Site Investigations
prescribed: for Foundations of Nuclear Power Plants."
the design feature and safety requirements 4. Regulatory Guide 1.138, "Laboratory
for the inner and outer piping, Investigations of Soils for Engineering
Analysis and Design of Nuclear Power
the effect of soil strain and wave passage, Plants."
assimilation of the necessary seismic and 5. Vesic, A.B., "Beams on Elastic Subgrade
soil data, and the Winkler's Hypothesis," Proceedings,
Fifth International Conference on Soil
inertial response calculation €or the inner Mechanics and Foundation Engineering,Vol.
piping, 1, 1961, pp. 845-850.
14. E.C. Goodling, "Flexibility Analysis of 25. ASME Code Committee for Pressure Piping,
Buried Piping", Joint ASMEICSME Pressure B31.3, 1987 Edition.
Vessels and Piping Conference, Montreal,
June, 1978. 26. ASME B3 1.1-1992 Edition, Appendix VII,
Nonmandatory Procedures for the Design of
15. -
E.C. Goodling, "Buried Piping An Analysis Restrained Underground Piping, 1992.
Procedure Update', International Symposium
on Lifeline Earthquake Engineering, Fourth 27. Welding Research Council, WRC Bulletin
U.S. National Conference on Pressure 379, Alternative Methods for Seismic
Vessels and Piping Technology, ASME, Analysis of Piping Systems, February, 1993.
Portland OR, 1983.
28. G. Stawniczy, W. R. Bak, and G. Hau,
16. ASCE, "Seismic Response of Buried Pipes "PipingStress-Strain Correlation for Seismic
and Structural Components", ASCE Loading," Journal of Pressure Vessel
Structural Division Committee on Nuclear Technology, Vol. 110, November, 1988.
Structures and Materials, New York, N. Y.,
1983 29. ASME Boiler and Pressure Vessel Code,
Code Case N-47-26, Class 1 Components in
17. Goodling, E.C., "Restrained Underground Elevated Temperature Service, 1992 Edition.
Piping-Some Practical Aspects of Analysis
and Design," Third U. S. Confcrence on 30. Ferritto, J.M., "Evaluation of Earthquake
Lifeline Earthquake Engineering, ASCE, Los Induced Ground Failure", Technical Report
Angeies, Aug. 22-24, 1991. of Subcommittee 7, Interagency Committee
on Seismic Safety in Construction, U.S.
18. Lin, C. W., "Seismic Analysis of Buried Department of Interior, Geological Survey,
Piping", Transactions SMIRT 11, Vol. K, 1982.
Page 429, Tokyo, Japan, August, 1991.
31. Seed, H.B. and Silver, M.L., "Settlements of
19. Lin, C.W., "Criteria for Scismic Analysis of Dry Sands During Earthquakes", Journal
Buried Piping". ASME PVP-Vol. 237-1, pp. Geotechnical Division, ASCE, April, 1972.
171-171, 1992.
32. Silver, M.L. and Seed, H.B., "Volume
20. A X E , Guidclincs for the Seismic Design of Changes in Sand During Cyclic Loading",
Oil and Gas Piping Systems, Committee on Journal SMFE Div., ASCE, Sept. 1971.
Gas and Liquid Fuel Lifelines of the ASCE
Technical Council on Lifeline Earthquake 33. U.S. NRC, Regulatory Guide 1.60, Design
Engineering, 1984. Response Spectra for Seismic Design of
Nuclear Power Plants, Rev. 1, December,
21. Winterkorn, H. F. and Fang, H. Y., 1973.
"Foundation Enginccring Handbook", Van
Transverse
Horizontal
Displacement
Axial wo
X Displacement
Figure 1 Soil Reaction to specific Ground Displacement Figure 2 Layout of A Typical Buried Transfer Line (Inner and
Outer Pipe are Shoun Side-By-Side far Clarity]
- Und~rgroundPiping
-
rt- Above Ground Piping
Not in Contact with SoiI Soil B u r i d
Soil Supportbd
Soil Friction
constrained
SoiI Strains
controlkd
Loose 3- 6 2 0 - 60
MediUrn 7 - I5 60- 300
Dense 16 - 30 300- loo0
PD/4t
0.8 + 2-4 + 5.3 = 8 5 < min (3 x 15.7,2 x 25.0) inner pipe
1.2 + 3.0 + 0.0 4.2 < min (3 x 15.0, 2 x 35.0)
E. Outer pipe
[(b
+ Fo )+ *]]/A + i [(Mwi M q i M T ] 5
~ min (3S,2Sy) (7.15) (7.16a)
0.0 + 0.0 + 0.0 + 1.7 + 5.0 i 12.8 = 19.5 <min (3x157, 2i25.0) inner pipe
0.5 + 0.0 + 0.0 + 7.7 + 14.4+. 0.4 = 23.0 < min (3 x 15.0, 2 x 35.0) outer pipe
(8+ e)
/A + i (Ms + M T ) I
~ min (3S,2Sy) (7.15) (7.16b)
0.0 + 0.0 + 3.9 + 12.8 = 16.7 < min (3 x 15.7, 2 x 25.0) inner pipe
0.0 + 0.0 + 6.7 + 14.4 = 21.1 <min (3 x 15.0,2 x 35.0) outer pipe