过渡词SIDD pipe bedding and Ontario provincial
standards
Zhao, J.Q.; Daigle, L.
NRCC-44694
A version of this paper is published in / Une version de ce document trouve dans: Proceedings of the International Conference on Underground Infrastructure Rearch, Kitchener, Ontario, June 11-13, 2001, pp. 143-152
1 INRODUCTION
Designs and installations of drainage pipes in On-tario are provided in various Ontario Provincial Standards Specifications (OPSS) and Drawings (OPSD). Specifications for rigid circular concrete pipes are given in OPSS 410, 421, 501, 514 and 1820; OPSD 802.030 to 802.034; OPSD 807.01, 807.03 and 807.04. The specifications (herein re-ferred to as “the OPSS”) reprent the best practice and have been widely and successfully ud in On-tario, Canada.
The OPSS on rigid pipes are bad on the tradi-tional indirect method. In this method, designs of rigid pipe rely on the Marston-Spangler theory for earth loads in trench and embankment installations (ACPA 1992; OCPA 1986). Recently, analys of pipe-soil interactions using the finite element method have resulted in the development of a direct design method. This method, commonly referred to as the Standard Installation Direct Design (SIDD) standard (ASCE/ANSI 1993), provides a different earth pressure distribution around the pipe for each of the four standard installation types. Becau of the due consideration of lateral earth support and re-duced reaction at the pipe invert, the SIDD method can provide significant savings in the pipe design and its installation. The SIDD method is not cur-rently included in the OPSS.
This paper prents comparisons between the OPSS and SIDD method in terms of technical de-velopment, construction specifications and costs. The results of a short survey on the SIDD method are also discusd.
2 DEVELOPMENT OF ONTARIO PROVINCIAL
STANDARDS
The traditional Marston-Spangler method is referred to as the “indirect method” (ACPA 1992; OCPA 1986) becau the strength of a pipe is determined by 3-edge bearing tests, and then related to the field strength by bedding factors (Marston and Anderson 1913; Heger et al. 1985; McGrath 1993). Bad on this method, four types of bedding for concrete pipe have been developed: types ‘A’, ‘B’, ‘C’ and ‘D’ (ACPA 1992).
convenient
In the early 1900’s, Marston and his co-workers developed the formulas for soil loads and superim-pod loads on buried rigid pipe installed in both trench and embankment conditions (Marston and Anderson 1913; Marston 1930). This work was bad on the silo theory for grain loads on silo walls and floors developed by Jansn in 1895 (Kellogg 1993; Jeyapalan and Hamida 1988). Marston’s trench theory was further expanded by Spangler and his co-workers (Spangler and Handy 1973). Th
ere-fore, the trench theory is commonly referred to as the Marston-Spangler method (Selig and Packard 1986).
佛学入门SIDD pipe bedding and Ontario Provincial Standards
高蛋白质食物
Jack Q. Zhao & Lyne Daigle
Institute for Rearch in Construction, National Rearch Council Canada, Ottawa, Canada
ABSTRACT: The current OPSS and OPSD that provide specifications for drainage pipes in Ontario, are shown to be a simplified and more conrvative version of the traditional Marston-Spangler method. The SIDD method was developed through extensive finite element analys. It improves the pipe installation prac-tice by lesning the requirements in bedding and compaction, and allowing the u of native backfill materi-als. There is potential cost-savings in using the SIDD method. The SIDD standard adopted by ASCE/ANSI, however, ems to have misd a few important pieces of design information and is inconsistent with the original SIDD rearch publication in the definition of soil types. The paper shows that an improved version of the SIDD method should be considered for adoption as an alternative in the OPSS, while the traditional Marston-Spangler method is maintained in the standards.
2.1 Loads on buried pipe
Design loads for rigid pipe have traditionally been determined using the Marston load theory (Marston 1930):
(1)
where W = backfill load per unit length; C = load factor; γ = unit weight of backfill material; and B = trench width, B d, at top of pipe for trench condition or pipe outside width, B c, for embankment condition. The formulas for calculating load factor, C , vary for different types of installation (Marston 1930).
2.2 Bedding factors In the Marston-Spangler method, bedding factors are ud to relate the calculated external loads to the 3-edge-bearing strength of pipe:
(2)
where S eb = 3-edge bearing strength; W = calculated
external load; and B f , = bedding factor.
Bedding factor, B f , is dependent on bedding an-gle, quality of contact between the bedding and the
pipe, the supporting lateral pressure on the pipe, and
the area over which this lateral pressure acts (Selig垃圾分类画
and Packard 1986). The higher the bedding factor,
the better the quality of the bedding. In general,
there are four class of bedding as shown in Table 1
for trench conditions and in Table 2 for embankment
conditions (ACPA 1988; Spangler and Handy 1973).
Shaped bedding is one of the options for class B
and C for both trench and embankment conditions.
Values of corresponding bedding factors given by
the OPSS are included in the tables for comparison.
Table 1. Bedding class and bedding factors, trench condition
Bedding factor, B f
Bedding class † ACPA (1988) OPSD 807.01 (depends on reinf. ratio) B 1.9 1.9
C 1.5 1.5
D 1.1 not available
† bedding class A= either a concrete cradle or arch, reinforced
or plain; B = either a shaped subgrade with granular founda-tion, or a granular foundation to springline; C = either a shaped subgrade, or a granular foundation to 1/6 of outside diameter; and D = Flat subgrade.
For trench conditions, the OPSS bedding factors
are the same as the industry standards for class B
and C. For class A, the low value that corresponds
to plain concrete cradle or arch is ud by the OPSS,
which yields more conrvative designs with this class.
Table 2. Bedding class and bedding factors, embankment condition
Bedding factor, B f
OPSD 807.03† Bedding class ACPA (1988) jection projection
B 1.9 2.02
C 1.5 1.70
D
B f
(Equations 3
and 4)
not available not available † k µ = 0.19, r sd
For embankment conditions, the OPSS u the conrvative but easy-to-u values. The ACPA manual gives a general mi-empirical equation for positive projection embankment condition:
(3) (4) where A = a constant corresponding to pipe shape and for circular pipe A =1.431; N = a parameter which is a function of the distribution of the vertical load and horizontal reaction; x = a parameter which is a function of the area of the vertical projection of the pipe over which active lateral pressure is effec-tive ; q = the ratio of the total lateral pressure to the total vertical load; k = the ratio of the unit lateral soil pressure to unit vertical soil pressure (Rankine’s co-efficient of active earth pressure and k = 0.33 will usually be sufficiently accurate); C c = the load coef-ficient for positive projection pipe; H = cover depth; and B c = pipe outside diameter. As shown in Equations 3 and 4, bedding factors for embankment conditions are not readily available since they not only depend on bedding class, but also depend on the ratio of cover depth to pipe out-side diameter, embankment type and the effective pipe area for active lateral pressure. Using the same conditions as the positive projection ca of the
燃气灶开关
OPSS, the bedding factors are calculated using
Equations 3 and 4, and are plotted in Figure 1 as a function of the ratio of cover depth to pipe outside diameter.
f eb B W S =2
B
C W γ=xq N A B f −=÷÷øöççèæ+=2m B H C mk q c c
Figure 1. Variation of bedding factor with ratio of cover to pipe outside diameter
In general, bedding factors for all four cas de-crea as the ratio of cover to pipe outside diameter (H/B c ) increas until it reaches about 20. For H/B c higher than 20, the bedding factors remain constant. The constant bedding factors for positive projec-tion conditions are the same as for zero projection conditions where lateral earth supports are ne-glected. Table 2 and Figure 1 show that the bedding factors for zero projection conditions in the OPSS agree with tho given by the equations. The values of the bedding factors for the positive projection in the OPSS, however, are less than tho constant val-ues shown in Figure 1. Nevertheless, tho OPSS bedding factors are quite cons
ervative becau the ratio of H/B c ldom exceeds 20 for most applica-tions.
2.3 Discussions of Ontario Provincial Standards Values of bedding factors given in the OPSS are the same as tho given in Spangler and Handy (1973) and the ACPA manual (1988). However, the OPSS have the following distinctive features:
• Bedding class ‘D’ is not included, indicating its u is discouraged.
• Soil types are defined in the Occupational Health and Safety Act (OHSA 1978), which classifies soils into four types, mainly for shoring purpos (health and safety concern).
• Trench definition is also as per OHSA (1978). • Compaction effort is more precily defined than ACPA (1992). It is defined in terms of maxi-mum dry density.
• Bedding factors are simpler and readily available for embankment conditions.
• Backfill materials are classified into five granu-lar types, namely OPSS granular ‘A’, granular ‘B’ type I, granular ‘B’ type II, granular ‘M’ and lected subgrade material. Engineering proper-ties and gradations of the granular materials are defined.
trench bedding factors.
While most of the unique features listed above are improvements over the industrial standards such as ACPA (1988), others require modifications to re-flect the state-of-the-art technologies. For instance, the OPSS refer to the OHSA trench and soil classifi-cations, which are meant for preventing occupa-tional hazards during constructions and protecting individuals working in and around the trenches. They are not meant for pipe design purpos.
It is worthwhile to note that the inherent differ-ence between rigid and flexible pipe is not reflected in the OPSS – they are treated in a similar way, al-though OPSS 421 provides definitions for rigid and fl
exible pipe in terms of pipe behavior. Recognizing the different load-carrying mechanisms of rigid and flexible pipe and identifying the means to achieve optimal design will result in cost-effective and dura-ble installations.
The above review shows that the OPSS are in-deed bad on the Marston-Spangler indirect method, but with simplifications. Furthermore, trenches and in-situ soil types are not defined in terms of the current engineering practice although backfill materials are well defined. 3 DEVELOPMENT OF STANDARD INSTALLATION DIRECT DESIGN
The SIDD method was introduced by the American Concrete Pipe Association (ACPA 1993) and was adopted as an ASCE Standard (ASCE/ANSI 1993). In this method, soil pressure distributions around a pipe are defined according to four standard types of installation. Vertical and horizontal arching factors are introduced to determine total vertical and hori-zontal loads. No bedding factors are ud; instead, pipe wall design is bad on the moments, thrusts and shears in the pipe wall due to the external and internal loads. This approach is referred to as the “direct design” method. The development of this method is reviewed in this ction to provide back-ground information for comparing it with the indi-rect method reviewed in the previous ction.
01
2
3
45
5
10
1520
25
30
Value of H/B c
B e d d i n g f a c t o r , B f
3.1 Analysis of pipe-soil interaction
In order to understand fully the interaction between the buried concrete pipe and its surrounding soils, the American Concrete Pipe Association sponsored an extensive multi-year rearch project carried out by Simpson Gumpertz & Heger Inc. (Heger et al. 1985). As a result, a Soil-Pipe Interaction Design and Analysis program (SPIDA) was developed. The analysis part of the software us the finite element method (FEM) and treats the pipe-soil system as a two-dimensional plane strain problem. The design part of the software is bad on a strength and crack control design procedure that was specifically de-veloped for circular reinforced concrete pipe (Heger 1982; ACPA 1993). Using SPIDA, a wide range of soil types and densities were studied and the results
were ud to develop the SIDD bedding standards (McGrath 1993).
Assuming symmetry about the vertical plane and uniform condition along the pipeline, SPIDA us one half of the installation in the FEM model and takes a unit slice of the installation that is in a state of plane strain. This numerical reprentation of the pipe installation is a common approach in studying the behaviour of the pipe-soil system. The features and limitations of SPIDA are summarized by Heger et al. (1985).
3.2 Standard installation direct design method Bad on a large number of SPIDA simulation re-sults
covering a wide variety of conditions, four standard types of standard installations, each with a unique pressure distribution, were introduced by ACPA (1993) and adopted by ASCE (ASCE/ANSI 1993) and by AASHTO (1996). It ems, however, to be a step backward becau the design practice has gone from a four-class bedding approach to FEM, and now back to another four-type installation approach. In this regard, Kellogg (1993) states: “With the popularity of numerical methods of analysis, classical approaches to the problems have been waning. However, in the everyday practice of engineering, even in the computer age, clod-form solutions bad upon classical me-chanics and solutions to boundary-value differen-tial equations are still needed. The time and cost to prepare sophisticated numerical models for analysis simply cannot be justified in the market-place, except for the most complex and critical civil engineering structures.”
Trench and embankment conditions and the terms ud in the SIDD standard installations are shown in Figures 2 and 3. Figure 2. Trench configurations ud in SIDD installations (ACPA 1993).
Figure 3. Embankment configurations ud in SIDD installa-tions (ACPA 1993).
Table 3 shows the four standard types of SIDD standard installation (ASCE/ANSI 1993) and soil types as defined in ASTM D 2487 (1993). Degree of
compaction is in terms of Standard Proctor Density.
except for Type 4
material and
compaction each side,
same requirements
as haunch
material and
compaction each side,
same requirements
as haunch
Table 3. SIDD Standard installation types† (ASCE/ANSI 1993)
Degree of compaction
Instal-lation type
Bedding thickness
(t b)
Haunch
and outer
bedding
Lower side
b0
and t b 75 mm.
For rock foundation:
t b D0/12,
and t b 150 mm
95% ML or
100% CL
2 t b D0/24,
and t b 75 mm.
For rock foundation:
t b D0/12,
and t b 150 mm 90% SW or
95% ML
85% SW,
90% ML or
95% CL
3 t b D0/24,
and t b 75 mm.
For rock foundation:
t b D0/12,
and t b 150 mm 85% SW,
90% ML or
95% CL
85% SW,
90% ML or
95% CL
4 No bedding re-
quired,
For rock foundation:
t b D0/12,
and t b 150 mm No com-
paction re-
quired, ex-
cept if CL,
u 85%
CL
No compac-
tion re-
quired, ex-
cept if CL,
u 85% CL
† D0 = outside diameter of the pipe. Soil symbols as per Uni-
fied Soil Classification System (USCS) (ASTM D 2487 1993).
Of the four standard installation types, type 4 is of the lowest quality that requires no compaction (except for CL backfill soils), whereas type 1 is of the highest quality that requires well graded materi-als compacted to a minimum of 95% of the Maxi-mum Standard Proctor density. Special care
is re-quired for the middle bedding (directly beneath pipe invert) where the soil needs to be kept loo in order to transfer the loads to the haunch area, thereby re-ducing stress concentration at the invert of the pipe (ASCE/ANSI 1993; Meyer et al. 1993). Excessive compaction of the middle bedding, and insufficient compaction of the haunch area can result in pipe cracking (Wilson 1985).
The ASCE/ANSI Standard (1993) provides the pressure distribution diagram, which is called Heger pressure distribution, and its coefficients (Figure 4); however, it falls short by not providing formulas and coefficients to determine maximum thrusts, bending moments and shears for each of the four standard in-stallations (ASCE/ANSI 1993). The formulas and coefficients are provided in ACPA (1993) so that designers can u them to calculate design forces. Simpson Gumpertz & Heger Inc. provides SPIDA analysis rvice at a cost of US$15 to US$75 per de-sign (Heger et al. 1985). Figure 4. Heger pressure distribution diagram and coefficients (ACPA 1993).
3.3 Field performance verification
Sargand et al. (1994) independently instrumented a 600 mm diameter concrete pipe installed using the SIDD Type 3 installation in laboratory. They pointed out that to that date only a few installations had been instrumented and monitored for perform-ance verification of pipe-soil systems that were lai
d according to the SIDD standards. Their monitored results showed that the SIDD method is good at pre-dicting moments before the formation of cracks, but not so good after the formation of cracks in concrete pipe. Furthermore, measured thrusts were consistent with the predictions of the SIDD only for low sur-face pressures. Two reasons were given for the mismatch between the experimental and theoretical results:
• The SIDD FEM model is not capable of simulat-ing the behavior of a cracked pipe ction ade-
quately. Although Heger et al. (1985) state that
SPIDA can model uncracked and cracked pipe
ction by using different stiffness for the
beam elements that reprent the pipe ction,
the work by Sargand et al. (1994) showed that
SPIDA needs improvement in modeling the pipe
ction so that formation of cracks can be ade-
quately accounted for.
• SIDD assumes symmetry in geometry and load-ing, which is a common assumption in order to
VAF
Installation
Type
1
2
3
4
VAF
1.35
1.40
1.40
1.45
HAF
0.45
0.40
0.37
0.30
A1
0.62
0.85
1.05
1.45
A2
0.73
0.55
圣诞礼物英语
0.35
0.00
A3
1.35
1.40
1.40
1.45
A4
0.19
0.15
0.10
0.00
A5
0.08
0.08
0.10
0.11
A6
0.18
0.17
0.17
0.19
a
1.40
1.45
1.45
1.45
b
0.40
0.40
0.36
0.30
c
0.18
0.19
0.20
0.25
e
0.08
0.10
0.12
0.00
f
0.05
0.05
0.05
-
u
0.80
0.82
0.85
0.90
部门职责模板v
0.80
0.70
0.60
-
