STRUCTURAL DESIGN CONSIDERATIONS ‘Sewers and culverts must have adequate structural strength to withstand external loading from construction equipment, earth backfill andtratfic, and must main- tain structural integrity for the de- signservicellife, The load carrying capability of any pipe is depen- dent upon the inherent structural strength of the pipe and the sup- port provided by the surrounding Liberal policies for corrugated steel pipe are sometimes promot- ed which are not equalto the gen- erally morestringent requirements for concrete pipe. When the vari- | ous design methods and criteria developed for the different ma- terials available for sewers and culverts are considered, policy decision procedures can become exceedingly complex. Satisfactory design methods must be based on theories, model and full-scale testing, and, most importantly, experience. Appro- priate criteria involves relating requirements for usage and per- formance with experience. Pipe policies generally have certain political and economic overtones, but they should also be based on experience. ‘The proper design of sewers and culverts requires considera- tion of the different, but interre~ lated, fields of hydrology, hydrau- lics, structural behavior, durability and economics, and construction procedures. Knowledge of the performance of a pipe material in each of these fieldsis essential for completeevaluation and compari- son. This Buried Fact reviews the structural behavior of flexible cor- rugated steel pipe and the devel- ‘opmentof design criteriaand pro- cedures, soil. Conerete pipe can be designed with the strength necessary to carry practically any load, and de- signs are basedon proven anduni- versally accepted principles of engineering mechanics. Concrete pipe can be tested, and its struc tural strength proven before pur- ‘chase and installation, Corrugated steel pipe is a thin, flexible structure. It has virtually 1 inherent strength to resist ex- ternal loads and handling stresses. Its supporting strength depends ‘on more stringent and difficult re- quirements for foundation prepa ration, bedding preparation, pipe handling and placement, backfill material and fill procedures. Val- es for all of these requirements must be assumed during the de~ sign phase, andstructural distress or failure may result when adesign value is not achieved during in- stallation. Because of flexibility and lack of inherent structural strength, corrugated steel pipe failures occur as aresult of detlec- tion, buckling of the pipe wall, splitting of the pipe wall seams, damage during installation, and other installation problems includ- ing flotation due to light weight. This light weight and lack of inhor- entstructural strength are directly related to the very thin wall thick ness of corrugated steel pipe. For most pipe sizes, more stee! is used in reinforced concrete fy pipe than is used for corrugated steel pipe. For example, as shown in Table 1,a60-inch diameter rein- Table 1. Comparison—Conerete Pipe vs. CSP. PIPE DiaMeTER, INCHES. INCHES. WALL THICKNESS, STEEL USAGE, POUNDS PER LINEAR FOOT [concaeTe Pire| SS CONCRETE PIPE | CSP 38 48 60 R 84 96 4 0.064 0.064 0.078 0.109 0.108 0.078 36 67 108 128 18 221 1. Coneretepipe—Wal 8, Ohio DOT Design Tables 2 €8P=3 7 corugstionsforced concrete pipe has 61 per ameter, 14-gage corrugated steel “pipe. In-addition, tire steel iri con- crete pipe is covéred with denge, protective concrete, DESIGN A flexible corrugated steel pipe has virtually no inherent flexural strength, and, in the buried condi- loads must bederived from active lateral pressures and passive lat- eral pressures of the soil Induced .. aS the pipe deflects andits sides move outviard. Sincethe pipe sup- porting strength depends on the sidefili material, its structural be- _Havjor ust be analyzed cansid- ering soil-structyre interaction ‘which is directly rélated to detlec- jon changes oT the ploe. Any ra- ‘tional strubtural design procedure {or bufied flexible pipe shoul clude amethod for predicting de- figction under specific installation conditions so as to prevent buck- ling and inverse curvature of the pipe. Inthe design of structural mem= ‘bers, the strain or deformation of ‘an element can. be ‘determined from the ratio of the load or stress. ‘on the member to its modulus of elasticity (strain = stress + médu- _lus of elasticity). The modulus of “elasticity for the material is either _known oF it can, be determined from laboratory tests. ‘The deflection 6fabuiried circu- lar pipecan be predicted in asimi- lar fashion. The cross-sectional ring deflects (detorms) according to the ratio of the load on the ring to the modulus of elasticity of the material, In this instance, the ma- terial modulusis acombination of ofthe pipe andthe modulus (stiff- ness) of the soll, so that: pipe deflection = load on pipe aa pipe stifiness * soil stiffness. tion, its ability to support vertical , This.is the form of the lowa = modulus of passive resis- cent more stee! than a 60-incl’di- ~ Formula, widely used for predict- tance, pounds per square ing deflections of buried.coru- inch per inch gated steel pipe. © aa Professor Spangler developed the formula from experimental = - stallations of buried flexible cor- IOWA FORMULA rugated steel pipe. A total of ten Professor M. G. Spangler, lowa pipe were tested with diameters: State University, published the eter eee eon lowa Formula i194; gg, under uniform height of fll of D/KWR? “Sect. = RET - + "= Spangler hypothesized that Et Q081eR! the lateral movement of various points on the pipe ring were known, the ditrioution of lateral AX = horizontal deflection ofthe pressures could be determined = pipe, iiches by multiplying the movement of Dy = Gelection lag factor any point by the modulus of pas- i= bedding constantwhichvar-_slveresistanes, e. For mathemati ieswiththeangieotitiebed- cal convenience, this lateral pres- wher ding sure was assumed to be a simple ‘Weearth load on the’pipe, parabolic curve embracing only ounds per linear inch ‘the middle 100 degree arc of the i= mean pipe radius, inches _—_pipe, Figure 1. He also assumed E=modulus of elasticity ofthe _thetotal vertical load was uniform- pipe material, pounds per ly distributed across the width of square inch the pipe, and the bottom vertical moment of inertia of the reaction, equal totheverticaload, pipe cross-section, inches was disiributeduniformly over the to the fourth power width of the pipe bedding sieThe Marston load theory was used to evaluate the total vertical load, W, on the pipe: .W = OwBt where: =~ C= load coetticient w=weight, of backfill material, pounds per cubic foot B. = outside width of pipe, feet The [oad coefficient varies with the type of backtill material and depth of burial, and, in the em- bankment condition, is also af- fected by the projection ratio and the settlement ratio. The settle- ment ratio is a rational concept, but impossible to evaluate in ad- vanoeof construction, and, there- fore, is considered asemi-empiri- cal constant. Recommended val- ues for the settlement ratio were determined from avery small num- ber of flexible pipe installations. In 1956, after discovering that the modulus of passive resistance could not possibly be a property ‘of soil because its dimensions ‘were not those of a true modulus, Spengler-and Dr. R. K, Watkins proposed a modulus of soil reac-- tion, E’, defined ass LOAD FACTOR The load fattor incorporates ‘the parameters that have to do with the magnitude and distribu- tion of thesoll pressures on a bur- ied pipe. The pipe deflection is directly proportional to the load factor and, yet, less is known about its components then any others in the lowa Formula, Changes in construction proce- dures or bedding could vary the actual load factor more than 100 percent from the load factor cal- culated in the design phase. Deflection Lag Factor Spangler originally stated, "The deflection lag factors observed in the experiments range from 1.38 to 1.46, and in no instance was equilibrium completely attained. Therefore, 1.5 Is suggested as a conservative value for design use for standard corrugated-pipe cul- verts installed without strutting or predeforming,” and after further investigation, “The deftection lag factor cannot be less than unity and Has been observed to range upward toward a value of 20. A normal range of values from 7.25, to 1.50 is’suggested for design purposes.” The U. S. Bureau of Reclamation; however, after re- view of many projects, concluded that,.“The actual value, however, depends on when the immediate deflection is measured, the vo ume change rate of the soil, and the load onthesoil. Dyis basically an empirical factor and ranges from 1 to 6 in observed tests.” - Er SeR Hae The modified lowa Formula, as used today, is therefore: ~ __" DrKWR® AXESOOIE if the lowa Formula is rear- ranged as: oyKw) EVR) + (0.081 E} The following terms can be troduced to_desoribe the three Ax= Bedding Constant The bedding constant, K, is dik mensionless, and varies with the bedding angle, « (Figure 1), as” presented in Table2. The angle of bedding describes theIcad resist ing area of the bedding under the pipe. As the angle of Dedding in- creases, the loaded areainreases and the pipe deflects less, No fur- ther study has been déne on this constant since its conceptién in 1941, even though it ean influ- ence the deflection predicted by ‘the lowa Formula by as much as 25 percent. Earth Load ‘The Marston theory is the most common method of calculating the earth load, W, on the pipe and is recommended by Spangler for the lowa Formula. In the Marston theory, the load depends on whether the pipe is in atrench or embankment (or combination), the type of backrilisoil, thesettle- ment of the pipe in relation to the backfill material, and the distance that the pipe projects into the nat- ural soll foundation. Virtually all corrugated steel pipe research and experience has been on high ‘way culverts constructed as posi tive projection embankment stallations, where good side fill ‘compaction is relatively easy to achieve. Conversely, there is a lack of research and experience on corrugated stée! pipe instatled in trench conditions. Table 2. Bedding Angle—Constant Relaticnships. _ separate factors that affect the pipe deflection: D:KW=loadfactor = = BeovING| EV/R? = ring stiffness factor 7 0.061 = soil stiffness factor ae Andthe modified lowa Formula, represented as: = Load Factor