Blast-Resistant Highway Bridges: Design and Detailing Guidelines (2010)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

30 3.1 Overview The experimental test programs for Phases I and II are presented in this chapter. Phase I included small-scale non- responding columns tested in Vicksburg, MS, by the U.S. Army Corps of Engineers. Phase II involved designing and testing half-scale reinforced concrete bridge columns. All of the instrumentation, construction, and testing for Phase II were completed at the Southwest Research Institute (SwRI) test site in Yancey, TX. 3.2 Phase I Load Characterization Study The test program for small-scale non-responding columns tested during Phase I of this experimental research program, the data acquisition and instrumentation plan, and the exper- imental test setup are presented in this section. All instrumen- tation, construction, and testing were completed at the Big Black Test Site (BBTS) at the Engineering Research and Devel- opment Center (ERDC) of the U.S. Army Corps of Engineers in Vicksburg, MS. The Phase I test program included eight small-scale blast tests at four sets of standoff. The objective of the small-scale tests was to characterize the structural loads on square and round bridge columns due to blast pressures. The experi- mental observations indicate how cross-sectional shape, standoff, and geometry between the charge and column po- sitions influence blast pressures on the front, side, and back faces of bridge columns. The data gathered from these tests also allow the assessment of classical methods used to pre- dict blast loads and a proposed method to modify pressures on the front face of a square column to obtain structural loads on a circular column. Although the blast loads acting on large flat panels such as walls and slabs are well charac- terized from past testing performed by the military and others, little is known about how blast pressures interact and engulf slender structural components such as bridge columns. 3.2.1 Experimental Setup Classical scaling laws were used to design model tests that met the research objectives and remained within practical ex- perimental limits of precision and accuracy. Phase I blast tests consisted of charges detonated on the ground surface to gen- erate a regular Mach front air blast load on the columns. A scaled standoff of three was selected for each test to minimize the fireball and detonation products loading on the non- hardened pressure gauges, while a scale factor of eight al- lowed the explosive charge to be within the BBTS air burst safety and noise limits. The three main test variables included the charge weight, standoff, and cross-sectional shape. Thus, while the scaled standoff was held constant at three, the ac- tual charge weight and physical standoff were varied in the test program. Table 4 lists the dimensions of each small-scale column. The test setup for the Phase I non-responding columns is illustrated schematically in Figure 22. Figure 23 shows the test site prior to one of the detonations. The test setup foun- dation consisted of a 20-ft × 6-ft × 6-in. smooth-surfaced polymer-fiber reinforced concrete slab reinforced with a 2-ft × 2-ft × 1-in. replaceable steel plate at the charge location (mid- way between the columns). The columns were anchored by flange plates cast directly into the slab on either side of the charge. Sensitized nitromethane was used as the high explosive material, and non-fragmenting plastic cylindrical containers with a charge diameter-to-height ratio of 1:1 approximated a spherical detonation. Two free-field pressure gauge mounts were also cast into the slab at each standoff on either side of the charge. PVC pipe cast into and underneath the slab pro- tected the instrument cables from the blast. The test setup allowed both columns to be loaded simultaneously with the C H A P T E R 3 Experimental Program

31 same explosive charge, which increases the accuracy of the comparison between round and square column loadings. 3.2.2 Data Acquisition and Instrumentation Plan Each non-responding column was instrumented with nine structure-mounted pressure gauges or instruments (SMI). Two free-field overpressure gauges were used during each test at the same standoff as the test column, one on either side of the charge. The locations of each gauge are shown in Table 5. The pressure gauge layout for a round and square column are illustrated in Figure 24. ERDC performed an extensive set of pre-test peak pressure analytical predictions to select pressure gauge sizes and ranges. High range Kulite HKS pressure gauges were specified. 3.3 Phase II Response Test The design of reinforced concrete columns tested during Phase II of this experimental research program, the instru- mentation and construction of each test specimen, and the experimental test setup are presented in this section. All of the instrumentation, construction, and testing were com- pleted at the Southwest Research Institute test site in Yancey, Texas. Table 4. Phase I column design. Column Type Cross-SectionalDimension (in.) Wall Thickness (in.) Area (in.2) Zx (in. 3)Ix (in.4) S (in.3) Round Pipe 4.5 4.5 0.674 8.1 15.3 21.7 6.79 9.63 9.97 Structural Tube* 0.500 8.0 12.1 * Shop Fabricated . . . . . . . . . Fiber Reinforced Concrete Round Model Square Model Charge Steel Plate 2'x2'x1" Compacted Sand Sub-grade, Loess Mounting Flange PLAN ELEVATION 20' 6" Free-Field Pressure Gauge 6' 8.25" 24.75" 16.5" Pressure Gauges .. . .. . .. . Figure 22. Phase I test setup schematic. Round Column Square Column Charge Figure 23. Phase I test setup.

32 The test program included ten half-scale, small standoff tests and six half-scale, local damage blast tests, as shown in Table 6. The goal of the small standoff tests was to observe the mode of failure (i.e., flexure or shear) for eight different column designs. In a local damage test, charges were placed very close to or in direct contact with the test column. The objective of the local damage tests was to observe the spall and breach patterns of blast-loaded concrete columns. Experimental observations were used to evaluate the perfor- mance of several design parameters and to determine the Table 5. Phase I pressure gauge locations. Type of Pressure Gauge*† Notes SMI Face-On, 0° 8.25 8.25 16.5 24.75 SMI Side-On, 90° 16.5 24.75 SMI Back-Side, 180° 8.25 16.5 24.75 Free-Field‡ 0 - - Normal toground plane *Pressure gauge layout for one pier †SMI = Structure Mounted Instrument (relative to blast source) ‡Free-field pressure gauge at range of front face of pier Height Above Ground (in.) Normal to Face (a) (b) Gauge 20 : Free -f ield ov erpressure gauge located at standoff equal to that at t he base of the fr ont of co lu mn 1 7 Gage Orientat io n (Plan View ) Si de Gauges Rear Gau ges Front Gauges 4.5" 16.5" 8.25" 33" Gauge Num bers 24.75" Concrete Surface Al l charges by tota l weight ar e NitroMethane (NM) sitting on surfa ce wi th the detonatio n poin t at charge center of m ass. Standoff = λ = 3 MODEL ROUND COLUMN 4 2 8 3 9 5 6 Gauge 20 : Free -f ield ov erpressure gauge located at standoff equal to that at t he base of the fr ont of co lu mn 1 7 Gage Orientat io n (Plan View ) Si de Gauges Rear Gau ges Front Gauges 4.5" 16.5" 8.25" 33" Gauge Num bers 24.75" Concrete Surface Al l charges by tota l weight ar e NitroMethane (NM) sitting on surfa ce wi th the detonatio n point at charge center of m ass. Standoff = λ = 3 MODEL SQUARE COLUMN 4 2 8 3 9 5 6 Figure 24. Phase I pressure gauge layout: a) round column, b) square column.

capacity and failure limit states of concrete highway bridge columns. 3.3.1 Background on Parameters Selected for Testing The test specimens designed for this program incorporated design and detailing standards commonly used in practice in various regions of the country. A survey of several states’ department of transportation (DOT) design and detailing standards is summarized below to provide the reader with a better understanding of each state’s design philosophy. The national DOT design survey and background theory on blast-resistant column design were used to determine the eight different column designs with five main test variables. A sum- mary of each column design is presented below to indicate how each design represents current national standards. 3.3.1.1 State Department of Transportation Design Survey The goal of the state DOT design summary is to illustrate similarities and differences among states’ design and detail- ing standards. Each state’s DOT maintains a bridge and high- way design manual or similar document that provides design standards for that state. The standards pertaining to concrete column design are summarized in Table 7. At a minimum, each state’s standards are designed to meet the AASHTO LRFD Bridge Design Specifications. However, the state standards are commonly more stringent than the AASHTO require- ments due to local design issues, such as earthquakes or highly corrosive environments. Parameters that were not varied in this test program were found to be consistent among state DOT design standards, such as concrete strength, concrete class, concrete clear cover, and reinforcement grade. The longitudinal reinforcement ratio can be calculated according to AASHTO specifications. This value was held constant at 1.0% for the test program, which is typical for current bridge column design. 3.3.1.2 Close-in, Blast-Resistant Column Design and Detailing Building components subjected to blast loads are designed to ensure adequate shear strength so that flexure is the control- ling mode of failure. In flexure, reinforced concrete structural components that are properly detailed possess good ductility, while in shear, failures occur in a brittle manner. Thus, it is de- sirable to have flexure be the controlling mode of failure. A plastic hinge analysis on a blast-loaded component considers all potential plastic hinge locations to ensure the maximum possible shear demand. A basic component design includes a nonlinear SDOF analysis that considers each stage of response. Figure 19 illustrates a beam that goes through three stages of deformation: purely elastic, a combination of elastic and plas- tic, and purely plastic. Typically in blast-resistant design, large inelastic deformations are allowed and play an important role in a component’s dissipation of energy. Due to the large uncer- tainty associated with determining blast loads, an SDOF analy- sis usually provides adequate results for design. 33 Test No. Column Designation Column Design Scaled Standoff (ft/lb1/3) Instrumentation Test Objectives/Notes Small Standoff Tests 1 1A1 gravity R/W 1/3 < 3 2 1A2 gravity R/W 1/3 < 3 3 2A1 gravity R/W 1/3 < 3 4 2A2 gravity R/W 1/3 < 3 5 1B gravity R/W 1/3 < 3 6 2B gravity R/W 1/3 < 3 7 2-seismic seismic R/W 1/3 < 3 8 2-blast blast R/W 1/3 < 3 9 3A gravity R/W 1/3 < 3 10 3-blast blast R/W 1/3 < 3 Local Damage Tests 1 2A2 gravity R/W 1/3 < 1 2 2A1 gravity R/W 1/3 < 1 3 1A1 gravity R/W 1/3 < 1 4 2B gravity R/W 1/3 < 1 5 2-blast blast R/W 1/3 < 1 6 3A gravity R/W 1/3 < 1 3 free-field pressure gages for explosive yield validation, 6 strain gages per specimen, and high speed camera To determine flexural response shapes and shear/flexure interaction for large round & square columns under blast loads, including seismic detailing and sections designed specifically to resist blast High speed camera and post-test measurement only To quantify spall/breach thresholds for round & square columns; to quantify extent of local damage Table 6. Phase II test matrix.

34 While design guidance exists for blast-loaded building columns, there are currently no standards on how to design and detail a bridge column to resist blast loads. Bridge columns behave differently than building columns exposed to blast loads, preventing the direct application of current design guide- lines for buildings. For example, bridge columns are easier to access than most building columns, which drastically changes the blast loads used in design. Typically, building columns are designed for the reactions from blast-loaded exterior walls or beams, while bridge columns are the primary members resist- ing blast loads. In addition, the applied axial load magnitude relative to the axial load capacity is usually much greater in building columns than in bridge columns, which affects the location on the axial force-bending moment interaction diagram and hence the behavior under laterally applied loads. There is little research on the blast-wave propagation around exposed structural members such as slender bridge columns. Therefore, all existing design guidelines need to be revisited for bridges specifically. However, many of the design principles used for blast-resistant building design should apply equally well to bridges. Current design and detailing requirements for petrochemical facilities subjected to blast loads emphasize the importance of ductility, which is similar to seismic design (ASCE, 1997). Accordingly, some general blast-resistant design concepts may still be helpful for designing blast-loaded bridge columns, and detailing requirements should focus on provid- ing adequate reinforcement for the formation of plastic hinges. 3.3.1.3 Test Variables For variables in Table 7 that show a range in design values depending on location within the United States, priority for the test program was given to those design details that were believed to be the most critical to reinforced concrete column performance when subjected to blast loads. The following five test program variables are discussed below: cross-sectional Bent Type Shape Min. Diam. D (f t.) f' c (psi) Clas s Cover c (in.) Fo oting (bottom ) Be nt Cap (top ) CA Sing le O/ Re 3600 2 f ixed fixe d S eism ic CO - - 3 4000 2 f ix ed Durability CT S/ M R ound 3 4000 F 3 fixe d DE Multi Round 3 4500 D 2 fixe d f ixed Seism ic FL S/ M R o/Re 3 3 f ix ed IL Multi Ro/Re 2. 5 2 f ix ed hing e ND S/ M R o/Re 2 3000 AE fixe d NJ S/ M R ound 3 A 2 fixe d NY Multi Ro/Re HP/ A 2 fixe d f ixed SC Sing le O/Ro 3 4000 2 f ix ed fixed Seism ic TX Multi Ro/Re 1.5 3600 C 3 f ix ed prop AASHTO 4000 A 3 Grade f y (ksi) Size Min . Min # Long. Bars Rei nf . Rati o L % Type Gr ade f yh (ksi) Size Min. Pitch ma x (in. ) CA 60 5 6 BW H 6 0 5 3.75 mi d hei ght CO 60 4 H oops 60 4 3 CT 60 4 S piral 60 4 DE 60 5 S piral 60 5 mi d hei ght FL 60 Hoop s 60 4 IL 60 7 S/ H 60 60 ND 60 5 NJ 60 Spiral 60 5 3 .5 mid hei ght NY 60 S/ H 60 4 1 2 mi d hei ght SC 60 8 6 BW H 60 6 1 2 mi d hei ght TX 40 6 6 1 S/ H 40 † 3 6 AASHTO 60 5 6 S/ H 60 3 1 2 bottom S/M = Single or Multi O = Oblong S/H = Spiral or Hoop Ro = Round Re = Rectangular BWH = butt welded hoops All fields left blank are controlled by AASHTO Specifications † Longitudinal Bars may be designed as Gr. 40 to reduce splice lengths Stat e Stat e De sign Concerns Column Geometry Concrete Standards Col. Connection Long. Reinf . Standards Shear Reinf . Standards Splice Location Table 7. State DOT column design summary.

shape, length-to-depth (L/D) ratio, type of transverse rein- forcement, volumetric reinforcement ratio, and splice loca- tion. These design parameters were varied in the test program to provide a broad representation of current bridge column designs nationwide. 3.3.1.3.1 Cross-Sectional Shape. There are three shapes typically used in bridge column design: circular, rectangular, and oblong. Circular columns are the most economical and commonly used cross-sectional shape, as shown in Table 7. Circular columns require less formwork and provide a greater ease of construction than other shapes. Rectangular-shaped columns, however, are also regularly used in six of the eleven states listed in Table 7. There is currently little research on how blast loads inter- act with bridge columns or slender members of any shape. Circular and square cross-sections were selected as test vari- ables to better understand how blast loads interact with slen- der members with different reflecting surfaces because load characterization is an important step the blast-resistant de- sign and detailing process. 3.3.1.3.2 Length-to-Depth Ratio. Another important design parameter for blast-loaded columns is the L/D ratio. The L/D ratio can be varied by changing the column cross- sectional depth while holding the length or height constant. The depth of the cross-section is directly proportional to the shear strength of a concrete column, as shown in Equation 6 (AASHTO, 2007), which is important in blast-resistant design. where: f ′c = specified compressive strength of concrete at 28 days (psi) fy = yield strength of reinforcing bars (psi) d = effective depth of cross-section (in.) b = effective width of cross-section (in.) s = spacing of stirrups (in.) Av = area of shear reinforcement with a distance s (in.) The minimum column dimension in Table 7 varies among the states depending on column type and shape. Most DOTs have adopted conservative minimum column dimension spec- ifications to allow design redundancy and safety against vehic- ular collisions with substructures, which are not considered explicitly in most cases. For example, TxDOT provides a table of minimum steel requirements for round columns with diameters between 18 in. and 60 in. for designers, as shown in Figure 25. V V V f bd A f d s N c s c v y = + = ′ +2 6( ) 35 Bars T Height No. of Bars Size % Size & Pitch "Y" Maximum 18"ϕ 6 #6 1.04 #3 @ 6" 18' 24"ϕ 8 #7 1.06 #3 @ 6" 24' 30"ϕ 8 #9 1.13 #3 @ 6" 30' 36"ϕ 10 #9 0.98 #3 @ 6" 36' 42"ϕ 14 #9 1.00 #3 @ 6" 42' 48"ϕ 18 #9 0.99 #3 @ 6" 48' 54"ϕ 18 #10 1.00 #4 @ 9" 54' 60"ϕ 22 #10 0.99 #4 @ 9" 60' Column Diameter Vertical Steel Bars V T Bars Column Section V BARS T BARS CO LU M N D IA M ET ER 3" TYP. CO LU M N H EI G H T PI TC H SPIRAL DIAMETER " H " O N E FL A T TU RN T O P & B O TT O M Figure 25. Minimum column steel requirements for round columns (TxDOT, 2001).

36 There are four main column types: solid wall, single- column, multi-column, and pile bents, as shown in Figure 26. The two most common column configurations, single-column and multi-column, were represented in this test program with two different column dimensions. A single-column configu- ration consists of either a hammerhead or tee-shape capbeam overhanging each side of a large column that supports the superstructure as a whole. With increasing column height and narrow superstructures, the single-column configuration becomes more economical than the multi-column config- uration by reducing the required amounts of material and formwork. Therefore, lower L/D ratios (i.e., larger column diameters) represent single-column configurations. A multi-column configuration consists of two or more columns connected by a capbeam that supports the super- structure at points between the columns. A multi-column bent is the best choice for wide superstructures. Therefore, higher L/D ratios (smaller column diameters) represent multi- column configurations. 3.3.1.3.3 Type of Transverse Reinforcement. The type of transverse reinforcement also varies from state to state. The majority of DOT manuals prefer discrete hoop reinforcement over continuous spiral reinforcement due to ease of construc- tion. However, spiral reinforcement increases confinement and rebar cage stability relative to the case of discrete hoops, which is critical to seismic designs that depend on this extra ductility. The use of spirals could also be helpful in blast- resistant design. The use of continuous spiral reinforcement requires fewer anchorages than the use of discrete hoops, which minimizes the probability of pullout failures. The use of discrete hoops or ties requires proper anchorage depending on the type of loading: typical or seismic. A typical load consists of tradi- tional gravity loads (dead and live load) plus wind load, which require the use of standard hooks. Section 5.10.2.1 of the AASHTO LRFD (2007) defines standard hooks for transverse reinforcement as one of the following: • No. 5 bar and smaller: 90° bend, plus a 6.0 db extension at the free end of the bar • No. 6, No. 7, and No. 8 bars: 90° bend, plus a 12.0 db exten- sion at the free end of the bar • No. 8 bar and larger: 135° bend, plus a 6.0 db extension at the free end of the bar where: db = nominal diameter of the reinforcing bar (in.) Recent work by Bae and Bayrak (2008) on the seismic per- formance of full-scale, reinforced concrete columns demon- strated the opening of seismic discrete ties using hooks with a 135° bend, plus an extension of 8.0 db. The AASHTO LRFD Section 5.10.2.2 defines seismic hooks as a “135° bend, plus an extension of not less than the larger of 6.0 db or 3 in.” Bae and Bayrak noted that, unlike the “full-scale concrete columns, the hooked anchorages often reach close to the center of the core concrete in small-scale column specimens.” Bae and Bayrak used a minimum hook length of 15.0 db for the remaining tests to prevent anchorage failures. The larger “hook length proved to be very effective, and opening of the 135° hooked anchor- ages of the ties was not observed in the other tests.” To avoid similar anchorage failures with the half-scale test columns constructed for the current research, specified hook lengths for the blast-loaded column designs were longer than those used successfully in the seismic tests by Bae and Bayrak. These “blast” hooks were specified to consist of a 135° bend with a 20.0 db extension at one free end of the bar, reversing direc- tions each spacing. Additionally, for the purposes of the test program, there was a desire to represent current national prac- tices while still taking into consideration recent research find- ings. Accordingly, both spiral reinforcement and discrete hoops or ties with standard and blast hooks were included as variables in the experimental program. 3.3.1.3.4 Volumetric Reinforcement Ratio. The volu- metric reinforcement ratio varies depending on the governing loading condition: typical or seismic. Equation 7 (AASHTO, 2007) specifies the minimum volumetric reinforcement ratio, ρs, for typical, gravity-loaded columns. For the states included in Table 7, seismic loads usually govern column design and detailing requirements in California and South Carolina. Equation 8, from the AASHTO LRFD seismic provisions, specifies a more stringent minimum volumetric reinforcement ratio than for gravity-loaded columns. (a) (b) (c) (d) Figure 26. Typical column types: a) solid, b) single- column, c) multi-column, d) pile bent (NYSDOT, 2006).

where: f ′c = specified compressive strength of concrete at 28 days (psi) fy = yield strength of reinforcing bars (psi) Ag = gross area of column (in.2) Ac = area of concrete core (in.2) The seismic provisions require sufficient transverse rein- forcement “to ensure that the axial load carried by the column after spalling of the concrete cover will at least equal the load carried before spalling and to ensure that buckling of the longitudinal reinforcement is prevented” (AASHTO, 2007). Thus, the spacing of transverse reinforcement is important for shear resistance and confinement in the plastic hinge regions (typically, the top and bottom) of a seismically loaded column. Shear resistance and confinement in plastic hinge regions are also important for blast-loaded columns. The minimum vol- umetric reinforcement ratio shown in Equation 9 was used for blast-loaded column designs in this test program. where: f ′c = specified compressive strength of concrete at 28 days (psi) fy = yield strength of reinforcing bars (psi) Ag = gross area of column (in.2) Ac = area of concrete core (in.2) The above equations apply to circular columns. For a rectan- gular column, the minimum total gross sectional area, Ash, of hoop reinforcement is specified in Equations 10 and 11 for seis- mic columns in the AASHTO LRFD seismic provisions (2007) and blast-loaded columns in this test program, respectively. where: f ′c = specified compressive strength of concrete at 28 days (psi) A sh f f sh c c y ≥ ′0 18 11. ( ) A sh f f sh c c y ≥ ′0 12 10. ( ) ρs c y f f ≥ ′0 18 9. ( ) ρs c y f f ≥ ′0 12 8. ( ) ρs g c c y A A f f ≥ −⎛⎝⎜ ⎞ ⎠⎟ ′ 0 45 1 7. ( ) fy = yield strength of reinforcing bars (psi) s = vertical spacing of hoops, not exceeding 4 in. (in.) hc = core dimension of column in the direction under consideration (in.) Essentially, 50% more confinement steel was specified in the design of blast-loaded columns over current seismic pro- visions to further investigate the benefits of additional trans- verse reinforcement to column strength, ductility, and energy dissipation. Therefore, a different volumetric reinforcement ratio for each type of loading (typical, seismic, and blast) was considered in the test program. 3.3.1.3.5 Splice Location. Another important design parameter is the splice location of longitudinal reinforcement. Ideally, splice zones of longitudinal steel should occur at a point in a member where the bending moment is equal to zero or is very small. The moment is zero at the inflection point of the bending moment diagram; therefore, a natural place to splice longitudinal reinforcement is near that point. The location of the inflection point depends directly on the assumed boundary conditions and load distribution, which varies for typical, seismic, and blast loads. Figure 27 illustrates the moment diagram and inflection points for each load case assuming fixed supports. Columns designed for gravity loads, including a horizontal wind load, have an inflection point near 0.2L from each column support. Wind load, however, does not typically control a gravity-loaded column design, and the typical splice near the column base allows for ease of construction. Seismic loads induce a high lateral displacement between the column supports, creat- ing an inflection point near mid-height (0.5L). Therefore, seismic column design restricts splices within 1.5D of each support to provide sufficient rebar cage stability and column ductility within these plastic hinge regions. A column exposed to a blast wave approximated with a uniform load results in inflection points near 0.2L from each support for an elastic column response. However, a blast-loaded column is expected to behave inelastically, causing the deformed shape and effective support conditions to change as a function of time. For example, the inflection points for a uniformly loaded inelastic column that is initially fixed at both ends occur around 0.15L from each support. To investigate the effects of splice location on the performance of blast-loaded bridge columns, the test program included two different splice configurations: near the base for a typical column or no splice to represent seismic columns. 3.3.2 Column Design The typical and seismic column designs used in the test program were based on current state DOT standards and 37

38 (a) (b) (c) Figure 27. Moment diagram and deflected shape for each load case: a) typical, b) seismic, c) blast. Figure 28. Plastic hinge analysis for blast-loaded column. the AASHTO LRFD specifications. To reduce the chances of a potential shear failure resulting from a small standoff test, blast-loaded columns were designed using a plastic hinge analysis similarly to current seismic design provisions and as illustrated in Figure 28. The boundary conditions correspond- ing to a propped-cantilever condition shown in the figure are consistent with the research objectives and with the reaction structure used in the test program. Additional information on the assumed boundary conditions and reaction structure design is given in the following subsections.

3.3.2.1 Blast-Column Design A blast-loaded column design considers all potential plas- tic hinge locations to determine the maximum possible shear demand on the column. The maximum shear demand is a function of the boundary conditions and load distribution. Boundary conditions depend on the assumed threat scenario and the position of the center of detonation relative to the orientation of the bridge under consideration. A propped- cantilever was assumed for this test program in accordance with Table 7 for two reasons. First, this condition is consis- tent with the design details used by various state DOTs, and second, because this assumption will usually result in a con- servative estimate of shear demand for a blast-loaded bridge column, which was expected to be the controlling mode of failure. A propped-cantilever is a conservative assumption for the experimental tests and most design scenarios because these boundary conditions lead to the largest shear demand at the base for all probable cases, except that of a cantilever, which is an unlikely configuration for a bridge column. Thus, when considering a uniformly loaded column, propped- cantilever boundary conditions produce a larger shear demand at the base of the column than that produced by the cases of simply supported and fixed–fixed columns. Accordingly, the reaction structure was designed to provide these support conditions. The reaction structure design and details are pre- sented in Section 3.3.3. Blast-load distribution on a column is a function of stand- off. As shown in Figure 29, columns with a large standoff can be approximated by a uniform load; however, columns with a small standoff are better approximated with a varying load in most cases because, as the blast wave propagates away from the blast source, the base of the column will be loaded first (assuming the center of detonation is near the ground, which is a common assumption for blast scenarios involving a vehicle-delivered explosive). For this research, a linearly varying blast load was assumed for the plastic hinge analysis. 39 Figure 29. Blast load distribution versus standoff.

The required pitch of transverse reinforcement can be determined by setting the maximum shear demand deter- mined in the plastic hinge analysis equal to the shear design equations from the AASHTO LRFD, modified to account for strain rate effects (ASCE, 1997), and solving for the spacing. The plastic moment (Mp), which is equal to the flexural capac- ity of the cross-section (MN), also accounts for the dynamic material strength, with dynamic increase factors for strain rate effects. Again, the maximum shear demand was used to deter- mine the required volumetric reinforcement ratio to ensure that adequate shear capacity was provided and to force the for- mation of plastic hinges (a flexural failure). 3.3.2.2 Column Design Summary Ten columns with eight different column designs were tested in Phase II of the experimental program. Typical designs for concrete columns designed in areas of high seismicity (designated Seismic) as well as those located in areas with a low seismic threat (designated A or B) were considered. In ad- dition, two columns designed specifically for the case of blast loads (designated Blast) were also included in the test matrix. Test variables for each column design are summarized in Table 8 and illustrated in Figure 30. Constants specified in this test program include 4000 psi, Type-A concrete, 1 in. concrete 40 Long. Reinf. Shape Size (in.) Splice Location Pitch/ Space (in.) Type Vol. Ratio s % 1 1A1 Round 18 0.25L 6.0 Hoops 0.82 2 1A2 Round 18 0.25L 6.0 Hoops 0.82 3 1B Round 18 0.25L 6.0 Spiral 0.82 4 2A1 Round 30 0.25L 6.0 Hoops 0.47 5 2A2 Round 30 0.25L 6.0 Hoops 0.47 6 2B Round 30 none 6.0 Spiral 0.47 7 2-seismic Round 30 none 3.5 Spiral 0.80 8 2-blast Round 30 0.25L 2.0 Spiral 1.40 9 3A Square 30 0.25L 6.0 Ties 0.47 10 3-blast Square 30 0.25L 2.0 Ties 1.40 Small Standoff Test No. Column Designation Column Geometry Transverse Reinf. Table 8. Test variable summary. Figure 30. Column cross-sections.

clear cover (which corresponds to 2 in. of clear cover in full- scale columns), GR 60 steel reinforcing bars, 1.0% longitu- dinal reinforcement ratio, and propped-cantilever boundary conditions. 3.3.2.3 Material Properties The concrete and steel specified in this project were selected to represent that which is typically used in the construction of reinforced concrete highway bridge columns. Due to the vari- ety of concrete strengths and types used in different states, the values specified by the AASHTO LRFD were selected for the test program. Concrete of 4000 psi strength and composed of Type-A cement with a maximum aggregate size of 3⁄8 in. (to accommodate the small rebar spacing) was selected for this test program. Also, standard deformed, uncoated, Grade 60 reinforcing bars were specified for all reinforcement except the continuous spirals that allow the use of a smooth bar (accord- ing to the AASHTO LRFD). 3.3.2.4 Axial Load Most bridge columns have much greater axial load capac- ity than axial load demand from gravity loads, and thus they typically experience service compressive loads that are well below the balance point of the column. Therefore, including a compressive axial load in the test setup would only increase both flexural and shear resistance of the specimens. Further- more, the construction of a test frame able to maintain an axial load on column specimens while surviving multiple blast tests was impractical and cost-prohibitive given the allocated funds. Consequently, given the expected increase in capacity an axial load would contribute, coupled with the cost of developing a test setup capable of applying an axial load over the entire test program, the experimental tests con- ducted for this research neglected the effect of a compressive axial load. Furthermore, while it may be possible for a scenario to exist in which a column will experience tensile forces due to uplift on the deck prior to experiencing peak flexural response, a survey of DOT design details showed that the majority of states use a configuration with the superstructure resting on a bent cap, which would not allow a column to experience tensile forces. Even for bridge configurations in which the superstructure is directly connected to a column, a column will reach its peak shear and flexural responses very early in time for the attack scenarios creating the most severe lateral loading conditions (i.e., small standoffs). The best use of allo- cated funds was to consider an experimental testing program that represented the most probable design threats combined with the most common bridge column configurations, and thus the Phase II experimental test program did not consider blast-loaded bridge columns with a tensile or compressive axial load. 3.3.3 Reaction Structure Design The steel reaction structure and reinforced concrete slab for the small standoff tests were designed by Protection Engineer- ing Consultants (PEC). The reaction structure was designed to be pre-fabricated offsite and delivered in one piece, while the slab was constructed onsite. The design and detailing of the reaction structure and slab are described below. 3.3.3.1 Steel Reaction Structure Design The steel reaction structure was designed to resist dynamic blast loads. A single-degree-of-freedom analysis was com- pleted for each frame member using SBEDS (U.S. Army Corps of Engineers, 2007) and assumed fixed end conditions. SBEDS (Single-degree-of-freedom Blast Effects Design Spreadsheet) is a commonly used program in the blast-resistant design community. The reaction structure consisted of HSS 8 × 8 × 5⁄8 mem- bers, 1⁄2-in. ϕ steel rods, and W10 × 88 anchor beams, as shown in Figure 31. The HSS shapes were A500 Gr B steel, while all other steels were specified to have a minimum yield strength of 50 ksi. All connections were welded with 1⁄2-in. fillet welds at perpendicular joints or 5⁄8-in. flared groove welds otherwise, as shown in Figure 32. 3.3.3.2 Reinforced Concrete Slab Design The reinforced concrete slab was also designed for dynamic blast loads. The majority of the slab was 2-ft deep, but it increased to a depth of 2 ft 6 in. around the 5-ft × 7-ft hole for the column footing, as shown in Figure 33. The slab thickness increased near the column footing to account for the large reaction forces expected there from the column and reaction structure. Figure 33 also illustrates the extensive reinforce- ment that is used throughout the slab. Casting of the slab took place onsite using a regular concrete truck. 3.3.4 Data Acquisition and Instrumentation Plan The test specimen and instrumentation in a blast test are subjected to extreme loading conditions (temperature and pressure). The type and location of instrumentation must be able to endure these conditions. This test program specified the use of strain gauges on reinforcing bars, free-field pres- sure gauges, and high-speed video cameras to gather experi- mental data and observations. All wires or cables within close proximity of a test specimen were bundled and buried under- 41

ground to prevent damage. Post-test measurements and pic- tures were taken to document damage conditions following each test. 3.3.4.1 Strain Gauge Installation on Reinforcing Bars Each column was instrumented with at least eight strain gauges. Data were collected from six of the strain gauges during each small standoff test. The installation of eight strain gauges provided redundancy in case any strain gauges were damaged during construction and allowed the use of all six channels provided by the data acquisition system. As shown in Figure 34, the gauges were located on transverse and longitudinal bars to measure the strain near the base of the column, represented by triangles and circles, respec- tively, in the figure. (Note that the column in Figure 34 was straightened prior to casting.) The strain gauges used in this research project were TML Strain Gauges from Tokyo Sokki Kenkyujo Co., Ltd., Type FLA-5-11-15LT, with a gauge length of 5 mm. 3.3.4.2 Free-Field Pressure Gauge Locations Three free-field pressure gauges were used during each small standoff test. Pressure gauges were located at 37 ft, 51 ft 8 in., and 76 ft from each charge. The free-field pressure gauges needed to be placed at least 30 ft from the test spec- imen to ensure the survival of the gauges over the ten tests. Free-field pressure gauges measure the side-on (i.e., free- field) pressure as the shock front passes the gauge. The aver- age free-field pressure and impulse were used to calculate TNT equivalency and efficiency for each blast test. 42 Figure 31. Reaction structure. Figure 32. Welded connections for the reaction structure. Figure 33. Slab reinforcement.

3.3.4.3 High-Speed Video Cameras One or two high-speed video cameras were used for each small standoff and close-in test. Video provided by these cameras was useful in observing aspects of behavior that were difficult to discern from the strain gauge data and other instrumentation. Often, however, the “fireball” from an explosion obscured the specimen response, and the high- speed video was of limited benefit. 3.3.5 Column Construction The construction of all columns and footings took place at the test site. A summary of the entire construction process and column fabrication is provided below. 3.3.5.1 General The construction process began with the delivery of materi- als to the test site. First, the formwork and rebar cage for each column footing were constructed. Figure 35 illustrates the reinforcement used in each footing, including a mat of #8 bars top and bottom, #4 closed stirrups in each direction, and #6 longitudinal bars anchored into the bottom steel mat. Essentially, a box of steel was provided in each 5-ft square-, 2-ft 6-in.-deep concrete footing. The footings were designed with sufficient capacity to ensure that failure would occur in the column for all blast tests. Next, the longitudinal and transverse reinforcements for each column were placed vertically into the footing cages, as shown in Figure 35. To decrease variability in the spacing of transverse reinforcement, wooden spacers were used, as shown in Figure 36. After each cage was completed, speci- fied bars were instrumented. After instrumentation, the footings were cast and allowed to cure for several days (Fig- ure 37) before formwork was placed around the column rebar cages. The columns were then cast in a second lift with a concrete pump truck and consolidated with a hand vibra- tor, as shown in Figure 38. The construction timeline allowed at least 22 days for the concrete to cure prior to testing, and concrete strength based on cylinder tests had reached 97% of the specified compressive strength at the start of the blast testing. 3.3.5.2 Column Constructability Column constructability was not an issue for the typical gravity-loaded column designs. The construction process was considerably more difficult, however, for the seismic-loaded and blast-loaded column designs due to the increased volume of transverse reinforcement and smaller spacing or pitch. The reinforcing cage for the square blast column shown in Fig- ure 37 was more difficult to construct than the round typical column, though clearly not impossible. Also, special care was taken when vibrating the tightly spaced rebar cages to ensure proper consolidation of the concrete. 3.3.6 Small Standoff Test Setup As stated previously, the desired boundary conditions for the Phase II close-in blast tests were those of a propped- cantilever column. Figure 39 shows an actual column placed into the test setup to illustrate the assumed boundary con- ditions. To create a fixed connection at the base, the large column footing was placed into the slab opening and grouted into place to help prevent rotation. The steel collar that wrapped around the top of the concrete column was then 43 SP LI CE H EI G H T 2'-6" COLUMN 3A LONG. REINF: 24 #6 EQUALLY SPACED TIE REINF: #4 @ 6"O.C. 1 2 3 9 10 11 5 6 7 4' -5 " 2' -6 " 1' -6 " 1' -3 " 6" 2' -1 0" Figure 34. Strain gauge layout (Column 3A).

bolted to the reaction structure to provide the pinned con- nection near the top of the column. It should be noted that over the course of the test program the slab of the reaction structure experienced significant damage and may have allowed some rotation to occur at the base so that the assumed boundary conditions were not fully realized. To address this concern, the analysis of the test data and evaluation of column response considered the effect of rotation at the base. Analysis of the data and evaluation of column behavior during the test pro- gram are described in Chapter 5. 3.3.7 Spall/Breach Test Setup The goal of the local damage tests was to observe the spall and breach patterns of columns subjected to close-in blasts. Therefore, end restraints against displacement or rotation were 44 (a) (b) Figure 35. Column footing and column rebar cage: a) column 1B, b) column 3-blast. Figure 36. Wooden spacers. Figure 37. Column rebar cage.

3.4 Summary This chapter presented the design and instrumentation of columns tested during Phases I and II of this experimental re- search program. The small-scale, non-responding blast test setup was summarized. All of the instrumentation, construc- tion, and testing for Phase I was completed in Vicksburg, MS, by the U.S. Army Engineer Research and Development Center. The setups for the Phase II small standoff and local damage blast tests were also summarized. All Phase II instrumentation, construction, and testing was completed at the Southwest Research Institute test site in Yancey, Texas. The following chapter presents the analytical research program. 45 Figure 38. Column casting. Figure 39. Phase II: Small standoff test setup. Figure 40. Phase II: Local damage test setup. not needed because only local damage was being observed. The local damage test setup consisted of a standalone column sub- jected to a close-in blast load in a clearing. The base and top were not supported or restrained from any rotation. Figure 40 illustrates a column just prior testing.