STEEL STRUCTURE DESIGN REQUIREMENTS
8.2 SEISMIC REQUIREMENTS FOR STEEL STRUCTURES
Table 8.6 presents nominal shear values for plywood and oriented strand board attached to steel stud wall assemblies. Design values are determined by multiplying the nominal values by a phi (N) factor as presented in Sec. 8.6.5. These nominal values are based upon tests performed at Santa Clara University (Serrette, 1996). The test program included both cyclic and static tests;
however, the values presented in Table 8.6 are based upon the cyclic tests as they are intended for use in seismic resistance. In low seismic areas where wind loads dominate, nominal values have been recommended for wind resistance by AISI based upon monotonic tests (AISI, 1996). The cyclic tests were performed using the assemblies that were determined to be the most critical from the static tests. The assemblies cyclically tested consisted of 3.5 x 1.625 inch C studs fabricated with ASTM A446 Grade A (33 ksi) with a minimum base metal thickness of 0.033 inch. Since the tests were conducted, ASTM A446 Grade A has been redesignated ASTM A653 SQ Grade 33. The test panels were four feet wide and 8 feed high, the sheathing material was applied vertically to only a single side of the studs, and there was no sheathing or bracing applied to the other side.
The cyclic tests were performed using a sequential phase displacement protocol under develop- ment at the time of the test by an ad hoc Committee of the Structural Engineers Association of Southern California. Nominal values were conservatively established by taking the lowest load in the last set of stable hysteretic loops. It is expected that subsequent testing of steel stud shear wall assemblies will reduce or modify some of the restrictive limits currently proposed for the use of the system such as the nominal maximum thickness of the studs of 0.043 inch, the aspect ratio of 2:1, and the ability to use sheathing on both sides of the wall.
8.7 SEISMIC REQUIREMENTS FOR STEEL DECK DIAPHRAGMS: Since the design values for steel deck are based on allowable loads, it is necessary to present a method of deriving design strengths. Two N values are presented — 0.60 for steel deck that is mechanically attached and 0.50 for welded steel deck. These factors are consistent with current proposals being
circulated for inclusion in updates of Ref. 8-5.
8.5 STEEL CABLES: The provisions of Sec. 8.5 are virtually unchanged from previsous editions. Although the provisions in Ref. 8-8 are dated, they are the only ones available and there was no sentiment to eliminate them from the Provisions. The allowable stress levels of steel cable structures specified in Ref. 8.8 are modified for seismic load effects. The value of 1.5T4 was chosen as a reasonable value to compare with increases given to other working stress levels.
REFERENCES:
Serette. 1996. Shear Wall Values for Light Weight Steel Framing. American Iron and Steel Institute.
CONCRETE STRUCTURE DESIGN REQUIREMENTS
9.1 REFERENCE DOCUMENT: The main concern of Chapter 9 is the proper detailing of reinforced concrete construction for earthquake resistance. The bulk of the detailing requirements in this chapter are contained in Ref. 9-1, Building Code Requirements for Reinforced Concrete, ACI 318-95. The commentary for ACI 318-95 contains a valuable discussion of the rationale behind detailing requirements that is not repeated here.
9.1.1 MODIFICATIONS TO REF. 9-1: The modifications noted for ACI 318-95 are: changes in load factors necessary to coordinate with the equivalent yield basis of this document; additional definitions necessary for seismic design requirements for structural systems composed of precast elements; and changes that incorporate certain features of the detailing requirements for
reinforced concrete that have been adopted into the 1997 Uniform Building Code.
Included as Sec. 9.1.1.5 is a statement on reinforced concrete structural systems incorporating precast concrete elements. One design alternative is emulation of monolithic reinforced concrete construction. The other alternative is the use of the unique properties of precast elements interconnected predominately by dry joints. For the first alternative Sec. 9.1.1.5, 9.1.1.6 and 9.1.1.9 define design procedures ensuring that the resulting structural systems have strength and stiffness characteristics equivalent to those for monolithic reinforced concrete construction. The existing code requirements for monolithic construction then apply for all but the connections. The second alternative, the Appendix to Chapter 9, is included for information and for trial design by users.
Procedures for structural system composed from precast elements interconnected predominately by dry joints are included as an appendix because the existing state of knowledge makes it
premature to propose code requirements based on that information. The complexity of structural systems, configurations and details possible with precast concrete elements requires:
1. Selecting functional and compatible details for connections and members that are reliable and can be built with acceptable tolerances;
2. Verifying experimentally the inelastic force-deformation relationships for welded, bolted, or grouted connections proposed for the seismic resisting elements of the structure; and 3. Analyzing the structure using those connection relationships and the inelastic reversed
cyclic loading effects imposed by the anticipated earthquake ground motions.
Research conducted to date (Cheok and Lew, 1991; Elliott et al., 1992; Englekirk, 1987; French et al., 1989; BSSC, 1987; Hawkins and Englekirk, 1987; Jayashanker and French, 1988; Mast, 1992; Nakaki and Englekirk, 1991; Neille, 1977; New Zealand Society, 1991; Pekau and Hum,
1991; Powell et al., 1993; Priestley and Tao, 1991; Stanton et al., 1986; Stanton et al., 1991) documents concepts for design using dry connections and the behavior of structural systems and subassemblages composed of precast elements both at and beyond peak strength levels for nonlinear reversed cyclic loadings, and provides the basis for the appendix.
Emulation of Monolithic Construction Using Strong Connections: For emulation of the behavior of monolithic reinforced concrete construction, Sec. 9.1.1.6 provides two alternatives. Sec.
21.2.2.6 in Sec. 9.1.1.6 covers structural systems with "wet" connections. Sec. 21.2.2.7 in Sec.
9.1.1.6 covers structural systems with "strong" connections.
For frame systems that use strong connections, Sec. 21.2.2.7 and 21.2.7, the different connection categories envisaged are shown in Figure C9.1.1-1. Considerable freedom is given to locating the nonlinear action zones (plastic hinges) along the length of the precast member. However, those hinges must be separated from the connection by a distance of at least three quarters of the member's depth. Wet-joint connections are permitted at the strong connection but not at the hinge location.
Provision 21.2.7.2 makes the strength required for a strong connection dependent on the distances hinges are separated from that connection, the strengths of those hinges and the nonlinear deformation mechanism envisaged. The conditions described by Sec. 21.2.7.2 for a beam to continuous column connection are shown in Figure C9.1.1-2, which is an adaptation of Figure R21.3.4 of Ref. 9-1. Because the strong connection must not yield or slip; its nominal strengths, Sn CONNECTION, in both flexure and shear must be greater than those corresponding to development of the probable strengths Mpr1 and Mpr2 at the hinge locations. Figure
C9.1.1-2b, illustrates the situation for flexure. Per Ref. 9-1 moments Mpr1 and Mpr2 are
determined using a strength reduction factor of 1.0 and reinforcing steel stresses of at least 1.25fy.
For columns above the ground floor, moments at a joint may be limited by flexural strengths of the beams framing into that joint. However, for a strong column-weak beam deformation mechanism, dynamic inelastic analysis and studies of strong motion measurements have shown that beam end moments are not equally divided between top and bottom columns even where those columns have equal stiffness. Elastic analysis predicts moments as shown in Figure C9.1.1-3a while the actual situation is likely to be as shown in Figure C9.1.1-3b. Accordingly, provision 21.2.7.3 is included for the midheight column connection.
Use of Prestressing Tendons: Sec. 9.1.1.7 defines conditions under which prestressing tendons can be used, in conjunction with deformed reinforcing bars, in frames resisting earthquake forces.
As documented by Ishizuka and Hawkins (1987), if those conditions are met no modification is necessary to the R and Cd factors of Table 2.2.2 when prestressing is used. Satisfactory seismic performance can be obtained when prestressing amounts greater than those permitted by Sec.
9.1.1.6 are used. For example, the connection of Figure C9A-3 has a satisfactory performance for a probable strength of 180 kips and a reversed deformation of 10 mm. However, as documented by Park and Thompson (1977) and Thompson and Park (1980) and required by the combination of New Zealand Standards 3101:1982 and 4203:1992, ensuring that satisfactory performance requires modification of the R and Cd factors.
Figure C9.1.1-1 Connection categories
Figure C9.1.1-2 Design forces for strong connections between beams and continuous columns.
Figure C9.1.1-3 Moments at beam-to-column connections.
9.1.1.12: The minimum thicknesses for concrete diaphragms reflect current usage in joist and waffle systems and topping slabs for precast floor and roof systems. Bonding of the top slab provides restraint against slab buckling which, for an untopped joist or waffle system, is provided by the webs.
9.1.1.14 Coupling Beams: Short-span coupling beams between shear walls, under reversing loads simulating earthquakes, have been experimentally investigated by many researchers (Barney et al., 1978; Bertero and Popov, 1975; Brown and Jirsa, 1971; Hirosawa et al., 1973; Ma et al., 1976; Paulay and Binney, 1974; Scribner and Wight, 1978; Shiu et al., 1978; Wight and Sozen, 1975).
The tests indicate that short flexural members with small clear-span-to-effective-depth ratios behave differently than slender flexural members. When short coupling beam specimens were subjected to inelastic load cycles, large flexural cracks formed at both beam ends. With increasing load reversals, cracks from each direction of loading interconnected, forming a vertical plane of weakness at each end of a beam. Thus, instead of a conventional truss mechanism, shear transfer across the plane of weakness was provided primarily by aggregate interlock and shear friction.
Under subsequent inelastic load cycles, the shear resisting mechanism deteriorated rapidly resulting in a loss of load capacity by "sliding shear."
It was found (Barney et al. , 1978; Paulay and Binney, 1974; Shiu et al., 1978) that increasing the number of hoop stirrups was not effective in improving resistance against sliding shear.
Therefore, various configurations of special shear reinforcement to improve seismic performance of short coupling beams were tested (Barney et al., 1978; Bertero and Popov, 1974; Paulay and Binney, 1974; Scribner and Wight, 1978). Full-length diagonal reinforcement was found to produce the greatest energy dissipation and deformation capacities.
In tests at the Portland Cement Association (Shiu et al., 1978), beam specimens with
clear-span-to-effective-depth ratio of 2.8 were able to sustain over 34 reversing load cycles. A maximum nominal shear stress of 12.5%ðfEQ \O(_,c) was recorded and a maximum imposed deformation of over nine times yield deflection was measured. In addition, the specimen was able to dissipate three times more energy than a comparable specimen without diagonal shear
reinforcement. No sign of sliding shear was observed at completion of testing. Similar results have been reported by Paulay and Binney (1974). In both investigations, diagonal shear reinforcement was designed by the proposed equation to resist the total shear force by truss action.
Based on test results, Paulay (1977) recommended that diagonal shear reinforcement be used to carry 75 percent of the induced shear in flexural members when nominal design shear stress under load reversals is larger than 3%ðfEQ \O(_,c) psi. When nominal shear stress exceeds 4.5%ðfEQ
\O(_,c) psi, he recommended that diagonal reinforcement should carry 100 percent of the induced cyclic shear forces.
For clear-span-to-effective-depth ratios greater than 4.0, PCA tests (Barney et al., 1978; Shiu et al., 1978) indicated that specimens with conventional longitudinal reinforcement resisted over 46
inelastic reversing load cycles. At the same time, maximum deformation of 13 times yield deflection was measured. Therefore, experimental data have shown that beams with
clear-span-to-effective-depth ratios greater than 4.0 do not require diagonal reinforcement. In addition, diagonal reinforcement is not very effective in beams with clear-span-to-effective-depth ratios greater than 4.0.
It is permitted to use reinforcement arrangements without the diagonal bars if it is assumed that the coupling beams may fail early in the design earthquake and the consequences of such failures are taken into account. The consequences include, but are not necessarily limited to, reductions in stiffness due to lack of coupling between walls as well as debris or deformed members blocking exits and damage or failure of nonstructural components and cladding and the connections thereto due to increased drifts. The vertical load carrying capacity of the structure must not be impaired.
Even if the coupling beams are expected to fail early in the earthquake, the full design strength of the coupling beams still must be provided if they are assumed to be part of the seismic force resisting system. This is to reduce the likelihood that the transition of shear wall behavior from coupled to uncoupled action will begin at unduly low seismic loads.
9.2 BOLTS AND HEADED STUD ANCHORS IN CONCRETE: The allowable loads on anchor bolts have been chosen to suit the capacity reduction factors in this document.
9.2.2 BOLTS AND HEADED STUD ANCHORS: These requirements follow those given in the Uniform Building Code modified by recent improvements in shear capacity calculation given in the PCI Design Handbook, Fourth Edition.
While the requirements do not prohibit the use of single anchor connections, it is considered necessary to use at least two anchors in any load-carrying device whose failure might lead to collapse.
The requirements generally relate to groups of anchors attaching a loaded steel plate to the concrete surface. The thickness of this connector plate also is a design consideration and must be adequate to allow the anchors to perform in group action if the calculated design strengths are to be realized. A plate thickness of not less than one half the diameter of the anchor shank is recommended.
These requirements are intended to provide proper strength of connections. To achieve adequate ductility for seismic or other dynamic loads, use of auxiliary reinforcement for confining concrete or for direct load transfer should be considered. A number of tests have shown that hairpins or similar reinforcement confining the concrete engaged by the anchors and running through the failure surface into the adjacent concrete provides enhancement of a connection's ductility under dynamic loading. No clear recommendations as to the design of such reinforcement have been suggested, but its use is highly recommended in all anchor connections and particularly those subject to seismic or other dynamic loading.
Tests have shown that there are consistent shear ductility variations between bolts anchored to drilled or punched plates with nuts and connections using welded, headed studs.
Recommendations for design are not presently available, but this should also be considered in critical connections subject to dynamic or seismic loading.