288 SEISMIC DESIGN OF REINFORCED-CONCRETE STRUCTURES. Maximum Shear Stress /7e.(MPa 1 Ductity, max a si ZS a toe a atin te 7 oo ado 40 Maximum Shear stress /WE.08) Figure 9-28. Variation of rotational cuctity with maximum aver- ‘ge shear stress in PCA isolated wall test. (From Ret. 9-48) ductility of the wall In walls with relatively stiff and ‘well-confined boundary elements, some amount of web Crushing ‘can occur without necessarily, limiting. the flexural capacity of the wal. Corley et al* point out that trying to avoid shear failure in walls, particularly walls with stiff and well-confined boundary elements, may be a questionable design objective. Thus, although ‘ACI Appendix A limits the maximum average shear stress in walls to 10/7 (a value based on monotonic tests) with the intent of preventing web crushing, web crushing occurred in some specimens subjected to shear stresses only slightly greater than 7)”. However, those specimens where web-crushing failure occurred were able to develop deformations well beyond the yield deformation prior to loss of capacity. 4, Sequence of large-amplitude load cles. Dynamic: inelastic analyses of isolated walls? have indicated that in a majority of case, the maximum or a near-max- imum response to earthquakes occurs early, with per- haps only one elastic response cycle preceding it. This contrasts with the loading program commonly used in {Quasistatic tests, which consists of load cycles of pro- gressively increasing amplitude. To examine the effect Of imposing large-amplitude load cycles early in the test, two nominally identical isolated wall specimens ‘were tested. One specimen was subjected to load cycles fof progressively increasing amplitude, as were most of the specimens in this series. Figure 9-29a indicates that specimen B7 was able to sustain a rotational ductility of. Slightly greater than 5 through three repeated loading ‘yeles. The second specimen (B9) was tested using a modified loading program similar to that shown. in Figure 9-6, in which the maximum load amplitude was {imposed on the specimen after only one elastic load yele. The maximum load amplitude corresponded to a rotational ductility of 5. As indicated in Figure 9-29b, the specimen failed before completing the second load cycle. Although results from this pair of specimens cannot be considered conclusive, they suggest that tests using load cycles of progressively increasing amplitude may overestimate the ductility that ean be developed tinder what may be considered more realistic earth- {quake response conditions. The results do tend to con- firm the reasonable expectation that an extensively cracked and “softened” specimen subjected to several previous load cycles of lesser amplitude can better ac- Commodate large reversed lateral deflections than a Virtually uncracked specimen that is loaded to near- capacity early in the test. From this standpoint, the severity of the modified loading program, compared to the commonly used progressively-increasing-amplitude loading program, appears obvious. ‘5. Reinforcement detailing. On the basis of the tests con isolated walls reported in References 9-47 and 9-48, Ossterle et al. proposed the following detailing requirements for the hinging regions of walls: ‘© The maximum spacing of transverse reinforce- ‘ment in boundary elements should be Sd, where dy, is the diameter of the longitudinal reinforce- ‘ment, ‘© Transverse reinforvement in the boundary ele- ‘ment should be designed for a shear Vay = Myy/lSloy where My, = nominal moment strength of boundary element 1, = width of boundary element (in the plane of the wall) + No lap splices should be used for crostes in segments of boundary clements within the hing- ing region. «+ recommendation on anchoring horizontal web reinforcement in the boundary elements, such as is shown in Figure 930, has been adopted by ‘ACL Appendix A. For levels of shear in the range Of Sf. to TO, the study indicates that alter nate 90° and\135° hooks, a8 shown in Figure 5-30b, can be used. ‘The specimens tested in this series had special con- finement reinforcement only over a length near the base equal to the width of the wall, ie, the approximate length of the hinging region. Strain readings as well as observations of the general condition of the walls after failure showed that significant inelasticity and damage ‘were generally confined to the hinging region. In view ofBEHAVIOR OF CONCRETE MEMBERS UNDER EARTHQUAKE-TYPE LOADING 200300 SEISMIC DESIGN OF REINFORCED-CONCRETE STRUCTURES Tuned vertlcaty (a) (b) Figure 9-30 Alternative detals for anchorage of horizontal web reinforcement in boundary elements. (From Ret. 951.) (a) Detail for walls subjected to iowto-moderae stress levels. (6) Osta for walls subjected to high shear stress levels this, it has been suggested that special confinement reinforcement for boundary elements need be provided only over the lengths of potential hinging regions. These are most likely to occur at the base and at points along the height of the wall where discontinuities, associated with abrupt and significant changes. in geometry, strength, or stiffness, occur. Coupled Walls As mentioned earlier, a desirable characteristic in an earthquake-resistant structure is the ability to respond to strong ground motion by progres- sively mobilizing the energy-dissipative capacities of an ascending hierarchy of elements making up the struc- ture, In terms of their importance to the general stability and safety of a building, the components of a structure may be grouped into primary and secondary elements. Primary elements are those upon the integrity of which depend the stability and safety of the entire structure or a major part of it. In this category fall most of the vertical or near-vertcal elements supporting gravity loads, such as columns and structural walls, as well as ong-span horizontal elements. Secondary elements are those components whose failure would affect only lim- ited areas or portions of a structure. The strong-column, weak-beam design concept dis- ‘cussed earlier in relation to moment-ressting frames is an example of an attempt to control the sequence of yielding in a structure. And the “capacity design” ap- proach adopted in New Zealand—which, by using even greater conservatism inthe design of columns relative to beams, seeks to insure that no yielding occurs in the columns (except at their bases) —is yet another effort to achieve a controlled response in relation to inelastic action. By deliberately building in greater flexural strength in the primary elements (the columns), these design approaches force yielding and inelastic energy dissipation to take place in the secondary elements (the beams) ‘When properly proportioned, the coupled-wal sys tem can be viewed as a further extension of the above design concept. By combining the considerable lateral stiflness of structural walls with properly proportioned coupling beams that can provide most of the energy-is- Sipative mechanism during response to strong ground ‘motions, a better-performing structural system is ob- tained, The stiffness of the structural wall makes it a desirable primary clement from the standpoint of dam age control (by restricting interstory distortions), while the more conveniently repairable coupling beams pro- vide the energy-dissipating secondary elements. Figure 9-31a shows two-wall coupled-wall system and the forces acting at the base and on a typical coupling ‘beam. A typical distribution of the elastic shear force in the coupling beams along the height ofthe structure due {o a statically applied lateral load is shown in Figure 9.31, Note thatthe accumulated shears at each end of the coupling beams, summed over the height of the structure, are each equal to the axial force atthe base of. the corresponding wall. The height tothe most critically stressed coupling beam tends to move downward as the coupling-beam stiffness (i.e, the degree of coupling. between the two walls) increases. Tn a properly designed earthquake-resistant coupled wall system, the critically stressed coupling beams should Yield frst—before the bases of the wall. In addition, they must be capable of dissipating a significant amount cof energy through inelastic action. These requirements call for fairly stiff and strong beams. Furthermore, the desire for greater lateral-load-resisting efficiency in the system would favor sti and strong coupling beams. Figure £31. Laterally toaded coupled wall systom. (a) Forces ‘on walls at base. (0) Typical distribution of shears in coupling ‘beams over height of situctur.However, the beams, should not be so stiff or strong flexurally that they induce appreciable tension in the walls, since a net tension would reduce not only the yield moment but also the shear resistance of the wall (recall that a moderate amount of compression im- proves the shear resistance and ductility of isolated walls). This in turn can lead to early flexural yielding and shear-related inelastic action at the base of the tension wall. Dynamic inelastic analyses of coupled-wall systems® have shown, and tests on coupled-wall systems under cyclic reversed loading" have indi- cated, that when the coupling beams have appreciable stiffness and strength, so that significant net tension is induced in the tension wall, a major part of the total base shear is resisted by the compression wall (ie. the ‘wall subjected to axial compression for the direction of loading considered), a situation not unlike that which ‘occurs in a beam. ‘The design of a coupled-wall system would then involve adjusting the wall-o-coupling-beam strength and stiffness ratios so as to strike a balance between these conflicting requirements. A basis for choosing an appropriate beam-to-wall strength ratio, developed from ‘dynamic inelastic-response data on coupled-wall sys- tems, is indicated in Reference 9-55. Once the appropri ate relative strengths and stiflnesses have been estab- lished, details to insure adequate ductility in potential hiinging regions can be addressed Because of the relatively large shears that develop in deep coupling beams and the likelihood of sliding shear failures under reversed loading, the use of diagonal reinforcement in such elements has been suggested (see Figure 9-32) Tests by Paulay and Binney®* on diago- nally reinforced coupling beams having span-to-depth ratios in the range of 1 to 1} have shown that this arrangement of reinforcement is very effective in res ing reversed cycles of high shear. The specimens exhib- ited very stable foree-deflection hysteresis loops with Figure 8-32 Diagonaly reinforces coupling beam. (Adapted ‘rom Ret 9-56) CODE PROVISIONS FOR EARTHQUAKE-RESISTANT DESIGN 301 significantly higher cumulative ductility than compara~ ble conventionally reinforced beams. Tests by Bamey et al." on diagonally reinforced beams with span-to- depth ratios in the range of 2.5 to 5.0 also indicated that diagonal reinforcement can be effective even for these larger span-to-depth ratios. In the diagonally reinforced couplings beams re- ported in Reference 9-57, no significant flexural rein- forcement was used. The diagonal bars are designed to resist both shear and bending and assumed to function at their yield stress in both tension and compression, To prevent early buckling of the diagonal bars, Paulay and Binney recommend the use of closely spaced ties or spiral binding to confine the concrete within each bun- dle of diagonal bars. A minimum amount of “basketing reinforcement,” consisting of two layers of small-diame= ter horizontal and vertical bars, is recommended. The aid should provide a reinforcement ratio of at least 0.0025 in each direction, with a maximum spacing of 12 in, between bars. 9.4 CODE PROVISIONS FOR EARTHQUAKE-RESISTANT DESIGN 9.4.1 Performance Criteria The performance criteria implicit in most earthquake code provisions require that a structure be able to: 1, Resist earthquakes of minor intensity without damage; a structure would be expected to resist such frequent but minor shocks within its elastic range of stresses, 2. Resist moderate earthquakes with minor structural ‘and some nonstructural damage: with proper design and construction, it is expected that structural damage ‘due to the majority of earthquakes will be limited to, repairable damage. 3. Resist major catastrophie earthquakes without col lapse. While no clear quantitative definition of the earth ‘quake intensity ranges mentioned above has been given, the criteria provide a general indication of what should be expected in terms of overall performance of struc tures designed in accordance with the provisions of codes. It is clear from the above criteria that the principal concern in earthquake-resistant design of buildings is t0 assure life safety, that is, to prevent collapse under the ‘most intense earthquake that may reasonably be ex pected at the site during the life of a structure. More recently, however, engineers and public officials have ‘become increasingly aware of the need to insure that buildings housing facilities essential to post-earthquake operations, such as hospitals, power plants, fire stations, and communication centers, not only survive without‘302 SEISMIC DESIGN OF REINFORCED-CONCRETE STRUCTURES collapse but remain operational after a severe earth- quake. This means that such buildings should suffer a ‘minimum of damage. Thus, for these facilities, damage ‘control has been added to life safety as a major design objective, Often, damage control becomes desirable from a purely economic standpoint. The extra cost of prevent ing severe damage to nonstructural components of a building, such as partitions, glazing, ceilings, elevators, and other mechanical and electrical /electronie systems, may be justified by the savings realized in replacement costs and from continued use of a building after a strong earthquake. 4.2 Code-Specitied Design Lateral Forces The availability of dynamic analysis programs (see Ref- erences 9-59 to 9-64) designed for use with large-capac- ity digital computers has made possible the analytical estimation of earthquake-induced forces and deforma- tions in reasonably realistic models of most structures, However, except perhaps for the relatively simple analy- sis by modal superposition using response spectra, such dynamic analyses, which can range from a linearly clastic time-history analysis for a single earthquake record to nonlinear analyses using a representative en- semble of accelerograms, are costly and may be eco- nomically justifiable as a design tool only for a few large and important structures. At present, when dynamic time-history analyses of a particular building are under- taken for the purpose of design, linear elastic response is generally assumed. Nonlinear (inelastic) time-history analyses are carried out mainly in research work. For the design of most buildings, reliance will usually have to be placed on the simplified prescriptions found in most codes. Although necessarily approximate in character—in view of the need for simplicity and ease of application—the provisions of such codes and the philosophy behind them gain in reliability as design ‘guides with continued application and modification to feflect the latest research findings and lessons derived from observations of structural behavior during earth- quakes. Code provisions must, however, be viewed in the proper perspective, that is, as minimum require- ‘ments covering a broad class of structures of more or less conventional configuration. Unusual structures must still be designed with special care and may call for procedures beyond those normally required by codes. The basic form of modem code provisions on earth- ‘quake-resistant design has evolved from rather simpli- fied concepts of the dynamic behavior of structures and hhas been greatly influenced by observations of the per- formance of structures subjected to actual earth- quakes." Tt has been noted, for instance, that many structures built in the 1930s and designed on the basis of more or less arbitrarily chosen lateral forces have successfully withstood severe earthquakes. The satisfac- tory performance of such structures has been atributed to one of more of the following"; (i) yielding in critical sections of members (yielding not only may have increased the period of vibration of such structures to values beyond the damaging range of the ground mo- tions, but may have allowed them to dissipate a sizable Portion of the input energy from an earthquake); (i) the greater actual strength of such structures resulting from so-called nonstructural elements which are generally jgnored in analysis, and the significant energy-dissipa- tion capacity that cracking in such elements repre- sented; and (ii) the reduced response of the structure due to yielding of the foundation ‘The distribution of the code-specified design lateral forces along the height ofa structure is generally similar ‘o that indicated by the envelope of maximum horizon- tal forces obtained by elastic dynamic analysis. These forces are considered serice loads, ie, to be resisted within a structure's clastic range of stresses. However, the magnitudes of these code forces are substantially smaller than those which would be developed in a structure subjected to an earthquake of moderate-to- strong intensity, such as that recorded at El Centro in 1940, ifthe structure were to respond elastically to such ground excitation. Thus, buildings designed under the present codes would be expected to undergo fairly large deformations (four to six times the lateral displacements resulting from the code-specfied forees) when subjected to an earthquake with the intensity of the 1940 El Centro. These large deformations will be accompa- nied by yielding in many members of the structure, and in fact, uch is the intent of the codes. The acceptance ‘of the fact that iti economically unwarranted to desigm buildings to resist major earthquakes elastically, and the recognition of the eapacity of structures possessing ade quate strength and ductility to withstand major earth uakes by responding inelastically to them, lies behing the relatively low forces specified by the codes. These reduced forces are coupled with detailing requirements designed to insure adequate inelastic deformation ex pacity, i.e, ductility. The capacity of an indeterminate structure to deform in a ductile manner, that is, t@ deform well beyond the yield limit without significant loss of strength, allows such a structure to dissipate ‘major portion ofthe energy from an earthquake without serious damage 8.4.3. Principal Earthquake-Design Provisions of ANS! ‘AS8.1-1982, UBC-85, and Appendix A of ACI 318-83, Relating to Reinforced Concrete ‘The principal steps involved in the design of quake-resistant castin-place reinforced concreteings, with particular reference to the application of the Provisions of nationally accepted model codes or stan- dards, will be discussed below. The minimum design loads specified in ANSI A58.1-1982, “Minimum Design Loads for Buildings and Other Structures” (ANSI- 82), and the design and detailing provisions con. tained in Appendix A of ACI 318-83, Building Code Requirements for Reinforced Concrete, will be used as bases for the discussion. Emphasis will be placed on those provisions relating to the proportioning and de- tailing of reinforced concrete elements, the subject of the determination of earthquake design forces being treated in Chapter 4, Where appropriate, reference will bbe made to differences between the provisions of these model codes and those of related codes. Among the ‘more important of these are the NEHRP Recommended Provisions for the Development of Seismic Regulations for ‘New Buildings (NEHRP-85),° which is based on ATC 3.06, and the latest edition of the Recom. ‘mended Lateral Force Requirements of the Structural Engineers Association of California (SEAOC-86).°29 ‘The ANSI-82 provisions relating to earthquake de- sign loads are based mainly on those found in the 1979 Edition of the Uniform Building Code (UBC-79; the requirements are essentially the same as those appearing in UBC-85), with a few elements taken from ATC 3.06 ‘The UBC earthquake-design load requirements are gen. erally based on the SEAOC Recommendations. Except for minor modifications, the design and detailing re- {quirements for reinforced concrete members found in UBC-85, NEHRP-85, and SEAOC-86 (UBC-88) are essentially those of ACI Appendix A. The earthquake Aesign provisions of the National Building Code of Canada—1985 (NBCC-85)°") are, except for minor differences, essentially the same as those found. in ANSI-82 and ACI Appendix A. Although the various code-formulating bodies in the United States tend to differ in what they consider the ‘most appropriate form in which to cast specific provi. sions and in their judgment of the adequacy of certain design requirements, there has been a tendency for the different codes and model codes to gradually take cer. tain common general features. And while many ques. tions await answers, it can generally be said that the ‘main features of the earthquake-resistant design provi- sions in most current regional and national codes have g00d basis in theoretical and experimental studies as well as field observations. As such, they should provide Teasonable assurance of attainment of the stated objec- tives of earthquake-resistant design. The continual re- finement and updating of provisions in the major codes to reflect the latest findings of research and field ob- servations” should inspire increasing confidence in the soundness of their recommendations ‘CODE PROVISIONS FOR EARTHQUAKE-RESISTANT DESIGN 309) ‘The following discussion will focus on the provisions of ANSI-82 and ACI Appendix A, with occasional references to parallel provisions of NEHRP-85 and SEAOC-86 (UBC-88), The design earthquake forces specified in ANSI-82 are intended as equivalent static loads, As its title indicates, ANSI-82 is primarily a load standard, defin- ‘ng minimum loads for structures but otherwise leaving ‘out material and member detailing requirements, ACI Appendix A, on the other hand, does not specify the ‘manner in which earthquake loads are to be determined, but sets down the requirements by which to proportion ‘and detail monolithic castin-place reinforced concrete members in structures that are expected to undergo inelastic deformations during earthquakes Principal Design Steps Design of a reinforced con- crete building in accordance with the equivalent static force procedure found in current U.S. seismic codes involves the following principal steps: 1, Determination of design “earthquake” forces: * Calculation of base shear corresponding to the computed or estimated fundamental period of vibration of the structure. (A preliminary design of the structure is assumed here.) * Distribution of the base shear over the height of the building. 2. Analysis of the structure under the (static) lateral forces calculated in step (1), as well as under gravity and wind loads, to obtain member design forces. Note that the code-specified forces are to be considered as serviceslevel loads (as against factored or ultimate loads), ‘The lateral load analysis can be carried out most conve. niently by using a computer program for frame analysis. Use of dynamic analysis to estimate the seismic design forces in members would take the place of steps (1) and. 2) above. ANSI-82, however, requires that the design forces obtained from a dynamic analysis shall not be less than 90% of the corresponding forces obtained using the equivalent statie procedure. 3. Designing members and joints for the most unfa- vorable combination of graviiy and lateral loads. The emphasis here is on the design and detailing of mem- bers and their connections to insure their ductile behav ‘The above steps are to be carried out in each princi- pal (plan) direction of the building. ANSI-82 allows the design of a structure in each principal direction inde ‘pendently of the other direction on the assumption that the design lateral forces act nonconcurrently in each principal direction. NEHRP-85 and SEAOC-86 (UBC- '88) require consideration of orthogonal effects for cer304 SEISMIC DESIGN OF REINFORCED-CONCAETE STRUCTURES tain building categories which may be sensitive to torsional oscillations or characterized by significant ir- regularities. For these cases, the design isto be based on the more severe combination of 100% of the prescribed seismic forces in one direction plus 30% of the forces in the perpendicular direction ‘Changes in section dimensions of some members may be indicated in the design phase under step (3) above. However, unless the required changes in dimensions are such as to materially affect the overall distribution of forces in the structure, a reanalysis of the structure using the new member dimensions need not be under- taken. Uncertainties in the actual magnitude and disti- bution of the seismic forces as well as the effects of yielding in redistributing forces in the structure would make such refinement unwarranted. Its, however, most important to design and detail the reinforcement in members and their connections to insure their ductile behavior and thus allow the structure to sustain without collapse the severe distortions that may occur during a major earthquake. The code provisions intended to in- sure adequate ductility in structural clements represent the major difference between the design requirements for conventional, non-earthquakeresstant structures and those located in regions of high earthquake risk Load Factors, Strength Reduction Factors, and Loatl- ing Combinations Used as Bases for Design Coes gen- erally require thatthe strength of load-esisting capacity of a structure and its component elements be at least ‘equal to or greater than the forees due to any of a number of loading combinations that may reasonably be expected to act on it during its life. In the United States, concrete structures are commonly designed using the ultimate-strengti? method. In this approach, struc- tures are proportioned so that their (ultimate) capacity is equal to or greater than the required (ultimate) strength. The required strength is based on the most critical combination of factored loads, that is, specified service loads multiplied by appropriate load factors. The capacity of an element, on the other hand, is obtained by applying a strengti-reducton factor @ to the nominal resistance of the element as determined by code-pre- seribed expressions oF procedures or from basic me- chanics, Load factors are intended to take account of the different variabilties in the magnitude of the specitied Toads, lower load factors being used for types of loads that are less likely to vary significantly from the speci- fied values. To allow for the lesser likelihood of certain types of loads occurring simultaneously, reduced load “since ACI 31871, the term “ultimate” his been dropped, 0 that, ‘what used 10 be felered to as “ultimatestengih deiga” 8 now ‘Smply called “stength design.” factors are specitied for some loads when considered in combination with other loads. ANSI-82 requires that structures, their components, and their foundations be designed to have strengths not Jess than the most severe ofthe following combinations of loads 14D 12D + 1.6L + 05(L, or Sor R) 12D + 16(L, or or R) + (OSL or 0.8) 12D + 13W + OSL + 0.(L, or Sor R) 12D + 15E + (0.5L or 0.25) 0.90 ~ (1.30 oF 1.5E) (91) For garages, places of public assembly, and all areas ‘where the live load is greater than 100 Ib/t2, the load factor on L in the third, fourth, and fifth combinations in Equation 9-1 isto be taken equal to 1.0. Also, where the effects of F, H, P or T, as defined below, are significant, these are to be considered in design as the following factored loads: 1.3F, 1.64, 1.2P, and 127. In Equation 9-1, U = required strength to resist the factored loads D = dead load live load roof live load snow load rain load wind load ‘earthquake load = load due to fluids with well-defined pressures and maximum heights H = load due to soil pressure P = load due to ponding T= load due to effects of temperature, shrinkage, moisture changes, creep, differential settlement, or combinations thereof mame mate 1 ACI 318-83 specifies slightly different load factors for some load combinations, as follows LAD + 1.1L O.75[1.4D + 1.1L & (1.7 oF 1.87E)] 09D + (1.3W or 1.43) LAD + 1.7L + (1H or LAF) 0.9D ~ (1.H or L4F) O.95(1.4D + LIL + 147) > 1.4(D +7) (9-2)the design of earthquake-resistant structures, UBC- as well as SEAOC-86 (UBC-88), modifies the second third equations in Equation 9-2 to read as follows ya (1D+b+6) \osp +148 aid NEHRP-85 requires consideration of the following combinations in addition to those normally re- by governing code: (11 +054,)D+ 1004105 4108 (0.9 -054,)D £108 Lo There A, is the effective peak acceleration forthe site 4 indicated_on the coresponding map given in NEHRP5, Values of A, corresponding to the seven Afferent map areas range from 005 t0 0.4, the lowest alue being assigned to map area 1 and the highest to map arca 7 ‘As mentioned, the capacity of a structural element is aleulated by applying a strength reduction factor 4 to the nominal strength of the element. The factor 9's intended to take account of variations in material strength and uncertainties inthe estimation ofthe nom: inal member strength, the nature ofthe expected failure mode, and the importance of a member to the overall Safety of the structure, For conventional reinforced concrete structures, ACI 318-83 spciis the fol Salues of the strength eduction factor ¢ 0.90 0.90 0.75 for flexure, with or without axial tension for axial tension for spirally reinforced members subjected to axial ‘compression, with or without flexure 0.70 for other reinforced members (tied columns) sub- jected to axial compression, with or without flexure (an increase in the $ value for members subjected to combined axial load and flexure is allowed as the loading condition approaches the cease of pure flexure) oss 0.70 for bearing on concrete for shear and torsion ACI Appendix A specifies the following exceptions to the above values of the strength-reduction factor as siven in the main body of the ACI Code 1. For structural members other than joints, a value @ = 0.60 is to be used for shear when the nominal shear Strength of a member is less than the shear correspond- ing to the development of the nominal flexural strensth ‘of the member. For shear in joints, = 0.85. ‘CODE PROVISIONS FOR EARTHOUAKE-RESISTANTDESIGN 305 2. For frame members subjected to flexure and am compression exceeding 4, ///10-—where the gross cross-sectional area ahd the transverse forcement provided does not conform to the more gent requirements for lateral reinforcement for quake-resistant structures (see below, “Code Provisions Designed to Insure Ductility in Reinforced Concrete Members, item 3, paragraph (0). 6 ~ 0.50. Exception 4 above applies mainly to low-rise walls or portions of walls between openings. Exception (2) is intended t0 discourage the use of ties as lateral reinforcement in columns designed for earthauake resistance Code Procisions Designed to Insure Ductilty in Rein- forced Concrete Members The principal provisions of Appendix A of ACI 318-83” will be discussed below. AAs indicated earlier, the requirements for proportioning and detailing reinforced concrete members found in UBC-85, NEHRP-S, and SEAOC-86 (UBC8) ate essentially those of ACT Appendix A. Modifications to the ACI Appendix A provisions found in NEHRDAS and SEOAC-%6 (UBC-88) will be referred to. where appropriate Special provisions governing the design of earth- ‘uake-esistant structures first appeared in the 1971 edition ofthe ACI Code. The provisions of Appendix A supplement or supersede those in the main body of the code and deal withthe design of ductile moment-esist- ing space frames and shear walls of castin-place rein- forced concrete. Neither the main body of the code nor Appendix A specifies the magnitude of the earthquake forces 10 be used in design, The Commentary to Appendix A states that the provisions of the Appendix are intended to enable structures to sustain series of oscillations inthe inelastic range without critical loss in strength. It is generally accepted that the intensity of shaking envie sioned by the provisions of Appendix A correspond to those of UBC seismic zones 3 and 4, In the 1983 edition of the ACI Code, a section (Section A.9) was added to cover the design of frames located in areas of moderate seismic risk, roughly corresponding. to UBC. seismic zone 2, For structures located in areas of low seismic Tisk (corresponding to UBC seismic zones 0 and 1) and designed for the specified earthquake forces, very little inelastic deformation may be expected. In these cases, the ductility provided by designing to the provisions contained in the main body ofthe code will gonerally be ‘sufficient. ‘A major objective of the design provisions in ACL Appendix A, as well a in the main body of the code, i to have the strength of a structure governed by a ductile type of flexural failure mechanism. Stated another way, the provisions are aimed at preventing the brittle or abrupt types of failure associated with i reinforced and overreinforced members failing in flex