In-Situ Evaluation of Two Concrete Slab Systems. I: Load Determination and Loading Procedure

Transcription

1 In-Situ Evaluation of Two Concrete Slab Systems. I: Load Determination and Loading Procedure Nestore Galati, M.ASCE 1 ; Antonio Nanni, F.ASCE 2 ; J. Gustavo Tumialan, M.ASCE 3 ; and Paul H. Ziehl, M.ASCE 4 Abstract: The primary objective of in-situ load testing is to assess the safety and serviceability of an existing structural system with respect to a particular load effect. At this time, the most appropriate loading level and procedure, as well as the associated evaluation criteria are being reconsidered in light of technological advances in construction methods, analytical tools, and monitoring instrumentation. The in-situ load test method for reinforced concrete systems described in the ACI Building Code Requirements for Structural Concrete, namely the 24 h load test method and its evaluation criteria, has been in use for several decades, but may no longer serve the needs of contemporary construction and engineering practices. As a result, other load test methodologies and associated evaluation criteria are under development. This paper and a companion paper describe the rationale and application of an alternative approach to the determination of load level, loading procedure, instrumentation requirements, evaluation criteria and outcomes for two field projects. The first case study is relative to a posttensioned concrete slab where many areas were characterized by tendon and reinforcement misplacement, resulting in inadequate flexural strength and inadequate shear/flexure transfer at column/slab intersections. The second case study is the structural evaluation of a typical floor bay of a two-way reinforced concrete slab system, presenting distributed cracking at the positive and negative moment regions. Finite-element-method models were created for both structures to aid the load test design. The numerical models validated the field observations. DOI: / ASCE :4 207 CE Database subject headings: Concrete structures; Field tests; Instrumentation; Load tests; In situ tests; Concrete slabs. Introduction In-situ load testing is relevant for a variety of reasons including assessment of the effect of design and construction omissions and deficiencies; novel strengthening and retrofit methods; capability of an existing structure to carry loads different from the original design; and, safety of structures that have experienced corrosion and degradation. Presently, the default method for in-situ load testing of concrete structures is that prescribed in Chapter 20 of the Building Code published by the American Concrete Institute ACI Committee This load test method and its evaluation criteria are widely referred to as the 24 h load test because the test load is held in place on the structure for a period of 24 h. 1 Design Engineer, Strengthening Division, Structural Group, Inc., 7455 New Ridge Rd., Ste. T, Hanover, MD ngalati@ structural.net 2 Professor and Chair, Dept. of Civil, Architectural, and Environmental Engineering, Univ. of Miami, 225 MacArthur Engineering Bldg., Coral Gables, FL nanni@miami.edu 3 Senior Staff Engineer, Simpson Gumpertz & Heger, Inc., 41 Seyon St., Ste. 500, Waltham, MA gtumialan@sgh.com 4 Associate Professor, Dept. of Civil and Environmental Engineering, Univ. of South Carolina, 300 Main St., Columbia, SC corresponding author. ziehl@engr.sc.edu Note. Discussion open until January 1, Separate discussions must be submitted for individual papers. To extend the closing date by one month, a written request must be filed with the ASCE Managing Editor. The manuscript for this paper was submitted for review and possible publication on September 10, 2007; approved on January 15, This paper is part of the Journal of Performance of Constructed Facilities, Vol. 22, No. 4, August 1, ASCE, ISSN /2008/ /$ There are two drawbacks associated with this approach. Recent work conducted by ACI Committee has shown that the existing evaluation criteria of ACI Chapter 20 were developed for simply supported members using working stress design principles and material properties consistent with technology available in the 1920s concrete strength in the range of 2,000 psi 13.8 MPa and low yield-strength reinforcing steel. Such criteria are therefore not directly applicable to the majority of modern structural systems and may not be relevant to today s construction and engineering practices. Further, the load factors and the resistance factors have changed with evolving building codes and minimum design load requirements, whereas a similar evolution has not been reflected in the test load levels and the corresponding evaluation criteria. The second drawback of the existing load test method is related to feasibility and economics. This arbitrary length of loading time 24 h is only an apparent simulation of longterm effects and is more probably the consequence of using dead weight e.g., sand, cement bags, water as the loading medium. In addition, because residual displacements are to be measured 24 h after the test load has been removed, the total test duration not including setup requires 48 h to complete. Making use of modern equipment, instrumentation, and analytical tools, researchers and practitioners have attempted to develop an alternative test method that is more economical and, more importantly, provides much improved information regarding the behavior of the structural system of interest. This alternative method is referred to as the cyclic load test CLT method Gold and Nanni 1998; Nanni and Gold 1998a,b; Mettemeyer and Nanni 1999; Galati et al. 2004; Casadei et al. 2005; ACI With the JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES ASCE / JULY/AUGUST 2008 / 207

2 Fig. 1. Loading profile for CLT method ACI 2004 CLT method, the load is typically applied with hydraulic jacks in stepped loading and unloading cycles that increase in magnitude as the test progresses. Deflection, strains and other parameters of interest are recorded continuously during the load test and the structural response is evaluated with criteria that compare the linearity, repeatability, and permanency of the structure. Apart from the use of more modern technology, the novelty of the CLT method is in its ability to query the structure with repeated load cycles rather than a constant load applied for a predetermined length of time. The CLT method also has drawbacks: Due to its recent introduction, there is a significant lack of historical data to provide confidence in the method. Only on rare occasions, this method has been used on structures up to failure and, therefore, a calibration of the residual strength and evaluation criteria is not readily available. Comparisons with the 24 h test method are scarce. This is due to the fact that owners cannot justify the expenses of research and funding agencies perceive this topical area as not sufficiently fundamental. This paper and its companion Ziehl et al describe the CLT in-situ evaluation of two facilities a parking garage facility and a library in order to introduce principles and outcomes of the load test method in the context of likely projects. These case studies represent an ideal test bed for the CLT procedure because the use of the traditional test method could have posed a safety threat. The first structure was deficient because of misplacement of posttensioning tendons and mild reinforcement, whereas the second facility presented diffuse cracking. The first paper focuses on the determination of the load level and the loading procedure for each structure. Special considerations related to the design and conduct of this type of load test are presented and critically discussed. The companion paper focuses on evaluation criteria and their significance, limitations and applicability. Research Significance There is a need for a safe and reliable in-situ evaluation methodology of concrete structures. The CLT method, which consists of the use of stepped loading and unloading, such that changes in behavior of the structure become the basis for its evaluation, may possess these attributes. With the CLT procedure, there are no requirements for an arbitrary load hold period and measurement of deflection recovery long after the test load has been removed. Fig. 2. Test area and location of loading points two-way PT slab parking garage The CLT method is also inherently safe as the load is applied progressively and a sudden movement of the structure would correspond to an immediate drop in the load as applied by hydraulic jacks. Determination of the test load level and loading procedure are not entirely straightforward and are here exemplified using two case studies recently undertaken. Background Research Load testing of concrete structures in the United States is a century old tradition with one of the earliest well-documented cases to be found in the 1890s Birkmire In the early days, in-situ load testing was the most direct proof of performance of proprietary and novel, at the time, construction materials and methods. The American Concrete Institute began formalizing load test procedures for concrete structures in 1920 ACI At that time, the evaluation criteria for passing the load test focused on maximum deflection under sustained load combined with the recovery of deflection after the test load was removed. Subsequent codes ACI 1936 defined the deflection evaluation criterion as a function of the span length squared and divided by the total depth of the member cross section. This form of the deflection criterion is still in effect ACI Notable investigations into load testing of concrete structures documenting the practice of the last decades can be found in the literature FitzSimons and Longinow 1975; RILEM 1984; Bungey The cyclic load test method described in this paper is a relatively recent development and therefore only a limited number of reported case studies exist Gold and Nanni 1998; Nanni and Gold 1998a,b; Mettemeyer and Nanni 1999; Galati et al. 2004; Casadei et al This method attempts to make use of advances in technology e.g., equipment, instrumentation and analytical tools to provide a safe and reliable procedure for structural evaluation consistent with contemporary construction and engineering practices and societal needs. Among the technological developments that may allow a quantum leap in the use of load testing for structural evaluation is acoustic emission AE ASTM AE evaluation is a standard practice in other applications, such as composite vessels ASME 2004a,b. As a passive means of nondestructive evaluation, AE requires some form of loading to create a sudden release of energy, such as that caused by crack growth. The transient surface waves that are caused by this sudden release of energy are 208 / JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES ASCE / JULY/AUGUST 2008

3 Table 1. Moment and Shear Demand and Capacity of Structural Members Two-Way PT Slab Parking Garage Test label Condition M u kn m kip ft CLT No. 1 Shear collar at face of shear collar M n kn m kip ft at face of shear collar and at face of exist. capital V u at d/2 from face of shear collar V n at d/2 from face of shear collar Objective Evaluate performance of shear collar increase of punching shear strength and reduction of flexural demand Evaluate crack widths at service CLT No h LT Shear collar Compare ACI 437 to ACI 318 procedure CLT No. 2 CFRP strengthened at face of exist. capital at face of exist. capital 1, at d/2 from face of exist. capital at d/2 from face of exist. capital Evaluate performance of CFRP strengthening increase of flexural and punching shear strengths detected and characterized with AE piezoelectric sensors. This monitoring technique happens to be a perfect match for the loading sequence of the CLT method. Evaluation of reinforced concrete RC with acoustic emission has been most widely applied in Japan in combination with a stepped loading procedure. Consequently, Japan has developed a recommended practice for AE evaluation of RC structures NDIS but without specifying a particular loading protocol. One very important aspect of acoustic emission evaluation is the observation that upon reloading to a certain level the emission is generally much reduced from that which occurred during the original loading. This effect is known as the Kaiser effect ASTM At higher levels of damage this effect begins to break down. When significant emission does begin during reloading at a lower level of load than was previously applied the Felicity effect is said to be present ASTM Because these effects are dependent upon both load history and load intensity most AE loading procedures use a pattern of loading, unloading, and then reloading either to the same level of load or a slightly lower level of load. This pattern enables a check for the Felicity effect. Description of Loading Procedures Two load test procedures are described in the following sections. The first is the 24 h monotonic uniform load test prescribed by ACI 318, whereas the second is the cyclic load test ACI 437.1R h Load Test This method is based on a relatively long-term duration of loading and it is used to evaluate whether a structure or a portion of a structure satisfies the safety requirements of the code. The total test load is maintained for a period of 24 h. The test load is then removed and a set of final readings is made 24 h after the removal of the test load. As the test load is generally applied similarly to the design load pattern, i.e., in a uniformly distributed manner, certain characteristics of the structure, such as load sharing and fixity of supports, do not need to be fully investigated before the load test begins as the structure will behave as it would under design conditions, and its ability to hold the design load will be determined directly by the load test ACI 437.1R-07. Preliminary calculations are typically done as a rough guide to correctly position the instrumentation where the maximum responses are expected. The downsides of this method are related to the application of a uniformly distributed load, which can be time consuming and difficult, especially when testing large areas or performing multiple tests within a structure. Also, the test duration is at least two days 24 h at maximum load and 24 h unloaded, assuming that retesting is not necessary. Cyclic Load Test The load is applied in cycles to discrete areas that have been selected to maximize specific responses investigated in the mem- Table 2. Planned Point Load P LL Values Two-Way PT Slab Parking Garage Test label Criteria Load combination P LL CLT No. 1 Moment at the outer face of shoring posts uniform direction ACI D W +1.6L ACI 318 b 1.15D+1.5L CLT No h LT a Moment at the outer face of shoring posts uniform direction ACI 318 b 1.15D+1.5L CLT No. 2 Moment at the face of the capital uniform direction ACI D W +1.6L ACI 318 b 1.15D+1.5L ACI 318-C c 1.2D W +1.6L a Equivalent load P LL was kept constant on the structure for 24 h. b Value under consideration by ACI 318. c Value proposed by ACI-318-C. JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES ASCE / JULY/AUGUST 2008 / 209

4 Fig. 3. Modification of the static scheme of the structure due to the installation of the shoring posts CLT No. 1 and 24 h LT two-way PT slab parking garage ber. In order to determine the required magnitude, quantity, and location of applied concentrated loads, a thorough understanding of the structure s characteristics is necessary, including the effects of load sharing and end fixity ACI 437.1R-07, which requires relatively complex modeling. The CLT typically makes use of hydraulic jacks controlled by hand or electric pumps, assuring that the load can be removed in a matter of seconds. Increasing the loading unloading cycles up to a predetermined maximum load level allows the engineer a real-time assessment of member characteristics e.g., linearity and repeatability of response, as well as permanency of deformations. The duration of the cyclic load test is a few hours. The procedure of a cyclic load test consists of the application of loads in a quasistatic manner in at least six loading/unloading cycles see Fig. 1 ACI 437.1R-07. CLT has acceptance criteria to be checked during and after the load test: repeatability, permanency, and deviation from linearity. All are related to the response of the structure and they are described in detail in the companion paper Ziehl et al Fig. 5. Loading and measuring equipment two-way PT slab parking garage Case Histories The following sections report two case studies on the application of the CLT method to assess the structural performance of two structures presenting different types of deficiencies. For both structures, numerical models based on the finite element method FEM were developed to predict the intensity of the concentrated forces that when applied to each structure would produce the same effect, in terms of bending moments and/or shear forces, resulting from the target factored, uniformly-distributed load combination. Two-Way Posttensioned Concrete Slabs in a Parking Garage Structure Fig. 4. Schematic of the load tests two-way PT slab parking garage Description of the Structure The garage decks are two-way, posttensioned PT concrete slabs, supported by circular and square columns with capitals. The concrete slab is mostly 165 mm 6.5 in. thick. The original drawings indicated a nominal concrete strength of 28 MPa 4,000 psi and minimum steel yield strength of 414 MPa 60 ksi for the mild reinforcement. The tendons consisted of low relaxation, 1,860 MPa 270 ksi, seven-wire strands subjected to an effective stress of 1,213 MPa 176 ksi after losses. The slabs showed cracking on the slab topside, extending between columns. A field investigation revealed that the tendon and mild reinforcement was misplaced at the negative moment regions, resulting in inadequate calculated flexural strength and in- 210 / JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES ASCE / JULY/AUGUST 2008

5 Fig. 6. DCDT and strain gauge layout two-way PT slab parking garage Fig. 7. AE sensor layout two-way PT slab parking garage adequate calculated shear/flexure transfer at the column/slab intersections. Two areas of a typical slab were load tested to determine the effectiveness of the two strengthening techniques selected as repair options. For brevity, only the results from one of the areas shown in Fig. 2 are described and further information can be found in Galati and Nanni At this location, the first test, labeled as CLT No. 1, was performed to evaluate the effectiveness of shear collars in increasing the punching shear strength and, at the same time, reducing the flexural demand by shifting the position of the critical section for bending analysis. Both the CLT and the 24 h techniques were used with this configuration. At the same location, a second test, coded as CLT No. 2, was performed with the purpose of evaluating the use of a carbon fiber reinforced polymer CFRP material system to increase the flexural strength and, indirectly, the shear strength as the effective depth of the reinforcement could be increased. Both load tests were intended for negative moment evaluation in correspondence of column F2 as shown in Fig. 2. The test procedure involved applying concentrated loads at predetermined locations of the floor slab and monitoring of its response in the vicinity of the applied loads. Load Intensity ACI 437.1R recommends that the load intensity as provided in Chapter 20 of be redefined. In this instance, as only part of the structure is to be engaged and moment redistribution occurs, the test load magnitude, TLM including dead load already in place is the largest of or or TLM = 1.3 D w + D s TLM = 1.0D w + 1.1D s + 1.6L L r or S TLM = 1.0D w + 1.1D s L r or S + 1.0L 3 where D w dead load due to the self-weight; D s superimposed dead load; L live loads produced by the use and occupancy of the building not including construction or environmental loads, such as wind load, snow load, rain load, earthquake load, flood load, or superimposed dead loads; L r roof live load produced during maintenance by workers, equipment, and materials or during the life of the structure by movable objects such as planters and people; and S snow load. For this building, the superimposed dead load is equal to zero, as are the snow and roof live loads; therefore, the test load magnitude is given by TLM = 1.0D w + 1.6L = 7.2 kpa 150 psf 4 The load was applied at four points distributed around the column of interest as shown in Fig. 2. The intensity of the applied load at each point was determined to produce the same effect in terms of negative moment resulting from the factored, uniformly distributed load defined by Eq. 4. Determination of Equivalent Loads Numerical models were made to determine the magnitude of the concentrated point loads that would produce the similar bending moment due to the factored uniformly distributed loads UDL 1 2 JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES ASCE / JULY/AUGUST 2008 / 211

6 Fig. 9. Load test area two-way RC slab building at the critical test section. A two-dimensional model was implemented using commercial FEM software Computers and Structures, Inc The model consisted of one-dimensional beam elements representing columns and a fine mesh of plate elements to represent the floor systems. The material properties of concrete were assumed to be isotropic and linear elastic using a concrete modulus of elasticity equal to E c =57,000 fc 24.8 GPa psi. The presence of a flexural crack along line F was modeled by reducing the flexural stiffness of the elements along this line. This assumption was validated with the experimental results collected in the field. The moment demands given in Table 1 were used as a reference for the test setup design. Shoring posts to simulate the presence of shear collars were used for CLT No. 1. After conducting CLT No. 1, the 24 h test was performed for comparison of the two methods. Table 2 summarizes the findings in terms of point loads P LL determined prior to testing. The equivalent point loads for Area 1 in its original configuration are significantly lower than when using the shoring posts. This is because their presence modified the static scheme of the structure Fig. 3 by acting as an elastic support next to the locations where the moments were to be calculated, resulting in much lower loads for those tests in which they were used. Fig. 8. Load cycles for the three tests two-way PT slab parking garage Load Testing and Measurement Apparatus Fig. 4 shows an overall schematic of the push-down test. Shoring was installed on one floor above the tested zone to provide contrast for the hydraulic jacks. Wood bearing pads were used between the spreader beams and the structural floor to protect the concrete from localized damage. Table 3. Moment Capacity and Demand Two-Way RC Slab Building Location Level B, column strip defined by Line 12 in correspondence of Column H M n kn m kip ft M u kn m kip ft V n V u Objective Evaluate performance of the slab to negative moments Table 4. Point Load P LL and Resulting Effects Two-Way RC Slab Building Test label P LL Strip width m ft M u,tlm kn m kft M u,test kn m kft V u,tlm V u,test Load Line Load Line / JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES ASCE / JULY/AUGUST 2008

7 Fig. 10. Negative moments generated by uniform dead load and concentrated load two-way RC slab building Fig. 12. DCVT layout two-way RC slab building The equipment used consists of four kn 220 kip hydraulic cylinder jacks and pump, direct current differential transducers DCDTs for measuring deflections, two load cells 900 and 1,800 kn 100 and 200 kip, and eight R6I acoustic emission sensors resonant in the vicinity of 60 khz see Fig A data acquisition system recorded values at a rate of 1 Hz, displaying the load versus deflection curves in real time. Acoustic emission data were recorded separately with an eightchannel, two-high-speed-board system. Deflection measurements were taken in 14 different locations. The layout of the DCDTs is shown in Fig. 6 a. For CLT No. 2, a total of eight strain gauges were also placed on the CFRP reinforcement distributed in correspondence with the repaired cracks Fig. 6 b. Because the slab was loaded in negative moment, the AE sensors were mounted on the compression face of the concrete slab underside in a grid pattern with outside dimensions of m 5 15 ft centered on Column F2 Figs. 7 a and b. High vacuum grease was used as couplant and contact was maintained with specially fabricated magnetic hold-down devices. The evaluation threshold used was 45 db. The wavespeed in the slab was determined on-site based on time-of-arrival with pencil lead breaks used as simulated acoustic emission sources. CLT Loading Procedure Figs. 8 a c show the applied load cycles. The actual load cycles may vary slightly depending on the performance of the system as monitored during the test and the minimum load that has to be maintained to eliminate slack in the system. The applied load cycles do not start from zero to account for the weight of the testing equipment that was measured to be 360 kg 800 lb per loading point. The 24 h load test profile is shown in Fig. 8 b. Two-Way Reinforced Concrete Slab in a Building Structure This section describes one of two load tests performed on Level B2 of a library building. The aim of the load test was to assess the structural performance of the two-way RC slab system for the original design loads by monitoring the negative bending moment capacity. The selected area for the test is shown in Fig. 9. Description of the Structure The structural floor is a two-way slab supported by rectangular columns. The concrete slab is mostly 265 mm 10.5 in. thick. The material characteristics indicated a nominal concrete strength of 20.7 MPa 3,000 psi and minimum steel yield strength of mild reinforcement of 275 MPa 40 ksi. Further information can be found in Galati and Nanni Fig. 11. Schematic of the load test two-way RC slab building Fig. 13. Loading equipment two-way RC slab building JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES ASCE / JULY/AUGUST 2008 / 213

8 Table 5. Experimental and Analytical Results CLT No. 1 DS10 Two- Way PT Slab Parking Garage Load cycle max mm in. max mm in. FEM Fig. 14. AE sensor layout two-way RC slab building Load Intensity The dead load was determined as the addition of the self-weight of the structure 193 Pa 130 psf and a super imposed dead load of 37 Pa 25 psf. The live load is 186 Pa 125 psf. The TLM as prescribed by ACI was used D + 1.7L = 543 MPa 365 psf 5 Table 3 lists the values of bending moment and punching shear capacities of the existing slab and the factored forces due to the design loads. These values were determined for a column strip of 3.2 m 10.5 ft width. In Table 3, M n and V n existing slab bending and punching shear capacities and M u and V u factored bending and punching shear demands. The capacities were calculated considering the material properties at the time of the construction, without accounting for degradation over time. The reason for the load test was to verify if this assumption would hold as diffuse cracking was observed in the entire structure. Determination of Equivalent Loads Numerical models were implemented in order to determine the magnitude of the concentrated point loads that would produce the same negative bending moments of the factored UDL at the critical cross section as that caused by the design loads. A twodimensional model consisting of beam elements representing columns and plate elements representing the slab was created and analyzed with SAP Table 4 summarizes the findings in terms of required point loads, P LL. The load P LL was applied at each of four locations Fig. 9. At each point, the force was distributed over a mm in. area to avoid punching through the slab. It was not possible to apply the load symmetrically with respect to Column H12 due to the presence of piping. For this reason, different loads were applied at the Loading Lines 1 and 2 Fig. 9. Fig. 10 shows the negative moment distribution corresponding to the application of the UDL and to the test loads, respectively. Fig. 15. Load cycles two-way RC slab building Equipment and Measurement Apparatus The load test was performed in a push-down method using a procedure similar to the one used in the parking garage Fig. 11 with shoring installed on one floor above to provide contrast. Deflection measurements were taken with DCDTs mounted on tripods supported on the level below the one being tested Fig. 12. The wires from the data acquisition system were taken to the DCDTs through the air conditioning ducts. The testing equipment consisted of four 290 kn 66 kip hydraulic cylinder jacks and two hydraulic pumps, DCDTs for measuring deflections, and two load cells 450 and 900 kn 50 and 100 kip, respectively Fig. 13. A data acquisition system was set to record data at a rate of 3 Hz from all devices, providing real-time display of collected data as well as the load versus deflection curves. The acoustic emission data acquisition system was the same as that used for the parking garage. Six sensors were distributed on a grid 214 / JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES ASCE / JULY/AUGUST 2008

9 Table 6. Experimental and Analytical Results CLT No. 2 DS10 Two-Way PT Slab Parking Garage Load cycle max mm in. max mm in. FEM uncracked max mm in. FEM cracked max mm in. FEM cracked+permanency with dimensions of m 4 8 ft centered on Column H-12 Fig. 14. The sensor coupling devices and evaluation threshold were as described previously. CLT Loading Procedure The testing procedure was conceptually similar to the one used for the parking garage. Once all instruments were connected, a preload was applied to seat the components and to eliminate slack in the system. Following that, the slab was loaded in eight cycles. Four steps for loading followed by two steps for unloading were used. Each load step was maintained for at least 2 min. Fig. 15 shows the applied load cycles for the tested areas. The maximum load reached in the first two cycles corresponds to the service load level. The maximum load reached in Cycles 5 and 6 corresponds to the load combination suggested by ACI , whereas the last two cycles corresponded to the load level prescribed by ACI Chapter The applied load cycles do not start from zero to account for the weight of the testing equipment that was measured to be 3.5 kn 800 lb per loading point. Model Validation Good agreement between experimental and analytical results at the service load level stages was observed in all cases. Two-Way PT Concrete Slab Table 5 shows a comparison between the experimental and the theoretical results at the most demanding location DS10 for CLT No. 1. The model fits well the experimental results for CLT No. 1 as the behavior of the structure was elastic with no residual deflection measured when the load was removed. In fact, no new cracks were observed when performing the cyclic load test; the effect of the applied loads was mostly to increase the size of the existing cracks. For CLT No. 2 the preexisting crack was repaired filled by gravity feed of epoxy before installing the FRP strengthening and, therefore, two models were considered: cracked and uncracked. Table 6 shows a comparison between numerical and experimental results at the location monitored by DCDT DS10 for CLT No. 2. These results indicate that deflection predictions for the uncracked slab condition matches well the first four cycles. The measured deflection is closer to the calculated deflection based on a cracked slab condition, accounting for permanency, for the later cycles Fig. 16. Two-Way RC Slab Table 7 compares the FEM predictions with the experimental results recorded during the load test. The cracking of the slab was introduced in the model as a smeared cracking as the location of the existing cracks and crack formation was not known. These results indicate that deflection prediction for the uncracked slab condition matches the first two cycles, whereas the measured deflection is closer to the calculated deflection based on a cracked Fig. 16. Cycles deflection diagram for CLT No. 2 two-way PT slab parking garage Table 7. Experimental and Analytical Results DS12 Two-Way RC Slab Building Load cycle max mm in. max mm in. FEM uncracked max mm in. FEM cracked JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES ASCE / JULY/AUGUST 2008 / 215

10 Fig. 17. Cycles deflection diagram two-way RC slab building slab condition for the last two cycles Fig. 17. The slab started to correlate better with the theoretical cracked slab behavior at the third cycle. The effect of the permanency was not considered as it was negligible compared to the recorded deflections. Conclusions This paper presented the loading procedure to conduct a CLT in-situ evaluation of two facilities: a posttensioned parking garage facility and a library. For both structures, the load test was conducted without disrupting the normal operation of the facilities. FEM models were created for both structures and were well correlated with experimental observations. The selection of test load magnitude and the use of point loads to simulate distributed loading effects are described in detail. Both case studies represented the ideal test bed for the CLT procedure due to its inherent safety. In fact, both structures, one deficient due to misplacement of posttensioning tendons and the other with diffuse cracking, could have presented safety hazards. Acknowledgments The ACI Concrete Research Council, the NSF Industry/ University Cooperative Research Center on Repair of Buildings and Bridges with Composites and the UMR University Transportation Center on Advanced Materials are gratefully acknowledged for their financial support to the research. Yizhuo Chen assisted with the load tests and data reduction and his assistance is greatly appreciated. References American Concrete Institute ACI Standard building regulations for the use of reinforced concrete, Standard Specification No. 23, Detroit. American Concrete Institute ACI Building code regulations for reinforced concrete. ACI T, Detroit. American Concrete Institute ACI Strength evaluation of existing concrete buildings. ACI 437R-03, ACI Committee 437, Farmington Hills, Mich. American Concrete Institute ACI Building code requirements for structural concrete. ACI , Farmington Hills, Mich. American Concrete Institute ACI Test load magnitude, protocol and acceptance criteria. ACI 437.1R-07, ACI Committee 437, Farmington Hills, Mich. ASME. 2004a. Section V Nondestructive examination. Boiler and pressure vessel code, New York. ASME. 2004b. Section X Fiber-reinforced plastic pressure vessels. Boiler and Pressure Vessel Code, New York. ASTM Standard terminology for nondestructive examinations. ASTM E a, West Conshohocken, Pa. Birkmire, W Skeleton construction in buildings, Wiley, New York. Bungey, J The testing of concrete in structures, 2nd Ed., Chapman and Hall, New York. Casadei, P., Parretti, R. Nanni, A. and Heinze, T. 2005, In-situ load testing of parking garage RC slabs: Comparison between 24-h and cyclic load testing. Pract. Period. Struct. Des. Constr., 10 1, Computers and Structures, Inc SAP 2000, Version II, Berkeley, Calif. FitzSimons, N., and Longinow, A Guidance for load tests of buildings. J. Struct. Div., 101 7, Galati, N., Casadei, P., Lopez, A., and Nanni, A Load test evaluation of Augspurger ramp parking garage, Buffalo, N.Y. Rep. No , Center for Infrastructure Engineering Studies, Univ. of Missouri-Rolla, Rolla, Mo. Galati, N., and Nanni, A Load testing of two post-tensioned concrete slabs at garage A. Windsor on the Plaza, Kansas City, Mo. Final Rep., Prepared for Simpson Gumpertz & Heger Inc., Univ. of Missouri-Rolla, Rolla, Mo. Galati, N., and Nanni, A In-situ structural evaluation of a twoway RC slab at the National Institute of Health, Bethesda, Maryland. Final Rep., Prepared for Simpson Gumpertz & Heger Inc., Univ. of Missouri-Rolla, Rolla, Mo. Gold, W. J., and Nanni, A In-situ load testing to evaluate new repair techniques. Proc., NIST Workshop on Standards Development for the Use of FRP for the Rehabilitation of Concrete and Masonry Structures, Tucson, Ariz., NISTR 6288, D. Duthinh, ed., Feb. 1999, 3-102/112. Mettemeyer, M., and Nanni, A Guidelines for rapid load testing of concrete structural members. Rep. No. 99-5, Center for Infrastructure Engineering Studies, Univ. of Missouri-Rolla, Rolla, Mo. Nanni, A., and Gold, W. 1998a. Evaluating CFRP strengthening systems in-situ. Concrete Repair Bull., Int. Concrete Repair Ins., 11 1, Nanni, A., and Gold, W. J. 1998b. Strength assessment of external FRP reinforcement. Concr. Int., 20 6, NDIS Recommended practice for in situ monitoring of concrete structures by acoustic emission. Japanese Society for Non- Destructive Inspection. RILEM Technical Committee 20-TBS General recommendation for statical loading test of load-bearing concrete structures in situ TBS2. RILEM technical recommendations for the testing and use of construction materials, E& FNSpon, London, Ziehl, P. H., Galati, N., Nanni, A., and Tumialan, J. G In situ evaluation of two concrete slab systems. II: Evaluation criteria and outcomes. J. Perform. Constr. Facil., 22 4, / JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES ASCE / JULY/AUGUST 2008