Case Study: Strengthening of Parking Garage Decks
`with Near-Surface-Mounted CFRP Bars
`J. Gustavo Tumialan, M.ASCE1; Milan Vatovec, M.ASCE2; and Paul L. Kelley, M.ASCE3
`
`Abstract: This paper describes a parking garage retrofitting project where near-surface-mounted carbon fiber-reinforced polymer 共FRP兲
`bars were used to strengthen the reinforced-concrete decks. The garage reportedly exhibited numerous signs of deterioration, such as
`excessive deflections of the decks at bay midspans, extensive concrete cracking, concrete spalling, etc. The results of the structural
`analyses indicated that several negative-moment slab regions were deficient to support the design loads. The amount of overstressing was
`up to 50% in some areas. The structural analysis considered as-built conditions and showed that deficiencies were predominantly due to
`misplacement of the negative-moment steel reinforcement. The paper summarizes the design approach for carbon FRP strengthening of
`the concrete decks and describes a load testing program used to evaluate their performance. Finally, a description of parameters and
`considerations used in development and implementation of the adopted strengthening strategy is also presented.
`
`DOI: 10.1061/共ASCE兲1090-0268共2007兲11:5共523兲
`
`CE Database subject headings: Fiber reinforced polymers; Load tests; Parking facilities; Concrete, reinforced; Case reports; Decks.
`
`Introduction
`
`light weight, unlimited available
`Their high tensile strength,
`lengths, and resistance to corrosion contributed to the emergence
`of fiber-reinforced polymer 共FRP兲 composite systems in structural
`strengthening and retrofit fields in the United States over the last
`10 years. After overcoming the initial growing pains 共insufficient
`experience, track record, and knowledge overshadowed by over-
`zealous euphoria with “magic” material capabilities兲, the manu-
`facturers, academia, and practicing engineers all contributed
`toward developing an ACI-sanctioned, design methodology
`backed by numerous experimental and material tests, extensive
`research, and friendly discussions. As a result, FRP is the go-to
`method for strengthening of concrete structures in many interior
`or exterior applications.
`This paper describes a project in Massachusetts, where near-
`surface-mounted 共NSM兲 carbon FRP 共CFRP兲 bars were placed on
`the top side of reinforced-concrete garage decks to repair cracking
`and to strengthen regions of the structure garage deck. NSM FRP
`bars have shown to be an effective method for strengthening con-
`crete and masonry structures 共Tumialan et al. 2002; Parretti and
`Nanni 2004兲. This paper discusses the structural analyses of the
`garage decks, based on as-built conditions, and a load-testing pro-
`gram used to evaluate their structural performance before and
`
`1Senior Staff Engineer, Simpson Gumpertz and Heger, Inc., 41 Seyon
`St., Building 1, Waltham, MA 02453.
`2Principal, Simpson Gumpertz and Heger, Inc., 1375 Broadway, Suite
`600, New York, NY 10018.
`3Senior Principal, Simpson Gumpertz and Heger, Inc., 41 Seyon St.,
`Building 1, Waltham, MA 02453.
`Note. Discussion open until March 1, 2008. 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 March 6, 2006; approved on May 16, 2006. This paper is
`part of the Journal of Composites for Construction, Vol. 11, No. 5,
`October 1, 2007. ©ASCE, ISSN 1090-0268/2007/5-523–530/$25.00.
`
`after strengthening with a NSM FRP system. The paper also sum-
`marizes the design approach for CFRP strengthening, and de-
`scribes parameters and considerations used in the selection and
`implementation of the strengthening system that practitioners can
`take into account when working on projects of this nature.
`
`Background
`
`The parking garage was constructed in 1983. It is an L-shaped,
`three-level concrete structure, with approximately 2,325 m2
`共25,000 sq ft兲 per level. Two-way stacked, concrete ramps, each
`approximately 465 m2 共5,000 sq ft兲, provide car access to the
`second and third levels of the garage. Fig. 1 shows the typical
`footprint and layout of the garage decks.
`The typical deck-to-deck height is 2.44 m 共8 ft兲. The bottom
`level of the garage is a slab on grade. The decks are supported by
`interior concrete columns with capitals and exterior concrete ma-
`sonry bearing walls. At each interior column, there is a 3.05
`⫻3.05⫻0.10 m 共10 ft-0 in.⫻10 ft-0 in.⫻4-1/4 in.兲 thick drop
`panel. The size of a typical garage bay is approximately 8.33
`⫻8.33 m 共27 ft-4 in.⫻27 ft-4 in.兲. The garage decks are two-
`way, 0.20 m 共8 in.兲 thick, mild-steel reinforced, concrete slabs.
`The steel reinforcement in the decks varies between different col-
`umn bays and between garage levels.
`The garage reportedly exhibited numerous signs of deteriora-
`tion, such as excessive deflections of the decks at bay midspans,
`extensive concrete cracking, etc., since the early 1990s. In 2003, a
`new owner commissioned a comprehensive study of the causes of
`deterioration and, as a result, implemented remedial actions.
`
`Condition Survey of the Garage Decks
`
`The objective of the condition survey was to determine param-
`eters that qualitatively can depict the condition of the garage
`decks, with the purpose of estimating the amount of necessary
`repair and structural strengthening. A review of those qualitative
`
`JOURNAL OF COMPOSITES FOR CONSTRUCTION © ASCE / SEPTEMBER/OCTOBER 2007 / 523
`
` J. Compos. Constr., 2007, 11(5): 523-530
`
`Downloaded from ascelibrary.org by University Of Miami on 03/12/20. Copyright ASCE. For personal use only; all rights reserved.
`
`Metromont Ex-1004, p.1
`
`

`

`Analytical Results: As-Built Condition
`
`The results of the structural analysis indicated that several
`negative-moment slab regions 共specifically column strips兲 of both
`garage decks were deficient to support the design loads required
`by the current code, as well as the code at the time of construc-
`tion. The overstress was up to 50% in some areas. The analysis
`considered as-built conditions and showed that deficiencies were
`predominantly due to misplacement of the negative-moment steel
`reinforcement. The positive moment strength and shear strength
`共one-way and two-way兲 of typical column bays on both levels
`were found to be adequate to support the code-prescribed loads.
`Also, the analysis showed that, even when long-term deflection
`effects were considered, the calculated expected midbay displace-
`ments were significantly lower than the ones observed in the field.
`The observed displacements were therefore likely a result of pre-
`mature formwork removal during construction.
`Even though the current building codes require that garage
`structures be designed for a 2.40 kPa 共50 psf兲 live load, most
`garages in service today only experience a live load of approxi-
`mately 1.44 kPa 共30 psf兲 due to car parking constraints. To evalu-
`ate this in-service condition, under the assumption that the decks
`would actually be exposed to lower live loads than those pre-
`scribed by the code, the garage decks were analyzed to support a
`live load of only 1.44 kPa 共30 psf兲 for the second-level deck and
`of 2.88 kPa 共60 psf兲 for the third level deck 共30 psf live load plus
`30 psf snow load兲. The results indicated that several negative-
`moment areas on both decks were still deficient.
`
`CFRP Threshold Strength Calculations
`
`The performance of a CFRP composite system rapidly deterio-
`rates during exposure to elevated temperatures, such as those
`caused by fire. Therefore, the use of a CFRP system in a strength-
`ening application dictated the demonstration that
`the original
`structure 共without CFRP兲 has enough reserve strength to resist the
`service loads in the short term without failure or collapse, in case
`the supplementary FRP reinforcement is lost due to fire. Accord-
`ing to recommendations provided by American Concrete Institute
`共ACI兲 Committee 440 共ACI 440.2R-02兲, a CFRP system can be
`used to strengthen a structure if it can be shown, either numeri-
`cally or through load testing, that the slab possesses the so-called
`“threshold strength” to resist a special factored combination of
`design dead and live loads 共1.2 DL+0.85 LL兲.
`To address this requirement, the second-level and third-level
`decks were analyzed for the “threshold” loads as defined by ACI-
`440 共2002b兲. The structural analysis showed that the typical, as-
`built negative-moment regions of the garage were deficient to
`support even the threshold loads. Therefore, a load-testing pro-
`gram was planned to demonstrate that the structure possessed the
`threshold strength.
`
`Selection of CFRP System
`
`The use of externally bonded CFRP sheets to strengthen the ga-
`rage slabs was considered early in the garage-retrofit plans. How-
`ever, abrasion, deicing salts, and oil from vehicular traffic may
`damage and cause deterioration of the CFRP laminates placed on
`the topside of the slab. To minimize premature deterioration risks,
`an epoxy overlay system can be utilized to protect the CFRP
`system, but this would significantly increase the cost of the CFRP
`repair.
`
`Fig. 1. Layout of garage deck and load-test
`=0.305 m兲
`
`locations 共1 ft
`
`parameters, along with the results of the structural analyses, was
`used to determine whether a particular garage bay required
`strengthening. The condition survey consisted of the following:
`• Performing a ground penetration radar 共GPR兲 survey of the
`location of the negative-moment reinforcing steel in all bays in
`both directions, and at some representative areas of the
`positive-moment steel. The average top cover for the negative-
`moment steel reinforcement ranged between 38 mm 共1.5 in.兲
`and 102 mm 共4 in.兲. The original drawings for the structure
`specified 38 mm 共1.5 in.兲 for concrete cover. Selective cores in
`the concrete decks confirmed that the top mat of steel bar
`reinforcement was misplaced during construction in the drop
`panel areas 共column head兲. The reinforcement cover was
`significantly larger than specified by the original structural
`drawings.
`• Measuring the existing vertical deflection due to dead load at
`midspan of each column bay. Deflections were as high as
`114 mm 共4.5 in.兲 in several garage bays.
`• Evaluating and qualifying a slab based on the amount of ob-
`served cracking and other distress on the top side of the slab.
`The survey focused only on cracks,
`larger than 0.25 mm
`共0.01 in.兲.
`• Sounding of concrete with steel chains dragged over the deck
`topside and with hammers from the underside to identify con-
`crete delamination.
`
`Structural Analysis of Concrete Decks
`
`Analytical Modeling
`
`The two elevated decks were modeled using a finite-element
`modeling software 共RAM Concept, RAM International, Carlsbad,
`Calif.兲. The objective for use of high-end analytical model was to
`capture the effects parameters such as geometry changes across
`the floor plan, boundary conditions associated with drop panels
`and columns, edge beams, walls, stair openings, etc.
`The code-required factored load combinations for the second
`level,
`included self-weight of the structure and live load of
`2.40 kPa 共50 psf兲. For the third level, the considered loads in-
`cluded self-weight and 3.84 kPa 共80 psf兲 of live load 关to account
`for additional 1.44 kPa 共30 psf兲 of snow required by the Massa-
`chusetts State Building Code 共1997兲兴.
`
`524 / JOURNAL OF COMPOSITES FOR CONSTRUCTION © ASCE / SEPTEMBER/OCTOBER 2007
`
` J. Compos. Constr., 2007, 11(5): 523-530
`
`Downloaded from ascelibrary.org by University Of Miami on 03/12/20. Copyright ASCE. For personal use only; all rights reserved.
`
`Metromont Ex-1004, p.2
`
`

`

`As an alternative, a NSM CFRP system placed in saw-cut
`grooves on the topside of the slab was considered. This system
`would eliminate the direct wear-and-tear concerns, plus it would
`allow for installation of a waterproofing membrane after strength-
`ening the decks. The deck waterproofing coating typically re-
`quires preparation of the surface by shot blasting, which would
`also cause problems if externally applied CFRP sheets were used
`for strengthening.
`Thus, the NSM CFRP system was selected. The high-strength
`CFRP bars would be used to supplement the existing negative-
`moment steel reinforcement, therefore providing sufficient calcu-
`lable strength to the slab to support the code-prescribed loads.
`The common geometrical-conflict concerns 共location of existing
`steel reinforcement兲 associated with NSM placement, damaging
`or cutting the existing steel reinforcement, was not an issue in this
`case as the steel reinforcement and the available concrete cover in
`most areas of the garage was deep.
`Two different commercially available materials, one epoxy-
`based and one cementitious-based, were mechanically tested to
`evaluate the bond between CFRP bars and concrete. The test re-
`sults were used to analyze the behavior and suitability of the
`embedding paste. A total of six specimens 共three for each kind of
`embedding paste兲 were tested as part of this pilot-test program.
`The test specimens consisted of two concrete blocks. A NSM
`CFRP bar was bonded to each face of the blocks in the longitu-
`dinal direction, connecting the two blocks together. Only one
`block was considered the test specimen, with the bar having a
`limited bonded length of 203 mm 共8 in.兲. The bar was fully
`bonded on the other block, to cause bond failure to occur in the
`test region. The test showed a better performance of the speci-
`mens with epoxy-based paste. As expected, the bars did not frac-
`ture in any of the tests. However, the failure in the specimens with
`epoxy-based paste was due to splitting in the concrete, whereas in
`the specimens with cementitious-based paste the failure occurred
`due to sliding of the CFRP bar within the paste.
`The primary function of the embedding paste is to transfer
`tensile stresses between the concrete substrate and the FRP rein-
`forcement, which is achieved by providing appropriate bond be-
`tween the two. The epoxy-based paste, if a part of the system,
`should only be used in the temperature range recommended by
`the manufacturer. To do otherwise will likely accelerate aging,
`causing the paste to become unstable, and, therefore, compromise
`the bond between the concrete and FRP reinforcement, and, even-
`tually, the effectiveness of the FRP-strengthening system.
`The manufacturers typically recommend that the paste not be
`exposed to temperatures exceeding 65°C for prolonged periods.
`Even though this possibility is remote in the Northeast region of
`the United States, a hygrothermal 共heat and moisture兲 analysis
`was conducted using a commercially available software 共WUFI,
`Fraunhofer Institute in Building Physics, Holzkirchen, Germany兲.
`The objective was to ensure that the embedding paste would have
`an adequate in-service performance by determining expected peak
`temperatures on the topside of the concrete deck during a typical
`year. The model included thickness of the concrete decks, mate-
`rial properties of concrete 共such as water-cement ratio and den-
`sity兲, and concrete storage and transfer capabilities of heat and
`moisture. The input for the model consisted of typical climatic
`conditions found in the Boston area during a typical year. Two
`conditions for deck exposure were evaluated, one considering a
`nonshaded condition 共third-level deck兲 and the other considering
`a shaded condition 共second-level deck兲. Fig. 2 illustrates the ex-
`pected peak temperatures on the concrete-deck topside, which are
`well below the maximum exposure temperature recommended by
`
`Fig. 2. Peak temperatures on concrete deck topside
`
`the manufacturer. Therefore, it was concluded that the epoxy em-
`bedding paste was suitable for use in this project.
`
`Design of CFRP Strengthening
`
`Prior to conducting the load tests, the CFRP-strengthening repair
`was designed for typical negative-moment regions. The CFRP
`reinforcement was designed to achieve moment strength as re-
`quired by the full code-prescribed loads.
`The flexural-strengthening concept consisted of ten #3 US cus-
`tomary 共9.5 mm diameter兲 CFRP bars, embedded into precut
`grooves evenly distributed within the column strip on the slab
`topside 共along the width of drop panel兲. The length of the CFRP
`bars was 4.27 m 共14 ft兲. Fig. 3 illustrates the typical CFRP
`strengthening configuration.
`The CFRP bars have the following mechanical properties re-
`* , of 206.9 MPa
`ported by the manufacturer: tensile strength, f fu
`共300 ksi兲, modulus of elasticity of 124.1 GPa 共18,000 ksi兲, and an
`ultimate strain of 1.7%. The bars have a dimpled and textured
`surface. The fiber content is 60% by volume in a vinylester resin.
`Currently, ACI-440 does not provide guidance on design of FRP
`NSM strengthening. However, as the ACI-440 equations for FRP
`laminates are based on principles of force equilibrium, strain
`compatibility, and constitutive laws of the materials, these can
`also be used for designing FRP NSM strengthening. Thus, as
`experimental values of ␬m have been found to vary between 0.60
`and 0.84 共Parretti and Nanni 2004兲, a conservative bond-
`dependent coefficient, ␬m, of 0.60 was used to determine the ef-
`fective stress limit in the CFRP bars. An environmental-reduction
`factor, CE, of 0.85 for exterior exposure was also used based on
`ACI-440 guidelines. The design ultimate strength,
`f fu, was
`106.9 MPa 共153 ksi兲, which was computed by multiplying f fu
`* by
`the previous “knock-down” factors. As a reference, Table 1 pre-
`sents and compares the existing flexural strength 共considering as-
`built conditions兲 with the moment demand and the CFRP-
`upgraded flexural capacity, which was computed using a
`procedure similar to the one included in ACI-440 for FRP
`laminates.
`
`Load Testing Program
`
`Load-Testing Rationale
`
`The next phase of the structural evaluation included development
`and implementation of a load-testing program. The objective of
`
`JOURNAL OF COMPOSITES FOR CONSTRUCTION © ASCE / SEPTEMBER/OCTOBER 2007 / 525
`
` J. Compos. Constr., 2007, 11(5): 523-530
`
`Downloaded from ascelibrary.org by University Of Miami on 03/12/20. Copyright ASCE. For personal use only; all rights reserved.
`
`Metromont Ex-1004, p.3
`
`

`

`Guidelines provided by ACI Committees ACI-318 共2002a兲 and
`ACI-437 共2003兲 stipulate that if the structure behaves linearly
`when loaded to 85% of the factored design loads 共in this case
`1.4 DL+1.7 LL兲, the structure can be considered adequate to sup-
`port the full design loads. This test, where the CFRP-strengthened
`structure is loaded to an equivalent of 85% of the factored design
`loads, was designated as Design-Load Test.
`In addition, another issue relative to the magnitude of test
`loads was considered. For the second-level deck, the magnitude
`of threshold loads 共1.2 DL+0.85 LL兲 was equal to 85% of the
`factored design loads 共1.4 DL+1.7 LL兲, if live load was consid-
`ered to be 1.44 kPa 共30 psf兲. Thereby, by load testing the second
`level of the garage to the CFRP threshold limits, while assuming
`2.40 kPa 共50 psf兲 live loads,
`the adequacy of unstrengthened
`decks to safely support the expected realistic in-service live loads
`of 1.44 kPa 共30 psf兲 would also be verified.
`
`Load-Test Matrix
`
`Two representative locations were identified for load testing on
`the second level deck 共Fig. 1兲. The selection of these areas was
`based on observed distress and measured steel-reinforcement
`misplacement. Area A exhibited the most distress in terms of de-
`flection, cracking, and concrete delamination, and top-steel-
`reinforcement placement was deepest at this location. The slab at
`Area B was representative of a typical second-level deck in terms
`of slab distress, steel-reinforcement misplacement, and degree of
`overstress. Table 1 presents the Load-Test matrix.
`
`Load-Test Setup
`
`A cyclic-load-test methodology for strength evaluation of existing
`concrete buildings, recommended by ACI-437, was used to load
`test the garage decks. The load-test setup was intended to maxi-
`mize the negative-moment demand on the slab at the two repre-
`sentative locations on the second level. The test setup applied four
`point loads, two on each side of the column. As the test loads
`were point loads, the loads were calibrated to create a moment
`and shear demand along the column strip similar to that of a
`uniformly applied load. Thus, the magnitude and location of the
`point loads generated by the load-test apparatus were determined
`using the finite-element modeling program. The analysis indi-
`cated that a pair of point loads at 3.05 m 共10 ft兲 away from the
`center of the column and spaced 1.525 m 共5 ft兲 apart, simulta-
`neously applied on opposite sides of the negative-moment region
`would achieve the objective. Table 1 shows the magnitude of the
`loads for each test and the associated load-test moments. It also
`shows the target moments for Threshold and Design-Load Tests.
`The loads were generated by hydraulic jacks that reacted
`against minipiles installed in the soil below the slab on grade. The
`load was transferred through high-strength steel rods into a steel
`reaction beam placed on top of the tested slab 共Figs. 4 and 5兲.
`For each load test, the intended maximum load was attained
`through six load cycles 共three pairs of two兲. The maximum load in
`each cycle represented a fraction of the total desired load for the
`test: The first two test cycles were taken to 50% of the final load,
`the second two were taken to 75% of the final load, and finally the
`last two cycles were taken to the desired load for that load test.
`Each load cycle consisted of a minimum of four approximately
`equal load steps, followed by approximately 2 min of constant
`load, and then at least two steps to unload the structure.
`During the tests, the loads and displacements were continu-
`ously monitored by load cells and linear variable displacement
`
`Fig. 3. CFRP strengthening configuration 共1 in=25.4 mm兲
`
`the garage decks had
`this program was to demonstrate that
`enough threshold strength reserve to “qualify” for strengthening
`with a CFRP system, as well as to verify the efficiency of the
`CFRP strengthening.
`A diagnostic cyclic load test method was used to evaluate the
`structure. The first series of tests was intended to simulate the
`threshold loads prescribed by ACI-440 共1.2 DL+0.85 LL兲; these
`tests were designated as Threshold-Load Tests. If the deck satis-
`fied the ACI-440 threshold requirement, and if the CFRP system
`qualified for strengthening, the load testing program would in-
`clude a second series of tests aimed at showing that the deck,
`when strengthened with CFRP, had sufficient strength to support
`the full code-prescribed loads. This test was needed because of
`the potential nonductile failure concerns caused by the presence
`of cracks in the critical shear region around the column and by
`calculated deficiencies in the negative-moment regions 共topside兲,
`and because of the use of NSM bars in this type of novel
`application.
`
`526 / JOURNAL OF COMPOSITES FOR CONSTRUCTION © ASCE / SEPTEMBER/OCTOBER 2007
`
` J. Compos. Constr., 2007, 11(5): 523-530
`
`Downloaded from ascelibrary.org by University Of Miami on 03/12/20. Copyright ASCE. For personal use only; all rights reserved.
`
`Metromont Ex-1004, p.4
`
`

`

`Table 1. Summary of CFRP Strengthening Design and Load-Testing Matrix
`
`␾Mnexist
`关kN m
`共kips ft兲兴
`
`Mu
`关kN m
`共kips ft兲兴
`
`a
`␾Mnstr
`关kN m
`共kips ft兲兴
`
`Area
`
`Test
`
`Load target
`
`Load testing
`
`Condition and
`load orientation
`
`MThreshold
`关kN m
`共kips ft兲兴
`
`0.85Mu
`关kN m
`共kips ft兲兴
`
`MLoad Test
`关kN m
`共kips ft兲兴
`
`A
`
`A
`
`A
`
`A
`
`B
`
`B
`
`381
`共280兲
`
`313
`共230兲
`
`381
`共280兲
`
`313
`共230兲
`
`291
`共214兲
`
`291
`共214兲
`
`525
`共386兲
`
`491
`共361兲
`
`525
`共386兲
`
`491
`共361兲
`
`404
`共297兲
`
`404
`共297兲
`
`NA
`
`NA
`
`526
`共387兲
`
`492
`共362兲
`
`NA
`
`415
`共305兲
`
`Unstrengthened slab
`Column strip along the
`east-west direction
`Unstrengthened slab
`Column strip along the
`north-south direction
`CFRP strengthened slab
`Column strip along the
`east-west direction
`CFRP strengthened slab
`Column strip along the
`north-south direction
`Unstrengthened slab
`Column strip along the
`north-south direction
`CFRP strengthened slab
`Column strip along the
`north-south direction
`aStrengthening configuration consisted of ten #3 共9.5 mm diameter兲 CFRP bars in all the cases.
`bLoad test was not conducted.
`
`EW-A
`
`1.2DL+0.85 LL
`
`NS-A
`
`1.2DL+0.85 LL
`
`EW-A-CFRP
`
`0.85共1.4DL+1.7 LL兲
`
`NS-A-CFRP
`
`0.85共1.4DL+1.7 LL兲
`
`NS-B
`
`1.2DL+0.85 LL
`0.85共1.4DL+1.7 LL兲
`
`NS-B-CFRPb
`
`0.85共1.4DL+1.7 LL兲
`
`381
`共280兲
`
`356
`共262兲
`
`NA
`
`NA
`
`294
`共216兲
`
`NA
`
`NA
`
`NA
`
`445
`共327兲
`
`417
`共307兲
`
`343
`共252兲
`
`NA
`
`374
`共275兲
`
`355
`共261兲
`
`443
`共326兲
`
`415
`共305兲
`
`340
`共250兲
`
`NA
`
`transducers connected to data acquisition equipment and by
`manual dial gauges 共Fig. 4兲. The vertical displacement of the slab
`was specifically measured at six locations, three on each side of
`the column 共one at the edge of drop panel, one on the slab beyond
`the drop panel, and one at midspan兲.
`
`Load-Test Results
`
`The following criteria for cyclic load testing were used to qualify
`slab performance as acceptable. The corresponding definitions are
`provided elsewhere 共ACI-437R-03兲:
`• Repeatability of structural response during the load test;
`• Cycle permanency; and
`• Deviation from linearity.
`Figs. 6 and 7 show the load-deflection curves 共envelope of
`cyclic loading兲 for tests designated as EW-A and NS-A, imple-
`mented on unstrengthened slabs in Area A 共Threshold Tests兲. It
`should be noted that the unloading portion of the load tests is not
`
`shown for clarity. Both tests show linear behavior. No significant
`residual deflection was observed, and the structure passed the test
`criteria recommended by ACI-437 共2003兲. The same figures also
`show the load-deflection results for Tests EW-ACFRP and NS-
`ACFRP, after the slab in Area A was strengthened 共Design Tests兲.
`It can be observed that the structure still behaved linearly. The
`residual deflections were acceptable, and the deck again passed
`the test criteria. The load versus deflection plots show a slight
`increase in stiffness 共slope of the load-deflection curve兲 after
`strengthening in each case.
`There was a difference in flexural stiffness 共slope of the load-
`deflection curve兲 between the east and west spans during the E-W
`tests in Area A. This difference in response can be attributed to
`the presence of a masonry wall at the far end of the west span,
`which may have stiffened that span. The difference in stiffness
`between spans for the N-W tests in Area A is not as pronounced,
`and it probably stems from a slightly larger bay length on the
`south side of the column.
`
`Fig. 4. Overall view of test setup—deck topside
`
`Fig. 5. Overall view of test setup—deck underside
`
`JOURNAL OF COMPOSITES FOR CONSTRUCTION © ASCE / SEPTEMBER/OCTOBER 2007 / 527
`
` J. Compos. Constr., 2007, 11(5): 523-530
`
`Downloaded from ascelibrary.org by University Of Miami on 03/12/20. Copyright ASCE. For personal use only; all rights reserved.
`
`Metromont Ex-1004, p.5
`
`

`

`Fig. 6. Load-deflection curves—Test Area A 共E-W direction兲
`
`Fig. 8. Load-deflection curves—Test Area B 共N-S direction兲
`
`Two load tests were initially planned at Area B, both in the
`same direction 共N-S兲, before and after CFRP strengthening. How-
`ever, based on the initial response during the Threshold Test, and
`the fact that the observed distress and steel placement were not as
`pronounced as at Area A, it was decided to test the unstrength-
`ened structure in this location to the full design loads 共Design
`Test兲. Therefore, only one test was performed at this location
`共Test NS-B兲. Fig. 8 shows the load-deflection curves for this test.
`The analysis of the test results based on ACI-437 criteria indi-
`cated that the unstrengthened slab passed the Design Test criteria
`at this location. The load versus deflection plots show that there is
`no difference in flexural stiffness between the slab on the south
`side and the slab on the north side of the column.
`
`Load-Test Discussion
`
`The load tests showed that the garage bays with most pronounced
`deficiencies, both in terms of reinforcing-steel placement and in
`terms of deflection and delamination 共Test Area A兲, were ad-
`equate to support the ACI-440 threshold loads, which made them
`suitable for strengthening with CFRP; the test showed that all
`garage decks 共regardless of the level兲 are capable of supporting
`expected in-service loads in the short term in case the CFRP
`system is lost due to fire, vandalism, or any other reason.
`
`Fig. 7. Load-deflection curves—Test Area A 共N-S direction兲
`
`The load tests at Test Area A also showed that the most critical
`garage bay on the second floor, in an unstrengthened configura-
`tion, was also able to safely support expected in-service loads of
`1.44 kPa 共30 psf兲 based on placement of passenger cars and light
`trucks on the garage decks. Based on the similarities in the
`amount of flexural-strength deficiency between the garage decks,
`the third floor, which was not load tested, is also likely adequate
`to support the 1.44 kPa 共30 psf兲 live load without strengthening
`with CFRP bars.
`The results of the second series of tests at Test Area A 共Design
`Load Test兲 showed that the most critical bay, when strengthened
`with CFRP bars, was adequate to support the full design loads of
`2.40 kPa 共50 psf兲 live load. Based on this result, it was expected
`that all typical bays on both the second- and third-level decks,
`even though found to be deficient analytically, can support the
`code-prescribed loads when strengthened with CFRP bars.
`The load tests at Test Area B, where the deterioration, deflec-
`tion, and steel-placement deficiencies were not as pronounced as
`in Area A, indicated that the unstrengthened slab decks in some
`areas were capable of supporting the full code-prescribed loads
`without application of CFRP bars.
`
`Strengthening Program
`
`Determination of Slab Areas Requiring Immediate
`Strengthening
`
`According to the load-test results, only certain portions of the
`garage were in immediate need of strengthening. However, it is
`difficult to directly extrapolate the test results from this area to
`other garage areas and determine whether or not strengthening is
`prudent. After making comparisons that considered other param-
`eters specific to each location, such as steel-reinforcement place-
`ment and the associated structural deficiency, midspan deflections
`in specific column bays, and other observed distress 共such as
`cracking and concrete delamination兲,
`it was determined that
`approximately 70% of the negative-moment regions 共column
`heads兲 of the second- and third-level decks were in need of
`strengthening.
`In addition, even though strength deficiencies were not large
`enough to require immediate repair, durability concerns drove a
`decision to repair less-stressed areas also. A remedy of only crack
`and waterproofing, which was part of the overall repair scope of
`
`528 / JOURNAL OF COMPOSITES FOR CONSTRUCTION © ASCE / SEPTEMBER/OCTOBER 2007
`
` J. Compos. Constr., 2007, 11(5): 523-530
`
`Downloaded from ascelibrary.org by University Of Miami on 03/12/20. Copyright ASCE. For personal use only; all rights reserved.
`
`Metromont Ex-1004, p.6
`
`

`

`work, would initially prevent water entry into the structure. How-
`ever, normal slab vibrations and deflections due to traffic will
`likely reopen some of the cracks over time, as well as create new
`ones, allowing water to continue to deteriorate the steel reinforce-
`ment, concrete, and, ultimately, the structure.
`Placement of the CFRP system in all cracked negative-
`moment garage areas would supplement strength, repair cracks,
`and prevent new crack development 共the CFRP bars will effec-
`tively “stitch” the cracks兲. Therefore, the decision was made in
`consultation with the owner to immediately repair/strengthen the
`negative-moment slab regions of the entire garage. This approach
`would allow not only for remediation of the calculable structural
`deficiencies, but would also provide for a more durable garage.
`
`Implementation of Strengthening Program
`
`2.
`
`NSM CFRP bars were used to strengthen the moment negative
`regions of all the column heads at the garage. Before strengthen-
`ing the concrete decks with NSM CFRP bars, delaminated con-
`crete areas were exposed and replaced with new concrete. Also,
`cracks on the top of the deck were filled with gravity-fed epoxy.
`The construction procedure and sequence used for the strength-
`ening of the garage decks was as follows:
`1.
`Layout of CFRP reinforcement: Before sawcutting the con-
`crete, the desired location of the grooves was marked on the
`surface. The spacing between grooves was approximately
`305 mm 共12 in.兲 on center. At certain column head locations,
`survey results indicated that some steel bars were originally
`placed closer to the surface. Therefore, GPR was used to
`locate the bars to avoid damaging them during sawcutting.
`The sawcut spacing in those locations varied between
`203 mm 共8 in.兲 and 406 mm 共16 in.兲.
`Sawcutting of grooves: A road saw was used to sawcut the
`concrete slab and