Load Rating a Prestressed Concrete1 Double-Tee …docs.trb.org/prp/15-1302.pdf14 to better determine...

Aguilar, Jáuregui, Newtson, Weldon, and Cortez 1

Load Rating a Prestressed Concrete Double-Tee Beam Bridge without Plans by Proof 1 Testing 2 3 Carlos V. Aguilar Graduate Research Assistant 4

New Mexico State University 5 Department of Civil Engineering 6 Hernandez Hall, Box 30001, MSC 3CE 7 Las Cruces, NM 88003 8 9

David V. Jáuregui, PhD, PE Professor 10 New Mexico State University 11 Department of Civil Engineering 12

Hernandez Hall, Box 30001, MSC 3CE 13 Las Cruces, NM 88003 14 15

Craig M. Newtson, PhD, PE Associate Professor 16 New Mexico State University 17

Department of Civil Engineering 18 Hernandez Hall, Box 30001, MSC 3CE 19 Las Cruces, NM 88003 20

21 Brad D. Weldon, PhD Assistant Professor 22

New Mexico State University 23 Department of Civil Engineering 24

Hernandez Hall, Box 30001, MSC 3CE 25 Las Cruces, NM 88003 26

27 Tamara M. Cortez Graduate Research Assistant 28 New Mexico State University 29

Department of Civil Engineering 30 Hernandez Hall, Box 30001, MSC 3CE 31

Las Cruces, NM 88003 32 33

34 35 36

Submission date: July 25, 2014 37

A paper submitted for presentation at the 38 TRB 2015 Annual Meeting and publication in the Transportation Research Record 39

(Washington D.C.) 40

41 42

43 44 45

Paper length: text (5,000) + figures and tables (10 x 250) = 7,500 words 46


1 Graduate Student, Department of Civil Engineering, New Mexico State University, Las Cruces,

NM 88003 2

Professor, Department of Civil Engineering, New Mexico State University, Las Cruces, NM

88003 3

Associate Professor, Department of Civil Engineering, New Mexico State University, Las

Cruces, NM 88003 4

Assistant Professor, Department of Civil Engineering, New Mexico State University, Las

Cruces, NM 88003

Load Rating a Prestressed Concrete Double-Tee Beam Bridge without Plans by Proof 1 Testing 2 3 Carlos V. Aguilar

1, David V. Jáuregui

2, Craig M. Newtson

3, Brad D. Weldon

4, and Tamara M. 4

Cortez1

5 6

Abstract 7 Bridges with no design plans are currently an issue in New Mexico as many exist throughout the 8 State. Conventional load rating techniques cannot be utilized since these bridges have limited or 9

no design documentation. This creates uncertainties regarding the load-carrying capacity of these 10 structures. Only a few states have formal procedures on how these particular bridges should be 11 load rated. A project was conducted for the New Mexico Department of Transportation to 12

develop a procedure for load rating bridges without plans, in particular prestressed concrete 13 bridges. In accordance with the AASHTO Manual for Bridge Evaluation, a prestressed concrete 14 double-tee beam bridge was evaluated using advanced analyses and experimental methods 15

(including load testing and non-destructive material evaluation techniques). A four step load 16 rating procedure was implemented that included estimating the prestressing steel by Magnel 17

diagrams, verifying the estimate with a rebar scanner, testing the bridge at both diagnostic and 18 proof loads based on strain measurements, and finally rating the bridge using the proof test 19 results. Rating factors and posting loads were determined for AASHTO and New Mexico legal 20

loads. Due to the poor condition of the shear keys (some of which were broken), it is shown that 21 the load distribution between beams was adversely affected and the bridge should be load posted. 22

23 Keywords: Load testing; Load rating; Load posting; Non-destructive material evaluation; 24

Prestressed concrete bridges; No plans. 25

26

27 28

29 30 31

32 33

34 35 36 37


INTRODUCTION 1 In the U.S., the load rating procedures are specified in the AASHTO Manual for Bridge 2

Evaluation (1) hereafter referred to as the AASHTO Manual. In New Mexico, the majority of the 3

bridges have documentation and conventional structural systems that can readily be load rated 4

solely by analytical means. Yet numerous bridges exist that are more difficult to evaluate such as 5

those without design plans. Many of the planless bridges in New Mexico and other states are 6

older off-system (e.g., city or county owned) concrete bridges. Typical timber and steel bridges 7

can be field measured to obtain the necessary information to perform an analytical load rating. 8

For concrete bridges, however, finding the location, size, and cover of steel reinforcement is a 9

major challenge. In the absence of plans, the bridge capacity may be estimated based on 10

engineering judgment and the current bridge condition without an actual structural analysis. 11

Ratings determined in this fashion result in much uncertainty. Consequently, a project was 12

conducted for the New Mexico Department of Transportation (NMDOT) to develop a procedure 13

to better determine bridge ratings and posting loads for concrete bridges without plans based on 14

the AASHTO Manual (1). This paper focuses on the evaluation of a prestressed concrete T-beam 15

bridge using diagnostic and proof testing methods, non-destructive material evaluation 16

techniques, and advanced analyses. 17

18

Literature Review 19 A literature review was performed on proof testing and load rating of concrete bridges without 20

plans. Prior to the year 2000, a significant number of proof tests were conducted in Alabama (2), 21 Florida (3), Michigan (4), and New York (5). In these states, the majority of the proof tests were 22

performed on reinforced concrete and steel bridges, with only a few on prestressed concrete 23 bridges. Within the last ten years, other states have also sponsored bridge testing research for 24

load rating purposes including Iowa (6), Delaware (7), and Vermont (8). In addition, a project 25 was conducted in Georgia that included an appraisal of the state-of-the-art for bridge condition 26 assessment (9); bridge evaluation through load testing and advanced analysis (10, 11); and 27

development of load rating guidelines (12, 13). It is important to note that diagnostic testing has 28 been the preferred approach in the latest work conducted by the states. A major investigation was 29 recently conducted in Europe entitled Assessment and Rehabilitation of Central European 30

Highway Structures (ARCHES) that focused on the use of soft (under ambient traffic), 31 diagnostic, and proof load testing for bridge evaluation (14). Outside of the ARCHES project, 32 just a few other studies were found related to load testing of concrete bridges without plans (15, 33 16, 17, 18). Again, the tested bridges were mainly reinforced concrete (non-prestressed) and 34

excluding Bernhardt and DeKolb (15), diagnostic testing was used. 35

36 Research Need and Approach 37 From the AASHTO Manual (1) and the literature review, it was recognized that the proof testing 38

procedures were developed primarily for reinforced concrete and steel bridges at the strength 39

limit state and operating load level. Specific guidance for proof testing of prestressed concrete 40

bridges is lacking. Due to the high likelihood of exceeding the cracking moment based on the 41

target proof load specified in the AASHTO Manual (1), the load tests done in this study were 42

planned based on the Service III limit state (i.e., concrete cracking) for prestressed concrete 43


bridges rather than Strength I. This reduces the target proof load which limits the amount of 1

damage (if any) imposed on the bridge. It is important to note that Section 6A.4.2.2 of the 2

AASHTO Manual (1) allows the legal load rating to be determined based on the Service III limit 3

state. 4

5

BRIDGE DESCRIPTION 6 NMDOT Bridge 7701, constructed in 1974, is a single-span, prestressed concrete bridge with a 7 31 ft. (9.45 m) span owned by Doña Ana County. The bridge consists of nine double-tee beams 8 and each beam has seven shear keys on the top flange. There is no wearing surface and the 9 roadway width is 23 ft. 5 in. (7.14 m) from curb to curb. The bridge is located near Mesquite, 10 NM and carries County Road B-31 over an irrigation canal near NM-478. Overall, the bridge is 11

in fair condition. The most recent inspection report from November 7th, 2013, indicates a rating 12

of 6 for the deck and superstructure, and a 6 for the substructure. The topside of the beam flanges 13

have moderate cracking with heavy abrasion and small spalls. There are longitudinal cracks 14 between the beam stems and flanges, and the middle interior beam (beam 5) has vertical cracking 15

originating from the flange-stem transition; however, these cracks are not flexural. The south 16 exterior beam (beam 9) has a few spalls up to 6 in. by 6 in. (152 mm by 152 mm) with four 17

inches of exposed shear reinforcement on one side of the stem. 18 The shear keys consist of two steel angles, welded together with a short piece of rebar 19

and vary in condition. The exterior keys are all covered with grout and appeared to be 20

functioning properly. Between beams 4 and 5, six of the seven shear keys are broken and not 21 functioning. In addition, the keys between beams 6 and 7 are damaged and not working well. 22

Photographs of the bridge are provided in Figure 1. The bridge carries an Average Daily Traffic 23 (ADT) of 500. 24

25

BRIDGE ASSESSMENT PROCEDURE 26 This section describes the procedure that was developed to evaluate NMDOT Bridge 7701 with 27

emphasis on proof load testing as prescribed in Section 8 of the AASHTO Manual (1). First, the 28

total number and eccentricity of the prestressing strands are estimated using Magnel diagrams 29

(19). Material properties are obtained from the AASHTO Manual (1), AASHO specifications 30

(20), and/or the State’s provisions (21). Second, a Hilti PS 250 Ferroscan is used to detect the 31

primary steel reinforcement and verify the estimate from the Magnel diagrams. Third, a 32

diagnostic load test is performed to measure the beam strains under a truck load approximately 33

60% of the target proof load. This test is done to determine the critical transverse truck paths 34

(i.e., ones producing the largest measured strains) and compare the measurements with analytical 35

predictions. The available strain capacity for the beams are determined using the prestressing 36

estimate and assumed material properties. Fourth, a proof test is conducted with the goal of 37

applying the largest midspan moment possible using the dump trucks (above the target proof 38

loads) without exceeding the available capacity based on concrete cracking. The target proof 39

loads are calculated based on the AASHTO Manual (1) for the AASHTO and New Mexico legal 40

load weights and configurations. The proof test results are ultimately used in conjunction with 41

the target proof loads to determine the final load ratings and posting limits for the bridge. 42


1 2

FIGURE 1 NMDOT Bridge 7701 (prestressed concrete T-beam) in New Mexico. 3 4

Estimate of Prestressing Strands 5 The amount of prestressing steel was estimated using Magnel diagrams based on the 6

serviceability criteria (i.e., allowable stresses) for compression and tension at transfer and 7

service. The stresses were evaluated at the beam midspan resulting in four equations that were 8

written in terms of the inverse of the initial prestressing force (1/Pi) as follows (16): 9

10

𝑃1(𝑒) =1

𝑃𝑖≥

1

𝐴∗(𝑓𝑏𝑖+𝑀𝐷𝐿

𝑆𝑏)

+𝑒

𝑆𝑏∗(𝑓𝑏𝑖+𝑀𝐷𝐿

𝑆𝑏) (Eq. 1) 11

12

𝑃2(𝑒) =1

𝑃𝑖≤

−1

𝐴∗(𝑓𝑡𝑖+𝑀𝐷𝐿

𝑆𝑡)

+𝑒

𝑆𝑡∗(𝑓𝑡𝑖+𝑀𝐷𝐿

𝑆𝑡) (Eq. 2) 13

14

𝑃3(𝑒) =1

𝑃𝑖≤ 𝑘 ∗ (

1

𝐴∗(−𝑓𝑏𝑓+𝑀𝑇𝑜𝑡

𝑆𝑏)

+𝑒

𝑆𝑏∗(−𝑓𝑏𝑓+𝑀𝑇𝑜𝑡

𝑆𝑏)) (Eq. 3) 15

16


𝑃4(𝑒) =1

𝑃𝑖≥ 𝑘 ∗ (

−1

𝐴∗(−𝑓𝑡𝑓+𝑀𝑇𝑜𝑡

𝑆𝑡)

+𝑒

𝑆𝑡∗(−𝑓𝑡𝑓+𝑀𝑇𝑜𝑡

𝑆𝑡)) (Eq. 4) 1

2

where P1, P2, P3 and P4 are the solutions that satisfy the serviceability criteria for the bottom 3

stress at transfer (fbi), top stress at transfer (fti), bottom stress at service (fbf), and top stress at 4

service (ftf), respectively. In addition, e is the eccentricity of the prestressing steel; Sb and St are 5

the section moduli at the bottom and top of the beam; A is the cross-sectional area of the beam; k 6

is the factor for prestress losses; and MDL and MTot are the dead load moment and total moment 7

(i.e., dead load plus live load moments), respectively. 8

Since the bridge was constructed in 1974, the 1973 AASHO Standard Specifications (20) 9

was used. The release strength (fci’) was assumed to be 4,500 psi (31.0 MPa), which slightly 10

exceeds the minimum of 4,000 psi (27.6 MPa) specified by AASHO for pretensioned members. 11

The 28-day strength (fc’) was assumed to be 5,000 psi (34.5 MPa) in accordance with the 12

AASHO and state specifications (21). The prestressing steel was assumed to be 0.5-inch (12.7 13

mm) diameter, Grade 270, seven-wire, stress-relieved strands and 25% prestress losses were 14

estimated using AASHO. The allowable concrete stresses from AASHO are equal to the 15

following values: 16

17

𝑓𝑏𝑖 = 0.6𝑓𝑐𝑖′ = 2,700 psi (18.6 MPa) 18

19

𝑓𝑡𝑖 = 3√𝑓𝑐𝑖′ = 201.2 psi (1.39 MPa) 20

21

𝑓𝑏𝑓 = 6√𝑓𝑐′ = 424.3 psi (2.93 MPa)

22

𝑓𝑡𝑓 = 0.4𝑓𝑐′ = 2,000 psi (13.8 MPa)

23

Magnel diagrams were generated considering dead load and HS-20, H-20, and H-15 truck 24

loading using the AASHO impact and live load distribution factors of 1.3 and 0.458, 25

respectively; Figure 2 shows the diagram for the H-20. The P1 through P4 lines represent Eqs. 26

(1) through (4). The area enclosed by the four lines indicates the acceptable combinations of the 27

initial prestressing force (Pi) and eccentricity (e). Table 1 summarizes the high prestress / low 28

eccentricity solution (i.e., intersection of lines P3 and P4) and low prestress / high eccentricity 29

solution (i.e., intersection of lines P2 and P3). The effective prestressing force (Pe) was 30

computed as kPi with k equal to 0.75. To determine the area of the prestressing strand (As), Pi 31

was divided by the initial stress in the strand at transfer (fpi) which was assumed equal to 70% of 32

the ultimate strand stress (fpu). The number of strands was finally computed by dividing As by the 33

area of a single 0.5-inch (12.7 mm) diameter strand (equal to 0.153 in2

or 98.7 mm2). 34


1 NOTE: 1 kip = 4.45 kN; 1 in = 25.4 mm. 2 3

FIGURE 2 Magnel diagram for H-20 design truck. 4 5

TABLE 1 Estimate of Prestressing Strands 6 7

Variable HS-20 H-20 H-15

High P/T Low P/T High P/T Low P/T High P/T Low P/T

1/Pi x 10-3

(1/kip) 2.14 6.60 2.14 8.56 2.14 14.5

Pi (kip) 468 152 468 117 468 69.1

e (in) 1.48 8.84 1.01 10.2 1.87 14.0

Pe (kip) 351 114 351 87.6 351 51.8

As (in2) 2.48 0.80 2.48 0.62 2.48 0.37

No. of strands 16.2 5.24 16.2 4.04 16.2 2.39

NOTE: 1 kip = 4.45 kN; 1 in = 25.4 mm; 1 in2 = 645.1 mm

2. 8

9

Figure 3 shows the standard beam section for NMDOT Bridge 7701. Assuming a cover of 3 in. 10

(76 mm) to the center of the bottom strand and a strand spacing of 1 in. (25.4 mm) based on 11

historical data provided by the NMDOT, the eccentricities are 10.3 in. (261.6 mm) for two 12

strands, 9.78 in. (248.4 mm) for four strands, and 9.28 in. (235.7 mm) for six strands. From 13

Table 1, it appears that NMDOT Bridge 7701 was designed for the H-20 truck since the 14

eccentricity of 10.2 in. (259.1 mm) agrees well with the possible strand patterns. For an 15


eccentricity of 9.28 in. (235.7 mm), the steel equivalent increased to 4.32 strands which would 1

require 6 strands (i.e., 3 per stem). 2

3 NOTE: 1 in = 25.4 mm. 4 5 FIGURE 3 Standard beam section of NMDOT Bridge 7701. 6 7 Detection of Steel Reinforcement 8 A Hilti PS 250 Ferroscan system was used to scan the location, size, and cover of the 9 prestressing steel strands. The Ferroscan is generally capable of providing accurate results only 10 for the first reinforcement layer and a minimum spacing-to-cover ratio of 2:1 is required. The 11

reinforcement must also be greater than 0.4 in. (10.2 mm) from the concrete surface and no 12 deeper than 2.4 in. (61.0 mm) to obtain an accurate depth measurement. The Ferroscan is best 13 suited for detecting mild steel reinforcing bars and thus, for NMDOT Bridge 7701, the quality of 14 the scans for the prestressing strand was compromised due to the seven-wire strand arrangement. 15

The Ferroscan system generates imagescans of 2 ft. by 2 ft. (0.61 m by 0.61 m) areas that 16

can be connected to produce blockscans of a larger area. Blockscans of the north exterior beam 17

and the center beam (beams 1 and 5) were generated and several other smaller areas were 18 scanned on the remaining seven beams as a spot check. The blockscans showed three 19 longitudinal strands per stem harped at approximately 40% of the beam length. In the middle 20 20%, the prestressing strands are tightly spaced which complicated the size estimate. Due to the 21 seven-wire strand arrangement, the bar size estimate ranged from a #11 to a #3. Note that the 22

equivalent strand diameter for a 0.5-in. (12.7 mm) strand falls between a #3 and #4 bar. 23 Furthermore, the imagescans showed three strands with spacings of 2 in. (51 mm) and 4 in. (102 24 mm) at the ends and 1 in. (25.4 mm) spacings at midspan (see Figure 4) which agreed with the 25

strand estimate from the Magnel diagrams. 26


1 2 FIGURE 4 Imagescans in (a) harped and (b) non-harped areas of interior beam. 3

4

Diagnostic Load Test 5 A diagnostic load test was first performed to determine the critical truck paths for the proof load 6 test. Due to the condition of the shear keys, the connection between the beams was uncertain and 7 caution needed to be taken to guard against beam cracking. Using the estimate of the prestressing 8

strands discussed earlier, the cracking moment, available moment, target moment, and test truck 9 moment for an interior beam were calculated. The cracking moment was determined using the 10

following equation: 11

12

𝑀𝑐𝑟𝑎𝑐𝑘𝑖𝑛𝑔 = 𝑆𝑏 ∗ [𝑃𝑒 ∗ (1

𝐴+

𝑒

𝑆𝑏) + 𝑓𝑐𝑟] (Eq. 5) 13

14 where fcr is the modulus of rupture (equal to 7.5 times the square root of fc’) and the remaining 15 variables were defined earlier. In addition, the cracking moment was found assuming a prestress 16

loss of 25%. For fc’ equal to 5,000 psi (34.5 MPa), fcr equals 530 psi (3.65 MPa) and the cracking 17 moment equals 148.4 kip-ft. (201.2 kN-m). The available moment was found by subtracting the 18

dead load moment (35.8 kip-ft. or 48.5 kN-m) from the cracking moment and represented the 19 additional moment the beam could resist before cracking (112.6 kip-ft. or 152.7 kN-m). Dump 20 trucks with a total weight of 42.5 kips (189.1 kN) (14.7 kip or 65.4 kN single front axle, 27.8 kip 21 or 123.7 kN tandem rear axle) were used which is 85% of the Type 3 legal load. A moderate load 22 was applied as a precautionary measure against cracking due to the suspect shear keys and 23

uncertain live load distribution. The dump truck provided a maximum test truck moment of 92.5 24 kip-ft. (125.4 kN-m). Note that if the live load distribution factor is equal to one, the test truck 25 and available moments are nearly equal. Using a factor of 0.458 as computed based on AASHO 26


(20), the test truck moment was 42.4 kip-ft. (57.5 kN-m). Due to the damage observed on many 1 of the shear keys, the actual moment resisted by the beam was expected to fall somewhere 2 between 42.4 kip-ft. (57.5 kN-m) and 92.5 kip-ft. (125.4 kN-m). Hence, the test truck moment is 3 less than or equal to the available moment and thus, the beams should not crack. 4

Using the moments given above, the available strain and the expected test truck strain 5 were computed 1.0 in. (25.4 mm) from the bottom of the beam stem (i.e., the transducer 6 location). These values were compared to the strains measured during the load test to monitor the 7 bridge behavior for potential cracking and also to check the live load distribution. The following 8 equation was used to determine the strain values: 9

10

𝜀 =𝑀

𝑆∗𝐸𝑑𝑒𝑠𝑖𝑔𝑛 (Eq. 6) 11

12

where M is the available or test truck moment, S is the section modulus at the transducer 13 location, and Edesign is the modulus of elasticity of the beam. The available strain was calculated 14

to be 532 με. The strains due to the test truck were expected to be a minimum of 200 με for a 15

distribution factor of 0.458 to a maximum of 436 με for a distribution factor of 1.0. 16 The diagnostic load test consisted of two phases. In the first phase, the front axle of the 17

dump truck (weighing 14.7 kips or 65.4 kN) was positioned along seven transverse paths. For 18 paths 1 and 7, the truck was positioned approximately 2 ft. (0.61 m) from the curb and for paths 19 2 through 6, the truck was centered about beams 3 through 7, respectively. Figure 5(a) shows the 20

front wheel locations of the dump truck for paths 5 and 7. For each path, the axle was moved 21 along the bridge length incrementally until it was 1 ft. (0.31 m) from midspan. The BDI (Bridge 22

Diagnostics, Inc.) structural testing system was used to measure the beam strains. Strain 23 transducers were placed on all the beam stems at midspan at distances of 1 in. (25.4 mm) and 7 24

in. (178 mm) from the bottom of the stem. The largest measured strains were 202 με at beam 5 25 and 152 με at beam 7 for paths 5 and 7, respectively. Consequently, in the second phase, two 26

dump trucks were placed side-by-side with one truck positioned in either path 5 or path 7 and the 27 second truck placed 2 ft. (0.61 m) over as shown in Figure 5(b). In these two runs, the trucks 28 were backed up onto the bridge and the load was applied by the rear axle (weighing 27.8 kips or 29

123.7 kN) rather than the front axle. The test measurements showed that paths 3 and 7 were the 30 most critical with strains as high as 312 με at beam 5; the highest measured strain from paths 1 31

and 5 was only 265 με also at beam 5. Note that the largest strain from paths 3 and 7 is 32 approximately 1.6 times the minimum test truck strain of 200 με. This is a clear indication that 33 the shear keys are not functioning as intended, which was a concern going into the test due to the 34 visible damage. 35


1 NOTE: 1 ft. = 0.305 m. 2

3

FIGURE 5 Transverse truck paths for (a) first phase and (b) second phase of diagnostic 4 testing. 5 6

Proof Load Test 7 The proof load test consisted of four phases. In the first phase, the tandem back axle of a single 8

truck (weighing roughly 42.0 kips or 186.8 kN) was applied in path 5 and then path 7. The path 9 causing the most strain was then repeated in phase three with two trucks placed back-to-back. In 10 the second phase, two trucks were placed side-by-side, first in paths 1 and 5 and then in paths 3 11 and 7. The more critical loading configuration was then used in phase four with all four trucks 12 placed side-by-side and back-to-back on the bridge. See Figure 5 given earlier for the truck 13

paths. Figure 6 shows the instrumentation setup for the proof test. All nine beams had at least 14 two transducers at midspan. The diagnostic test results showed that beams 4 through 7 had the 15 greatest strains. Therefore, more gages were installed on these four beams at midspan. 16


1 2

FIGURE 6 Instrumentation layout for proof load testing. 3 4 To ensure that no shear cracking would occur in the bridge beams, a thorough shear evaluation 5

was done. Assuming a shear distribution factor of 1.0 6 , the maximum shear for a single beam caused by 42.0-kip (186.8 kN) tandem axles placed back-7

to-back was computed as 21.0 kips (93.4 kN). The web-shear cracking force, Vcw, was taken as 8 the beam shear capacity and estimated using the following formula (22): 9 10

𝑉𝑐𝑤 = 𝑓𝑡√1 +𝑓𝑝𝑐

𝑓𝑡𝑏𝑤𝑑 + 𝑉𝑝 (Eq. 7) 11

12

where ft is the tensile concrete strength; fpc is the compressive concrete stress after all losses at 13 the beam centroid; bw is the beam web width; d is the distance from the compression face to the 14

centroid of the tension reinforcement (not less than 80% of overall beam depth); and Vp is the 15 vertical component of the effective prestressing force. The concrete tensile strength varies 16

between 2 and 4 times the square root of fc’. The former value provides the better estimate for 17 reinforced concrete members or prestressed concrete members with a low level of prestressing 18

while the latter value is more accurate for end regions of fully prestressed members (22). 19 Applying Eq. (7) with these two values, Vcw ranged between 27.2 kips (121.0 kN) and 41.9 kips 20 (186.4 kN), which exceeded the sum of the ultimate live load shear of 21.0 kips (93.4 kN) and 21

dead load shear of 5.0 kips (22.2 kN). Thus, concrete cracking due to shear was not expected. 22 Table 2 lists the maximum measured strains for the four proof test phases at the 18 23

transducers located at the bottom of the beams at midspan. The beams and stems are numbered 24 from north to south and are labeled B#S# (e.g., B1S1 corresponds to beam 1, stem 1). Recall that 25 beams 4 through 7 had two transducers attached to the bottom of each stem; however, only the 26 larger of the two recorded strains are reported in the table. Strain transducers were installed on 27

opposite sides of the stems of beams 4 through 7 in the proof test to more closely monitor the 28 bridge response under heavy loading. In addition, the diagnostic test measurements showed the 29

shear key performance between these beams to be the most critical. The strain measurements in 30 the proof test did in fact show noteworthy differences across the beam stems, possibly due to 31 lateral bending. Consequently, the larger strain was used as the limiting criteria to prevent 32 cracking. 33

In phase one, the truck was able to reach the final position without exceeding the 34

available strain. Figure 7(a) shows the final longitudinal truck position. Table 2 shows that path 35 5, centered along beam 6, produced the greater strain of 508 με compared to 348 με for path 7. In 36 phase two, the test was stopped once the rear axles reached 10 feet (3.05 m) (see Figure 7(b)) 37 when the measured strains at one of the transducers approached the available strain of 532 με. 38 Table 2 shows that paths 3 and 7 produced a greater strain, 528 με compared to 474 με, for paths 39


1 and 5. In phase three, two trucks were placed back-to-back in the most critical single loaded 1 lane (i.e., path 5). In phase four, all four trucks were placed on the bridge. On both ends, two 2 trucks were placed in paths 3 and 7, simulating the most critical multiple lanes loaded condition. 3 Table 2 shows the measured strains for these test phases. The trucks were stopped when the rear 4

axles were at 6.5 feet (1.98 m) in phase three (see Figure 7(c)) and 5.0 feet (1.52 m) in phase four 5 (see Figure 7(d)) since the maximum measured strain was close to the available strain. 6

The strains measured during the proof test (see Table 2) gave strong evidence that the 7 shear keys were not functioning properly as shown by the extreme differences in strain, 8 particularly between beams 4 and 5. When a single truck was placed in path 5 during phase one, 9

the measured strains for beams 4 and 5 were 38 µε and 508 µε, respectively, a difference of 470 10 µε, indicating a break down in the shear key connection. Similar behavior was observed when 11 two trucks were placed back-to-back in path 5 during phase three which showed a strain 12

difference between beams 4 and 5 of 414 µε. Also, the measured strains from the diagnostic test 13 exceeded the strains calculated using the distribution factor from the 1973 AASHO 14 Specifications which is another indication of poor shear transfer between these two beams. 15

16

TABLE 2 Maximum Measured Strains from Proof Load Test 17 18

19 20

Transducer

Location

Strain (με)

Phase One Phase Two Phase Three Phase Four

Path 5 Path 7 Paths 1 and 5 Paths 3 and 7 Path 5 Paths 3 and 7

B1S1 0 -2 149 62 18 84

B1S2 1 -6 235 80 11 115

B2S1 4 2 228 83 17 111

B2S2 6 -3 255 224 28 195

B3S1 3 10 248 191 24 161

B3S2 21 9 341 197 55 203

B4S1 21 21 363 182 63 196

B4S2 38 33 474 158 82 205

B5S1 508 94 448 528 496 494

B5S2 372 157 391 470 407 456

B6S1 339 172 367 455 432 450

B6S2 279 269 239 316 348 344

B7S1 275 348 245 443 360 471

B7S2 213 189 233 227 242 289

B8S1 223 186 247 220 280 279

B8S2 87 277 105 266 146 288

B9S1 79 243 99 250 151 258

B9S2 56 175 68 140 93 228


1 NOTE: 1 ft. = 0.305 m. 2 3

FIGURE 7 Final longitudinal truck positions for (a) first, (b) second, (c) third, and (d) 4 fourth phase of proof testing. 5 6

RATING FACTORS AND POSTING LOADS 7 Based on the measured strains given in Table 2 and corresponding truck positions given in 8

Figure 7, load rating factors were determined from the proof load test. For all four phases, the 9

moments applied to the bridge from the test truck loading were used to compute the final load 10 ratings for AASHTO and New Mexico legal loads according to the AASHTO Manual (1). Recall 11

that the available strain of 532 used to monitor the beams during the proof test was based on 12 7.5(fc’)

1/2. To determine the final load ratings, however, the test truck moments associated with 13

an available strain of 507 based on an allowable concrete stress of 6(fc’)1/2

, were used. Based 14 on Section 8.8.3.3 of the AASHTO Manual (1), the adjusted target live-load factor (XpA) and the 15 target proof load (LT) were first determined using the following equations: 16

17

𝑋𝑝𝐴 = 𝑋𝑝 (1 +𝛴%

100) (Eq. 8) 18

19

𝐿𝑇 = 𝑋𝑝𝐴𝐿𝑅(1 + 𝐼𝑀) (Eq. 9) 20

21 where Xp is the initial target load factor, Σ% is the sum of the Xp adjustment factors, LR is the 22 unfactored live load due to the rating vehicle, and IM is the impact factor (equal to 1.33). The 23

AASHTO Manual (1) recommends a minimum value of 1.4 for Xp; however, since the proof test 24 was executed based on serviceability (limiting concrete tensile stress at service load) rather than 25 strength, a factor of 1.0 was used which is permitted for legal load rating (based on Section 26 6A.4.2.2 of the AASHTO Manual (1)). Also note that LT and LR were determined in terms of the 27 maximum moment at midspan of the bridge rather than load since the rating vehicles and proof 28 test trucks do not have the same axle configuration. The applicable adjustment factors from 29 Table 8.8.3.3.1-1 of the AASHTO Manual (1) included one-lane loading (+15%), non-redundant 30


load path (+10%), in-depth inspection performed (-5%), and ADTT less than 1000 (-10%). 1 Accordingly, XpA was 1.1 for one lane loaded and 0.95 for multiple lanes loaded from Eq. (8). 2 Target proof moments were then computed using Eq. (9) for the AASHTO legal loads (Type 3, 3 Type 3-3, and Type 3S2) and the New Mexico legal loads (NM 2, NM 3A, and NM 5A) which 4

are listed in Table 3 along with the vehicle weight, W. 5 6

TABLE 3 Target Proof Moments, Rating Factors, and Posting Loads 7 8

NOTE: 1 kip = 4.45 KN: 1 kip-ft. = 1.357 KN-m. 9 10 The target proof moments were then compared to the moments applied by the dump trucks 11

during the actual proof test. For phases one and three (single lane loaded), the applied moments 12 were 279 kip-ft. (378 kN-m) and 368 kip-ft. (499 kN-m), respectively, while for phases two and 13

four (multiple lanes loaded), the applied moments were 256 kip-ft. (347 kN-m) and 300 kip-ft. 14 (407 kN-m), respectively. The operating level (OP) capacity for the bridge was computed using 15

the following equation from the AASHTO Manual (1): 16 17

𝑂𝑃 =𝑘0𝐿𝑃

𝑋𝑝𝐴 (Eq. 10) 18

where k0 is a factor that depends on how the proof test was terminated (1.0 if the target proof 19 moment is reached and 0.88 if the test is stopped due to signs of distress) and LP is the applied 20 moment. Although the proof test was ended before the truck axles reached the final position, k0 21

was taken as 1.0 since the concrete design strength was used rather than the actual strength to 22 determine the available strain and prevent cracking. Consequently, the OP moments equaled 253 23 kip-ft. (343 kN-m) for phase one, 270 kip-ft. (366 kN-m) for phase two, 335 kip-ft. (481 kN-m) 24

for phase three, and 316 kip-ft. (428 kN-m) for phase four. The final load ratings were computed 25 based on phase one which had the smallest moment. Recall that the truck was stopped in this 26 phase when the back axle reached 14 ft. (4.27 m) resulting in a measured strain of 508 με which 27 equaled the available strain. The rating factors, RFO, for the legal load were computed using the 28

following equation: 29

𝑅𝐹𝑂 =𝑂𝑃

𝐿𝑅(1+𝐼𝑀) (Eq. 11) 30

Legal

Load

W

(kips) LR (kip-ft.)

LT (kip-ft.)

RFo

Safe

Posting

Load

(Tons) One Lane Multiple Lanes

Type 3 50.0 235 344 297 0.81 18.2

Type 3S2 80.0 231 338 292 0.83 30.0

Type 3-3 72.0 190 278 240 1.00 N/A

NM 2 33.6 179 262 226 1.06 N/A

NM 3A 46.3 235 344 297 0.81 16.9

NM 5A 80.6 247 361 311 0.77 27.3


The safe posting loads were then determined based on Section 6A.8.3 of the AASHTO Manual 1 (1) as follows: 2

𝑆𝑎𝑓𝑒 𝑃𝑜𝑠𝑡𝑖𝑛𝑔 𝐿𝑜𝑎𝑑 =𝑊

0.7[(𝑅𝐹𝑂) − 0.3] (Eq. 12) 3

Table 3 lists the rating factors and safe posting loads for each legal load. The table shows that 4 Bridge 7701 needs to be posted for all legal loads excluding the Type 3-3 and NM 2. It is 5 reiterated that the rating factors and posting loads were determined for the Service III limit state 6 (concrete cracking) and not Strength I (ultimate flexure) based on the large strains measured 7

during the proof test, which were significantly influenced by the poor shear key performance. 8 Before testing, the bridge had inventory and operating rating factors of 0.85 and 1.3 for HS-20 9 design truck loading and was not posted since the operating rating exceeded 1.0. The original 10

ratings were determined by the NMDOT based on historical information for off-system bridges 11 in Doña Ana Country, New Mexico but the poor condition of the shear keys was not considered. 12 Based on the results of this study, shear transfer between beams was shown to be an issue which 13

caused reductions in the load ratings and ultimately, the bridge was load posted by the NMDOT. 14

15

CONCLUSIONS 16 A four step load rating procedure for prestressed concrete double-tee beam bridges without plans 17 was developed. The implementation of this procedure yielded the following conclusions: 18

19

The literature review provided little guidance on proof testing of prestressed concrete 20 bridges without plans. Although testing has been conducted for load rating purposes, the 21 majority of recent tests were diagnostic rather than proof. In addition, proof tests have 22

been performed mainly on reinforced concrete bridges rather than prestressed concrete. 23

Conducting the load test based on serviceability, or the Service III limit state (concrete 24 cracking), was considered necessary to ensure the bridge remained uncracked. It was 25 important to take precaution in the testing, particularly in cases where the bridge beams 26 have damaged shear keys, since the actual live-load distribution factor was unknown. 27

A diagnostic load test was helpful in planning the proof test due to the missing and 28 damaged shear keys and uncertain load distribution. The diagnostic test provided 29

information on the actual bridge behavior under truck loading below the estimated 30 cracking load. 31

The monitoring of strain was effective in controlling the proof test to avoid overloading 32 the bridge. Preparatory calculations were required to determine an available strain that 33 the bridge could endure without cracking, and the proof test was monitored by ensuring 34

that the measured beam strains did not exceed the available strain. 35

Based on the dump truck positions that induced the available strains, the applied 36 moments were computed and compared to the target moments for the AASHTO and New 37 Mexico legal loads. These comparisons were then used in accordance with the AASHTO 38 Manual (1) to develop rating factors and safe posting loads. The load testing exposed the 39

true behavior of the bridge and showed the original load ratings (determined based on 40 historical information and assuming complete shear transfer between beams) were 41 overestimated. Consequently, the bridge was load posted by the NMDOT. 42

43


This study provided valuable information for future evaluations of prestressed concrete bridges 1 without plans. For instance, if an in-depth inspection is performed and no excessive signs of 2 damage or distress are found, a diagnostic test may be excluded if there is confidence that the 3 bridge behavior can be predicted. Furthermore, if high quality images of the prestressing strand 4

can be obtained from a rebar scanner and the bridge is in good condition, load testing may be 5 avoided altogether since the bridge may be modeled and rated analytically based on the bridge 6 properties determined from the design specifications and the Magnel diagrams. 7 It is important to note that even though the AASHTO Manual (1) was generally used in 8 this study, there are no specific guidelines in the manual that address the load rating of 9

prestressed concrete bridges through proof load testing. The manual provides guidance for proof 10 testing based on strength but not serviceability. Although common testing procedures were 11 employed herein, the procedures were adapted to the service limit state for concrete tension, 12

which required a unique approach not given in the AASHTO Manual (1). In addition, the 13 literature review done prior to this evaluation uncovered just a few publications (mainly reports 14 from state DOTs or consultants) that gave minimal direction for the load rating of bridges via 15

proof testing, in particular prestressed concrete bridges without design plans. 16

17 ACKNOWLEDGMENTS 18 The authors thank Mr. Ray Trujillo (NMDOT State Bridge Engineer), Mr. Jeff Vigil (NMDOT 19 Bridge Management Engineer) and Mr. Gary Kinchen (NMDOT Bridge Rating Engineer) for 20

their support to conduct this investigation. Mr. Scott McClure (NMDOT Research Bureau Chief) 21 and Mr. David Hadwiger (NMDOT Research Staff Manager) are gratefully acknowledged for 22

their oversight of this project. Also from NMDOT District 1, the assistance provided by Mr. 23 Ernest Sedillo and other personnel from the Anthony Patrol Yard in the load testing of NMDOT 24

Bridge 7701 is greatly appreciated. 25

26

REFERENCES 27 1. American Association of State Highway and Transportation Officials (AASHTO). (2011). 28

Manual for Bridge Evaluation, 2nd

Edition, Washington, D.C. 29

2. Conner, G. H., Stallings, J. M., McDuffie, T. L., Campbell, J. R., Fulton, R. Y., Shelton, B. 30 A., and Mullins, R. B. (1997). “Bridge Load Testing in Alabama.” Transportation Research 31

Board. 32 3. Shahawy, M. A. (1995). “Nondestructive Strength Evaluation of Florida Bridges.” 33

Proceedings of SPIE – The International Society of Optical Engineering (Nondestructive 34 Evaluation of Aging Bridges and Highways), Society of Photo-Optical Instrumentation 35 Engineers, Vol. 2456, pp. 101-123. 36

4. Saraf, V., Sokolik, A. F., and Nowak, A. S. (1996). “Proof Load Testing of Highway 37

Bridges.” Transportation Research Record, No. 1541, pp. 51-57. 38 5. Kissane, R. J., Beal, D. B., and Sanford, J. A. (1980). “Load Rating Of Short-Span Highway 39

Bridges.” Interim Report on Research Project 156-1, Research Report 79 (US DOT/FHWA), 40

Engineering Research and Development Bureau, New York State Department of 41 Transportation. 42

6. Wipf, T. J., Phares, B. M., Klaiber, F. W., Wood, D. L., Mellingen, E., and Samuelson, A. 43 (2003). “Development of Bridge Load Testing Process for Load Evaluation.” Final Report 44 on Iowa DOT Project TR-445 / CTRE Project 00-65, Research Report TR-445, Iowa 45 Highway Research Board, Iowa Department of Transportation. 46


7. Chajes, M.J., and Shenton, H.W. (2006), “Using Diagnostic Load Tests for Accurate Load 1 Rating of Typical Bridges,” Journal of Bridge Structures, Vol. 2, No. 1, pp. 13-23. 2

8. Jeffrey, A., Breña, S. F., and Civjan, S. A. (2009). “Evaluation of Bridge Performance and 3 Rating through Non-destructive Load Testing.” Final Report, Research Report 2009-1, 4

Vermont Agency of Transportation. 5

9. Wang, N., Ellingwood, B. R., Zureick, A.-H., and O’Malley, C. (2009). “Condition 6

Assessment of Existing Bridge Structures: Report of Task 1 – Appraisal of State-of-the-Art 7

of Bridge Condition Assessment.” Report of GDOT Project No. RP05-01, Georgia 8

Department of Transportation, Atlanta, GA. 9

10. O’Malley, C., Wang, N., Ellingwood, B. R., and Zureick, A.-H. (2009). “Condition 10

Assessment of Existing Bridge Structures: Report of Tasks 2 and 3 – Bridge Testing 11

Program.” Report of GTRC Project No. E-20-K90 and GDOT Project No. RP05-01, Georgia 12

Department of Transportation and Georgia Institute of Technology, Atlanta, GA. 13

11. Wang, N., O’Malley, C., Ellingwood, B. R., and Zureick, A.-H. (2011). “Bridge Rating 14

Using System Reliability Assessment. I: Assessment and Verification by Load Testing.” 15

ASCE Journal of Bridge Engineering, Vol. 16, No. 6, pp. 854-862. 16

12. Ellingwood, B. R., Zureick, A.-H., Wang, N., and O’Malley, C. (2009). “Condition 17

Assessment of Existing Bridge Structures: Report of Task 4 – Development of Guidelines for 18

Condition Assessment, Evaluation and Rating of Bridges in Georgia.” Report of GTRC 19

Project No. E-20-K90 and GDOT Project No. RP05-01, Georgia Department of 20

Transportation and Georgia Institute of Technology, Atlanta, GA. 21

13. Wang, N., Ellingwood, B. R., and Zureick, A.-H. (2011). “Bridge Rating Using System 22

Reliability Assessment. II: Improvements to Bridge Rating Practices.” ASCE Journal of 23

Bridge Engineering, Vol. 16, No. 6, pp. 863-871. 24

14. Casas, J. R., Olaszek, P., Sajna, A., Znidarie, A., and Lavric, I. (2009). “Sustainable 25

Development, Global Change and Ecosystems (Sustainable Surface Transport): Deliverable 26

D16: Recommendations on the Use of Soft, Diagnostic, and Proof Load Testing.” Document 27

No. ARCHES-02-DE16, Assessment and Rehabilitation of Central European Highway 28

Structures (ARCHES) Management Group. 29

15. Bernhardt, R. P. and DeKolb, S. P. (2003). “Inspection and Load Testing of Three Bridges.” 30 Final Report, Project No. DE-01739-2P, US Army Signal Center, Fort Gordon, GA. 31

16. Hag-Elsafi, O. and Kunin, J. (2006). “Load Testing for Bridge Rating: Dean’s Mill over 32

Hannacrois Creek.” Report FHWA/NY/SR-06/147, Transportation Research and 33

Development Bureau, New York State Department of Transportation. 34

17. Shenton III, H. W., Chajes, M. J., and Huang, J. (2007). “Load Rating of Bridges Without 35 Plans.” Final Report, Research Report DCT 195, Delaware Center for Transportation. 36

18. Commander, B., Valera-Ortiz, W., Stanton, T. R., and Diaz-Alvarez, H. (2009). “Field 37

Testing and Load Rating Report, Bridge FSBR-514, Fort Shafter, Hawaii.” Final Report No. 38

ERDC/GSL TR-09-9, US Army Corp of Engineers, Arlington, VA. 39

19. Krishnamurthy, N. (1983). “Magnel Diagrams for Prestressed Concrete Beams.” ASCE 40 Journal of Structural Engineering, Vol. 109, No. 12, pp. 2761-2769. 41


20. American Association of State Highway Officials (AASHO). (1973). Standard 1 Specifications for Highway Bridges, 11

th Edition, Washington, D.C. 2

21. New Mexico State Highway Department (NMSHD). (1970). Standard Specifications for 3 Road and Bridge Construction, Santa Fe, NM. 4

22. Hawkins, N. M., Kuchma, D. A., Mast, R. F., Marsh, M. L., and Reineck, K.-H. (2005). 5 “Simplified Shear Design of Structural Concrete Members.” Final Report, NCHRP Report 6 549, National Cooperative Highway Research Program, Washington, D.C. 7

Date post:	22-Mar-2018
Category:	Documents
Upload:	vulien
View:	220 times
Download:	3 times

Load Rating a Prestressed Concrete1 Double-Tee …docs.trb.org/prp/15-1302.pdf14 to better determine...

Documents