Load-Rating Procedures and Performance Evaluation of Metal Culverts

A nondestructive method for load rating of bridges without structural properties and plans

This paper summarizes the results of applying load rating process that has been applied to Halfaya Bridge over the main Nile in Khartoum, Sudan. Load rating procedures have reviewed as per Manual of Bridge Evaluationof AASHTO Standards. Three levels of H/Y live loadings have been considered. AASHTO HL-93 is used for design load rating. For Legal load rating the Sudanese legal loads - as per Sudan National Highway Authorityare considered. Permit load rating process has been made using actual special heavy truck used in Sudan during the last 20 years .The rating process has been made on the assumption that the capacity of the main structural element of the bridge, which is Girder type II Composite Steel Ibeams, had been decreased with different percentage from 10%up to 20%, due to different factors. Suitable software, CSi Bridge, has been used for calculations and the results are presented. The paper predicted the load rating factors for the composite steel I-beams of the bridge- moment and shear wise- for considerable capacity loss in future .The paper also checked the deflection of the composite I-Girder .Results showed that deflection is the governing factor when stiffness is lost by 8%, but still the momentcapacity is the governing factor –design load rating –when the loss is 5%of the capacity. Suggested several recommendations to save such vital structure are given

The increasing use of buried corrugated steel plate (CSP) bridges and culverts in transportation infrastructure contributes to looking for the most optimal solutions regarding durability and cost. The paper presents a finite element analysis and experimental testing under static live loads of a shallow-buried CSP bridge with a span of 10.00 m and a height of 4.02 m. The depth of the soil cover (0.65 m) did not meet the minimum requirements in this respect. Therefore, a reinforced concrete (RC) relieving slab was placed above the shell crown. In the parametric analysis, it was decided to use various types of expanded polystyrene (EPS) geofoam as an alternative to RC slab. Undoubtedly, using EPS is cheaper and faster than using an RC slab. Nevertheless, it is essential whether it gives a similar (or better) effect as the RC slab. Five models of the bridge were used in the analysis, i.e. without RC slab (model I), with RC slab (model II) and with various types of EPS (models III–V). The numerical analysis results were compared with experimental tests for the buried bridge with the RC slab. The obtained results can help design buried CSP bridges and culverts with a small depth of soil cover.

The structural response of a 600 mm inner diameter reinforced concrete pipe buried in a dense, well-graded sand and gravel soil and subjected to surface load from a single design truck axle with 0.3, 0.6, and 0.9 m of soil cover above the pipe crown is quantified using full-scale physical testing. The pipe did not crack at its minimum burial depth of 0.3 m under working CL-625 and CL-800 single-axle highway design loads as the largest tensile strains were only 50%–60% of those at the onset of cracking. Application of the fully factored CL-625 single-axle load at a burial depth of 0.3 m resulted in a tensile crack and a maximum circumferential bending moment of 6 kN·m/m; however, no limit state was reached as the crack width was around one-half the value used to define pipe serviceability and the maximum moment was around 70% of the theoretical ultimate capacity. The decrease in pipe demand from surface load with increasing soil cover is also quantified. At 400 kN of single-axle force, the crown moment decreased to 65% and 35% of the value at 0.3 m burial when the depth of soil cover was increased to 0.6 and 0.9 m, respectively.

Weight, speed, axle space, and following distance of the vehicles are important factors that determine the live load for simply supported beam bridges. In this study, the effect of live load on simply supported bridges was investigated by considering these parameters. The parameters namely weight, speed, longitudinal following space, and vehicle occupancy ratio on different driving lanes are considered using different numerical models. Then, the random traffic flow is simulated using Monte Carlo method based on Nan-xi Yangzi River Bridge’s weigh-in-motion system data. Finite element models are established for six types of simply supported beam bridges with the span ranging from 6 to 40 m and transient analysis is simulated under the traffic loading. Subsequently, the maximum value distribution of live load for lifetime is obtained for the collected data by the method of extrapolation. This study shows that the mean value of the live load for the simply supported beam bridges under a random traffic flow is lower than the permissible design load as per American Association of State Highway and Transportation Officials; however, it is higher than the permissible design load as per BS5400, and it is 0.82 times the Highway-I standard load.

Buried offshore pipes are prone to upheaval and lateral buckling because of the high temperature and pressure of the materials being transported. An attempt has been made in the present study to evaluate uplift and lateral capacity of buried offshore pipes in a soft clay seabed using finite-element software. The numerical model used in the present study has been validated against results available in the literature. Finite-element analyses are carried out to study the upheaval and lateral buckling behavior of pipes for both the no-tension (NT) condition, i.e., when separation occurs between the pipe and soil, and the full-tension (FT) condition, i.e., when the soil beneath the pipe remains attached to the pipe. The variation of both uplift and lateral normalized capacity factors with soil cover depth ratio is presented for different combinations of soil strength and effective weight. Both uplift and lateral capacity factors have been found to increase and reach a constant peak value with increasing depth of soil cover. A transition from global and local failure mechanisms of a pipe is observed when the capacity factors attain their peak values at a limiting soil cover depth. The critical cases for the design of buried offshore pipe in homogeneous clay are established for both NT and FT conditions.

Full-scale field tests were conducted on two 8.5-m (28-ft) span reinforced concrete arch culverts. The culverts were subjected to service live loads at depths of cover from 0.3 to 0.9 m (1 to 3 ft). At 1 ft of fill, the culverts were loaded with a simulated single-axle live load greater than 980 kN (220,000 lb). Field data collected included arch deflection, interface pressure, concrete strain, and reinforcing strain. The computer modeling of the live-load tests is reported. Two- and three-dimensional finiteelement analyses were conducted. Nonlinear material models were used for the backfill soil and the reinforced concrete in both the two- and three-dimensional analyses. Results from the finite-element analyses are compared with the field test data. Results of this study indicate that the two-dimensional models are limited in their ability to predict the longitudinal spreading of live load forces in the culverts and to predict maximum moments larger than those in the actual structure. The three-dimensional finite-element deflection predictions were dependent on the properties used to define the concrete material model; however, the design forces (moment, thrust, and shear) did not show the same sensitivity. This indicates that for typical structural design, material parameters may be approximated without seriously degrading the accuracy of predicted design forces. The test structures met AASHTO design criteria for strength and serviceability.

A large number of flexible pipeline structures exist under roadways. When the cover thickness over the pipeline is limited, effects of live loads can be more pronounced on the structure, especially when a relatively heavy vehicle is moving at low speeds or comes to stop over it. There is a need to address the topic of this paper because only a small number of publications made about buried flexible pipes have paid attention to their performances under static live loads. In this paper, first a few published field tests are examined to find out how shallowly buried flexible pipes respond differently to live loads than to the dead loads. In the second part of the paper, new equations are derived so that the deflection responses of a shallowly installed flexible pipe can be estimated under a static live load. The new equations fill a gap that currently exists in the AASHTO LRFD Bridge Design Specifications. The new deflection equations produced mixed results when their estimates were compared to the field test pipe deflection measurements reported in the literature. This may be partially because of the way the loading tests were carried out in the field. The new equations point out the importance of having a stiff layer above the pipe when the cover thickness is very limited.

The combination of seismic and vehicle live loadings on bridges is an important design consideration. There are well-established design provisions for how the individual loadings affect bridge response: structural components that carry vertical live loads are designed to remain well within the linear-elastic range while lateral load carrying components are designed to yield under large seismic excitations. The weight of the bridge superstructure is taken in to account as dead load in structural analysis for seismic loads; however, the effects of additional mass and damping of live loads on the bridge deck are neglected. To improve the design of highway bridges for multi-hazard effects of seismic plus live load, many questions arise and are addressed in this project via numerical simulations of short span bridges. Further extensions of this research can be extended to long span bridges whose seismic response is more heavily influenced by vehicle mass on the bridge deck.

Many in-service Reinforced concrete box culverts (RCBCs), designed according to older design codes using lighter design truck weights, may require restrictive load posting when rated for current design and permit loads although no structural distress is observed. Most load rating engineers typically utilize 2 dimensional (2D) structural frame software, which leads to fast but highly conservative results. The ratings are negatively affected by the conservative live-load attenuation through soil fill, along with the inability of the 2D method to account for any lateral load distribution. The present study aimed to load rate RCBCs based on strains obtained from field testing. The testing evaluated strains induced by live-load distribution through soil fill when load rating eighteen (18) cast-in-place RCBCs in Kentucky. RCBCs with one-, two-, three-, and four-cells were load tested and the effects of truck live loads analyzed. Fill heights varied from 0 m to 3.6 m, cell spans from 3.0 m to 6.7 m, and culvert spans from 6.7 m to 18.6 m. Strain gauges were attached to the interior surfaces of culverts and the data recorded corresponding to known truck loads and positions. Irrespective of the number of cells in the RCBC, strain variations in the top slab were similar with change in fill height. Maximum strains in the top slabs measured less than 10 microstrain, a value at which live-load effects could be considered negligible, at fill heights greater than 3.05 m. The load rating based on field tests improved both Inventory and Operating rating factors for all tested RCBCs.

Large-diameter pipes, as well as heavy vehicles, have become increasingly prevalent, which imposes uncertainties on pipe design. This paper describes the procedure and results of a series of geotechnical centrifuge tests performed on a large 1400 mm diameter reinforced concrete pipe subjected to heavy traffic loading up to 850 kN. The influence of soil cover depth, as well as the positions and magnitude of traffic loads, on the bending moments of the pipe were investigated. The centrifuge test results were found to be in reasonable agreement with those from full-scale tests. The pipe would experience the most unfavorable conditions when the heaviest axle of the traffic vehicle was located directly above the pipe crown. A deeper soil cover would lead to higher initial stresses in the pipe, as well as reduced influence of traffic load. However, even for a soil cover depth of 4 m, there is significant bending moment induced by the heavy truck loading, which cannot be ignored during pipeline design. A comparison was made between the centrifuge test results and several widely adopted design methods, and unconservative calculation results were noticed for large-diameter rigid pipes lying at a shallow soil cover depth subjected to heavy traffic loading.

Global as well as local behavior of prestressed girder bridges made continuous by adding continuity diaphragms with a recently proposed positive moment continuity detail were investigated in this study. The focus of the investigation is on the positive moment caused by temperature gradients, time dependent effects such as creep and shrinkage, and some live load positions, and on the force transfer mechanisms through the diaphragm. The study utilized different approaches including analytical models for temperature evaluations and finite element models for structural assessments. Field data from a bridge using the new detail were used to validate the developed models. The temperature field of the bridge at different times of the year was estimated using an analytical method. The computed temperature profiles, actual recorded temperatures at the bridge site, and AASHTO specified design gradients are presented and compared. Primary as well as secondary thermal stresses were calculated and restraint moment caused by temperature gradient was quantified. A 3-D finite element model capable of predicting the long term behavior of prestressed girder bridges is presented. A temperature independent creep model was adopted and calibrated using early age data. Construction sequence was considered in the analysis. The FE restraint moment predictions were compared to results obtained from other commonly used analytical method. A parametric study was conducted using the analytical method to investigate the creep coefficient values. Performance of the continuity detail under live load effects was investigated. A live load test was carried out at the bridge site using two loaded trucks. A full bridge 3-D finite element model was also developed and validated with the field data. The validated FE model was also used to investigate the efficiency of the continuity detail. A more detailed 3-D FE model that zooms in on the joint was also built accounting for critical behavioral aspects of the continuity details under service conditions. Contact between cast-in-place concrete and precast concrete, transfer length of prestressing strands, and actual 180º-hook hairpin bar detail were included in the detailed model. Force transfer mechanism, stress distribution and the effective gross moment of inertia at end of girder were investigated.

Link slabs in bridges can be a cost-effective solution to increase the service life of bridge structures. The partial continuity condition in link slab bridges, however can be challenging to quantify. In this work link slab performance is studied by means of a parametric study, a long-term Structural Health Monitoring (SHM) System, and a detailed finite-element model. The parametric study was conducted using mathematical and a two-dimensional (2D) finite element model. The study considers bridge design factors such as number of spans, span length, span length ratio, girder separation, and temperature settings, in conjunction with the corresponding girder and support characteristics to explore the relationship between the different factors that affect link slab performance, as well as develop a simplified expression to represent Link Slab behavior within a system. The effects of live load and temperature gradient were taken into account during the analysis and bearing pad support design recommendations were made. A long-term behavior of a newly constructed bridge was also investigated. Data from a SHM system installed on the Ouachita River Bridge was used to assess the performance of link slabs under varying continuity and support conditions. The performance of the bridge was assessed under live load, temperature gradient and uniform temperature loading conditions. Cracking of the link slab was observed regardless of whether crack control grooves existed or not. It was found that the link slab force are directly affected by link slab cracking. By comparing link slab forces from field data and from a 2D line model, it was found that support conditions are the major factor affecting the magnitude the link slab forces, and hence its performance. Therefore, special attention should be given to bearing pad design under link slabs and to avoid unintentional locking at these location (e.g. due to debris accumulation). A detailed three-dimensional (3D) finite element model was also developed using commercial software package ANSYS to better understand the force transfer mechanism and interactions between the link slabs, the girders and the supports. The results from these analyses contribute toward having a much better understanding of link slab performance, as well as the aspects of bridge design that are affected by the presence of a link slab. The 3D model showed that bearing pads are subjected to higher demands in partially continuous bridges (link slabs) in comparison to fully continuous bridges.

Effect of live load on the seismic design of single-story storage structures under unidirectional horizontal ground motions

The paper presents the static loading of four long-span deep-corrugated steel box culverts with spans of 14 and 8 m. The tests, which were part of a series of full-scale tests, involved strain and displacement measurements carried out at different cover depths under the load of a truck. Two of the culverts were stiffened at the crown. The test results showed that the stiffening applied on the culverts is quite effective and that plain structure is more sensitive to cover depth compared to stiffened structure. The crown stiffening is more effective under shallow soil covers. The maximum displacement of plain structure is approximately twice the maximum displacement of the stiffened structure measured at 45-cm depth of cover. The Swedish and Canadian design methods are conservative when estimating live load moments but they underestimate live load thrusts. The measured maximum thrusts are up to four times larger than the calculated design values where the variation depends on the size of the culverts and the depth of soil. The culvert size has considerable influence on the thrusts and moments generated at the crown centerline, which could be implemented in the design models.