Load Rating of Segmental Bridges (2024)
Chapter: 3 Literature Synthesis and Gaps
CHAPTER 3
Literature Synthesis and Gaps
The following sections provide a synthesis of findings from the literature review conducted in Chapter 2. A review and synthesis of current specifications for the design and evaluation of segmental bridges are critical in identifying knowledge gaps and establishing a coherent definition for what structures are covered by the load rating provisions developed in the proposed research. This work is a critical step in the continuing development of an appropriate analytical program.
3.1 AASHTO MBE—Segmental Bridge Sections
In the current MBE, Section 6A.5.11.3 states that for the transverse operating load ratings of the top slab of segmental concrete box girders, the factor of 1.20 specified in AASHTO LRFD Design Table 3.6.1.1.2-1 for one loaded lane shall be limited to a maximum of 1.00 (AASHTO 2020a). Since AASHTO LRFD (2020a) was calibrated based on the presence of two loaded lanes, the multiple presence factor (MPF), m, for one loaded lane is 1.20 rather than 1.00 (AASHTO 2020a). In AASHTO MBE C6A.5.11.3, it is explained that the transverse design of the top slab of segmental bridges is governed by axle loads. The amplification of individual axle loads for the single-lane condition is not appropriate. Maximum credible axle loads are less uncertain than the maximum credible vehicle loads as axle loads are limited by the bending resistance of vehicle axles. Therefore, it is limited to a maximum of 1.00.
Also, AASHTO MBE Section 6A.5.11.4 suggests that for operating rating of the design load at the service limit state, the number of live load lanes may be taken as the number of striped lanes (AASHTO 2020b). However, loads shall be positioned to create maximum effects, for example, on shoulders if necessary. The reason to consider striped lanes is to calibrate the service limit states and distinguish the operating rating from the inventory rating. It is also stated that while the use of number of striped lanes results in lower reliability for ratings at the service limit states, the resultant increment in βT is unknown. It was concluded that the use of the number of striped lanes is appropriate for operating ratings based on a brief study of existing bridges. However, the use of the number of striped lanes, or alternative solutions for the operating rating, requires further investigation and calibration.
Another important provision is related to system factors (φs). Since the previous LRFR did not sufficiently address the characteristic behavior of post-tensioned segmental box girder construction in the areas of longitudinal continuity (Corven Engineering 2004), continuum of the closed box girder in terms of damage resistance, multiple-tendon paths, and internal redundancy, system factors are established in the current AASHTO MBE for segmental bridges, considering some factors different from other types of bridges. NCHRP Report 406 (Ghosn and Moses 1998) and FDOT’s New Directions for Post-Tensioned Bridges (Corven Engineering 2004) provide findings for the system factors. Further investigation and results are provided in Section 6.5.9.
In the calibration of the current factors, the probabilistic approach shall be used. Data collection might be performed via as-built design plans, construction documents, inspection reports, repair or rehabilitation documents, and field survey and measurements.
3.2 Calibration of Concrete Bridges for Serviceability
The serviceability of the segmental bridge is of great concern. The minimum limit states include the flexure tensile stress, flexural compressive stress, and principal tensile stress for both longitudinal direction and transverse direction.
A successful reliability calibration requires rational limit state functions and target beta (βT), rigorous calibration procedures, and reliable resistance and load models. Both the limit state function and target beta depend on the failure type (i.e., limit state). Once the resistance equation is determined for each limit state, the limit state can be formed. Regarding target beta, the original calibration work for AASHTO LRFD appoints βT to be 3.5 based on a series of existing girder or slab bridges (AASHTO 2020a). The corresponding βT for LRFR is 2.5 for the 5-year rating level with a minimum allowed β of 1.5. The proper βT should also be determined for segmental bridges, which were not considered in the previous LRFD/LRFR calibrations. It is important to know the reliability level of the existing segmental bridge. The research team adopted the methodology from NCHRP Project 12-83 to determine the βT. From state agency feedback and industry experience, determining whether a new βT should be developed depends on the existing segmental bridge’s satisfactory performance. If the existing bridges have a satisfactory reliability level, then the target beta can be directly determined from the existing βs. Minor adjustments may be needed for a certain limit state if the β is not uniform. However, if the existing bridges exhibit unsatisfactory safety levels (e.g., cracks in early age), a new βT could be determined from other specifications (e.g., European codes, Canadian codes).
The statistical load models for the segmental bridges are expected to be much different from those for girder bridges in the following aspects. First, the previous load models developed for girder bridges consider the longitudinal directions only, whereas the segmental bridge shall be evaluated in both longitudinal direction and transverse direction. Second, in the longitudinal direction, the bending and shear were the main concerns for the girder bridge, whereas the torsion and time-dependent effects in addition to the bending and shear should also be considered for segmental bridges. Third, while simulating the live load effects, most of the previous calibration works assume the bridges span from 30 ft. to 200 ft., while the segmental bridges may exhibit a longer span length. Additionally, the live load analysis for girder bridges is typically done by simple beam analysis, whereas the research team discovered the need for more refined analysis, such as the 2D planar frame analysis and influence surface. Finally, the segmental bridges are expected not to be as sensitive to heavy trucks as the girder bridges because the entire cross section distributes the loads. As a result, the live load distribution on a normal probability paper may also be different from what girder bridges show.
There is a need to develop new statistical live load models for segmental bridges. The effects of the return period and the average daily truck traffic (ADTT) need to be determined. The WIM data should be requested nationwide based on live load. For this research, the typical segmental bridge configurations (e.g., span length, width) were chosen from the segmental bridge inventory.
3.3 Deterioration of Segmental Bridges
Consideration of structural deterioration is an essential topic for the bridge load rating as it is directly related to the assessment of the component’s load-carrying capacity. Condition factors are typically used to account for the reduction in load-carrying capacity. Based on the literature,
deterioration of segmental bridges is mostly due to corrosion of tendons. A correlation of various factors with corrosion of tendons might be standardized. These factors might include the location of the bridge, the type of structure, the type of tendon in terms of bonded/unbonded or grouted/ungrouted, the corroded section properties, such as cracked/uncracked, and the location of it through the member. In the literature, there are a few studies that reveal the relation between corrosion rate and section losses over time. This correlation can be used to derive the condition factors in the load rating analysis.
Regarding the research feasibility, corrosion of tendons can be detected by various methods during inspection of post-tensioned segmental concrete bridges. However, during routine inspections, the deterioration detections are mostly visual without advanced techniques. It may be difficult for an inspector to judge the severity of the tendon corrosion. Because of the lack of unbiased judgment of the actual tendon corrosion, the research team proposes to collect the tendon corrosion data from the other references, such as experiment-based reports and articles.
3.4 Condition Factors in Load Rating
The relationship between National Bridge Inventory (NBI) rating and φc can be found in MBE-3-E1 October 2018 Errata, Section 6A.4.2.3. Table 6A.4.2.3-1 provides the condition factor φc based on structural condition of member (see Table 3-1). Also, based on C6A.4.2.3, if condition information is collected and recorded in the form of NBI condition rating only (not as element-level data), then the following approximate conversion may be applied in selecting φc (see Table 3-2.).
Table 3-1 and Table 3-2 show the correlation between NBI Ratings and φc. However, as it is stated in MBE C6A.4.2.3, this relation is an approximation. Furthermore, it might not be accurate enough for the post-tensioned segmental concrete bridges. More investigation is needed. Also, FDOT’s New Directions for Florida Post-Tensioned Bridges offers illustrative examples for φc. These examples are created by considering the typical conditions of Florida concrete segmental bridges. However, there is no available reliability analysis to support the condition factors. Therefore, a calibration might be needed. In the calibration of condition factors, a probabilistic approach shall be followed by using the NBI database and inspection reports of segmental concrete bridges. However, it is worth noting that, although the section losses may be obtained using nondestructive testing/nondestructive examination (NDT/NDE), the condition factor
Table 3-1. Condition factor: φc.
| Structural Condition of Member | φc |
|---|---|
| Good and Satisfactory | 1.00 |
| Fair | 0.95 |
| Poor | 0.85 |
Table 3-2. Approximate conversion in selecting φc.
| Structural Condition Rating (SI&A Item 59) | Equivalent Member Structural Condition |
|---|---|
| 6 or higher | Good and Satisfactory |
| 5 | Fair |
| 4 or lower | Poor |
should still be applied, because not only does the condition factor account for reduced load-carrying capacity, it also considers the increased variability of the resistance. In addition, the accuracy of the data obtained for the section losses from NDT/NDE is a challenge for engineers to use given the data’s variability and lack of reliability. It should also be mentioned that the determination of post-tensioning tendon section loss (wire cross section reduction, wire break(s), loss of strands) is the subject of specialized inspections that are not typically part of biennial inspections. Use of general condition factors based on external visual assessments seems appropriate when specialized inspection results are not available. There is merit in the biennial inspection team becoming familiar with the segmental bridge post-tensioning details before conducting the inspection. Overt service condition defects particular to segmental bridges (joint openings, web cracking, transverse flexural cracking) and their post-tensioning systems (duct defects, cracked or spalled anchorage protection, signs of water migration through joints in the vicinity of tendons) could drive the condition rating of the superstructure.
The load rating engineer, in consultation with the bridge owner, should determine the effects of tendon section loss when the results of specialized inspections are available. Consideration should be given to items such as the following:
- Controlling load rating behavior (without section loss).
- The critical nature of the tendons with section loss (e.g., highly draped external span-by-span tendons that are typically few in number versus relatively flat cantilever tendons that are greater in number).
- Level of stress in the tendons—section loss may be such that it may increase the stress in remaining strands of a tendon without negatively impacting the ability to achieve service limit or strength limit stress requirements.
3.5 General Rating Equation for Load Rating
The MBE incorporates specific provisions for load rating of segmental concrete bridges in Chapter 6A.5.11. Longitudinal and transverse direction load rating capacities are to be checked. The general rating equation is as follows:
| (3-1) |
For strength limit states:
For service limit states:
where
| RF = | Rating factor. |
| C = | Capacity. |
| fR = | Allowable stress specified in the LRFD code. |
| Rn = | Nominal member resistance (as inspected). |
| DC = | Dead load effect due to structural components and attachments. |
| DW = | Dead load effect due to wearing surface and utilities. |
| P = | Permanent loads other than dead loads. |
| LL = | Live load effect. |
| IM = | Dynamic load allowance. |
| γDC = | LRFD load factor for structural components and attachments. |
| γDW = | LRFD load factor for wearing surfaces and utilities. |
| γLL = | Evaluation of live load factor. |
| γP = | LRFD load factor for permanent loads other than dead loads = 1.0. |
| ϕs = | System factor. |
| ϕc = | Condition factor. |
| ϕ = | LRFD resistance factor. |
FDOT currently provides the most detailed manual for segmental bridge load rating. In Chapter 6 of FDOT’s Manual for Bridge Load Rating (2021), FDOT modifies the current AASHTO MBE by removing and/or adding several specifications to achieve the load rating analysis in accordance with their minimum requirements.
| (3-2) |
where
| RF = | Rating factor. |
| C = | Factored capacity. |
| DC = | Dead load effect due to structural components and attachments. |
| DW = | Dead load effect due to wearing surface and utilities. |
| EL = | Permanent locked-in erection forces. |
| FR = | Bearing friction or frame action. |
| TU = | Uniform temperature. |
| γ = | LRFD load factor. |
| CR = | Creep. |
| SH = | Shrinkage. |
| LL = | Live load effect. |
| IM = | Dynamic load allowance. |
FDOT has included a table for live load factors for several limit states in the analysis of transverse and longitudinal directions (see Table 3-3). FDOT also includes stress limits for several conditions in the segmental bridges (Corven Engineering 2004), as shown in Table 3-4.
Table 3-5 summarizes the main differences between the specifications for load rating of segmental bridges in the current MBE and the Bridge Load Rating Manual (FDOT 2021).
Table 3-3. FDOT Table 6A.5.11-1—LRFR live load factors for segmental bridges.
| Direction and Limit | Inventory | Operating1 and FL1201 | EV1 | |
|---|---|---|---|---|
| Longitudinal | Strength, Flexure | 1.75 | 1.35 | 1.30 |
| Strength, Shear | 1.75 | 1.35 | 1.30 | |
| Service III, Flanges | 0.80 | 0.90 SL2 | 0.90 SL2 | |
| Service III, Web | 0.80 | 0.90 SL2 | 0.90 SL2 | |
| Transverse3 | Strength, Flexure | 1.75 | 1.35 | 1.30 |
| Service I | 1.00 | 1.00 | 1.00 | |
1. Apply the MPF to all loaded lanes, per LRFD 3.6.1.1.2, except make the single-lane MPF 1.00 for operating emergency vehicle (EV) and FL120 permit levels.
2. “SL” means the number of striped lanes; consider 1 ≤ lanes loaded ≤ SL.
3. For transverse limits, omit all lane loading.
Table 3-4. FDOT Table 8.2.B—allowable stresses for concrete bridges.
| At the Service Limit State after losses | Stress Limit INVENTORY Rating | Stress Limit OPERATING Rating | Source of Criteria |
|---|---|---|---|
Compression (Longitudinal or Transverse):
|
0.60f′c | 0.60f′c | |
|
Seg Guide Spec 9.2.2.1 | ||
|
Longitudinal Tensile Stress in Precompressed Tensile Zone: (Intended for Pre- and Post-Tensioned Beams and similar construction). For components with bonded prestressing tendons or reinforcement that are subject to not worse than for:
|
3√f′c psi tension | 7.5√f′c psi tension | LRFD Table 5.9.4.2.2-1 and FDOT FDOT no distinction for environment |
| No Tension | No Tension | LRFD Table 5.9.4.2.2-1 | |
|
Longitudinal Tensile Stress through Joints in Precompressed Tensile Zone:
(Intended for Segmental and similar construction)
|
3√f′c psi tension | 7.5√f′c psi tension | LRFD Table 5.9.4.2.2-1 Seg Guide Spec 9.2.2.2 FDOT has no distinction for environment |
|
No tension | No tension | Ditto and FDOT Seg. Rating Criteria |
|
100 psi min comp | No tension | Seg Guide Spec 9.2.2.2 FDOT Seg. Rating Criteria |
Transverse Tension, Bonded PT:
|
3√f′c psi tension |
6√f′c psi tension |
Seg Guide Spec 9.2.2.3 LRFD Table 5.9.4.2.2-1 FDOT has no distinction for environment FDOT Seg. Rating Criteria |
| At the Service Limit State after losses | Stress Limit INVENTORY Rating | Stress Limit OPERATING Rating | Source of Criteria |
|---|---|---|---|
Tensile Stress in Other Areas:
|
No tension |
No tension |
Seg Guide Spec 9.2.2.4 LRFD Table 5.9.4.2.2-1 |
|
6√f′c psi tension |
6√f′c psi tension |
Seg Guide Spec 9.2.2.4 LRFD Table 5.9.4.2.2-1 |
Principal Tensile Stress at Neutral Axis in Webs (Service III):
|
3√f′c psi tension |
4√f′c psi tension |
FDOT LRFR Rating Criteria |
|
* Principal tensile stress is calculated for longitudinal stress and maximum shear stress due to shear or combination of shear and torsion, whichever is greater. For segmental box, check neutral axis. For composite beam, check at neutral axis of beam only and at neutral axis of composite section and take the maximum value. Web width is measured perpendicular to plane of web. For segmental box, it is not necessary to consider coexistent web flexure. Account should be taken of vertical compressive stress from vertical PT bars provided in the web, if any, but not including vertical component of longitudinal draped post-tensioning - the latter should be deducted from shear force due to applied loads. Check section at H/2 from edge of bearing or face of diaphragm, or at end of anchor block transition, whichever is more critical. For the design of a new bridge, a temporary principal tensile stress of 4.5√f′c may be allowed during construction—per AASHTO Seg. Guide Spec. Initial load ratings for new design should be based on specified concrete strength. Load rating of an existing bridge should be based on actual concrete strength from construction or subsequent test data. |
|||
Source: Corven Engineering (2004).
Table 3-5. Summary of load rating specifications.
| MBE | FDOT | |
| General Rating Equation | Load factors γp = 1.00 for permanent loads. | Load factors for locked-in erection loads, bearing forces, and creep and shrinkage effects. Stress limit defined for longitudinal and transverse direction analysis. |
| Live Load |
MPFs are in accordance with AASHTO LRFD Table 3.6.1.1.2-2 (2020a). MPF = 1.0 for transverse operating rating with one lane loading. |
MPF as per AASHTO 3.6.1.1.2 except MPF = 1.00 for operating emergency vehicle (EV) and FL120 permit levels SL: Striped Lanes |
| Limit States | Strength I Service I Service III: Principal Tensile Stress A5.9.2.3.3 For operating rating live load lanes may be taken as # of striped lanes |
Longitudinal:
Strength: Flexure & Shear Service III: Web/Flanges Transverse: Strength: Flexure Service I For transverse limits omit lane loading. |
| Service Limit State | Legal Load Rating Permit Load Rating |
Inventory Operating Legal Emergency Vehicles Stress Limits for Joint Types |
| System Factors | Table 6A.5.11.6-1 for Flexure ϕs = 1.0 for Longitudinal shear and torsion, transverse flexure, and punching shear |
Conservatively apply a system factor of 1.00 |
MPF: Multiple presence factor.
