Link slabs are used over the piers developing jointless decks while adjacent bridge spans remain simply supported. The Michigan Department of Transportation incorporates link slabs during deck replacements and deep resurfacing. Field performance assessment documented full-depth cracking of most of the link slabs. These cracks allow surface water infiltration, which leads to accelerated deterioration. This study was conducted to address link slab design and performance issues. The literature is inconsistent with the influence of design parameters on link slab performance. The objective was to document the link slab behavior of its design parameters, to propose a method to calculate the link slab moment and axial force, and to propose recommendations for updating current design details and construction procedures. Single-girder, two-span, finite element assemblage models under various types and levels of loads in conjunction with the link slab design parameters were used to evaluate the moments and axial forces developed in the link slab. Analysis showed that support conditions underneath the link slab greatly influence the link slab moment and axial force. Use of moment interaction diagram is recommended for the design. A detailed analysis and design example is presented incorporating live load, temperature gradient load, and the support configurations.

Side-by-side box-beam bridges are often used at sites with tight under-clearance requirements and specified for accelerated construction. However, longitudinal reflective deck cracking is a recurring problem for these bridges and it raises concern for their durability and long-term safety. North American practices of the transverse connection design of this particular bridge are discussed in NCHRP Synthesis 393, and the Michigan design is presented as the preferred procedure. The most recent design in Michigan is an empirical procedure that incorporates the majority of the Synthesis 393 recommended best practices. Yet reflective deck cracking persists. A rational design procedure for which an analysis model was developed is presented here. In the rational design procedure, the analysis model is utilized to calculate the moment demand along the transverse joints. A two-stage transverse posttensioning procedure is recommended following AASHTO load and resistance factor design stipulations based on the moment demand calculated from the analysis model. The second stage of posttensioning precompresses the cast-in-place concrete deck, ensuring crack control. The objective of this study is to demonstrate the effectiveness of the two-stage posttensioning design and implementation procedure to mitigate reflective deck cracking. The effectiveness of the recommended design is demonstrated by utilizing a multistep finite element simulation of the construction process under construction and service loads. The cracking potential of the deck is evaluated under both current Michigan Department of Transportation and the proposed two-stage transverse posttensioning schemes. It is demonstrated that the two-stage posttensioning process can eliminate tensile stresses developed under gravity loading and reduce the temperature load effects to abate reflective deck cracking.

In the seismic analysis practice, the calculation of modal response has traditionally been limited to a cutoff frequency of about 33 Hz based on United States Nuclear Regulatory Commission (US NRC) Regulatory Guide (RG) 1.60 [1] response spectra. The structural response in higher modes is calculated as a missing mass correction by static analysis. Seismic ground motions at several sites (such as Central and Eastern United States) exhibit high frequency content, up to about 100 Hz. Additionally, the reactor building vibratory (RBV) loads that result from the suppression pool hydrodynamic loads due to loss of coolant accident (LOCA), and the annulus pressurization (AP) load from a postulated pipe break at the reactor pressure vessel (RPV) safe ends and shield wall generate peaks at frequencies in excess of 100 Hz. The qualification of safety equipment supported in the reactor building needs to reflect these high frequency motions. Extracting frequencies and mode shapes up to zero period acceleration (ZPA) frequencies in these cases may not be practical or economical. Therefore, the cutoff frequency criteria for these types of high frequency loads need to be evaluated so that the analysis produces a representative and a reasonably conservative response. In this study, the equipment response is described in terms of stress quantities, member forces, and moments resulting from the solution up to a cutoff frequency. The responses are compared to the full solution up to the ZPA frequency under hydrodynamic and AP loads using the Response Spectrum Method. The cutoff frequency is deemed adequate if the ratio of the truncated response considering missing mass to the full response is 90% or greater. The internal strain energy (or its surrogate kinetic energy) for all modes with frequencies below the cutoff is also studied to assess the missing strain energy in modes in excess of the cutoff. The evaluation presented also examines how well the strain energy correlates with calculated stresses.

In-line valves are qualified for static as well as dynamic loads from seismic and hydrodynamic (HD) events. Seismic loads are generally characterized by frequency content less than about 33 Hz whereas HD loads may exhibit a broad range of frequencies greater than 33 Hz. HD loads may also result in spectral accelerations significantly in excess of those due to the design basis seismic events. Current regulatory guidelines do not specifically address the evaluation of equipment response to high frequency loading. This paper investigates the response of skid and line mounted valves of piping systems under HD loads by using several independent rigorous finite element analysis solutions for various piping system segments. It presents a hybrid approach for the evaluation of the response of valves to HD and seismic loads. The proposed approach significantly reduces the amount of individual analysis and testing needed to qualify the valves. First, valve responses are evaluated on the basis of displacements since HD loads are generally characterized by high frequencies and small durations. Second, the damage potential of the loads on the valve actuators is represented by the energy imparted to the actuator quantified in terms of Arias intensity. The rationale for using the energy content is based on the fact that damage due to dynamic loading is related not only to the amplitude of the acceleration response but also to the duration and the number of cycles over which this acceleration is imposed.