Riveted Hull Joint Design in RMS Titanic and Other Pre—World War I Ships

2003 ◽  
Vol 40 (02) ◽  
pp. 82-92
Author(s):  
Richard Woytowich

Beginning with an overview of riveted joint construction, this paper shows that the efficiency of riveted joints in pre-World War I ships decreased as plate thickness increased. In the case of the RMS Titanic, some of the joints involved in the iceberg impact were only about 27% as strong as the plates they connected. A finite element model is used to show how such a joint would respond to the sort of out-of-plane load that the iceberg would have applied. For one possible load configuration, the joint failure is recreated. Finally, although Titanic and her sisters were not built to class, the design of the riveted joints is examined in the context of relevant Lloyd's Register of Shipping Rules.

Author(s):  
Daniel E. Jordy ◽  
Mohammad I. Younis

Squeeze film damping has a significant effect on the dynamic response of MEMS devices that employ perforated microstructures with large planar areas and small gap widths separating them from the substrate. Perforations can alter the effect of squeeze film damping by allowing the gas underneath the device to easily escape, thereby lowering the damping. By decreasing the size of the holes, the damping increases and the squeeze film damping effect increases. This can be used to minimize the out-of-plane motion of the microstructures toward the substrate, thereby minimizing the possibility of contact and stiction. This paper aims to explore the use of the squeeze-film damping phenomenon as a way to mitigate shock and minimize the possibility of stiction and failure in this class of MEMS devices. As a case study, we consider a G-sensor, which is a sort of a threshold accelerometer, employed in an arming and fusing chip. We study the effect of changing the size of the perforation holes and the gap width separating the microstructure from the substrate. We use a multi-physics finite-element model built using the software ANSYS. First, a modal analysis is conducted to calculate the out-of-plane natural frequency of the G-sensor. Then, a squeeze-film damping finite-element model, for both the air underneath the structure and the flow of the air through the perforations, is developed and utilized to estimate the damping coefficients for several hole sizes. Results are shown for various models of squeeze-film damping assuming no holes, large holes, and assuming a finite pressure drop across the holes, which is the most accurate way of modeling. The extracted damping coefficients are then used in a transient structural-shock analysis. Finally, the transient shock analysis is used to determine the shock loads that induce contacts between the G-sensor and the underlying substrate. It is found that the threshold of shock to contact the substrate has increased significantly when decreasing the holes size or the gap width, which is very promising to help mitigate stiction in this class of devices, thereby improving their reliability.


Author(s):  
Mostafa Habibi ◽  
Ramin Hashemi ◽  
Ahmad Ghazanfari ◽  
Reza Naghdabadi ◽  
Ahmad Assempour

Forming limit diagram is often used as a criterion to predict necking initiation in sheet metal forming processes. In this study, the forming limit diagram was obtained through the inclusion of the Marciniak–Kaczynski model in the Nakazima out-of-plane test finite element model and also a flat model. The effect of bending on the forming limit diagram was investigated numerically and experimentally. Data required for this simulation were determined through a simple tension test in three directions. After comparing the results of the flat and Nakazima finite element models with the experimental results, the forming limit diagram computed by the Nakazima finite element model was more convenient with less than 10% at the lower level of the experimental forming limit diagram.


1998 ◽  
Vol 1624 (1) ◽  
pp. 184-195 ◽  
Author(s):  
Thomas C. McCavour ◽  
Peter M. Byrne ◽  
Timothy D. Morrison

A comprehensive investigation of soil–metal structure interaction for long-span deep-corrugated reinforced steel box culverts was carried out in a project sponsored by the National Research Council of Canada in 1996. Two 12-m span box culverts were erected at a Dorchester, New Brunswick, test site using two backfill densities, one structural steel plate thickness, and a minimum cover of 300 mm. These structures are the largest steel box culverts erected to date. One structure was reinforced using continuous deep-corrugated crown stiffeners, and the other was intermittently reinforced using composite concrete metal-encased stiffeners. Strain and deflection of the structure were monitored in response to static axle loads positioned at six locations on the test surface. A finite element model was then used in numerical simulations of the soil–metal structure system. The measured culvert response was then compared with results from the finite element model. A nonlinear soil-structure interaction program (NLSSIP) was used to analyze the two long-span box culverts. NLSSIP was developed specifically for long-span soil–metal culverts and has been used for structures with and without stiffeners. The box culvert test provided a definitive relationship between soil stiffness and metal structure stiffness. The test was the first that evaluated intermittently reinforced composite concrete metal-encased stiffeners relative to conventional continuous reinforcement. The performance of the two types of stiffeners is evaluated and recommendations are made for future design and installation of long-span deep-corrugated steel box culverts.


2013 ◽  
Vol 351-352 ◽  
pp. 169-173
Author(s):  
Kuan Tang Xi ◽  
Jin Li ◽  
Tie Gang Zhou ◽  
Qing Xing Xu

Two kinds of finite element model which can reflect the effects of different loading positions were constructed with Beam 188 and Shell 181. Effects of different restraints, load models and rise-span ratios on out-of-plane buckling were studied by comparing results of fixed arches with that of pin-end arches under three loading models. It is conservative to design by employing results of radial loading. As for out-of-plane stability, pin-end arches are better than fixed arches when rise-span ratio is big. Compared with U-section pin-end circular arches with diaphragm, those with batten plates have batter out-of-plane stability, and they are more economical and easier to construct.


2004 ◽  
Vol 31 (3) ◽  
pp. 440-452 ◽  
Author(s):  
Ilian Zinoviev ◽  
Magdi Mohareb

A methodology for the analysis and design of laterally unsupported portal frames is proposed. A finite element model is developed to predict the elastic critical load and associated buckling mode. Regression analysis is then conducted to find lateral displacement and rotation field expressions that closely approximate the buckled configurations predicted by the finite element analysis. The obtained functions are then substituted into the total potential energy expression, and the stationarity conditions are evoked. The resulting eigen-value problem is solved for the out-of-plane buckling loads that are then compared with those based on the finite element model. The agreement between the two solutions provides an indication of the accuracy of the simplified energy solution. The member destabilizing effects induced by axial forces are separated from those induced by strong axis bending. The separation of these two effects is subsequently exploited in a two-step eigen-value procedure, aimed at determining the key member resistances defined in the interaction check of the standard CSA-S16-01, while accurately modeling the boundary conditions of the member. These are (i) compressive resistance of the member in the absence of bending effects and (ii) flexural resistance of the member in the absence of axial force effects.Key words: portal frames, lateral buckling, finite element analysis, wide flange sections, frame design, principle of stationary potential energy.


2013 ◽  
Vol 405-408 ◽  
pp. 781-785
Author(s):  
Kuan Tang Xi ◽  
Jin Li ◽  
Tie Gang Zhou ◽  
Tao Lin

Finite element model which can reflect the effects of different loading positions were constructed with Beam 188. Effects of different restraints, load models and rise-span ratios on out-of-plane buckling were studied by comparing results of fixed arches with that of pin-end arches under three loading models. It is conservative to design by employing results of radial loading. For ideal restraints, out-of-plane stability of pin-end arches is better than fixed arches when rise-span ratio is big. Effects of different loading positions on out-of-plane buckling were studied. Buckling loads of arches which are loaded at arch-axises are bigger than those of top flanges, but smaller than those of bottom flanges.


2017 ◽  
Vol 26 (3) ◽  
pp. 096369351702600
Author(s):  
Feng Zhang ◽  
Lei Gao

The debonding of the FRP plate from concrete and crack-propagation processes are complex and the current research studies regarding this debonding mechanism are insufficient and not comprehensive. This work proposes a plane stress model along with equal width and different width FRP to concrete models to simulate the debonding and crack-propagation processes are presented. The longitudinal and horizontal stress distributions were analysed and the FRP to concrete width effect and FRP thickness parameters were also studied by means of the proposed three-dimensional finite element model. The results show that the different width 3D model is optimal for analysing the spatial interfacial debonding of FRP from concrete. The concrete surface horizontal stress distribution along the length of the concrete substrate could judge the effective bond length. Both the normal stress and shear stress are mainly divided into the following two small central stress regions under the PRP plate: a high stress gradient region near the FRP plate edge and a stress-free region near the concrete edge. The debonding strength and the stiffness of the bonding interface increase with the width of the FRP plate and the FRP plate thickness. The stress range and magnitude are strongly dependent on the width of the FRP plate. Debonding begins at the FRP plate edge; the thicker FRP plate more easily exhibits debonding.


Sign in / Sign up

Export Citation Format

Share Document