No abstract

No abstract

A computational modeling effort was undertaken to combine finite element analysis (FEA) and computational fluid dynamics (CFD) methods to simulate the closing of the blind shear rams of a blowout preventer (BOP) under flowing conditions. The objective of this effort was to develop a high-fidelity fluid-structure interaction (FSI) simulation methodology that reliably assesses the combined mechanical and hydrodynamic forces acting on BOP shear rams that could potentially impact the rams’ ability to safely shut-in a well. BOP shear ram designers typically consider the material properties and geometry of the drill pipe to be sheared and the hydrostatic flowing pressure during ram closure. Fluid hydrodynamic effects on the rams are difficult to simulate and currently cannot be produced in a laboratory setting due to complexity and personnel safety in conducting such tests under high transient conditions, and thus are often neglected. To determine the best computational approach in terms of complexity and accuracy, a novel simulation methodology was developed by coupling the fluid interaction with the BOP as it shears a drill pipe using LS-DYNA® and ANSYS® Fluent®. The conditions and assumptions made for this analysis are presented herein, and initial simulation cases are compared against validation data to confirm model accuracy. For a high-pressure, high-flow well scenario in the Gulf of Mexico (GOM), a one-way coupling of the FEA and CFD simulations was determined to be the best approach for modeling the closing of blind shear rams under flowing conditions. This investigation also confirms that as long as the fluid is single phase, the ram forces due to fluid flow effects are small in comparison to the mechanical shearing force. It is noted, that highly dynamic flow events such as slugging flow or the potential erosive effects of sands or solid particles present additional risks, and the analysis methodology described here can serve as the basis for additional investigations into more complicated flow scenarios.

Abstract. The development and design of a cryogenic liquid quantity gauge for zero-g applications is described . The gauge, named the Compression Mass Gauge (CMG), operates on the principle of slightly changing the vo lume of the tank by an oscillating bellows. The resulting pressure change is measured and used to predict the volume of vapor in the tank, from which the volume of liquid is computed. For each gauging instance, pressures are measured for several different bellows frequencies to enable minor real-gas effects to be quantified and thereby to obtain a gauging accuracy of ± 1 % of tank volume. Southwest Research Institute™ and NASA-GRC have developed several previous breadboard and engineering development gauges and tested them in cryogenic hydrogen and nitrogen to establish the gauge capabilities, to resolve several design issues, and to formu late data process ing algorithms. The CMG has been selected by NASA's Future X program for a fli gh t demonstration on the USAFlBoeing Solar Thermal Vehicle Space Experiment (SOTVSE). This paper reviews the design trade studies needed to satisfy the SOTVSE limitations on CMG power, volume, and mass, and describes the mechanical design of the CMG.

This paper presents results of a computational study conducted to assess the multiscale resolution capabilities and limitations of different Detached Eddy Simulation (DES) multiscale turbulence models in unsteady flow predictions for internal axial flow in a single rod channel configuration. Two different DES models are compared in the present analysis. The DES models are based on the Spalart-Allmaras (S-A) one-equation model and the two-equation realizable k-ε model. A detailed assessment of the DES turbulence model coefficient for the S-A based DES model is presented. The predicted time-averaged mean velocity and turbulent stresses are compared with the available experimental results. Flow unsteadiness, which is important for determining heat, momentum, and mass transfer in the gap region, is presented through time histories and spectra of flow quantities. The unsteady spectra for the velocity fluctuations are also compared with the experimental observations. The results demonstrate that the DES turbulence model coefficient significantly influence the predicted solution. The realizable k-ε-model-based DES model is found to be numerically more stable than the one-equation S-A-based DES model. Predicted results demonstrate that the modifications need to be incorporated in the current DES model formulations for proper prediction of wall bounded internal turbulent flows.