Publications

Simmermacher, Todd W.; Tipton, David G.; Cap, Jerome S.; Mayes, R.L.

Abstract not provided.

Tipton, David G.

Abstract not provided.

Rouse, Jerry W.; Walsh, Timothy W.; Tipton, David G.; Stasiunas, Eric C.; Lawry, Matthew W.

Abstract not provided.

Rouse, Jerry W.; Walsh, Timothy W.; Tipton, David G.; Stasiunas, Eric C.; Lawry, Matthew W.

Abstract not provided.

Cap, Jerome S.; Reese, Garth M.; Tipton, David G.; Walsh, Timothy W.

Abstract not provided.

Walsh, Timothy W.; Tipton, David G.; Reese, Garth M.

Abstract not provided.

Tipton, David G.; Bhardwaj, Manoj K.; Wilson, Christopher R.; Dohrmann, Clark R.; Reese, Garth M.; Walsh, Timothy W.; Rouse, Jerry W.

Abstract not provided.

Conference Proceedings of the Society for Experimental Mechanics Series

Garcia, Ricardo M.; Tipton, David G.

This paper investigates the effectiveness of modeling thin composite structures with hex shell elements for structural dynamics simulation. The current finite element modeling method for an existing three-layer composite aerospace structure uses solid 8-noded hex elements. It is relatively expensive in terms of the number of degrees of freedom and element count. A finer mesh typically results in a more accurate solution, however, the computation time increases. Modal analysis was used to test if a single layer of hex shell elements for each material could replace multiple layers of solid hex elements, enabling computational savings. Element aspect ratio was varied on a solid hex model of a frustum part to optimize the technique. The hex shell modeling technique was then applied to the existing three-layer composite structure. The analysis results, when compared to validation data obtained from tests performed on the actual hardware, exhibit very satisfactory agreement. A single layer of hex shell elements are capable of providing solutions that are equivalent to multiple layers of hex elements. A considerable savings in element count and solution equations result. A broader understanding of modeling options for future, more efficient methods of modeling composite shell structures is also obtained. ©2010 Society for Experimental Mechanics Inc.

Tipton, David G.; Hunter, Patrick H.

Abstract not provided.

Cap, Jerome S.; Tipton, David G.

Abstract not provided.

Tipton, David G.

Abstract not provided.

Tipton, David G.; Hunter, Patrick H.

Abstract not provided.

Cap, Jerome S.; Tipton, David G.

Abstract not provided.

Sandia journal manuscript; Not yet accepted for publication

Tipton, David G.

This article is the second of two that consider the treatment of fluid-solid interaction problems where the solid experiences wave loading and large bulk Lagrangian displacements. In part-I, we presented the formulation for the edge-based unstructured-grid Euler solver in the context of a discontinuous- Galerkin framework with the extensions used to treat internal fluid-solid interfaces. A super-sampled L² projection was used to construct level-set data from the Lagrangian interface, and a narrow-band approach was used to identify and construct appropriate ghost data and boundary conditions at the fluid-solid interface. A series of benchmark problems were used to verify the treatment of the fluid-solid interface conditions with a static interface position. In this paper, we consider the treatment of dynamic interfaces and the associated large bulk Lagrangian displacements of the solid.We present the coupled dynamic fluid-solid system, and develop an explicit, monolithic treatment of the fully-coupled system. The conditions associated with moving interfaces and their implementation are discussed. A comparison of moving vs. fixed reference frames is used to verify the dynamic interface treatment. Lastly, a series of two and and three-dimensional projectile and shock-body interaction calculations are presented. Ultimately, the use of the Lagrangian interface position and a super-sampled projection for fast level-set construction, the narrow-band extraction of ghost data, and monolithic explicit solution algorithm has proved to be a computationally efficient means for treating shock induced fluid-solid interaction problems.

Sandia journal manuscript; Not yet accepted for publication

Tipton, David G.

Here, this paper is the first of two that consider the treatment of fluid-solid interaction problems under shock wave loading, where the solid experiences large bulk Lagrangian displacements. This work addresses the issues associated with using a level-set as a generalized interface for fluid-solid coupling where unstructured overlapping grids are used for the fluid and solid domains. In part-I of this work, we outline the formulation used for the edge-based unstructured-grid Euler solver in the context of the discontinuous-Galerkin method. The identification of the fluid-solid interface on the unstructured fluid mesh uses a super-sampled L² projection technique, that in conjunction with a Lagrangian interface position, permits fast identification of the interface and the concomitant imposition of boundary conditions. The use of a narrow-band approach for the identification of the wetted interface is presented with the details of the construction of interface conditions. A series of computations are presented to demonstrate the validity of the current approach on problems with static interfaces. In part-II, we present the coupled dynamic fluid-solid system, and present an explicit monolithic algorithm for the treatment of the fully-coupled system. The interface conditions associated with moving interfaces is considered, and a comparison of moving vs. static reference frames is used to evaluate the dynamic interface treatment. Finally, a series of two and and three-dimensional projectile and shock-body calculations are presented.

Sandia journal manuscript; Not yet accepted for publication

Tipton, David G.

Here, this paper considers the issues central to fast level-set construction for the general treatement of moving interfaces in coupled fluid-solid interaction problems where the Lagrangian solid experiences large bulk motion. The central idea is based on a super-sampled L² projection, that in conjunction with a Lagrangian interface position, permits rapid identification of the solid interface in the fluid mesh and enables the imposition of boundary conditions for the fluid. A series of convergence studies are presented in terms of numerical quadrature and mesh refinement to illustrate the effectiveness of the super-sampled projection on unstructured grids. The extraction of the interface location based on distance functions is compared to the super-sampled projection method. Finally, it is shown that the extraction of an interface location based on a zero level-set converges as O(h) when compared to the exact interface location - suggesting that the availability of a Lagrangian interface description is always preferred.

Mayes, R.L.; Miller, A.K.; Holzmann, Wilfried A.; Tipton, David G.

Abstract not provided.