Sandeep Menon Address: Multiphase Flow Simulation Lab 212 Marston Hall Department of Mechanical and Industrial Engineering University of Massachusetts Amherst Amherst, MA 01003 Telephone: (413) 545 4267 (413) 313 8260 Email: smenon .at. ecs.umass.edu

| Home | Research | Departmental of Mechanical Engineering | University of Massachusetts Amherst |

I am a graduate student in Mechanical Engineering at the University of Massachusetts Amherst, working primarily in the area of Computational Fluid Dynamics. I work with Prof. David P. Schmidt and Prof. Jonathan P. Rothstein - investigating the use of CFD techniques to study non-Newtonian droplet formation and behaviour. Prior to this, I worked with Prof. Blair Perot on the use of Graphics Processors for CFD.

On this page you will find descriptions of my research interests, updates on recent work, plus a few links to interesting web sites.

If you're looking for stuff related to OpenFOAM, click here.

Multiphase Flow Simulation using Interface Tracking Methods

The Lagrangian approach to interface tracking in Multi-phase flows demonstrates several advantages over reconstruction techniques like Level-Set methods and Volume-of-Fluids (VOF). The absence of interface smearing and improved accuracy of surface tension calculations has led to the success of Lagrangian methods, particularly at low Weber numbers. However, owing to the strong deformations that are characteristic of this technique, the quality of the underlying mesh must be maintained to ensure accuracy throughout the duration of the simulation process. This work incorporates modifications to connectivity structures in situations where vertex motion is no longer sufficient to account for large deformations, using a combination of bisection, contraction and flipping operations to edges in the mesh. Moreover, this approach is particularly attractive as it is applicable to general cases involving dynamic meshes, like wing-flapping and in-cylinder engine simulations.

Graphics Processors for CFD

High performance streaming processors have achieved the distinction of being very efficient and cost-effective in terms of floating-point capacity, thereby making them an attractive option for scientific algorithms that involve large arithmetic effort. Graphics Processing Units (GPUs) are an example of this new intiative to bring vector-processing to desktop computers; and with the advent of 32-bit floating-point capabilities, these architectures provide a versatile platform for the efficient implementation of such algorithms. To exemplify this, the implementation of a Conjugate Gradient iterative solver for PDE solutions on unstructured two- and three-dimensional grids using such hardware has been implemented. This would greatly benefit applications such as fluid-flow solvers which seek efficient methods to solve large sparse systems. The implementation has also been succesfully incorporated into an existing object-oriented CFD code, thereby enabling the option of using these architectures as efficient math co-processors in the computational framework. If you'd like to read the thesis, you can get it here.

S. Menon, and D. P. Schmidt, Conservative interpolation on unstructured polyhedral meshes: An extension of the supermesh approach to cell-centered finite-volume variables, Computer Methods in Applied Mechanics and Engineering, Accepted for publication. [Download]

K. Mooney, S. Menon, and D. P. Schmidt, A Computational Study of Viscoelastic Droplet Collisions, 21st Annual ILASS-Americas Conference, Cincinnati, Ohio USA, May 2010. [Download]

S. Menon, J. Rothstein and D. P. Schmidt, A Numerical Study of Axi-symmetric Droplet Formation Using A Moving Mesh Approach, ICLASS 2009, 11th Triennial International Annual Conference on Liquid Atomization and Spray Systems, Vail, Colorado USA, July 2009. [Download]

S. Menon, J. P. Rothstein, D. P. Schmidt, Z. Tukovic, Simulating Non-Newtonian Droplet Formation With A Moving-Mesh Method, 21st Annual ILASS-Americas Conference, Orlando, Florida, May 18–21, 2008.

S. Menon and J. B. Perot, Implementation of an Efficient Conjugate Gradient Algorithm for Poisson Solutions on Graphics Processors, 15th Annual Conference of the CFD Society of Canada, Toronto, May 2007. [Download]

Most of my work these days has something to do with OpenFOAM, an open-source CFD library.

MultiThreading class: A general multi-threading class for use with OpenFOAM, using the POSIX threads library (pthreads). A feature of this class is that it uses thread-pooling to avoid instatiation costs. This class is already a part of OpenFOAM-extend. You can access it at this location on the source-tree: $FOAM_SRC/OSspecific/POSIX/multiThreader

MultiPrecision class: A multi-precision scalar class that wraps around the GNU Multi-Precision Floating-point Reliable (MPFR) library. This allows arbitrary precision arithmetic and can act as a replacement for the OpenFOAM scalar class. [Download]

Incompressible 6-DOF solver: A general 6-DOF solver derived from shipFoam by Mark Couwenberg. Although this is demonstrated using an icoDyMFoam variant, it should be general enough to extend to any flow solver that uses dynamic meshes. [Download]

This is where I get to show-off. All work is done on OpenFOAM.

IC Engine Applicability: This animation depicts the applicability of the tetrahedral reconnection algorithm for Internal Combustion Engine simulations. Smoothing was achieved using the Mesquite mesh optimization library. [Download (2.6 MB)]

Parallel Mesh Adaptation: Extension of the adaptation scheme to an distributed-memory (MPI) paradigm, using the same IC-engine domain shown above. The case was run on four processors, to obtain an average speedup of about 3. This video also demonstrates dynamic load-balancing in parallel using parMetis. [Download HiRes (20.7 MB)] [Download LowRes (5.6 MB)]

3D Simulations of Drop-on-demand Inkjets: This method of inkjet printing relies on pressure modulation to achieve droplet formation. Droplet volume depends on the duration of pressure pulses. Elongated tails eventually leads to satellite droplet formation which affects print-quality - an undesirable effect. [Download (13 MB)]

Offset Droplet Collision: The animation depicts the collision between two water droplets at an offset angle of approx. 40 degrees. Adaptive thread-parallel tetrahedral re-meshing was employed to achieve this result. [Download (1.6 MB)]

Multiple droplet formation: A case showing the formation of multiple droplets from a pressure modulated ink-jet. This is a two-dimensional axisymmetric simulation. Mesh motion is prescribed by a Laplacian motion solver. [Download (2.7 MB)]

2D Drop-on-demand Inkjets: Axi-symmetric simulations of the drop-on-demand inkjet case. [Download (1.4 MB)]

Continuous Inkjets: High speed printing applications frequently require a continuous inkjet approach. In such cases, the formation of instabilities along the length of the jet cause it to break up into drops (and smaller satellite droplets). Most continuous inkjet methods employ thermal modulation to create spatial gradients of surface-tension. The resultant shear-stress at the interface results in Marangoni flow along the jet, eventually leading to instabilities and droplet formation. [Download (3.0 MB)]

Norway: An Eclipse color theme that replicates the Norway Today scheme from Netbeans.

Reddit: My daily fix.

Comics:
Calvin!: My all-time favourite.
PhD: The quintessential grad-school portrait.