Application Note: One Person’s Application can be Another’s Inspiration
It might be thought that a software tool that purports to do many things cannot be exceptional for any particular application. On the contrary, as observed in this article, having a multitude of capabilities can often lead to unexpected solutions for diverse and difficult problems.
How Open Channel Flows Translated to Bottle Filling
Experimental vs. simulation results from P & G's
air entrainment during bottle filling project.
Consider the addition to FLOW-3D a few years ago of a modeling capability for entraining air in open channel flows. This was a feature that numerous users working in hydraulics (e.g., water power generation) had requested. When released in a code update one of the first users was P&G who was simulating the filling of bottles with liquid detergent, which is not exactly open channel flow. For one thing, the bottle filling was not turbulent whereas the entrainment model assumed that fully developed turbulence was the principal mechanism for entraining air. A small addition to the model was able to overcome this limitation. Another early adopter of the new air entrainment model was a metal caster interested in the amount of air entrained when he was pouring a metal casting. This application also has little in common with open channel flow, yet it worked well because the entrainment model was founded on basic physical processes contributing to air entrainment.
A second example further highlights the general applicability that often exists in the physical models included in FLOW-3D. At the request of a paper manufacturer a special model development was undertaken to simulate the drying of paper. The model simulated the flow of hot air and water vapor through a porous material (i.e., paper) that contained some amount of absorbed water. The development was completed and that was the extent of it, until some time later BMW expressed interested in removing water from inorganic sand cores used in their metal casting business.
A comparison made by BMW between simulation and
experiment of the drying of an inorganic core.
From the physical point of view, the passing of hot air through porous sand to remove water contained in the sand binder is not conceptually different from removing water from wet paper. A test case to confirm this thought proved to work well. The accompanying figures show a comparison between simulation and experiment of the moisture remaining in a core after about half the initial moisture had been removed. This case was a little more complicated than it appears because some water that evaporated early in the drying process then condensed on colder portions of the core further from the hot air inlets. You can read more about BMW's use of the core drying model in a recent article published in Foundry Technology & Management.
If there is a moral in these stories it is that adding well-crafted modeling capabilities to a general purpose software often leads to undreamed of possibilities. By the way, users doing hydraulic simulations have also found air entrainment to be a useful addition to their modeling activities.