When a new-model clothes dryer began singeing a few delicate materials, a company engineer demanded that his department start modeling combustion.
"That will pinpoint the problem," he proclaimed. And although he did not have a good picture of how hot and cool air streams mixed in the appliance ducts, he insisted that modeling combustion was key to solving the singeing problem.
"Fluid flows are tricky," says Ed Williams, president of Blue Ridge Numerics Inc., a developer of flow-simulation software. "It's because they are not easily visualized outside the realm of CFD software. And without simulations, many engineers suspect the source of their problem is in one place when it's actually in another."
The clothes dryer, for example, had to mix a small stream of hot air from a gas burner together with a much larger flow of room-temperature air, and do it over a short distance. This "mixed" air then went into the clothes dryer. Fluid-flow simulation, however, showed little mixing before the air hit the clothes, so there were hot spots. If delicate cloth moved over a hot spot, it would indeed singe.
Once that was understood, the engineers quickly came up with a simple solution: place a screen in the airflow just after cooler air is introduced to the hot stream to create a pressure drop and a little turbulence that mixed the streams. Problem solved. No combustion modeling needed. Williams says similar scenarios play out frequently. For instance, a chemical company decided it had to model chemical reactions to get a better picture of what was happening in its vats. But after a preliminary investigation, the company determined that modeling chemical reactions is difficult, time consuming, and error prone. "With some advice from our team, they determined the issue was more of a mixing problem, which was solved quickly," says Williams.
Cavitation, another phenomena that frequently causes problems, comes from gas dissolved in liquids that forms bubbles in sudden pressure drops. Their collapsing can be destructive over long periods. "Lots of engineers insist that modeling cavitation is the only way to understand their problems," says Williams. "But the issue can be clarified with one question: Do you want to model it or reduce its effects? Usually the goal is to reduce or eliminate the effect by modifying the design. In most cases, CFD software, set to calculate pressures easily show users areas in the model that are likely to produce cavitation. They can then make model changes to reduce or eliminate problem areas. Without fluid-flow simulation that lets users 'see' into models, redesigns are often time consuming and ineffective," he says.
And lastly, a few engineers think CFD eliminates testing. It doesn't. But it can refine or guide testing, which is often limited by time, costs, and prototype sizes. "Without understanding or insight into what is happening and where, it is difficult to gather enough experimental information," says Williams. "Flow simulations can augment and even drive testing strategies, and reduce the number of prototype tests. CFD simulations also provide valuable insight into design behavior which translates into more design innovation and creativity," he adds.
Blue Ridge Numerics Inc.