Chasing the Dolphins (Tech Trends Feature)

Building a biomimetic underwater vehicle using CAD/CAM.

Ted Ciamillo is a few months shy of his 40th birthday, a justifiable time for a midlife crisis. Like many others confronting this milestone, he's going on a prerequisite soul-searching adventure, but he's not doing any of the clichéd activities, such as climbing Mt. Everest or hitchhiking across Asia. Instead, he's going to design and build an underwater vehicle and pedal it across the Atlantic.

In this article

Capturing the Fluke

Ciamillo dropped out of engineering school in his third year, but that didn't stop him from becoming an engineer in his own right. In his 6,000-square-foot cypress-framed workshop in Nicholson, Georgia, he and his staff develop and market a series of bicycle components using various CNC (computer-numeric controlled) machines. The 0G (Zero Gravity) TI, Ciamillo's invention and the company's signature product, is described as "a precision-crafted, superlight–road braking system" that is produced from "superior materials, optimized geometry, and a patented cam lever system."

As a machinist, Ciamillo has developed an interest in biomimetics, which is the study of mechanical devices that mimic the biological construction of animals. To design the propulsor of the Subhuman, his 15-foot-long human-powered submarine, he's looking to the shapes and movements of aquatic creatures for inspiration. To begin with, he has derived the geometry from a CAT scan of a dolphin's flukes.

"I tabulated the flukes' [x,y,z coordinates] and started using the Extruded Lofting feature in Solid Edge to string together a series of cross-sections to replicate the foil geometry of the dolphin," Ciamillo said (figure 1).

Figure 1. Based on a CAT scan of a sapinner dolphin s fluke, Ted Ciamillo constructed the Subhuman underwater vehicle in Solid Edge.

Siemens PLM Software (formerly UGS), the makers of Solid Edge, listed STL import as one of the new features available in v20, which began shipping in August 2007. If Ciamillo were starting his design today, he could have directly imported the CAT scan data as a binary STL file into Solid Edge. Regardless, he said he still would prefer to create the solid model from the ground up using the numerical data of the scan.

"When the [STL] file came in, I would still have to take measurements from it then reconstruct the shape, so [the import] would be just another additional step," he explained. In addition, "[working directly with the numerical data] allowed me to filter the errors in the data and average the values."

Trailing the Moon

The dolphin also guided Ted's earlier invention, mono-fin scuba equipment called the Lunocet. Ciamillo came up with the name by joining the Latin moon-oriented prefix luno with the whale-related suffix cet (from cetacea).

"Most foils in marine industry — for instance, sailboat rudders — have symmetrical shape[s] in cross-section, but no camber [concavity on either side]," Ciamillo pointed out. "The dolphin's flapping foil is symmetrical at rest, but [the animal] uses its fluke to induce camber during the swim, which adds efficiency."

Sharing his invention on the community forum at Deeper Blue in January 2005 (http://forums.deeperblue.net), Ciamillo wrote, "[The Lunocet] weighs in at 2.75 lbs (1,247.4 grams). I was able to produce 150 lbs of static thrust. Top speeds have not been accurately measured yet. The production model will have a thin translucent urethane skin and have a better chord-wise flexibility gradient."

Dr. Frank Fish (yes, that's his real name), the research scientist who first suggested the morphology of the dolphin for Ciamillo's projects, remarked, "The reason we're interested in the dolphins is that they have a highly efficient propulsion system. What they do is oscillate their tails, which are like wings. As they move up and down, they create a forward-directed lift force, which generates the thrust to propel the animal forward in a way that really doesn't require a lot of additional or wasted energy at all."

The performance of this novel propulsion system depends on a number of variables, including the ratio of body surface area to fin surface area, fluke flexibility, camber, peduncle hinge and pivot, and angle of attack. Fish and Ciamillo will need to conduct a series of real-world tests to figure out the best configuration for the Subhuman's propulsor and for the Lunocet.

Center of Gravity

When Ciamillo felt the Lunocet model in Solid Edge was ready for production, he exported the parts as IGES files into Mastercam software for generating tool path. Soon, the first physical prototype was born (figure 2).

Figure 2. The Lunocet, another of Ciamillo s inventions, mechanically mimics a dolphin s flapping tail.

"It was made of carbon fiber core with molded skin," Ciamillo said. "The skin was created in CAM (computer-aided manufacturing) using the IGES file. The core was essentially a piece of foam sandwiched between two pieces of carbon fiber. It was quite buoyant."

Ciamillo found out just how efficient the Lunocet was when he dropped the model into the 10,000-square-foot test lagoon behind his shop. "When we pushed it flat to the bottom and let it go, it left the bottom immediately, with the leading edge pitching up, the buoyant force provided the transverse motion similar to a flapping fluke. The Lunocet, generating lift, shot horizontally across the tank."

When he conducted a similar test on the 1,000-lb Subhuman, he discovered that the buoyancy was off by merely 3 lbs. He attributed this accuracy to Solid Edge's ability to calculate an assembly's weight and center of gravity. "If I didn't have Solid Edge, there would have been a lot of guesswork," he admitted. "I would have to put [the parts] in the water, see how negative or positive they were, then adjust their buoyancy."

The Dangerous Rift

In biomimetics, Ciamillo has found a way to bridge marine mammals' physiognomy and kinematics to mechanical engineering, but he's still struggling to bridge his real-world design expertise to CAD.

"I think CAD, CAM, FEA [finite-element analysis], and all these tools were originally developed to help people like me — people working by hand, building things from real-world experience. But looking at the tools these days, they're so advanced that I'd have to become a specialist to use them," he observed. "What may end up happening is, there may be a rift between people coming from my world looking for practical tools and others who have spent the majority of their career mastering CAD tools."

Ciamillo owns a copy of ALGOR, an FEA package, but "I haven't even opened the package," he confessed. "My workflow is skewed toward real-world, hands-on studies. I guess it comes from being a machinist."

In the recent years, relying on the hardware advances and increasingly sophisticated visualization algorithms, CAD vendors have begun advocating the use of digital prototypes over the costly physical mockups for testing and simulation (see "Autodesk Gets Digital," Cadalyst, December 2007, pp. 13–16). Ciamillo is characteristic of the user segment that vendors may have a tough time convincing to go along with the proposed switch.