February 2006 Mechanical Engineering Departments: Tech Focus, Pt. 2

This section was edited by
Associate Editor Jeffrey Winters.

Instrumentation and Control

MEMS Sensor
May Track Single
Cells
by Alan S. Brown

Sometimes, the best engineering solution comes after turning a problem on its head. That's just what University of Missouri-Columbia professor of electrical engineering Hui Tang and doctoral student Yuanfang Gao did in developing a small sensor to count individual blood cells.

Particle counters have been around since 1948, when Wallace Coulter used changes in an electrical field to detect the number of particles suspended in an electrolyte as they passed between opposing electrodes. Since each particle created a voltage pulse proportional to its volume, he found that he could measure their size as well. Medical professionals quickly established the Coulter Counter as the most accurate way to count red and white blood cells, which are of different sizes.

Greater flexibility in electrode design allows a new sensor to identify a wider range of particles.

Reducing Coulter Counters to microelectromechanical systems promises several benefits. Their small scale improves their resolution of similarly sized cells and biomolecules. MEMS are small enough to use in handheld field or emergency room devices. Depositing hundreds or even thousands of MEMS on a single wafer could make them cheap enough to use and throw away, eliminating issues with maintenance and fading accuracy.

The problem, until now, has been simplifying counter designs. Most previous designs required fabrication of liquid reservoirs and microchannels on the chip. Some placed the electrodes at the inlets and exits of these channels, or along the channels themselves. These designs were costly to fabricate and required complex fluid management.

Tang and Gao's solution turned this approach on its head by flowing the suspended particles through holes running through the thickness of the chip. This does more than merely eliminate complex channels and electrode deposition schemes, said Tang. It simplifies the deposition of electrodes and other devices on top of the chip. This may lead to cheaper and more portable particle counters. Applications could range from medical field use to assessments of chemical processes, ecological health, and even particulates in machine oil.

Hui Tang's prototype MEMS sensor may one day enable tracking of changes in individual cells.

Tang sees the new MEMS as a technology platform that will change how we study cells. By simplifying fabrication, he believes he can create more sensitive and varied electrodes, or use electrical forces to control the movement of suspended cells.

Intriguingly, Tang thinks he may be able to study individual cells by trapping them inside MEMS holes. He envisions examining cell metabolism by adding microsensors to measure changes in oxygen and carbon dioxide concentrations. Another concept is to excite cells at different electrical frequencies to assess the workings of their inner components.

"We don't want to just duplicate the capability of a Coulter Counter," he said. "We want to go beyond that and maybe one day measure a cell's DNA contents by its electrical reactivity."

Spill the Wine, and Take That Data
by Michael Abrams

When earthquakes hit California, winery floors can end up looking like it's the first day of Prohibition. With 90 percent of the state's $45 billion wine industry within barrel-cracking distance of a fault line, it might be useful to know how best to stack barrels to keep them from tumbling.

In an effort to solve this problem, researchers at California Polytechnic State University, in San Luis Obispo, put stacks of wine barrels through a battery of earthquake simulations. They outfitted 59-gallon French Bordeaux oak barrels with lateral transducers to measure lateral displacement, and horizontal transducers to measure rocking.

"The French oak could have had a different friction coefficient than American oak, but we didn't check that," said engineering student Jeremy Stanley, part of the research team. On the upper barrels, on stacks of two or four, they attached a string to measure displacement. The entire setup sat atop a concrete slab on Cal Poly's "shake table."

Having learned from earlier tests that sloshing had minimal effect, the researchers considered the barrels a "homogenous mass" and filled them with water. "If we had used real wine, it wouldn't have lasted too long," said Stanley. They then applied a series of pulses. A single pulse would send the table forward and back once, in imitation of a real "fault normal" pulse, which tends to be far more destructive than "fault parallel" pulses. The pulses were also run at a variety of frequencies. With nine degrees of freedom allotted to each barrel, their computer had to solve nine equations simultaneously, 200 times a second for each pulse.

After each harvest of data the research team members fine-tuned their computer model, then tried another pulse, continuing till the data matched the model. Once that happened, they would subject the stack of barrels to a pulse made from actual earthquake ground data. These tended to be the most destructive, sometimes sending barrels to the floor.

Although the goal is eventually to have a model that could be taken to specific sites and tested with simulations of earthquakes likely to hit the area, the study has already offered some insights into what barrel configurations might be best.

"In certain ground motions you will notice that a four-barrel-high stack is more stable than a two-barrel-high stack," said Stanley, "but if you change the frequency, then that can alternate." This jibes with reports from the real world: When earthquakes hit Sonoma Valley in 2003, wineries with four-barrel-high stacks reported less damage than their otherwise configured competitors.

Strong Vibes
by Harry Hutchinson

The mechanical engineering series of titles issued under the CRC Press imprint of Taylor & Francis Group is almost up to two dozen titles. The latest from the publisher in Boca Raton, Fla., is called Vibration and Shock Handbook, and is edited by a professor of mechanical engineering at the University of British Columbia who is a Fellow of both ASME and IEEE.

It is not an introduction to the subject, but a compendium of expertise. Editor Clarence de Silva and 48 other contributors from around the globe provide 45 chapters and a glossary. The first four chapters, written by de Silva, are "Time-Domain Analysis," "Frequency-Domain Analysis," "Modal Analysis," and "Distributed-Parameter Systems." As are all the chapters of the book, each is broken down into discrete parts to deal with different aspects of the topic. A fifth chapter, "Random Vibration," rounds out the "Fundamentals and Analysis" section.

Six chapters are devoted to "Computer Techniques." Another four fall under the heading "Shock and Vibration." Other sections include "Instrumentation and Testing," "Vibration Suppression and Control," and "Design and Applications."

Although the publisher calls the volume a handbook, it is an armful. The pages are not numbered sequentially throughout the text, but run chapter by chapter instead.

We didn't count them all, but judge that there are more than 1,500 pages between the two covers. On our office postal scale it weighed in at 7.5 pounds (more than 33 newtons). At $179.95 for the volume, that comes to about $24 a pound for an extensive treatment of the subject.

It joins 22 other books in the series on topics ranging from distributed power generation to composite materials and energy audits.