July/August, 2017

Remember the DEW Line?

The Evolution of Ground-Level Vibration Criteria from the Cold War Years to the Era of Nanoelectronics

By Richard D. Woods, PhD, PE, D.GE, NAE, Dist.M.ASCE

Over a span of about 70 years, geotechnical engineers have dealt with increasingly more demanding ground-level vibration limits for foundations ranging from a few ips to less than 1 μips.

Most of my graduate students today do not know about the DEW (Distant Early Warning) Line that was the cornerstone of the United States’ air defense arsenal during the Cold War (1947-1991). The DEW Line was the northernmost of three lines of radar stations across North America designed to provide early detection and characterization of Soviet bombers and land or sea invasions. An advanced design was used for the radar antennae that came with stringent ground motion (i.e., vibration) tolerances.

In the 1950s, when the DEW Line was created, the subject of modern foundation dynamics was in its infancy. At that time, recognized ground motion (i.e., allowable vibrations) criteria were adopted from research that had been conducted in the 1930s and 1940s. Tolerable human vibration response criteria ranged from 0.01 to 1.0 in./sec (ips) peak particle velocity (PPV), depending on whether the person was standing or lying down, or if it was day or night. For buildings, machine foundations, and sensitive instrument foundations, 2 ips was generally accepted as an upper limit for ground vibrations. Ground vibrations can be expressed interchangeably as displacement (in.), velocity (ips), or acceleration (in./sec2). The need for more appropriate ground motion or vibration levels criteria and dynamic foundation design methods was revealed by the DEW Line project, which fostered an intense period of research in soil dynamics spearheaded by Frank Richart, Jr., and Robert Whitman.

Later developments in electronics and microscale measurements associated with medical and scientific research during the transistor revolution of the 1960s required a new level of refinement in ground motion criteria. For the crystal-growing and chip manufacturing processes that supported the semiconductor industry, tolerable ground-vibration levels in the range of 1 to 10 Hz were specified by IBM, for example, to limit displacements to 3 millionths of an inch. However, sites that met this requirement were generally in remote areas away from industry, railroads, and highways. Therefore, better methods of vibration isolation were sought so that chip manufacturing could be conducted in industrial regions. The following decade witnessed significant progress in the development of vibration isolation and damping methods.

In the 1970s, the world of high-speed communications was tooling up. Fast, high-volume conduits were needed to handle the increasing volume of electronic data traffic. Optical fibers provided a solution. One popular process for producing high-quality optical fiber required that fine threads of special glass, about the size of human hair, be pulled from a gob of hot glass and forced through a die to render a continuous fiber with a uniform diameter of 50 microns and a length of a kilometer or more. Deviating even by one micron in the diameter caused internal reflections of light in the fiber and degraded the quality of the fiber. To achieve such uniform diameter, the die required dynamic stability. Ground motions required to achieve sufficiently stable glass draws were determined to be on the order of 0.02 ips in the frequency range of 5-12 Hz.

By the 1980s, the importance of frequency in ground motion criteria was recognized by the Office of Surface Mining (OSM). For example, new OSM guidelines for the boundary between cracking and non-cracking of plaster walls in houses started at 0.2 ips at 1 Hz, rose linearly to 2 ips at 30 Hz, and contained a plateau of 0.7 ips in the frequency range of 3.5-11 Hz.

During that same decade, clean room facilities for electronic chip manufacturing required very stable platforms with PPV < 0.002 ips at all frequencies. In some situations, sites that met this requirement were not available, so labs were built unknowingly, or even intentionally, at sites that exceeded the criteria. In these cases, clean room facilities were made functional by using supplementary vibration isolation measures like pneumatic isolation systems or electronic vibration canceling devices, which were developments spawned by the limitations identified during chip manufacturing in the 1960s.

The shift to more and more limiting ground motions from the 1950s to the 1980s is clear and continues into the new millennium. Two of several organizations that have developed criteria for different levels of sensitivity are the International Standards Organization (ISO) and the National Institute for Standards and Technology (NIST), based on work at Bolt, Baranek, and Newman, Inc. ISO and NIST criteria, published in the 1990s, generally cover the frequency range of 1 Hz to 100 Hz with a constant allowable vibration limit. In common terms, these criteria range from 2 ips to 30 μips PPV. Included in sites covered by ISO and NIST vibration criteria are medical operating rooms, electron microscopes, and laser-based E-beam lithography systems operating in the nanometer range.

To accommodate Micro-Electro-Mechanical-Systems (MEMS) research and manufacturing, NIST instituted another more stringent criterion called “NIST A.” NIST A requires ground motions to be less than 1 μips at 1 Hz, increasing linearly to 125 μips at 20 Hz, and then remaining constant to 100 Hz. Sometimes this criterion is impossible to meet, so it becomes necessary to resort to expensive electronic vibration cancelling devices to operate successfully.

Over a span of about 70 years, geotechnical engineers have dealt with increasingly more demanding ground-level vibration limits for foundations ranging from a few ips to less than 1 μips. Improved knowledge of soil properties from in-situ measurement of shear modulus, soil improvement methods like compaction and chemical grouting, soil densification with deep dynamic compaction and vibroflotation, and advancements in methods of dynamic analysis have allowed geotechnical engineers to address ever-increasing industry demands.

The new millennium is already well into MEMS research and manufacturing while we search for means of providing new, very low vibration-level platforms. NIST admits that its NIST A criterion may be impossible to achieve at many sites, so that other non-geotechnical solutions may be required. Precisely what controls are needed remains to be seen; but, history suggests that geotechnical engineers will play a vital role in providing sufficiently stable platforms for future research and manufacturing facilities.

RICHARD D. WOODS, PhD, PE, D.GE, NAE, DIST.M.ASCE, is emeritus professor of geotechnical engineering in the Department of Civil and Environmental Engineering at the University of Michigan in Ann Arbor, MI. Much of his research work has focused on foundation dynamics, vibration measurement, blasting and other vibration damage, site characterization for seismic mitigation, and applications of geophysics in geotechnical engineering. He is co-author with F.E. Richart Jr., and J.R. Hall, Jr., of the textbook Vibrations of Soils and Foundations. Woods can be reached at rdw@umich.edu.