ARCHAEOLOGIST FINDS THAT 21st CENTURY FEA SOFTWARE CAN HELP TO SOLVE PREHISTORIC ENGINEERING MYSTERY

A Unique Paleoindian Arrowhead Known as a "Folsom Point"

Petroleum engineer and archaeologist Tony Baker used ALGOR'S FEA software to model the formation of glass flakes similar to those chipped from this Folsom point, a unique arrowhead that was used to hunt bison in North America's western plains. He believes that additional research with ALGOR's FEA software can help to explain the arrowhead's construction, which will further educate anthropologists and archaeologists about prehistoric engineering capabilities. (Folsom point replica courtesy of Bob Patten, Paleoindian artifact replicator.)

January 25, 1999, Pittsburgh, Pennsylvania -- About 10,000 years ago, a group of prehistoric North Americans developed a distinct process for creating artistic yet highly functional arrowheads for hunting bison. Researchers have been unable to fully understand the engineering technique used to create these particular Paleoindian arrowheads. Although they agree that the process involved chipping flakes from glass-like materials, they have been unable to determine the exact method required to consistently produce this type of arrowhead.

Tony Baker, a full-time petroleum engineer and part-time archaeologist in Denver, Colorado, was fascinated by this prehistoric mystery. He decided to use finite element analysis (FEA) software from ALGOR, Inc. to model the formation of flakes from plate glass, a material similar to those used for the arrowheads. Based on the analysis results, Baker believes that additional research with ALGOR's FEA software could help to explain the arrowhead's construction, which will further educate anthropologists and archaeologists about prehistoric engineering capabilities, and demonstrate how modern mechanical engineering analysis methods can contribute to archeological research.

Flaking Technique Critical to Successful Arrowhead Manufacture

For nearly one million years, prehistoric humans created tools by striking materials, such as obsidian, basalt or chalcedony, with rocks, bones and other objects. They chipped flakes from these natural glass materials that fracture in a concentric pattern with a sharp edge and smooth surface. The Folsom people, a group of Paleoindians who were first excavated near Folsom, New Mexico, populated the great western plains from Texas to Canada. They had a precise flaking technique for engineering their arrowheads, known as Folsom points. Opinions vary about their methodology because modern replicators still cannot copy them consistently, despite their success in copying other Paleoindian arrowheads.

A life-long student of Paleoindians, Baker was intrigued when he learned about one archaeologist's series of physical experiments that were conducted to better understand flake formation. For his doctoral dissertation at the University of Pennsylvania, Dr. Andrew Pelcin had created hundreds of glass flakes by dropping a steel ball onto cross sectional pieces of one-half-inch-thick plate glass. His experiments revealed that the size and shape of the resulting flakes corresponded to the location of the ball's impact on the glass' edge, the angle at which the glass was hit ("angle of blow") and the core's platform angle. Pelcin identified two possible flake types: a long, thin "cone" flake and a short, wide "bending" flake.

Baker wanted to mathematically explain the formation of Pelcin's two flake types by replicating his experiments on the computer using ALGOR's linear FEA software package. If the computer-simulated flakes matched Pelcin's, he knew that ALGOR's software could be used to predict other flake shapes that would result from slightly different conditions. He would then uncover the angles of blow and platform angles that the Folsom people most likely used to create their arrowheads.

Replicating Flake Formation with FEA Software

Baker chose to simulate the experiment in which the ball hit the edge of a glass core with a 55-degree platform angle at a 70-degree angle of blow. He first created a 2-D model of the plate glass using ALGOR's finite element model-building tool, Superdraw III. An arbitrary one-pound force was used to model the ball. He applied boundary conditions to the lower left portion of the model to represent a clamping device that attached the glass core to a table and he removed elements within the device's grip to reduce analysis time.

Baker then analyzed the model with ALGOR's linear static stress analysis processor. After viewing the analysis results, he created a crack one element long at the model's point of maximum tensile stress. He repeated the process of analyzing the model and adding a crack at the adjacent point of highest tensile stress until the flake broke away. About 60 - 70 iterations were required to replicate one flake removal.

Paleoindians Chipped Glass Flakes Like the One in This ALGOR Model to Form Arrowheads Tony Baker used ALGOR's linear static stress analysis software to simulate a physical glass flake formation experiment. The location of impact in this analysis was set to a length known to produce a long, thin "cone" flake. In this 2-D ALGOR model, Baker is about to draw the final crack, which will cause this cone flake to break away from the core.

After he had created many flakes, Baker realized that two additional elastic boundary conditions, or "springs," were necessary at the point of impact when producing the long, thin cone flakes. They represented the steel ball's elasticity in real world conditions, which would cause a reduction in force over time. Baker removed the springs when the deflection, or stored potential energy, no longer caused vertical cracking. Their removal increased the horizontal force and caused the flake to break away. The elastic springs were not required to form the wider bending flakes because the location of impact is closer to the core's dense center, which requires the addition of potential energy to propagate a crack.

Software Successfully Replicates Experiment

ALGOR's software analysis results correlated closely with Pelcin's results. The cone and bending flakes' size and mass matched those created in the physical experiment.

"ALGOR's FEA software provided accurate stress and strain values in the glass core that can only be guessed in physical experiments," said Baker. "This enabled me to uncover the mechanics behind Dr. Pelcin's results."

Baker now plans to model a 3-D plate glass core in ALGOR's Superdraw III to represent a true Folsom point core from which Folsom people removed flakes. He will apply varied angles of blow and platform angles to identify more flake sizes and shapes. Baker is considering using EAGLE, ALGOR's parametric design and analysis programming language, to automate the crack propagation.

Predicting flake formation on the computer using ALGOR's FEA software will provide researchers with more detailed information about Folsom point manufacturing and prove that FEA software is a practical, valuable tool for archaeological research.