APPLICATION OF FINITE ELEMENT ANALYSIS TO THE UNDERSTANDING OF FLAKE FORMATION

Tony Baker and Andrew Pelcin

November 14, 1998
Modified September 30, 1999 -- see note #1 for reason

Over the last year, I (Tony) have been attempting to understand the mechanics of flake formation. My interest in this subject was first kindled at the Folsom Workshop and then later heightened by Andrew Pelcin,2 who did his Ph.D. research on a similar subject. In his research for his dissertation, Controlled Experiments in the Production of Flake Attributes, he experimentally created many glass flakes under different platform angles, angles of blow and other conditions. Of particular interest to me was the apparent existence of two different types of flakes occurring under what appeared to be similar conditions.

The methodology I used in this search for an understanding was to attempt to replicate Pelcin's work with computer software call Finite Element Analysis (FEA). This type of software is generally used by mechanical engineers and the particular package I selected was ALGOR.

My replication effort, to date, has been very successful and informative although it is far from finished. I believe I have discovered a possible explanation for the formation of two types of flakes Pelcin created with his work. Pelcin and I believe these two different flake types are the result of two different propagation modes. Contterell and Kamminga (1987:692) referred to these modes as "stiffness-controlled" and "compression-controlled". For reason that will become clear in this document they will be referred to as energy-rich and energy-poor, respectively. My explanation for these two types is new and exciting and I believe other knappers will also be interested. Since I never could keep a secret for long, I want to share these findings at this time in this status report.


I have divided this document into a number of sections in an attempt to make the presentation more understandable. Some sections are nontechnical while others are highly technical. I have identified each in the following descriptions. Just click on the link to read the section.

Return to Paleoindian & Other... Home Page

To contact us: Tony Baker (tabaker@ele.net) - Andrew Pelcin pelcin@erols.com


Notes

#1 Since this status report was first written in November 1998, I have continued almost non-stop with the research. I now know that the two flake types are not the cone and bending flakes that was first proposed in November 1998. They are the result of two different propagation modes that will be termed energy-rich and energy-poor in this document. So the modification of this document is only correcting the names of the flakes types and some information about the propagation speed of each.

#2 I have listed Andrew Pelcin (Andy) as a co-author for this document because he has been part of this FEA research from the first day. He created the data, real flakes, that I have replicated. He shared his data with me and it was in digital form which was like a gift from heaven. He freely shared in personal knowledge and I have talked to Andy almost weekly over the last year, asking questions similar to: "What did the flakes look like? What was the sound at the time of impact? Where did you get the glass?" I have bounced every harebrain theory about flake formation off him and there were many over the last year. Finally, he did the lion's share of the editing on this document.