Procedures to simulate fatigue cracks in steel bridge specimens for use in a Probability of Detection study

Luke Raymond Snyder, Purdue University

Abstract

An inspector's ability to correctly locate and identify surface defects in steel bridge components is critical to protecting public safety. Inspectors must be properly trained and possess the required skills so that the likelihood such defects are detected with a high probability. However, at present the probability that a given inspector or group of inspectors will be able to detect a given type of surface defect on a given bridge type or detail is unknown. To address this, Probability of Detection (POD) studies are commonly conducted in other industries to establish an inspector's ability and detection methodology to find certain defects in realistic inspection environments. To conduct such studies, specimens with known defects (size, location, type, etc.) are required. The realism of the bridge components, details included in the study, environmental conditions, and methods of access when compared with what inspectors normally deal with on an in-service bridge is essential. This study is focused on the creation of specimens to be used for a study that examines the POD associated with visual detection of cracks at selected steel bridge details. Creating these bridge components in a laboratory setting allows for the cracking methods to be monitored, manipulated, and duplicated for the crack sizing needs, as well as the adjustment of the crack detail locations. Knowing the location and sizes of the cracks within these bridge components will in effect provide the "answer key" for the POD inspections, and help in the evaluation of visual inspection of steel bridges. As a result of the research, methods and procedures for creating a variety of fatigue crack types in a number of common steel bridge details were developed. These procedures were developed and refined for laboratory replication of actual fatigue cracks in common steel bridge details for use in a probability of detection study and are reported herein. Additionally, the techniques developed to accurately simulate weathering and general coating failure provided the desired realism when compared to in-service bridge components.

Degree

M.S.C.E.

Advisors

Connor, Purdue University.

Subject Area

Civil engineering

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS