| dc.description.abstract | The number of highway guardrail systems with posts of unvalidated depth is unknown to the Wisconsin Department of Transportation. These guardrail posts provide the support for barrier systems designed to protect motorists and infrastructure in the event of a crash. Non-Destructive Testing in the form of stress-wave propagation and magnetic characterization is explored as a method of length analysis for guardrail posts.
Longitudinal waves introduced to the post through an impulse were evaluated for natural frequency and wave velocity to estimate the post length. Field testing revealed length prediction error of 0.8 m for wood (length range: 1.2 m to 2.1 m) and 0.3 m for steel (length range: 1.8 m) when the posts are in-situ. Lab testing was conducted on a variety of wood posts under conditions of free, rubber-fixed, and fixed boundaries. Predicted length error varied from 0.2 m to 0.4 m (length range: 1.2 m to 3.1 m). To increase understanding of waveform characteristics through the wood medium, the attenuation, frequency dependent attenuation, and phase were analyzed. In the field, wood and steel posts were analyzed when attached and unattached to the guardrail. The information from this analysis further enhanced the understanding of energy loss, reflected waves, and the wave velocity, and the effect of frequency on the waveform with respect to the limitations imbedded in the stress-wave propagation technique for length prediction.
Characterization of the magnetic anomalies produced by the interaction between the magnetic field and ferrous materials was also evaluated as a methodology to determine the length of guardrail posts. The geometry, material characteristics, and distance of the material from the instrument affected the recorded magnetic anomaly. Magnetic models of expected anomalies compared to measured results provide a method to evaluate the expected length to the real length of the post.
Improving the methodical collection of data may help reduce the intrinsic error associated with magnetic surveying and further increase the relative impact of the base of the post. A methodology was devised to analyze wood posts with steel base plates to demonstrate the potential implementation.
Under the current state of the devised techniques, neither stress-wave propagation of posts nor the magnetic characterization of magnetic material can accurately predict length within the specified error tolerance of 5 to 10 cm. | en_US |
