Files in this item



application/pdfChristopher_Senalik.pdf (6MB)
(no description provided)PDF


Title:Detection and assessment of wood decay – glulam beams and wooden utility poles
Author(s):Senalik, Christopher
Director of Research:Reis, Henrique M.
Doctoral Committee Chair(s):Reis, Henrique M.
Doctoral Committee Member(s):Hall, Wilfred B.; Kuchma, Daniel A.; Sreenivas, Ramavarapu S.
Department / Program:Industrial&Enterprise Sys Eng
Discipline:Systems & Entrepreneurial Engr
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Non-Destructive Evaluation
Brown Rot Decay
Finite Difference
Utility Poles
Glulam Beams
Wave Propagation
Abstract:This report documents the development of acoustic and ultrasonic inspection methods on wooden glulam beams and utility poles. All beams and poles examined where composed of Douglas-fir. The report begins with a description of the structure of trees, the mechanical behavior of wood subjected to rot, and current utility pole inspection methods. Background regarding inspection methodology on wood structures is provided. Cross-sectional images of the glulam beams and the utility poles used in this study were obtained through the use of computerized axial tomography (CAT or CT) scans. Areas of decay were identified using the CT scan images. Two methods of detecting defects within glulam beams are described: ultrasonic through-transmission and impact-echo. The analysis of the wooden utility poles starts with the development of a two-dimensional, finite difference time domain (FDTD) simulation to model wave propagation through the pole. The simulation is validated against empirical results. The through-transmission technique used on the glulam beam locates rot through the use of the highest magnitude frequency and area under the power spectrum density curve. A 100 kHz signal was sent through the glulam beam. In areas devoid of decay, the received frequency was approximately 100 kHz. In areas where decay was identified, the frequency of highest magnitude shifted lower towards 85 kHz. Also, the area under the power spectrum density curve of the received signal was greatly diminished in areas of decay. The impact-echo method used on the glulam beams locates rot through the use of the attenuation rate. An accelerometer was affixed to the surface of the beam. A ball bearing was dropped from 200 mm above the surface next to the accelerometer. The signal was recorded. A spectrogram of the received signal was developed, and the mean rate of attenuation of the frequency range 500 Hz to 20 kHz was calculated. An attenuation rate of 1.17 Nepers per millisecond was found to be the threshold indicative of the presence of rot. Attenuation rates greater than the threshold indicated the presence of rot; lower indicated sound wood. The threshold had an overall error rate of 7.2%. The report then shifts to developing a two-dimensional, finite difference time domain simulation that can model wave behavior through a wood pole cross-section. The model incorporates several features that have not been included in previous analyses. These features include: a frequency dispersive model of wave velocity and attenuation, cross-sectional density and geometry information collected directly from CT scans of the utility poles, a perfectly matched layer used to model the behavior or rot, and a center point formulation that allows waves to pass through the center of a cylindrically orthotropic medium. The simulation is validated against the waveform behavior predicted by an analytical model and against experimental data collected from impact through-transmission testing of three actual utility pole specimens. Defects of various sizes and locations are then simulated in order to identify associated changes in wave behavior. The results of the simulation are used to develop metrics to determine the size, depth, and general location of internal defects within a wooden utility pole. The metrics are then applied to data collected from the wooden utility poles with known internal defects for validation. The predicted defect areas are accurate to within 2.0% of the total cross-sectional area and have a positional accuracy within 17% of the cross-sectional radius.
Issue Date:2013-05-24
Rights Information:Copyright 2013 Christopher Adam Senalik
Date Available in IDEALS:2013-05-24
Date Deposited:2013-05

This item appears in the following Collection(s)

Item Statistics