Reliability of Techniques for Nondestructive Testing Over Coatings for MODU Inspection

Abstract

ABSTRACT A study was made of the reliability and sensitivity of nondestructive (NDT) techniques for locating flaws under coatings commonly found on offshore structures. Coating thicknesses ranged from 4-16 mils (0.1-0.4 mm). Ultrasonics were found to be more sensitive than magnetic-particle and eddy-current methods. INTRODUCTION The objective of this program was to determine the set of conditions required for the nondestructive detection of flaws in structures coated with paint or other coating systems. A successful inspection technique has the potential for substantial cost savings, particularly offshore. A number of instances have arisen in which coatings could not be removed, yet inspection of the underlying structure, and particularly welds, was required. The inspection methods evaluated in this study were ultrasonics (UT), magnetic-particle (MT), and eddy-current (ET). These techniques are commonly used during the structural inspection of platforms and MODU's. Two types of coatings were selected for evaluation based upon their wide use on offshore structures. These were the epoxy-mastic type paint system and the urethane system (four-part coating system). The coating thicknesses varied from 4 to 16 thousands of an inch (0.10-0.40mm). To simulate flaws sometimes found in welded connections, welds were made and purposely designed to produce shallow/tight cracks, deep/tight cracks, and shallow/wide cracks. In addition to these flaws, varieties of samples containing known defects were used in this investigation. All test samples were inspected with the various NDT techniques before and after application of coatings. TEST PROCEDURES AND SAMPLES Three types of test samples containing flaws were used in this investigation. The first type was in the form of steel plates 40 inches long and 12 inches wide (101.6 by 30.5 cm) containing defective welds. The second type comprised calibration samples containing known defects (depth and width of machined slits). The third type contained fabrication-type defects (toe cracks, laminations, and undercut) from real structures. Descriptions of test samples follow. 1. Artificially Produced Flaws in Plates Shallow/very tight cracks were made by making short tack welds on high-strength steels with a cellulose-coated electrode (E70l0-Al) [1]. The accompanying undercut of these tack welds simulated shallow/wide cracks. The deep/tight cracks were made by improperly welding the root-pass of a double butt weld joint. The improperly deposited downhill pass resulted in a center bead crack for the full length of the deposit. 2. Calibration Samples Magnetic-field indicator with slit widths of 0.040 inch (1.0 mm). Burmah-Castrol magnetic-field indicators, Type I and Type III. Steel block with three EDM notches. The notches are one-inch long and 0.010 inch in width (25.4 by 0.25 mm). Depths of the three notches are 0.020 inch (0.5 mm), 0.040 inch (1 mm), and 0.080 inch (2 mm). This block is typically used in eddy-current testing to calibrate the equipment.