Visual Techniques in NDT

Introduction

Visual techniques are widely used to ensure product reliability during manufacturing and to examine any gross discrepancies on the surface of operating components. These techniques involve illumination of object surface with light and examination of the reflected light using visual aids, usually at magnification (McIntire P and Moore P O, 1996). Visual examination can reveal gross surface defects, cleanliness, foreign objects, surface condition, mismatches and any other discrepancies (Baldev Raj et al 1997). Visual techniques are easy to apply and are considered to be the most effective and the least expensive NDT techniques.

Visual techniques are as old as the telescopic devices used for human organs without operative procedures. Phillip Bozzini was the first to develop cytoscopes way back in 1806 for the purpose of medical research. They were modified for examination of holes such as gun bores and hence, called borescopes. Since then, a variety of flexible and rigid borescopes and more powerful and efficient optical aids have been developed for quick examination of products during manufacturing and other real life situations. Pioneering work in this direction was carried out by John Lang, who developed closed circuit television based borescopes for inspecting inner surface of helicopter blades, jet engines, wings, turbine blades, etc (McIntire P and Moore P O 1996). Recent advances in microelectronics, computer technology and artificial intelligence have popularised the concepts such as machine vision for realising automated visual examination techniques and unmanned inspection stations. This article discusses the details of visual techniques for examination of surfaces. Typical instruments, testing methods, industrial applications and latest developments are also covered.

1. Instruments for Visual Testing

The human eye is an excellent sensor and with that, it is possible to easily perceive many material characteristics such as shapes, colours, gloss, shades, speeds, perspective etc. and discontinuities in them. The human eye is an important component for performing visual NDT. Visual examination carried out by an experienced inspector can reveal the general condition of the component. Usually, visual techniques are used for examining cleanliness, misalignments and other mismatches, foreign objects etc.

Optical aids are usually recommended for visual examination, essentially for magnification purpose and also for inspecting the inaccessible areas. For the examination of inside surfaces of tubes, bores and chambers, boroscopes, endoscopes, telescopes are used (Baldev Raj et al 1997). The length and diameter of the borescope can be varied depending on the dimensions of the object. Extension sections are available in 1, 2, 3 m lengths, permitting assembly of borescopes upto 10 m. Various designs of borescopes are used for different conditions. These include angulated, calibrated, panoramic, wide field, ultraviolet, waterproof, and gas cooled designs.

In recent times, with the availability of flexible fibre-optic borescopes, charge coupled device (CCD) cameras, and computer based image processing software, it is possible to examine corners, bent surfaces, and inaccessible surfaces. Using these instruments, it is possible to take sharp and clear images of parts and interior surfaces and make quantitative evaluations. Most of the flexiscopes possess a wide-angle objective lens that provides a 100? filed of view, and adjustable focus. Usually, for industrial use, they are more ruggedly constructed by wrapping the fibre optic systems with flexible steel lining. The diameter and length of the flexiscopes are usually adapted depending on the requirements. Selection of a visual instrument mainly depends on factors such as the object geometry and the access, expected defect size and resolution requirements.

The five basic elements in a visual test are the test object, the inspector, the optical instrument, illumination and recording. Each of these elements interacts with the others and affects the test results. The objective distance, object size, discontinuity size, reflectivity, entry port size, object thickness and direction of view are all critical aspects of the test object that affect the visual test. Reflectivity is another factor affecting illumination. Dark surfaces such as those coated with carbon deposits require higher levels of illumination than light surfaces do.

In many situations, in order to aid vision, magnification with power ranging from 1.5X to 2000X is employed. Depending on the working distance and the field of view various lower, medium and high power magnification systems (microscopes) are used. With high power systems, it would be possible to achieve resolution of a few microns. The defect size usually determines the magnification and resolution required for visual testing. For example, greater resolution is required to detect hairline cracks in welds than to detect an undercut.

2. Surface Examination Using Visual Techniques

Visual techniques are probably the simplest, quick, and widely used NDT techniques for the examination of material surfaces (McIntire P and Moore P O 1996). They are also used to verify the presence or absence of cracks, corrosion and other forms of in-service material degradation. Typical industrial applications of visual techniques are given in Table 1. In many situations, quantitative evaluation as regards to type, location and orientation are also possible. The unique advantage of many visual techniques is their ability to yield quantitative information more readily than any other NDT tests.

Visual testing is performed in accordance with applicable codes, standards, specifications and procedures. For example, visual testing of a nuclear reactor vessel and its internal components is performed according to the rules of the plant’s in-service test program and special requirements of regulatory agencies, e.g. American nuclear regulatory commission. Majority of the tests meet the requirements of the ASME Boiler and Pressure Vessel code, which forms a part of the in-service inspection program. For example, Section XI recommends visual testing for the examination of condition of a part, component or surface, for identification of leaks and for the examination of mechanical and structural conditions. The code also gives detailed test procedures. Qualified personnel are required to carry out the visual tests.

In most situations, it is specified that the test surface should be free of slag, dirt, grease, weld spatter or other contaminants. Before visual testing, personnel are usually given basic near vision acuity and colour recognition screening tests. Near vision measurements are recorded for each eye and for both eyes. Similarly, the angle of view is very important during visual testing, especially when quantitative information is to be obtained. It is essential that the inspectors attempt to observe the object surface at the center axis of the eye. The angle of view should not be more than 45? from the normal. Similarly, the period of time during which a human inspector is permitted to work is usually limited to about 2 hours on continuous basis to avoid errors concerning visual reliability and discrimination. There are several integrated visual testing variables beyond equating near vision acuity to performance including lighting, knowledge of crack pattern recognition, orientation of test object, psychological factors and test instructions.

The data produced from almost all types of NDT tests are usually recorded and interpreted visually. For this reason, almost any NDT test could be considered a visual test, particularly at the detection or interpretation stages. With magnetic particle, liquid penetrant, radiography, and some leak tests, the link is easily evident. The same is the case with the in-situ metallography and other microscopy methods. Visibility criteria are specified for magnetic particle tests and liquid penetrant tests, especially when fluorescent systems that use black light are used for achieving enhanced detection sensitivities.

3. Latest Developments

The basic design of the borescopes, which has been in use for many decades, has been modified accommodating the state-of-the-art advances in video, illumination, robotic, optical and computer technologies. Developments in image processing, artificial intelligence, video technology and other related fields have significantly improved the capability of visual techniques (Forsyth et al. 1998). Figure 2 shows typical application of visual technique to examination of wheels for detection of very fine cracks originate due to residual stresses. Liquid penetrant tests could not detect secondary cracks. Visual technique using video-microscope with imaging processing capability has clearly revealed a secondary crack as depicted in Fig. 2.

Present day demand for higher performance and faster production exceed the abilities of visual tests by humans. Consequently, visual tests made by human eye are being replaced by automated visual testing using optical instruments and unstaffed inspection stations. Such aspects are usually referred to as machine vision. In essence, the machine vision acquires processes and analyses images to reach conclusion automatically. A typical machine vision system consists of a light source, a video camera, digitiser, a computer and an image display. Usually, the test object is illuminated and the image is captured using a video camera for processing by computer. The computer first enhances the contrast of the image with a procedure called image enhancement (Gonzalez R G and Wintz P, 1987) and later, the image is segmented for feature extraction and finally for classification using the power of artificial intelligence. Fourier analysis, multivariant analysis and statistics are increasingly being applied to evaluate invariant parameters from the image data for their use in automated object recognition and machine vision (Dougherty E R, Giardina C R 1988). Further, laser based scanning systems are being developed for on-line measurement and evaluation of volume and average size of wood chips or iron ore pellets and the detection of cracks in asphalt or wood planks. Other recent developments include D-sight, edge of light (Forsyth et al. 1998) techniques. While later is yet in developmental stages, the former method has already found its way to practical use.

Concluding Remarks

Visual techniques are the simple, quick, and widely used NDT techniques to examine the material surfaces for qualitative as well as quantitative assessment of gross discrepancies. Surface examination using visual techniques encompassing not only optical considerations and image processing but also the peripheral technologies such as electronics, computers, process control management (refer Fig.1). Driven by the demand for higher performance and faster industrial production, advancing trends in automated visual testing are expected to continue into the future.

Bibliography

McIntire P and Moore P O, 1996, Visual and optical testing, Vol 8, ASNT

Baldev Raj, Jayakumar T and M.Thavasimuthu, 1997, Practical NDT testing, Narosha, New Delhi

Forsyth DS, Komorowski J P, Gould R W and Marincak A, 1998, Automation of enhanced visual NDT techniques, Proc. 1st Pan-American Conf. for NDT, Toronto, Canada, pp 107-117

Gonzalez R G and Wintz P, 1987, Digital image processing, Addison-Wesley Publishing Co.

Dougherty E R, Giardina C R, 1988, Mathematical methods for artificial intelligence and autonomous systems, Prentice Hall Inc.