New developments in advanced welding
Weld quality assurance
There are four types of observations of weld quality: visual examination, destructive analysis, non-destructive examination and in-situ observations. These are discussed separately below. Which types of weld inspection procedures are used may depend on the weld procedures, the weld qualification procedures, and/or codes for welding. For example, all welds on pressure vessels, or on valves, gauges and fittings that are used on pressure vessels, are subject to Section IX of the American Society of Mechanical Engineers (ASME) welding code, or local Boiler-and-Pressure Vessel codes.
A visual examination gives information about the surface regularity of the weld and the presence or absence of cracking, surface porosity or undercuts. Restrained welds in higher carbon steels have a tendency towards cracking due to the hardness of the weld metal and heat-affected zone, particularly at the weld close-out. The width and bead profile of welds can be compared with that observed in trial welds under similar conditions taken before the commencement of production operations. If the weld is fully penetrating, and there is access to the underbead, the presence or absence and the regularity of the underbead ensures that complete penetration did occur. Visual examination by the welder or weld machine operator is often the first step in ensuring the correct operation of the equipment and ensuring there is no variation in the preparation of materials, cleanliness or shielding gas.
Destructive analysis is the process of cutting up the weldment to observe the depth and shape of the weld metal and the heat-affected zone. Often the material pieces containing the weld are mounted in a standard size mount of epoxy or plastic, metallographically polished and chemically etched to show up microstructural features. The section of the weld that is displayed can be a cross-section of the weld or can be longitudinal along the length of the weld. This process provides the best possible indication of the quality of the weld in the location of the cross-section; however, it does destroy the welded structure.
In some production operations, the destructive analysis is carried out on a sampling of parts (e. g. 1 in 50, or 1 in 1000), is carried out on parts that are rejected for some other reason unrelated to the welding operation, or is carried out on less expensive simulated parts that would have a similar heat flow pattern to the real weldment. The destructive analysis is time consuming and expensive, in that work pieces are destroyed, and is therefore used judiciously. Nevertheless, there is no substitute for the destructive analysis during the establishment and qualification of welding procedures, and the calibration of non-destructive analysis and visual analysis equipment.
More sophisticated destructive analysis can include hardness testing, tensile testing, bend testing, dynamic-tear, impact and fatigue testing of the weld metal and heat-affected zone. Tailor blank welding for the automotive industry utilizes a cupping test, where the material including the weld must be significantly deformed with a semi-circular punch, without breaking the weld metal. Structural welds for submarines are subjected to an explosion bulge test. Here plates containing a weld are deformed by an explosion to produce a significant bulge which the weld must survive.
4.2.16 Non-destructive examination
Non-destructive examination (NDE) procedures include the use of X-rays, ultrasonic examination and acoustic emission monitoring. X-rays can detect weld porosity, lack of side-wall fusion, missed seam defects and inclusions. In some cases, for example in the nuclear industry, 100% inspection of parts is required by welding codes. Ultrasonic inspection can detect the boundaries of the weld material and porosity that is larger than the ultrasonic wavelength, and a missed seam if the seam is perpendicular to the direction of propagation of the ultrasound. Often a hermiticity check, using, for example, a helium leak detector, is used in production environments.
A skilled manual welder, using a hand-held welding torch, carefully observes the size and position of the weld pool, the light emitted from the weld pool and the sound emitted from the arc. On the basis of these observations, he or she judiciously adjusts the speed and position of the torch, and perhaps the power of the arc. It is the intent of in-situ observation equipment to duplicate and extend the operations carried out automatically by the skilled worker. In the last few years, there have been extensive investigations of various visual and audio signals from a laser weld pool with the aim of using these signals for detection of weld quality and ultimately for the control of the welding process.
Electronic signals have been derived from the incident laser light that is reflected from the weld pool, the ultraviolet emission from the weld pool, the infrared emission, the audio emission carried through the atmosphere and ultrasonic emission carried away from the weld pool by the weldment itself. In some cases, the light emitted from the weld pool has been spectroscopically analyzed to indicate shield gas contamination and the sound emitted from the weld pool has been frequency analyzed. There were indications that the dominant frequency in the sound emitted may be related to the depth of the keyhole, in much the same way that the dimensions of an organ pipe control the notes that are emitted. These observations have not yet resulted in wide acceptance of such technology as a control measure. The types of observation best suited for control may depend on the materials being welded and the geometry of the weldment.
