New developments in advanced welding
Safety in laser beam welding
Experts in laser safety divide the potential hazards into two categories, beam hazards and non-beam hazards. The non-beam hazards include factors such as the glare of the welding process, which may contain significant ultraviolet radiation, and fumes from welding. For the most part, the non-beam hazards are similar to those encountered in other welding processes and are discussed in Chapter 10, Occupational health and safety. One additional non-beam hazard that may not be present in other welding processes is the high voltage used for carbon dioxide and flash lamp pumped Nd:YAG lasers. All industrial lasers should be packaged so that operators and engineers cannot possibly have access to the high voltage under normal operating conditions. It is essential that repair technicians be trained to understand high voltage.
Beam hazards is a term used to describe possible eye and skin damage due to contact with a laser beam. All industrial laser processing equipment should be inside a safety enclosure so that welding machine operators cannot under any circumstances come into contact with a beam, hence beam hazards are negligible. In the USA and certain other jurisdictions, this is called a ‘Class I’ enclosure. Electrical interlocks are located so that if doors to the enclosures are opened, for example to load or unload a part to be welded, laser action is inhibited. It is essential that operators and other staff be trained not to bypass the interlocks.
The wavelength of the output of carbon dioxide lasers is not transmitted through ordinary optical materials such as glass or plexiglass, hence safety enclosures for carbon dioxide lasers can be made of materials that allow the operator and spectators to view the welding process. The wavelengths of the output of Nd:YAG and diode lasers are however transmitted through normal window material, so operators and spectators could potentially come into contact with laser radiation that is reflected from the weldment. Consequently, Class I enclosures for these laser systems normally are made from sheet metal, with either a closed-circuit television camera to ensure the beam is aligned on the seam to be welded, and to allow observation of the weld, or special glass windows through which the beam is not transmitted, or both.
When it is necessary for service personnel to operate the laser without the enclosure, for example for aligning the mirrors in the beam path to ensure the laser beam is centered in the beam path, special precautions must be undertaken. These precautions include the use of wrap-around safety glasses that do not transmit the laser wavelength and safety curtains to ensure that nobody other than the service personnel can encounter the beam. Any laser facility with lasers that are not Class I should have a Laser Safety Officer to evaluate the safety of the laser installation, educate staff that are in contact with lasers, and ensure that adequate administrative and engineering controls are installed to limit the possibility of accidents.
4.6 Future trends
Fiber lasers and fiber laser amplifiers were originally developed for the telecommunications industry. As the power output capability increased, fiber lasers were packaged for industrial applications and are finding uses in cutting, welding and drilling. Just as neodymium ions can be doped into
YAG rods and slabs for applications in industrial Nd:YAG lasers, these and other ions can be doped into silica or other substrates that can be drawn into long thin fibers. The active element doped into the fiber can be ytterbium, in which case the wavelength of the output is very close to the wavelength of the Nd:YAG laser. Alternatively, the active element could be erbium, giving a wavelength of 1545 nm. The surface layer of the fiber, called cladding, is doped to give it a higher index of refraction so that it acts like a mirror; the laser radiation is reflected off the walls of the fiber but can be transmitted down the core of the fiber in a nearly loss-less fashion. The core of the fiber is typically 50 mm in diameter; the entire fiber including the cladding is 125 mm in diameter. Outside the cabinet in which the beam is generated, the fiber lies within an armored sheath which provides protection during handling.
The beam transmitted from the fiber is usually of very high quality and can be focused to a small spot size. Fiber lasers can be either continuous or repetitively pulsed. The ‘wall plug’ efficiency of fiber lasers (ratio of electrical power in to optical power out) is 20% or higher and is the highest of all industrial lasers. Fiber lasers are commercially available at average powers up to the kilowatt range; these high power lasers are made by coupling together the output of many smaller lasers. The fiber laser, because of its exceptionally fine focusing powers, promises to find applications in laser welding. At the time of writing, however, it is too early to evaluate the results.
Combined welding, also referred to as hybrid welding, laser-assisted arc welding, or arc-assisted laser welding, uses the energy input from a laser source as well as the energy input from a gas metal arc torch, a gas tungsten arc torch, or a plasma arc torch. The process was first investigated prior to 1980 but has found commercial applications in manufacture of automobiles and ships in the early part of the twenty-first century. It has been used with both carbon dioxide and Nd:YAG lasers. This process is discussed in more detail in Chapter 6.
Since the combined process makes more efficient use of the relatively expensive laser power, and since the addition of filler metal allows gap tolerances to be relaxed and provides a method of controlling the metallurgy of the weld metal, increasing use of the combined process is expected in the future.
In passing through the junction between two regions of a semiconductor that have different dopants, an electron loses energy and in some cases can emit this energy in the form of a coherent laser beam. Such laser diodes have a high divergence because of their small size. They can, however, be packaged together in such a way that the output of individual diodes adds to produce a kilowatt level, with specialized optics in the package to control the divergence.
In most cases, laser welding requires moving a laser focusing head, complete with gas shielding nozzles, over the part to be welded in a controlled manner with a relatively precise standoff distance. This is accomplished through computer numerically controlled workstations which control the movement of the beam over or around the part, or the movement of the part over or under the stationary beam.
In recent years, laser welding systems utilizing galvanometric scanners have been developed by a number of suppliers. Two fast scanners deflect the beam in orthogonal directions; the beam passes through focusing optics and is focused onto the part. A relatively small angular deflection of one of the scanning mirrors results in a relatively large deflection over the part being welded. One commercially available system uses a 1.25 m focal length mirror and can weld over an area of 0.61 m by 1.22 m. It produces 5 spot welds per second with a weld nugget of 3 to 4 mm in diameter. In this case, the target market is the automotive industry. In addition to the speed of welding, another advantage is the long standoff between the scanning and focusing head and the workpiece, allowing parts to be rapidly loaded and unloaded. This technology requires a laser of very good quality, to produce enough intensity to make the weld using the long focal length lens. Since it is impractical to apply a shielding gas over such large weldments at high speed, the process can only be used on material that can be welded in air. Moreover, unless multiple scanning systems are used or the part is manipulated, the spot welds can be applied from one side only, by line-of-sight from the focusing mirror.
Often, welding processes including laser welding require that a part be preheated to produce the desired metallurgical properties of the welded materials, or to prevent cracking. For example, laser weld repair of valves was accomplished in a four station work system where the parts were loaded or unloaded in station one, subjected to preheat with a gas torch in position two, laser welded in position three, cooled down in position four, and then rotated back to position one where the parts were unloaded and replaced. The timing was controlled by the length of the welding time in position three with the heat input in position two adjusted so that an adequate preheat was obtained. In other cases, inductive heating has been used to provide preheating for the laser welding process.
A related area in which preheat is beneficial is that where a significant amount of filler metal is required. Laser welding is not an efficient process when the laser energy is used to melt a large amount of filler metal. When it is desirable to weld with a filler metal, there are potential advantages to using a ‘hot wire’ feed, where the wire is externally resistively (and hence inexpensively) heated to a temperature close to its melting point prior to the wire being injected into the weld pool created by the laser. In spite of its potential advantages, this process has not been extensively investigated or utilized up to the present time.
There have been a number of instances where the use of two laser beams simultaneously has increased the laser beam welding capability. The laser welding had been limited at high speeds due to an instability of the molten pool known as the ‘humping’ instability (Albright and Chiang, 1988) which was observed as a regular bulging and constriction of the weld face. Banas (1991) developed the process of twin-spot welding using a spherical mirror hinged along its center so that relative angular deflections of the two halves produced two focal regions with an adjustable gap between them. With the laser power split between the two spots aligned along the direction of motion, laser welding of steel could be carried out at higher speeds without the occurrence of the humping instability. For example, in welding 1.5 mm stainless steel, the onset of the humping instability was shifted from 15m/min to almost 30 m/min using the twin-spot technique. The method was initially developed for tube welding but has been adopted for tailor blank welding and other high speed welding applications.
In welding tee-joints in thick section steel, for example in assembling stiffeners for deck plates in shipbuilding, one approach is to laser weld with the beam at a glancing incidence along one side of the joint, then repeat the weld from the other side of the joint to form a completely penetrating joint. It is found, however, that the first weld produces some distortion of the metal, due to the triangular shaped partial penetration fusion zone. It might be thought that the second weld produces similar but opposite distortion, resulting in straightening, but in fact some residual distortion remains. This distortion problem was eliminated when the weld was executed from both sides simultaneously, using a split beam. A feature of the two-sided welding is the linking together of the two keyholes, resulting in a fusion zone that maximally overlaps the seam.
The disadvantage of the two-beam welding techniques is the need for special optics to split the beam, a challenge due to the high powers involved. Moreover, careful alignment of the beam on the beam-splitting mirror is required to ensure approximately equal power in the two beams. In the case of split-beam welding of sheet metal, some work has been done using a focusing optic that produces an elongated beam, rather than using two distinct spots.
