Laser and Laser-Arc Hybrid Welding in European Shipbuilding

David J. Howarth
Lloyd's Register, 71 Fenchurch Street, London EC3M 4BS United Kingdom

Paper presented at the 11th CF/DRDC International Meeting on Naval Applications of Materials Technology, held at Halifax (Nova Scotia, Canada) on 7-9 June 2005.

Abstract

The use of lasers in the shipbuilding industry is in its infancy. In 1992, the European shipbuilding industry initiated work to look at the use of CO₂ lasers and the information that would be necessary to gain approval from the commercial regulatory bodies, i.e. the ship classification societies. At that time, only one yard had industrial experience of such techniques and only for the manufacture of pre-fabricated panels intended for cruise liner decks and internal walls. The research investigated applications on panel lines, both for butt and fillet welds. The driving force was an expected major reduction in distortion which would, in turn, make possible a main reduction in the amount of re-work, currently estimated to be as much as 15-30% of the total labour cost for new ship construction. The work resulted in the publication of guidelines for procedures for the approval of CO₂ laser welding in ship construction in 1996 by the European ship classification societies. Further work since that date has moved to laser-arc hybrid processes, employing both CO₂ and Nd:YAG lasers with MAG/GMAW, and culminating in two further revisions of the guidelines. This paper briefly describes the successes and pitfalls that have occurred over this 10-year period and the philosophies behind the guidelines.

1. Introduction

Very few ship construction yards use laser welding so it is clear that in shipbuilding the technology is still in its infancy. In June 1992, the European shipbuilding industry published a feasibility study concerning the use of laser welding in shipbuilding. ^[1] At that time, only one yard - Meyer Werft in Germany - had industrial experience of such techniques and even then only for the manufacture of pre-fabricated box panels intended for non-structural cruise ship decks and internal walls. The initial study raised a number of issues that would need to be overcome before laser welding technology could be used in production. The main barrier was seen to be rigid requirements of the regulatory authorities (in the case of shipbuilding, classification societies such as Lloyd's Register), who publish a series of rules and regulations, standards in layman's terms, according to which ships are constructed. Therefore, a second collaborative project ^[2] was initiated, and this time, it included the classification societies. This project aimed to look at applications on panel lines, both for butt and fillet welds. The obvious driving force was the anticipated major reduction in distortion resulting in improved accuracy of fabrication and a reduction in the excessive amounts of re-work seen in shipyards and which result from such problems. It was expected that the high capital cost of the laser welding equipment could be justified by savings in man-hours associated with the re-work.

This research was completed in 1996 and one of the major outcomes was the publication of guidelines for procedures for the approval of CO₂ laser welding in ship construction by the European classification societies and later by Lloyd's Register. ^[3] Further European-funded work under the acronym SHILWACQ ^[4] then looked at the wider issues of quality control and non-destructive examination of laser welding in shipbuilding. Other research initiatives looked for example at Nd:YAG lasers, ^[5] the beam of which can be delivered at the workpiece through flexible fibre optic cables. However, these lasers were limited to lower individual power (~4kW) than so far used with the CO₂ lasers, although experiments where the output of up to three lasers was combined optically were also investigated.

Meanwhile, development in industry was moving extremely quickly and due to the problems of achieving fit-up accuracy, in particular gap size, required for autogenous laser welding, the industry has moved into laser-arc hybrid welding, where a traditional arc welding process and a laser beam are combined in a single weld pool. Further EU6 and privately sponsored research has looked at these applications and the technology lessons learnt have resulted in further revisions of the classification societies guidelines, ^[7] which are still in force today. This paper briefly describes the technologies applied, the successes and pitfalls that have occurred over the 10-year period, and the philosophy underlying the published guidelines.

2. Fundamentals of laser welding

2.1 Use of high-power lasers for welding

Quickly after the practical realisation of Einstein's theory of stimulated emission was achieved in the first laser in the 1960s, lasers were being used as energy sources for different materials processing applications such as drilling, cutting and welding. Specific benefits of the use of lasers for these types of applications include the fact that the laser energy, which can be from milli-watts to multiple kilowatts in power, can be concentrated in a spot of typically less than a millimetre in diameter, giving a very high power density. When using lasers for welding applications in metals, this high energy density allows the so-called keyhole welding mechanism to be used, which is illustrated schematically in Figure 1.

In this mechanism, the focused laser energy in effect drills a vapour column in the workpiece. This keyhole is surrounded by molten metal flowing around it as the laser beam and thereby the keyhole are traversed along the workpiece. When the keyhole has passed, the molten metal solidifies, creating the weld. With this mechanism, welds of a very high depth-to-width ratio can be made with a very low heat input, leading to much narrower heat-affected zones and less distortion than is common with traditional conduction limited welds.

Although originally mainly applied to thin sections (i.e. up to a couple of millimetres in thickness), as the available laser powers increased, the interest in using lasers for welding of thicker sections naturally increased, particularly because with the keyhole mechanism, the laser could potentially realise in a single pass what would have to be done in a multi-pass procedure using traditional (arc) welding techniques. This would not only render benefits in terms of reduced heat input and distortion, but for example also in terms of easier weld preparation (e.g. straight edges for a butt weld), reduced or eliminated need for filler metal addition and increased productivity. Naturally, as welding is the main construction technique in the shipbuilding industry, some shipbuilders have looked at laser welding as a means of gaining (some of) these benefits.

2.2 Laser-arc hybrid welding

Even with the higher laser powers available, it proved less straightforward than expected to extend the application of lasers to welding of thicker and thicker sections. The main challenges for these applications were to get the joint fit-up good enough for (autogenous) laser welding (meaning a joint gap limited to a couple tenths of millimetres maximum), as well as achieving acceptable weld properties, even at the high cooling rates in thicker sections. One way of overcoming these issues is through the use of filler wire addition, but this generally leads to a considerable reduction in the achievable travel speed and/or penetration, as a large part of the available laser energy is used to melt the filler wire, making the process less efficient.

Another development, first investigated in the late 1970s, but of much renewed interest since the late 1990s, is laser-arc hybrid welding. In this hybrid process, the laser beam is combined with an arc welding process in a single weld pool. Although many variants are feasible (for example with GTAW, GMAW and PAW), of most interest is the use of GMAW (MAG) in a hybrid arrangement, because with this variant, addition of molten filler metal to the weld pool is inherent to the process. A schematic of this process is depicted in Figure 2. The main benefits of laser-arc hybrid welding over laser welding tend to be in terms of the increased tolerance to joint fit-up (joint gaps up to approximately 1mm can be coped with) through the addition of molten filler wire (when GMAW is used). In addition, higher travel speeds and joint completion rates can sometimes be realised, as well as improvements to weld quality and properties through the larger, more slowly solidifying weld pool.

2.3 High-power laser sources

There are several types of laser sources that have the power levels required for welding the sort of thicknesses common in shipbuilding (typically 6mm and over). The main one, which has been the workhorse of the laser industry for many years, is the CO₂ laser, in which gaseous CO₂ is the laser medium. Its benefits include good reliability, reasonable electrical efficiency (in comparison to other lasers), the relative ease with which high powers can be achieved (sources up to 45kW in power have been realised, although commercially readily available sources currently go up to some 20kW in power) and their good beam quality, which means a small focused spot can be realised at a reasonable stand-off distance from optics to workpiece. The main disadvantage linked to the CO₂ laser is the fact that the beam (at the primary wavelength of 10.64µm, i.e. in the infra-red) needs to be guided to the workpiece via mirrors, which often limits its use to systems based on Cartesian,gantry-driven operation.

The other main high-power laser source used in industry at the moment is the Nd:YAG laser, in which the laser medium is generally a rod of yttrium-aluminium-garnet (YAG), doped with neodymium (Nd) atoms. These lasers are typically available up to 6.5kW in power, though the commercial standard is 4kW. Main advantage of these lasers is that due to their 10-times shorter wavelength (at 1.064µm still in the infra-red), the beam can be guided via a flexible fibre-optic cable, making robotic, three-dimensional processing more easily achievable. Main disadvantage is their limited power and their poor efficiency, although this can be improved with the use of diodes rather than flash lamps for pumping.

The most recent addition to the high-power laser family is the Yb fibre laser, which - although not yet used in industry - is being investigated at different research institutes at power levels up to 17kW. In these lasers, the beam is generated in long, thin glass fibres with a suitable dopant, usually Yb for high powers. The wavelength at ~1.070µm is very close to that of an Nd:YAG laser, and it can also be guided via fibre-optic cable. In addition, it benefits from a high electrical efficiency, good beam quality and easy scalability.

3. Properties of laser welds

3.1 Strength and ductility

Laser keyhole welds are narrow welds when compared with normal, conduction-limited arc welds. The narrowness of laser welds creates difficulties in the measurement of the mechanical properties of the weld zone. For example, standard arc welding procedure qualification tests ^[8] and classification society rules ^[9] require an all-weld metal tensile test to determine the strength and ductility of the weld metal. Clearly, with such a narrow weld zone in laser welds, a standard all-weld metal tensile specimen would sample a greater proportion of parent and heat-affected zone material rather than weld metal. This makes the results at best difficult to interpret and at worst totally meaningless. However, tensile strength can be estimated from hardness for which good correlations exist.

Therefore, the real issue that needed addressing at the start of the developments for application in ship construction was the ductility of laser welds and how to measure it. Within a standard arc weld, a cross bend test is often used to measure ductility but even for these, problems arise with laser welds. The low heat input and the narrowness associated with the weld produce welds of high hardness; if a standard root or face bend test is performed across such a weld, yielding would take place preferentially next to the weld in the lower strength parent material and no deformation of the weld would take place. Consequently, the ductility of the weld zone is not determined. To overcome this problem, a longitudinal bend test was developed where the former diameter is sufficient to provide the minimum required percentage elongation in the outer surface of the bend specimen.

The determination of the weld metal yield properties was a more difficult problem to solve. Investigations were carried out into the use of narrow tensile tests and large-scale longitudinal tensile tests, but these failed to provide a practical answer. However, within the projects it soon became obvious that the use of autogenous laser welding was not a practical proposition in ship construction, because of the maximum joint gap that can be tolerated (of the order of a couple tenths of a millimetre) before an underfilled weld results. The use of filler wire was thus found necessary, which increased the interest in laser-arc hybrid welding. A move to laser-arc hybrid welding also meant that existing consumable approval methods, as used for arc welding, could be utilised.

3.2 Macro- and micro-structure

The macro-section in Figure 3a shows a typical autogenous CO₂ laser butt weld and that in Figure 3b an Nd:YAG laser-MAG hybrid weld. Three distinctive zones can be seen in both pictures:

• weld metal/fusion zone
• heat-affected zone (HAZ)
• parent plate

Fig. 3. Macro-sections of a) an autogenous CO₂ laser weld b) an Nd:YAG laser-MAG hybrid weld

Depending on the chemical composition of the materials being welded, high cooling rates associated with low heat input and/or high travel speed produce hard microstructures, whereas slower cooling rates associated with higher heat inputs and/or slower travel speeds give softer and generally tougher microstructures. With laser welds, the weld metal is typically in a molten state for a very short period of time and then solidifies and cools rapidly. As a result, the weld zone microstructure can be very hard and strong compared to the surrounding plate material. For steels with carbon contents in excess of 0.12% - that is most steels used in ship construction - the microstructure, when rapidly cooled, is fully martensitic. Although not always the case, generally such hard microstructures are associated with a relatively low toughness and are therefore unacceptable in ship construction.

A further result of such hard microstructures is that the weld is considerably overmatched in strength when compared to the surrounding parent material. In fact, it is not unusual for such overmatch to be of the order of a factor of two. However, this hard zone is very narrow; in Figure 3a the weld metal zone has an average width of approximately 1.5 mm. It is now generally believed that strength overmatching can be beneficial as it promotes failure in the parent plate which may be lower in strength but higher in toughness, due to the crack driving force being higher in the lower strength material. Cracks initiated in weld metal of an overmatched weld do normally deviate into the softer and tougher base metal. Such behaviour is known as Fracture Path Deviation (FPD), which can be seen in Figure 4 in a Charpy V-notch impact specimen. It should, however, be recognised that it is possible for cracks to propagate through weld metal and not undergo FPD in the event of truly brittle or defective welds.

The addition of filler wire (as for example in laser-GMA hybrid welding) can help to improve the weld metal toughness through alloying additions. With thick section laser-arc hybrid welding, two distinct zones can sometimes be identified in the weld metal (cf. Figure 3b); the upper zone near the weld cap mainly associated with the arc welding process and the lower zone near the weld root mainly associated with the laser welding process. Different microstructures are observed in these zones. The upper zone of the weld often has a microstructure dominated by grain boundary ferrite, acicular ferrite, bainite and Widmannstätten ferrite. An example of the microstructure in this zone is seen in Figure 5a. The lower zone of the weld is often dominated by martensite and grain boundary ferrite; an example is shown in Figure 5b. In contrast to the weld metal, the HAZ has the same microstructure in the upper and the lower zone. Close to the fusion line, the HAZ microstructure is martensitic with some grain boundary ferrite; further away from the fusion line the microstructure consists of ferrite and carbides. Figure 5c is typical of the microstructure in the HAZ close to the fusion line.

Fig. 5. Microstructure in the a) upper weld zone (near the weld cap) b) lower weld zone (near the weld root) c) heat-affected zone

3.3 Hardness

Current shipbuilding standards ^[8] restrict weld zone hardness levels for arc welds in constructional steels to a maximum of 350HV5, which was also considered applicable to laser welds. Obviously, however, with the low heat input and high cooling rates of laser welds, control of the hardness within this level can be difficult. The hardness of the weld zone is mainly determined by the steel composition and the cooling time. For the latter it is commonly accepted to use the cooling time from800°C to 500°C ( Δt _8-5 ) as a measure. Based on this assumption, it is possible to estimate the hardness from various models, which typically are based on experiments with a number of steels followed by regression analysis; the following uses the model by Terasaki. ^[10] Obviously, the prediction of the models gives an estimate of the hardness only. As an example, the influence of the travel speed on the hardness (through its effect on the cooling rate) is shown in Figure 6.

Practically, for a given welding process, there are two ways in which the weld zone hardness can be controlled, namely through the hardenability of the material, and through the cooling rate. In practice, this gives three methods that can be used:

1. Limit the travel speed and thereby cooling rate, so that the microstructure and consequently also the hardness become acceptable for the relevant steel composition.
2. Control the chemical composition and thereby hardenability, so that the hardness is acceptable at all travel speeds.
3. Apply preheat, in order to decrease the cooling rate and thus the hardness.

To allow the maximum benefit to be derived from the low heat input, fast laser welding process, control of the chemical composition beyond that agreed for arc welding is most suited. Standard composition limits for steels used in ship construction allow for carbon levels up to 0.21%, but even at 0.12% carbon, a fully martensitic microstructure as may result from a low heat input laser weld will exhibit a hardness level of 400HV5. The effect of the carbon content on hardness is illustrated in Figure 7.

The compositional controls agreed3 for laser welding steels are shown in Table 1. Two steels are specified, designated L24 for normal strength ship steel applications and L36 for higher tensile ship applications. The mechanical properties are those shown in the Rules. ^[9]

Table 1 Chemical composition requirements for laser weldable shipbuilding steels

C	0.12% max.
Mn	0.9 to 1.6% (see Note 1)
Si	0.50% max
S	0.005% max.
P	0.010% max.
Al (acid soluble)	0.015% min. (see Note 2)
Nb	0.05% max.
V	0.10% max.
Ti	0.02% max.
Cu	0.35% max.
Cr	0.20% max.
Ni	0.40% max.
Mo	0.08% max.
N	0.012% max.
NOTES Manganese may be reduced to 0,70% for the L 24 grade consistent with the lowest values used in the weld procedure test. The total aluminium content may be determined instead of the acid soluble content. In such cases the total aluminium content is not to be less than 0.020%.

With these compositional controls, however, difficulties were still experienced with limiting the hardness in all cases to a maximum of 350HV5. Based on the understanding that laser welding is a low hydrogen process (<5ml H₂ per 100g of weld metal), the requirement could be raised to 380HV5, as also current practice for guaranteed low hydrogen arc welding.

Even with laser-arc hybrid welding, although it incorporates an arc and generally filler wire addition, control of weld zone hardness still can be a challenge, because of the low heat input and high travel speed. In addition, when using laser-arc hybrid welding, often a desire exists to use a general structural steel composition suitable for arc welding, which allows higher levels of carbon and the carbon equivalent value than the levels recommended by the classification societies ( Table 1) for laser welding. In those cases, the use of pre-heat may be the only option to limit the hardness. The effect of pre-heating may also be calculated using the predictive models; for example, applying a pre-heat of the order of 120°C is considered necessary to limit the hardness to 380HV5 at a speed of 2m/min for a steel with 0.12% C. However, preheat is often not the preferred option as it is costly and time consuming and can limit the gains achievable in the reduction of distortion; control of the steel composition as described above (with the effect illustrated in Figure 7) then may provide a more sensible approach.

3.4 Charpy V-notch impact properties

It was clear from the start of the developments that a standard Charpy V-notch impact test sample taken with the notch located in the centre of a laser weld did not provide a result truly representative of the properties of the weld metal. This results from a phenomenon known as Fracture Path Deviation (FPD), which was already mentioned and illustrated in Figure 4. Since the strength of the weld metal is highly overmatching when compared with that of the surrounding parent material, a crack running through such a narrow region will automatically deviate into the softer surrounding material. A Charpy impact specimen exhibiting FPD will therefore measure the toughness of a mixture of the weld zone and the parent material, thus giving what could be considered an invalid result, since the true toughness of the weld metal was therefore not determined. Research did, however, suggest that a fracture toughness Crack Tip Opening Displacement Test (CTOD) could (if the weld metal has low toughness) accurately determine the toughness of the weld metal. However, under normal conditions and even with such a searching test, FPD would still occur. In addition, the test technique is extremely expensive and therefore not suitable as a routine testing method. Although initially providing encouraging results, alternative methods of utilising the Charpy test by way of side grooving and by a technique that became known as the 'Three Weld Technique' failed to fully satisfy the requirements for a simple, low cost quality control test.

Subsequent research ^[11] suggested, however, that the standard Charpy test is suitable to determine if the weld metal has acceptable toughness This research was carried out on laser welds that were artificially made brittle. It showed that in the case of a low toughness weld, the crack during the Charpy test would propagate through the weld, and FPD would not occur ( Figure 8). This was found to be the case where the energy value was below that required by the Rules. Therefore, the Charpy test with all its failings is still considered the only suitable test, and FPD must be recorded, but is not automatically considered an invalid result.

3.5 Fatigue performance

A considerable body of data relating to the fatigue strength of laser welds in steel now exists. ^[12] The results are encouraging as in no valid measurement has the endurance been less than the design endurance for a similar type of arc welded joint (examples can be seen in Figures 9 and 10 for butt and T-butt joints, respectively). However, a word of caution is in place since a number of the tests carried out were on specimens that were too small to contain realistic residual stress levels and thus the endurance in the test is likely to have overestimated the performance of the joint geometry in a structure. Additionally, only three joint geometries have been investigated so far namely, butt joints, T-butt joints and cruciform joints. Also, all welds were made under optimal conditions, which is not necessarily representative of real applications.

Nonetheless, on the basis of the information generated to date, although insufficient to produce design rules for laser welded joints, it is reasonable to assume that the design principles established for arc welded butt and T-butt joints can be applied with confidence.

Although much of the work reported on has been for butt welds, by far the greatest part of the total weld length in ship construction consists of fillet welds. The design strength of a fillet weld is calculated from the throat thickness (or by conversion from the leg length). These parameters are specified in classification rules and can be easily confirmed non-destructively by the surveyor during construction. Fillet welds can also be made with lasers, but because of the deep penetration of the laser, the majority of the weld throat is then within the confines of the thickness of the material being joined, making it impossible to externally confirm the adequacy of the minimum design throat thickness (this difference is schematically illustrated in Figure 11).

For this reason, all laser T-joint welds were initially made as full penetration T-butt welds, since T-butt welds can be ultrasonically tested to confirm penetration. However, with experience, with the advent of wider hybrid welds and with automated control and recording of weld parameters, the confidence and assurance needed to accept (partial penetration) laser fillet welds was reached. An example of such a joint can be seen in Figure 12.

5. Common weld imperfections

As with all fusion welding processes, there are a number of imperfections that can form during laser welding. Full descriptions of possible imperfections (both internal and at surface) and their acceptance levels are adequately covered within the International standard ISO 13919-1:199713 and are only briefly mentioned here.

The methods of examination for imperfections in laser welds require careful consideration. Firstly, visual examination is used to assess laser welds for external imperfections. The standard imperfections that can thus be detected are:

• Surface-breaking cracks and porosity
• Lack of penetration (in the case of a full-penetration weld)
• Undercut
• Excess weld metal and excessive penetration (excessive root reinforcement)
• Drop through and sagging
• Linear misalignment
• Incompletely filled groove
• Root concavity shrinkage groove

Conventional radiography or ultrasonics generally form a bottleneck in production due to the high speed of the laser process. The answer would seem to lie in the real time control and recording of key parameters used in the process. Deviations outside the qualified range would be flagged for a later, more conventional inspection. Radiography and ultrasound may then be used to examine laser butt welds for internal imperfections. (Radiography is not effective at monitoring internal defects in T-joints. For this reason only ultrasound inspection is required for T-joints.) The standard imperfections that can thus be detected are:

• Cracks
• Porosity and gas pores
• Shrinkage cavities
• Solid inclusions
• Lack of fusion
• Lack of penetration

The defects listed above can also be found in conventional arc welds. However, the early work on CO ₂ laser welding highlighted what is the major problem with laser welding of thick section ship steels, namely that of solidification flaws.

Solidification flaws are internal crack-like flaws formed at the weld centreline ( Figure 13). The causes of such flaws have been attributed to steel composition, heat input and plate thickness. As a result, careful control of these parameters (and in particular of the composition) is required to avoid them during production.

Fig. 13. Solidification flaw in a) a CO₂ laser weld in 15 mm plate b) an Nd:YAG laser-MAG hybrid fillet weld

Solidification flaws were also experienced in the early development of the electron beam welding of steels. This problem was found to be strongly linked to the sulphur and phosphorous content of the steel. Empirical relationships were developed as a guide to demonstrate the steel compositions that were readily weldable and those that would be prone to the occurrence of solidification flaws. The relationship, known as a cracking index, gave a number based on the composition and it would have to exceed a specified value in order to be accepted for electron beam welding. However, the empirical relationships established for electron beam welding are not valid for laser welding. Indeed to date, no exact relationships for laser welds have been found. The best two so far are listed below (however, it would be unsafe to try and determine an acceptable steel composition from either of the equations):

C x [10Mn - 200S - 400P] - 12Zn + 0.3 = Cracking Index F ₁     (1)

50C + 5Mn + Si + 7Cu - 100S - 200P + 1 = Cracking Index F ₂     (2)

A more practical method was developed within the SHILWACQ project, ^[4] where the susceptibility of a particular steel composition to the occurrence of solidification flaws was determined by constructing a weldability lobe as part of the welding procedure. A weldability lobe is constructed by making bead on plate welds at a number of different combinations of welding powers and travel speeds. The resulting welds are then assessed visually and by radiography and categorised as acceptable, defective (i.e. containing solidification flaws) or as showing drop-through or incomplete penetration. These parameter combinations are then plotted as shown in Figure 14.

The weldability lobe indicates the welding conditions under which welds free from solidification flaws are to be expected for a particular steel composition. However, welding under conditions that have been shown to be free from solidification flaws does not remove the need to inspect production welds. It does, however, allow the consideration of steel compositions outside those given in the original Classification Society Guidance Notes. ^[3]

6.1 Introduction

The main reason for using lasers for shipbuilding applications tends to be the increased manufacturing accuracy that can be realised, mainly resulting from a reduction in thermal distortion. This can have a direct financial benefit during assembly, through a reduced need for rework and straightening, the cost of which has been estimated to be as high as 15 to 30% of the labour costs for new ship hull production. ^[13-16] Still, today, there are only about a handful of shipbuilders worldwide using lasers in production. Some of these use laser or laser-arc hybrid welding as a direct substitute for a more traditional welding process, whereas others use the specific characteristics and opportunities of lasers, for example resulting from the keyhole welding mechanism, to allow new designs or thinner materials to be employed. Furthermore, in some shipyards the laser is not just used for welding, but also for other materials processing techniques such as cutting, marking and primer removal. In the following, the main shipyard applications of lasers in Europe will be briefly discussed.

6.2 Meyer Werft

Meyer Werft of Papenburg, Germany, specialises in large cruise vessels, as well as river cruisers and freight ships, and is without doubt the yard making most intensive use of lasers for welding applications. ^[17-24] Firstly, there are the so-called I-Core ^TM panels, which are metallic sandwich panels, in which the internal stiffening is provided by individual metallic slats at right angles to the cover plates. Welding is performed using a keyhole stake weld through the cover plates into the stiffeners, resulting in very flat panels and therefore easy subsequent use in their ships. Meyer Werft started developing I-Core in 1994 and set up a separate welding shop for these panels, currently equipped with two 12kW CO₂ lasers. The I-Core sandwich panels are for example used in decks, walls, bulk heads and staircase landings in their cruise ships, but they are also finding increased application outside the shipbuilding industry, for example for railway rolling stock, park houses and tipper trucks.

The other laser application at Meyer Werft is on a panel line. Installed in early 2002 and thought to be the largest laser welding installation in the world, it is equipped with four further 12kW CO₂ lasers. On this fully automated panel line, butt and T-joints of up to 20m in length for 'traditional' stiffened panels are made using laser-GMA hybrid welding. Panels and stiffeners are milled before welding, introducing a small groove (of the order of 6°). The benefits realised are mainly in terms of increased productivity (also allowing a reduction in the number of welding stations) and improved accuracy/reduced distortion of the sub-assemblies. It is estimated that currently over half the entire weld length in Meyer Werft's ships is welded using a laser.

6.3 Blohm+Voss

Blohm+Voss specialise in building frigates, corvettes, fast cruise liners and large yachts, and therefore much of the steels it welds are only 4-5mm in thickness. In 2000 - after more than 10 years of research - they introduced a line for laser welding of butt and T-joints of 'traditional' panels at its yard in Hamburg, Germany. ^[15,25-27] The separate panel line was introduced as part of a drive towards precision manufacturing and a streamlining exercise of the shipyard, and is equipped with two 12kW CO₂ lasers.

Blohm+Voss uses its lasers as a 'flexible tool', meaning that they can be used for welding, cutting, marking and primer removal. At the moment, a laser is used for cutting and then laser butt welding of panels of up to 12m in length(in thickness typically 3 to 8mm), although laser-GMA hybrid welding is being investigated for the butt welds, because of the increased tolerance to joint gaps. For welding of T-joints for stiffener attachment, it uses a simultaneous double-sided autogenous laser welding procedure. To prevent underfilling, the underside of the stiffeners is milled flat to give a tight fit-up. With this approach, the yard can create a full-penetration, double-sided T-butt weld in stiffeners of between 5 and 12mm in thickness.

6.4 Odense Steel Shipyard

Odense Steel Shipyard (OSS) near Odense, Denmark, is mainly active in the market for non-passenger ships, such as large container ships and tankers. Its investigations into the use of lasers began in the late 1980s, leading to a pilot cell processing sub-elements, equipped with a 12kW CO₂ laser in the 1997. ^[14,28-29] Similar to Blohm+Voss, the laser is used for cutting, marking, primer removal and welding (mainly of T-joints). From 2000 onwards, OSS started investigating laser-GMA hybrid welding, the development of which was accelerated in2003, when it upgraded and increased the use of its laser cell for welding sub-assemblies for some small navy support vessels.

In addition, OSS has been involved with development projects looking at the suitability of using flexible optical fibre-delivered Nd:YAG laser beams for robotic welding of more complex components. There are plans to introduce Nd:YAGlasers in a production cell.

6.5 Other yards and activities

Fincantieri started its developments for laser welding at its shipyard in Monfalcone, Italy, where it builds large cruise vessels. ^[30,31] There, it installed a seam welder equipped with a 17kW CO₂ laser as part of the traditional panel line. Initially, it was envisaged to autogenously laser weld butt weld the milled plates on the panel line, but due to variation in joint fit-up, filler wire addition was incorporated. A further change into laser-GMA hybrid welding is being investigated.

Other yards where laser and laser hybrid welding have been heavily researched, or are close to being used in production include the Aker Kvaerner Masa yard in Helsinki, Finland, where a laser-hybrid welding system is being incorporated on an existing panel line system. ^[32] Similarly, the Aker Warnow Werft near Rostock, Germany, is retro-fitting two 4.4kW diode-pumped Nd:YAG lasers onto an assembly plant. ^[33] This is in line with the increased interest in flexible fibre-delivered solid-state lasers because of the easier beam guidance and thereby increased opportunities for three-dimensional processing. Other yards are also looking at solid-state (fibre or Nd:YAG) lasers include Meyer Werft, IZAR and others. ^[34,35]

7. Conclusions and future prospects

It is clear from all the development that has taken place over the years that laser and laser-arc hybrid welding have applications within the shipbuilding industry. The benefits that can be realised are primarily in the area of improved joint completion rates and reduction of thermal distortion, particularly in thinner sheets. Early involvement of the regulatory bodies and the classification societies in this development has helped to ease the path of introduction.

For successful introduction, the characteristics of the low heat-input keyhole welding process raises certain requirements. For example, excellent joint fit-up and dimensional control are necessary, although these requirements can be relaxed somewhat by the employment of laser-arc hybrid welding, rather than laser alone. In addition, the fast cooling rates realised in the weld zone and the narrow weld profile put certain constraints on the allowable chemical composition of the materials used.

Nonetheless, successful introduction has been realised by some yards in Europe. To help facilitate further introduction, the Classification Societies have produced rules/guidelines that allow the approval and successful introduction of laser welding into commercial shipbuilding.

References

1. S. T. Riches, J. Klaestrup Kristensen et al, 'Laser Welding in Ship Construction Phase 1 - Feasibility Study', TWI -UK/Force Institute-Denmark, June 1992.
2. 'Laser Welding in Ship Construction', Final project summary report, compiled by TWI, L-SHIP/PSG995).16, April 1995.
3. 'Guidelines for the Approval of CO₂ Laser Welding', Lloyd's Register, March 1997.
4. EU BRITE-EURAM Project BE 97-4223: SHILWACQ 'Shipbuilding Laser Welding: An Integrated System for Assurance and Quality Control.
5. 'Exploitation of High Power Nd:YAG Laser Processing', TWI Group Sponsored Project 88277, May 2002.
6. EU Framework 5 - GROWTH Project GRD1-10486 ShipYAG 'Shipbuilding Low Cost, Versatile and Safe Welding by YAG Laser Application', May 2004.
7. 'Guidelines for Approval of Laser Welding', Lloyd's Register, May 2005.
8. 'Specification and qualification of welding procedures for metallic materials. Welding procedure test. Arc and gas welding of steels and arc welding of nickel and nickel alloys', ISO 15614-1
9. 'Rules and Regulations for the Classification of Ships, Part 2, Manufacture, Testing and Certification of Materials', Lloyd's Register, July 2004.
10. T. Terasaki, T. Akiyama and M. Serino, 'Chemical Compositions and Welding Procedures to Avoid Cold Cracking', Proc. of Second International Conference on the Joining of Materials (JOM-2), Elsinore, Denmark, April 1984.
11. I. Hadley, J.D. Russell, 'Investigation of Fracture Path Deviation in Laser Welds', TWI Report 221238/1/98, July 1998.
12. G.S. Booth, 'Fatigue of Laser Welds for Ship Construction - A Review', TWI Report 620924/2/95, September 1995.
13. ISO 13919-1:1997, Welding - Electron and laser Beam Welded Joints - Guidance on Quality Levels for Imperfections, Part 1 Steel.
14. G. Sepold and P. Seyffarth, 'Laserstrahltechnologien für den Schiffbau (Laser beam technologies for shipbuilding)', Proc. Conf. Schweissen und Schneiden 1994, DVS Berichte 162, 159-164, Bremen, Germany, 28-30 September 1994.
15. M. Sellerup, 'Experiences of laser welding and cutting in hull production', Proc. Int. Inst. of Welding on Welding in Shipbuilding, DVS Berichte 195, 25-27, Hamburg, Germany, 17-18 September 1998.
16. Anon., 'Brochure: Laser beam welding and cutting at Blohm+Voss GmbH', Blohm+Voss, Hamburg, Germany.
17. P. Seyffarth, 'The use of high-power-lasers in shipbuilding', Proc. Int. Conf. Advanced Metallic Materials and their Joining, Bratislava, Slovakia, 25-27 October 2004.
18. F. Roland, H. Lembeck, S. Jaudas and C. Lankenau, 'Trends, problems and experiences with laser welding in shipbuilding', Proc. Int. Inst. of Welding Shipbuilding Seminar, Odense, Denmark, 17-19 April 1996.
19. Anon., 'Brochure: Laser welding and sandwich panels', Meyer Werft, Papenburg, Germany, 1999.
20. P.A. Blomquist, 'Lasers in U.S. shipbuilding', Industrial Lasers Solutions for Manufacturing, 5-8, 2002.
21. F. Roland, 'Laser welding in shipbuilding - chances and obstacles', Proc. 8th Laser Application Symposium, KIMM, November 1997.
22. F. Roland, 'Stake welding of lightweight structures for shipbuilding using a high power laser', Proc. Convergent Energy Focal Spot Seminar, Orlando, Florida, U.S., 26-29 March 1995.
23. F. Roland and T. Reinert, 'Laser welded sandwich panels for the shipbuilding industry', Proc. Conf. Lightweight Construction - Latest Developments, Paper 8, London, U.K., 24-25 February 2000.
24. F. Roland and H. Lembeck, 'Laserschweissen im Schiffbau - Erfahrungen und Perspektiven auf der Meyer Werft (Laser welding in shipbuilding - experiences and perspectives at the Meyer Werft)', Proc. Int. Aachen Welding Conf., Aachen, Germany, 3-4 May 2001.
25. F. Roland, T. Reinert and G. Pethan, 'Laser welding in shipbuilding - and overview of activities at Meyer Werft', Proc. Int. Inst. of Welding Conf. on Advanced Processes and Technologies in Welding and Allied Processes, Paper B-II, Copenhagen, Denmark, 24-25 June 2002.
26. A. Kahl, 'Compact Shipyard', Proc. Int. Inst. of Welding Conf. on Welding in Shipbuilding, DVS Berichte 195, 60-63, Hamburg, Germany, 17-18 September 1998.
27. Anon., ''Revolution' in Halle 3 (Revolution in Hall 3)', Euro Laser, 30-34, December 2001.
28. Anon., 'Blohm+Voss GmbH - first shipyard in the world to implement laser beam welding and cutting', Tribon Newsletter, 42, 4-5, July 2001.
29. M. Sellerup, 'Laser-welded ships are afloat', Industrial Laser Solutions for Manufacturing, 14 (4), 20-22, 1999.
30. S. Lembæk and M. Elvang, 'Laser applications at Odense Steel Shipyard', 4th Wismar Seminar on Modern Joining Technologies in Shipbuilding, Wismar, Germany, 28 September 2004.
31. L. Manzon, 'Welding in the shipbuilding industry: the experience at Fincantieri', Welding Review International, 17 (2), 6-8, 1998.
32. L. Manzon, 'Laser welding in shipbuilding: application and prospects at Fincantieri', Proc. conf. ASPOW, Paper 18, Geesthacht, Germany, 4-5 February 1999.
33. P. Jernström, 'Research on welding techniques in the integrated project InterShip', 4th Wismar Seminar on Modern Joining Technologies in Shipbuilding, Wismar, Germany, 28 September 2004.
34. P.Seyffarth, U. Jasnau and R. Reipa, 'Einsatz von Lasertechnik bei Aker Ostsee/Nachrüstung von Fertigungsanlagen mit Festkörperlasern (Employment of laser technology at Aker Ostsee/Retro-fitting of solid-state lasers on assembly stations)', 4th Wismar Seminar on Modern Joining Technologies in Shipbuilding, Wismar, Germany, 28 September 2004.
35. H. Staufer, 'Laser-hybrid welding of ships', Welding Journal, 83 (6), 39-43, 2004.
36. G. Pethan and F. Roland, 'Docklaser', 4th Wismar Seminar on Modern Joining Technologies in Shipbuilding, Wismar, Germany, 28 September 2004.