LEARN MORE

Related Suppliers

Trumpf Inc.



Report on LASERS

A quick look at some of the things that are going on in what is becoming an increasingly hot technology...

The activity remains brisk in laser tailored blank welding. The predominant type of lasers used to make the blanks—for now, anyway—are CO2 lasers. In the case of one laser equipment vendor, TRUMPF Inc. (its Laser Technology Center is in Plymouth, MI), Frank D. Brennan, its Laser Sales Manager, says that 8-kW tends to be the size selected for the task. He explains that there doesn't seem to be much of a gain in throughput at higher power for the simple reason that steel can't be welded much faster: it needs time to melt and to solidify.

However, he points out that some people believe that high-powered YAG lasers—"high-powered" in this context meaning 4 kW—are the way to go. The reason why they're partisan to this approach: the ability to snake the beam through an optical fiber.

And the reason why there may be an increase of YAG lasers for blank welding relates directly to this fiber optic capability. There is an increase in the amount of non-linear welding for tailored blanks.

Consider a simple blank for a door. The main area of the door is made of 1.2 gage material. But on the side where the door is to be hung on the body, greater strength—which translates to greater thickness—is required. Say 2.0 gage.

Historically, a blank of this type would simply involve welding two rectangular pieces of material together with a straight seam. But the need for strength is actually localized in the area where the brackets are to be attached to the door. There's no need for an entire rectangular shape. This gets right to one of the rationales for tailored blanks: instead of having a component made entirely with the thickest material required (or entirely with corrosion-protective coatings), the component is made so that requirements are met only in the area where they need to be met.

So in the case of the door, a U-shaped blank would be more appropriate. This U-shape requires non-linear cutting and welding. This is facilitated by the ability to run a fiber optic along the arm of a robot, which can readily move in the required space. So to the extent that design engineers begin to really tailor their blanks, there may be a move away from simple X-Y coordinate-based blanks to more complex shapes. Which, in turn,would lead to a greater implementation of YAG lasers in blank welding applications. (However, there may be a continuation of CO2 lasers in this area thanks to a flexible arm development—see Flexibility Comes to CO2)

Another driver of increased YAG use is the increased use of aluminum for body panels. The issues are material reflectivity and coupling characteristics. Simply put, the YAG wavelength (1.06 mm) is more conducive for welding aluminum than the CO2 wavelength (10.6 mm). (Brennan points out that it is possible to take a high-power CO2 laser, direct it right onto an aluminum alloy...and the material doesn't melt.)

Doubling Up. A recent development for CO2 welding, one that's been available for YAG welding (through the development of a Canadian company, Automated Welding Systems, Markham, Ontario), provides two spots on the workpiece surface at the same time.

In the case of the YAG setup, this is accomplished by directing the output of two lasers through one welding optic that outputs two spots, which can be side-by-side or overlapping. One benefit of this approach is that if two 4-kW YAG lasers are used, there's 8 kW on the workpiece surface. Which facilitates welding.

In the case of the CO2 setup, it is one laser and one set of optics. Yet two beams emerge. The spatial relationship of the two spots is a function of how the optics are fabricated.

One advantage of this two-spot arrangement relates to blank welding. If the edges of the blanks aren't sheared with completely straight edges, the ability to overlap two spots means that there is some compensation for gaps, so the two pieces can be joined.

Design Flexibility. Hydroforming is huge. And it promises to get even bigger. General Motors' bet on hydroformed, laser-cut rails for the Chevy Silverado and GMC Sierra pickups has paid off smartly. According to Frank Brennan, every major automotive frame supplier is working on hydroforming projects and he indicates that programs in the works are likely to result in orders of some 200 lasers by the fourth quarter of 1999.

The benefits of hydroforming, especially as compared with traditional fabricating methods, are multiple. Not only are part counts slashed, but time, material, and operators are all reduced, as well, as welding individual pieces gives way to forming tubes.

Hydroformed rails require an assortment of holes, slots, and other openings. Fundamentally, there are two ways that these can be produced. One is via hydropiecercing. It requires fixed tooling. The other method is to use a laser. Because of the nature of the hydroformed rail cutting operation, which necessitates moving up and down along the curved tube, a robot-laser combination is the norm. Which means YAG lasers. Brennan says that for most hydroformed rail cutting, 700- to 1,500-W lasers are the norm. Not only is the need for fixed tooling eliminated, but robot vendors such as Fanuc and Motoman have been optimizing their hardware and software for laser cutting, with an especial emphasis on cutting hydroformed tubes. All of which is to say that the use of lasers for cutting hydroformed tubes is allowing design engineers with greater freedom, and even the ability to make changes latter in the program than are economically allowed with hard tooling.

As people who are familiar with frame plants know, those places tend to be heavy-metal environments: tough, to say the least. Arc welding is a natural in that setting. . .but lasers? Yes, we're talking about "industrial" lasers, but still . . . Brennan says that lasers have been proven to perform reliably in frame plants. The key issue is to keep the amount of time that the laser cavity is opened (to change flash lamps, for example) to a minimum. The reason: if contaminants get on the reflective surfaces, things can go very bad in short order.

(TRUMPF's Haas lasers, such as the model HL 1003D, a 1-kW unit that's widely used in hydroforming applications, has a design that facilitates protecting the pumping chamber. The flash lamps are in the top portion—or lid—of the laser resonator. So when the top is opened, it can be a simple matter of placing a clean towel over the pumping chamber, thereby protecting the mirrors.)

No Lamps. Imagine a situation, however, where there is not really an issue with changing the lamps for a YAG laser. Imagine that the life of what is standing in for lamps is on the order of 10,000 hours. (In the context of one-shift operation, that translates to about five years without having to open up the chamber.) This isn't imaginary, as Rofin-Sinar Inc. (Plymouth, MI) is now offering an array of diode-pumped YAG lasers up to 2.5 kW output. As Richard Walker, the firm's general manager, puts it, modestly, "Lamps are no longer the best solution"—at least as long as you're looking for CW, not pulsed, operation. You turn a diode on and it runs continuously; you can't pump a diode. "In an industrial environment," Walker says, "you don't need to open the top."

Diodes are solid-state devices. They are used in such things as CD players. In order to get the sort of output used for industrial lasers, there are stacks of diodes built up.

One of the issues with diodes is that compared with flash lamps, they are pricey, even though during the last decade the price per watt has plummeted. But this price penalty relates to the initial investment cost. A diode-pumped system can have a 50% higher capital investment cost. But Walker points out that taking into account the beam quality provided, the operating cost (the diode pumped laser is much more efficient with regard to transferring the power out of the wall plug for lasing) and the maintenance cost ("solid-state" means no moving parts), the diode-pumped laser is more economical than the standard YAG laser.

Rofin-Sinar is even going beyond diode-pumped YAG lasers to straight diode lasers: the diode output is used.

After the issue of achieving enough power out of a diode laser was overcome—you could get 2.5 kW—there was still an issue of beam quality. But not only has Rofin-Sinar achieved the required improvement in the beam, they've also managed to put the beam through a 1.5-mm fiber. While the output isn't what's needed for cutting, anything a YAG laser can weld, a diode laser can weld, too (with the exception of deep penetration welding).

Compared to any other type of 2.5-kW laser—YAG or CO2—the diode laser is an extraordinarily compact device, which is an important aspect in increasingly compact work areas. With the low maintenance requirements, the 30% power efficiency, and the long-life of the diodes, this type of laser is now positioned to become highly competitive in the industrial marketplace.

The Leading Edge
surface treating with a laser
Surface treating with a laser is an area that Dr. Mazumder thinks will grow in importance. (Photo courtesy of TRUMPF).

According to Jyoti Mazumder, director of the University of Michigan's Center for Laser Aided Intelligent Manufacturing, (he's also Robert H. Lurie Professor of Engineering, Mechanical Engineering and Applied Mechanics, College of Engineering), one of the most significant areas of development that is making its way out of the research setting and into commercial applications is solid free-form production.

Basically, this permits the creation of things like tooling—initially for prototype runs, but eventually for production applications—for injection molding and diecasting.

While some people might look at this process, which involves melting materials with a laser (type and power are related to parameters of process speed and surface quality achieved), as being analogous to rapid prototyping processes like stereolithography, Mazumder points out that rather than melting polymers, this process works with materials like H13 tool steel. "If you can do H13, you can do anything," he quips.

"A key advantage of this process is that you can design the molds or dies so that the cycle time can be reduced by 40 to 50%. That's where the money is: the time it takes to produce parts," Mazumder says. The reason why these fast cycles are possible, he explains, is because (1) the design of the cooling channels can be optimized (the tooling build is driven by a CAD design, so the channels can conform to workpiece shapes more readily than is possible with conventional tool-making processes) and (2) the process permits the introduction of other alloys into the mold or die where needed, so, for example, more conductive materials can be strategically located within the overall structure.

One area that he thinks is ripe for growth is the use of lasers for surface modification. Although laser cutting and laser welding are fairly well accepted world wide, when it comes to this application, "People are only scratching the surface," he puns. Not only does this mean treating the surfaces of things like dies for better performance, but also for cladding components, like aluminum engine features (e.g., valve seats).

Some advanced work that Mazumder and his colleagues at Michigan are working on is actually creating tailored materials for specific applications. For example, they might create a material that's strong or light. Or they might optimize for things like creep or coefficient of thermal expansion. This would allow engineers to develop the parts they want, then create the materials that would provide the characteristics they are looking for.

Flexibility Comes to CO2. The history of 3D beam delivery systems for high-power CO2 lasers is a short and comparatively unhappy one. Those who remember some of the systems from the late 1980s would undoubtedly prefer to forget them.

But for the past several years, a simpler arrangement has been used in Europe. Generally described as a "flexible arm," it can conceivably stem what seems to be an inexorable tide of YAG units in non-linear welding applications. (Why Europe and not here? One reason, Walker suggests, is that in the U.S., there was greater emphasis on transmission welding than on bodies; in Europe, the reverse was true.)

Simply, there are two tubes—the length of which are purpose determined—united with a joint; each tube has another joint at its open end. One joint is coupled to a laser resonator. The joint on the other tube is attached to a laser head that is fitted on the end of a robot arm. The joint in the middle is suspended from the ceiling with a cable that has an extension/retraction mechanism.

Within each of the joints are sealed optical elements. What this means is that there is generally no need to do anything with the mirrors, which was a problem with some of the early systems. Also, the beam path is always the same, so there is reliable consistency.

While more limited in range of motion than a flexible cable attached to a laser head on the end of a robot arm as is the case with a YAG laser, this arrangement does allow three dimensional welding. Although the system is comparatively new here, there is North American use welding exhaust components in production.

The Remote Approach. One potential breakthrough development for the use of lasers in body welding applications is through the use of remote welding systems. At Rofin-Sinar, for example, they've developed a system that employs its diffusion-cooled laser. The laser is mounted on a platform. The work zone is below the laser such that there is open access for conventional tooling for things like door panels; the footprint of the structure is 12 × 14 ft.

Ordinarily, the standard focal length for a 2.5-kW laser that's used for welding would be on the order of 250 mm. But with the beam quality provided by the diffusion-cooled laser and a clever optical arrangement that includes high-speed, high-precision linear (for the focus lens) and rotary (for the copper mirrors) stages, they are able to provide a focal length of 1.4 m. With a 3.5-kW laser, the focal length is 1.6 m. Which means that the work envelope is greater than 1 × 1 m for either one of these systems.

Realize that this remote welding system is able to make moves between weld locations at a rate in excess of 2 m/sec. Looked at from the standpoint of comparing it with a robotic spot welding system, it can produce four to 10 times as many welds in the same amount of time. Which means that it is possible to do a whole lot more work in a much smaller area, to say nothing of the advantages of laser welds versus resistance welds.

According to Walker, "The proof of principle for this system is over. Now it is a matter of finding the right applications." He suggests that within a year there will be applications both at OEM and supplier plants.