's Forum Archives - General

Archive Home >> General(1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )

friction stir welding - any applications in bikes, yet?(12 posts)

friction stir welding - any applications in bikes, yet?DougSloan
Sep 11, 2003 9:25 AM
Very intersting; anyone know of bike applications, yet?

Friction Stir Welding Emerges as High-Performance Alternative to Fusion Welding

Technical innovations continue to mark the evolution of a highly effective solid-state welding process that has been limited by high costs of capital equipment.

By Mark W. Shortt

Experienced users of friction stir welding (FSW) invariably describe the process as a significant advancement in aluminum joining, capable of producing stronger, more durable joints with less distortion than conventional fusion welding. The technique uses the high rotational speed of a tool and the resulting frictional heat created from contact to crush, stir together, and forge a bond between two metal alloys. Since being patented in 1991 by The Welding Institute (TWI), Cambridge, England, friction stir welding has been the focus of considerable research and development that continues to produce innovations in machinery, tooling, and process development.

Many benefits of friction stir welding derive from its being a solid-state process, in which joining occurs without fusion (below the melting point) of the metal alloys to be welded. The absence of melting eliminates many of the problems normally associated with conventional metal welding, such as fumes, spatter, porosity, solidification cracking, and shrinkage. As a result, the process is uniquely suited to join materials, such as aluminum alloys used in aerospace work, that are difficult to fusion weld. These include the 2000 and 7000 series aluminum alloys and the 2195 aluminum-lithium alloy, used on the external fuel tank of the NASA Space Shuttle.

"The process joins materials that are generally considered unweldable, with high reliability and properties that approach the parent material," says Jack Thompson, chief engineer, General Tool Company (GTC), Cincinnati, Ohio. "Friction stir welding is penetrating critical applications for welded aluminum because it is an extremely robust, reliable process. If the process parameters are achieved, the weld will exhibit very predictable properties."

In conventional friction welding, heat is generated as the two surfaces to be joined are rubbed together. However, friction stir welding goes one step further by employing a third piece—a profiled, wear-resistant pin tool— to create added heat and friction in the weld joint. During FSW, the pin of a shouldered tool is slowly plunged into the joint between the two materials to be welded, and rotated at high speed. The resulting friction creates a plasticized shaft of material around the pin. As the pin moves forward in the joint, it "stirs," or crushes, the plasticized material to create a solid-phase weld with a fine-grain microstructure. The process requires extensive use of rigid clamping to prevent movement of the workpiece, as high pressures are needed to plasticize the metal. Friction stir welding uses a non-consumable tool that requires no filler wire or gas shielding for welding aluminum.

Aluminum alloys are not the only materials that can be joined by friction stir welding. Other low-temperature metals—such as copper, lead, and magnesium—are currently being welded, and the joining of higher-temperature alloys, such as steel, titanium, and Iconel® has also been demonstrated.

"Theoretically, almost any metal or thermoplastic can be friction stir welded," says Brent Christner, friction stir welding manager, Eclipse Aviation Corp., Albuquerque, New Mexico. "The key to the higher-temperature alloys is finding a pin tool material that retains stiffness, strength, fatigue, and wear resistance properties at the temperatures at which the materials being welded become plastic. Significant progress has been made in high-temperature tool materials over the last several years, and will soon be transitioned to production."

Stronger, Lighter Suspension Links

Last November, Tower Automotive Inc. (Grand Rapids, MI) announced that it had worked in partnership with Ford Motor Company to successfully develop a lightweight, replacement suspension link with the increased durability necessary to withstand required loads. According to Tower Automotive, its testing confirmed the ability of friction stir welding to reduce weight, lower costs, and increase joint efficiency and fatigue life over gas metal arc welding, the method traditionally used by the automotive industry. During the design validation and process validation phases of the FSW link development, the company conducted tensile, compression, and fatigue tests to compare the FSW link design to tubular steel and stamped steel link designs.

"The fatigue test is where the FSW link outperforms the other two," says Art Scafe, advanced new product development leader for body and suspension structures. "On average, the tubular steel link failed around three lives, or 450,000 cycles, and the stamped link didn't make one life cycle. The friction stir-welded link lasted one million cycles, and when the load was increased by 15%, it went another 200,000 cycles without failure."

According to Scafe, the greatest challenges to producing friction stir-welded suspension links were in refining a manufacturing process and building the machine. Tower Automotive worked with ESAB to develop what it calls "the first synchronized, two-sided machine for producing lightweight suspension links." The synchronized, two-sided machine reportedly permits faster welding of 12.7-mm material (welding over 20 mm/sec), produces a stronger, double-sided weld with no void, and is less susceptible to variation in extrusion tolerances.

"Tower Automotive saw the market shifting to low-weight suspension components and found a suitable approach to produce low-weight components and still maintain good tolerance control," says Scafe. "The friction stir-welded link is 43% lighter than a comparable steel tubular link."

Generally, the level of quality generated by a friction stir weld is superior to other types of welding, says Scafe, who adds that product design "needs to be created with FSW in mind." Until recently, however, the process had been hindered by two limiting factors: the requirement for pin tools of different lengths when welding materials of varying thickness; and the reliance on a single-piece pin tool that leaves a "keyhole" when retracted at the end of the weld.

RPT Adjusts to Variable Thickness, Prevents Keyholes

These limitations were overcome when Jeff Ding, a welding engineer at the NASA Marshall Space Flight Center (Huntsville, Ala.), developed an automatic, retractable pin tool (RPT) that uses a computer-controlled motor to automatically retract the pin into the shoulder of the tool at the end of the weld. By permitting automatic retraction and extension, the design allows the pin tool to adjust to tapered weld depth and eliminates keyholes in the weld joint. Both pin angle and length can be adjusted for changes in material thickness, making it possible to use friction stir welding on materials that change in thickness during the weld.

"These welds are typical on the Space Shuttle External Tank, where thickness will transition from 0.310 inch to 0.650 inch to 1.00 inch," says Ding. "The RPT also 'closes out' the keyhole left by the departing pin at the end of the weld. An example would be circumferential welds where the pin retracts after it passes the starting point of the weld. The RPT is also required for the self-reacting pin tool currently being developed for NASA manufacturing programs."

A simple, hand-crank RPT was originally designed and fabricated to demonstrate the concept, according to Ding. The prototype made use of a manually turned wheel to retract the pin; its success led to more robust RPTs, which currently weld up to 0.650-inch-thick material, he says. Already, the innovation has contributed to the development of customized FSW that has proven to provide routinely reliable welds.

"This feature allows for variable-thickness materials to be completely joined, as well as contoured surfaces with variable thickness, when used in machines with sufficient degrees of freedom to track and orient the tool to the surface," says Jack Thompson of General Tool Company. In the friction stir welding equipment that General Tool developed for NASA /Lockheed Martin to create longitudinal welds in the barrels of the Space Shuttle's external tank, GTC devised what Thompson calls a "significant improvement." The patented GTC system employs a fused quartz measuring rod, inserted down the center of the RPT pin, to measure the length of the RPT pin independently of external loads and thermal expansion.

"Fused quartz exhibits a coefficient of thermal expansion 1/30 of the pin material," says Thompson. "An LVDT, affixed to the RPT spindle housing, bears on the measuring rod string and lets the machine track the end of the pin to ± 0.001 inch, regardless of loads and expansion of the pin. This compensation system keeps track of the exact length of the pin as its length grows thermally about 0.025 inch, and the pin is compressed by welding loads by about 0.010 inch. Prior to implementing this system, these pin length errors were compensated by trial and error, open-loop offsets."

One of the first companies to commercialize NASA's patented auto-adjustable pin tool last year was MTS Systems Corporation (Eden Prairie, Minn.), a supplier of integrated, computer-based testing and simulation systems and products for automating manufacturing processes. According to MTS Business Development Manager, Mike Skinner, the automatic, retractable (auto-adjustable) pin tool has "made it possible to employ friction stir welding to achieve continuous welds (the ability to close the keyhole) on large components, using FSW equipment that features a relatively small work envelope." Skinner says that the innovation "improves the process by allowing FSW to be utilized on larger components without expensive upsizing of equipment."

MTS Systems has been awarded a patent for a Self-Reacting FSW technology, one of three FSW modes (the others are the fixed pin tool and adjustable pin tool modes) that can be implemented by means of the company's Adaptable Adjustable Pin Tool (AdAPTTM) System. Self-Reacting FSW is a key capability of a universal friction stir welding system that the company is building for a project of the National Center for Advanced Manufacturing (NCAM)-Louisiana Partnership, a research center focused on building the technology base for manufacturing next-generation launch vehicle systems. The project, made possible through cooperative funding from NASA and the State of Louisiana, will seek to advance the state of the art in complex curvature welding. Its objective is to demonstrate the ability to fabricate large-scale, complex contour aluminum panels using the Self-Reacting technology. For more on the technology, see Tech Updates ("Self-Reacting Process Employs Unique Weld Head Technology").

Thin-Gauge Lap Welds for Lightweight Jet

MTS Systems has also been working closely with Eclipse Aviation Corp. to develop, validate, and certify the friction stir welding process for lightweight materials used in the Eclipse 500 jet. The six-person, twin-engine jet is currently undergoing a 16-month testing program that is expected to culminate with FAA certification in December of this year. Reported to be the first company to use FSW in production on thin-gauge aircraft aluminum, Eclipse Aviation is using the process to lap weld stiffeners (stringers and frames) to the aircraft skins of the Eclipse 500. Friction stir welding is slated to replace rivets in most major assemblies of the aircraft, including the cabin, aft fuselage, wings, and engine mounts.

For this project, the company has had to develop the technology for application to thin-gauge lap welds. Its objectives include corrosion protection of the mating surfaces, control of distortion, material property characterization, and the development, with MTS Systems, of a friction stir welding system capable of welding complex contours. Eclipse received U.S. Federal Aviation Administration (FAA) approval of its friction stir welding process specification in May 2002, one year ahead of schedule.

At press time, some 9,488 inches of production welding had been completed for assembly of test/certification aircraft. Use of the technology has eliminated nearly 7,000 fasteners and associated hole drilling, and increased joint strength by up to three times, Christner says. It is said to have resulted in joining speeds four times faster than automated riveting and 20 times faster than manual riveting.

"On the alloys and thickness that Eclipse is using, tensile strength is up to three times greater than mechanical fastening," Christner says. "Fatigue life is comparable to or better than mechanical fastening. Testing of a full-size barrel simulating the aircraft fuselage showed fatigue life in excess of 23 aircraft lifetimes."

General Tool recently replaced a cast panel with a fabricated panel composed of extruded plate and extruded shapes, a substitution that was made practical by the use of FSW. While weld strength was never an issue due to the high quality of the welds, the lack of distortion ensured consistent flatness and straightness of the weldment, thereby eliminating significant post-weld machining. Thompson also credits the "fast and predictable" nature of the process.

"While the operator of the FSW work station needs to be careful and consistent in loading material into the weld fixture, once the start button is pushed, the process is not dependent on the technique and undivided attention of a skilled artisan," he explains. "This material and process substitution saved General Tool over $1 million in producing about 400 of these highly finished, 72-in x 89-in and 72-in x 59-in panels. These savings were realized after building a FSW machine, buying a process license, and developing a proven set of process parameters."

According to NASA's Jeff Ding, the cost savings associated with friction stir welding are a result of the reliability and repeatability of the process. "When the proper weld parameters are established for a particular thickness and a specific aluminum alloy, the process as compared to fusion welding produces fewer defects," says Ding. "This, in turn, reduces the costs of Material Review Board dispositions. (The Material Review Board evaluates specific defects and determines if a repair is required, and if so, how the repairs are to be completed.) Also, as important to the External Tank Program as the cost savings, is the reliability issue. The friction stir welding process improves hardware safety margins."

Tools, Materials, & Process Developments

The recent emergence of industrial robots as integral components of friction stir welding systems has produced another option for companies seeking to reduce costs of capital equipment, operations, or outsourcing. Greater flexibility and operating efficiencies, as well as major cost reductions, are among the advantages that robotic friction stir welding can bring to providers and users of friction stir welding services. For more on robotic friction stir welding, see the Shop Spotlight.

Ding says that a design concept for a hand-held friction stir welding device was recently introduced, and is undergoing further development. Additional work is focusing on pin tools: one redesigned retractable pin tool is said to be capable of welding and closing out material two inches thick. Other activity includes the development of a new process—thermal stir welding—that separates the primary elements of friction stir welding (heating, forging, and stirring) for independent control.

Significant advancements have also been made in the development of five-axis, gantry-style FSW machines and general-purpose FSW machinery, according to General Tool's Thompson. Currently in development are machines that can handle up to 35,000 pounds of thrust within a working envelope of 8 ft x 10 ft x 26 ft. The machines, which typically include retractable pin tool heads, can weld a general path in space, thereby enabling the welding of aircraft skin shapes. They also have tool-tilt servo axes, which allow the tool to be oriented to the work to meet path tangency and process control requirements.

Thompson says that the development of friction stir weld tool materials, such as tungsten-rhenium and polycrystalline cubic boron nitride (PCBN), allow the FSW process to "broaden into applications involving steel, nickel base alloys, and stainless steels." Although these tool materials are still in the research phase, the success demonstrated to date "indicates that the process will soon be employed all across the materials spectrum," he says.

Challenges to Overcome

Despite its ability to achieve strong and repeatable welds with low distortion and excellent mechanical properties, friction stir welding is far from reaching its enormous potential. According to Thompson, the process is somewhat limited by the scarcity of FSW machinery in industry today and by the significant tooling and fixturing required to rigidly hold, clamp, and back a workpiece. Also, without the benefit of referring to long-established experience and handbook guidance that is widely available for traditional welding processes, engineers need to do job-specific process development and qualification for friction stir welding.

However, Thompson says that FSW process modeling and experimental studies that reveal flow behavior around the tool have been correlated to the point that accurate computer predictions of the weld process are feasible. A successful analytical approach requires unique computer codes, test material properties, and specialized expertise, he says.

"The emerging understanding that comes from developing useful models points the way to the development of standardized methods and relevant material data that can remove or significantly reduce empiricism from process development," says Thompson. "Technical presentations in the past have often been based on finding a suitable set of process parameters and then characterizing properties of the welds in terms of tensile, ultimate, fatigue, crack growth, and corrosion. In the future, we would expect materials to be characterized in a way that would support a weld engineer's identifying a set of successful weld parameters for a material of particular thickness from material properties and a set of 'weld design equations.' "

Companies that use the friction stir welding process for any purpose, including R&D or production, are required to pay a licensing fee to The Welding Institute. NASA's Jeff Ding believes that friction stir welding will be unable to realize its full potential until the intellectual property rights expire and there is greater availability of lower-cost welding systems and tooling. "The TWI license is too expensive for the 'mom and pop' shops to secure a license," says Ding. "Investment costs are also high: tooling is expensive for the smaller operations to be able to afford it. These costs need to be reduced."

Some of the R&D work currently being done involves the application of the process to higher-temperature materials, thin-gauge lap welding, and process modeling. Further work is under way in thick-section welding, designing pin tools to increase welding speeds, and pre- and post-weld thermal treatments to improve corrosion resistance.

Tower Automotive has researched the use of friction stir welding to produce tailor-welded blanks, and is currently evaluating the potential for using the process to spot-weld aluminum components. "Friction stir welding applications are limitless, so growth potential is enormous," says Scafe. "Things that were not viewed as a potential application are being considered today."
Thermoplastics: 25 years ago, a child's toy...Spunout
Sep 11, 2003 9:33 AM
I had a construction set with plastic girders. It came with a sortof 'dremel' tool that rotates a plastic rod that would create friction, melt the rod (filler)) and weld plastic girders into place.

Anyone else remember these?

It was the same concept. Neat!

I think the problem will also occur with very thin bicycle tubing and the need to deposit alot of material to make the weld.
At Va Tech, we used it to seal plastic vials for ...Humma Hah
Sep 11, 2003 10:19 AM
... carrying specimens into a reactor for neutron activation analysis. They modified a drill press to spin the vial caps on, and it worked great. You don't want stuff leaking into your nuke.

I later had some sample welds made to seal thermal energy storage containers for a solar power project. Worked good, but the project was unworkable for other reasons.

I've never tried it with metals. It could work, if someone takes the time to perfect it. But I'd not like to buy one of the first production run ... I'd bet on process control problems.
forget itdoped
Sep 11, 2003 10:31 AM
the last thing we need is more automation. skilled tradesmen (tig welders etc...) are where it's at. people need jobs.
forget Tig ...Humma Hah
Sep 11, 2003 10:39 AM
... guys with torches brazing lugs. THAT'S what we need! ;-)
forget Tig ...doped
Sep 11, 2003 10:46 AM
tig, brazing it don't matter. hands. skilled people. jobs.
and bin laden. that's what we need.
why stop there?DougSloan
Sep 11, 2003 11:01 AM
Why not step back to carving and bending wood, if you want real craftsmanshop...? ;-)

Can you chrome-plate wood?Humma Hah
Sep 11, 2003 12:36 PM
For the 100th anniversary of the first motorcycle, a Dalmier-Benz that made a hobby-horse look lightweight, somebody build a reproduction.

I think a reproduction wooden bicycle would give us all something to be greatful for in modern bikes.
We have used this and similar technologyboyd2
Sep 11, 2003 1:30 PM
in the Aerospace buis. It is great stuff. I think that the problems using it on bikes would be the interface pressure is very high. The tubes may not survive the axial pressure. The other big problem is that one of the parts must spin and that part must be axi-symetric. So you could only weld one end of the tube and you would have to do something else (Tig) on the other end. Problem 3 this method produces alot of flash and that would have to be machined away more labor + additional process + additional machines = lots more money.

On the plus side you could weld dissimilar metals. You could have an Inconel rear triangle with an aluminum main triangle. That would be real cool!
I was thinking about that ...Humma Hah
Sep 11, 2003 3:19 PM
... it will only work for one end of the tube, so you can't close a triangle with it. Schwinn faced that problem with the "electroforged" welds on my cruiser and other classic Chicago Schwinns. They upset-welded the top and down tubes onto the head tube, which was itself upset welded from two forged halves. It made an incredibly strong joint, and also quite beautifully radiused, but they couldn't do the same thing at the seat tube and BB.

I think they decided it counted the most at the head tube.

Electroforging also required hand-dressing to remove the flash. It is not a feature of later-day imported Schwinns.

Inconel. The last time I presented that to a machinist and asked him to machine it, he chased me out of the shop with a large wrench. I once tested a pair of 7/8" bolts made of it, and one of them failed to break at the full capacity of a 120,000 lb Baldwin test machine. Those bolts were being tested for a space application.

Now, inconel and an aluminum/lithium alloy (such as the latest design for the space shuttle external tank), THEN you got some rocket science in your bike!
Machining Inconelboyd2
Sep 12, 2003 4:08 AM
I have designed and built lots of stuff out of inconel here. The secret to machining it seems to be use no lubricant and go really, really slow. Go figure that one out! I don't know why I am writing this here, just board this morning I guess.
It will most certainly replace rivetingbigrider
Sep 12, 2003 5:51 AM
One of the beauties of the process is the ability to stir weld wery thin pieces of material. The welds are perfect and strong and remain flush to the material so in essence it is a weld that looks like it is glued rather than welded. The only sign is the swirls in the metal caused by the stirring. The airplane industry will most certainly apply this technology and I would venture to say it will be widespread. I think the automation and small production numbers will prohibit the near term use of this technology in the bike industry.

You probably wouldn't gain anything except a cleaner looking line on the weld and a little less weight from eliminating the extra material needed on the welds. Not a big payoff for the costs.