Friction stir welding (FSW) is a method of joining metals without melting them. A rotating tool plunges into the seam between two pieces of metal and moves along the joint, generating enough frictional heat to soften the material and physically stir the two sides together. Invented by The Welding Institute (TWI) in 1991, it has become a go-to technique for industries that need strong, defect-free joints in aluminum and other lightweight alloys.
How the Process Works
The core of FSW is a specially designed tool with two parts: a broad, flat shoulder and a smaller protruding pin (sometimes called a probe). The tool spins at high speed and presses into the joint line between two metal workpieces. As it moves forward along the seam, friction between the tool and the metal generates intense heat. The shoulder alone produces roughly 70% of that heat, while the pin generates the rest. This heat softens the metal into a plasticized, clay-like state without ever reaching its melting point.
The spinning pin then stirs the softened material from one side of the joint into the other, mixing them together at a molecular level. Behind the tool, the stirred material cools and consolidates into a solid joint. Because the metal never melts, FSW avoids the porosity, cracking, and distortion problems common in conventional welding. The shoulder’s diameter and downward pressure largely determine how the softened material flows, making tool design one of the most important variables in producing a quality weld.
What Makes It Different From Traditional Welding
Traditional methods like MIG and TIG welding are fusion processes. They melt the base metal and often add filler material to create a joint. That melting introduces a range of potential problems: gas pockets (porosity), incomplete fusion between layers, and large zones of heat damage on either side of the weld. In a comparative study of aluminum alloy 5083, radiographic inspection of friction stir welds revealed defect-free joints, while TIG and MIG welds showed internal flaws including lack of fusion, lack of penetration, and porosity.
The performance gap shows up clearly in the numbers. FSW joints in that same aluminum alloy reached 94.1% welding efficiency, meaning the joint retained nearly all the strength of the original unwelded metal. MIG joints reached 82.2%, and TIG joints managed just 70.1%. Hardness improved too. FSW joints were 46 to 50% harder than the base material, compared to 31 to 35% for TIG and 24 to 29% for MIG. The reason comes down to what happens inside the metal: FSW refines the grain structure and introduces strain hardening in the stir zone, while fusion welding’s high heat input degrades the surrounding material.
Zones Inside a Friction Stir Weld
A cross-section of a friction stir weld reveals three distinct zones, each with different properties. Understanding these zones helps explain why FSW joints behave the way they do.
The stir zone (also called the nugget) sits at the center of the joint where the pin passed through. Here, the combination of heat and intense mechanical deformation creates a refined grain structure through a process called dynamic recrystallization. Grains in this zone can be as small as about 5 micrometers, far finer than the original metal. The result is a dense, strong core.
Surrounding the nugget is the thermomechanically affected zone (TMAZ). This region experienced both heat and some mechanical deformation from the tool, but not enough to fully recrystallize the grain structure. It typically shows a drop in hardness compared to the nugget because the heat dissolves some of the strengthening particles within the alloy.
Beyond the TMAZ lies the heat-affected zone (HAZ), where the metal felt heat from the process but no mechanical stirring. The elevated temperature causes the alloy’s hardening particles to coarsen and weaken. In tensile tests, fractures most often occur in the TMAZ or HAZ rather than in the nugget itself, which tells you the stir zone is consistently the strongest part of the weld.
Which Metals Can Be Friction Stir Welded
Aluminum alloys are by far the most common FSW material, and the process is especially valuable for high-strength aerospace-grade alloys (like the 2000 and 7000 series) that are notoriously difficult to fusion weld without cracking. The international standard ISO 25239, now in its second edition, specifically covers FSW of aluminum and its alloys.
Copper is another strong candidate, particularly when it needs to be joined to aluminum. In electrical and refrigeration industries, aluminum-to-copper joints are essential, and fusion welding these two metals creates brittle compounds at the interface that weaken the bond. FSW’s lower temperatures and vigorous stirring action reduce this problem significantly, though controlling the formation of those brittle intermetallic layers remains an active challenge. Successful joints have been demonstrated across a range of alloy combinations, from pure copper joined to 1000-series aluminum up to 5083 aluminum joined to high-purity copper alloys.
Steel, titanium, and magnesium alloys can also be friction stir welded, though the higher temperatures needed to plasticize steel and titanium place extreme demands on the tool material, which must resist wear at temperatures that would soften conventional tool steel.
Where FSW Is Used in Industry
Aerospace was the first industry to adopt FSW at scale. The fuel tanks for Delta II and Delta IV rockets were among the earliest major applications, replacing fusion welding to produce lighter, stronger joints. NASA later used FSW to manufacture a 39-meter-long liquid hydrogen tank for its Space Launch System, relying on a purpose-built facility standing 52 meters tall. India’s space agency, ISRO, launched a rocket in 2018 that was the first to fly with propellant tanks built entirely with FSW, citing improved productivity and payload capacity.
In aviation, the Eclipse 500 business jet was one of the first aircraft to use FSW extensively, applying it to upper and lower wing skins, cabin panels, cockpit skins, the engine beam, and aft fuselage sections. One of FSW’s biggest advantages in airframe construction is that it can join stringers (stiffening ribs) directly to skin panels, eliminating thousands of rivets and the overlapping material those rivets require. That translates directly into weight savings.
The automotive sector uses FSW for battery trays and structural components in electric vehicles, where long, straight aluminum joints need to be airtight and distortion-free. Shipbuilding is another natural fit: studies on aluminum alloy 5083, a marine-grade alloy, have shown that FSW produces joints with higher quality and efficiency than MIG or TIG welding for hull panels and superstructures.
Common Defects and How They Form
FSW is not immune to defects, but the types of flaws differ from those in fusion welding. The most common problems are flash (excess material squeezed out from under the shoulder), voids, wormholes (tunnel-like cavities running along the weld), and joint line remnant, where traces of the original interface remain visible because the two sides weren’t fully stirred together.
Nearly all of these trace back to an imbalance between heat input and material flow. Too much heat, often from spinning the tool too fast relative to its travel speed, causes excessive softening and flash. Too little heat, from spinning too slowly or traveling too fast, leaves the material too stiff to flow properly, resulting in voids and tunnel defects. Wormholes and voids tend to form on the advancing side of the weld (the side where the tool’s rotation direction matches its travel direction) when the travel speed is too high. Kissing bonds and joint line remnants appear when the tool rotates slowly and advances quickly, leaving insufficient stirring to break up the original joint surface.
Getting the parameters right is the key to avoiding these problems. For aluminum alloys, typical industrial settings use rotation speeds around 1,200 rpm and feed rates in the range of 15 to 21 mm per minute, along with a controlled downward force of several kilonewtons. Optimizing these three variables together can push weld reliability above 0.87 on a 0-to-1 scale, according to predictive modeling studies on aluminum alloys.
Key Process Parameters
Three main variables control the quality of a friction stir weld. Spindle speed (how fast the tool rotates) determines how much frictional heat is generated. Feed rate (how fast the tool moves along the joint) controls how long the material is exposed to that heat and stirring action. Axial load (the downward force pressing the tool into the workpiece) keeps the shoulder in firm contact with the surface and ensures plasticized material doesn’t escape.
These parameters interact with each other. A higher spindle speed paired with a slow feed rate creates a “hot” weld with a wide heat-affected zone and risk of flash. A fast feed rate with low rotation produces a “cold” weld prone to voids. The tool geometry adds another layer of complexity: the shoulder diameter determines the heat footprint, while the pin shape (threaded, tapered, or fluted) influences how aggressively the material is stirred. Finding the right combination for a given alloy and thickness is part engineering, part empirical testing, though predictive models using neural networks are increasingly used to narrow the window before any metal is cut.