FSW Flash Formation in Aluminum: Causes, Effects, Prevention
Table of Contents
- The Mechanics of FSW Flash Formation
- Primary Causes of FSW Flash
- Excessive Shoulder Plunge Depth
- Heat Input from Rotation Speed and Travel Speed
- Joint Fit-Up and Fixture Deflection
- Effects of Flash on Weld Quality and Cycle Cost
- Parameter and Tool Decisions That Reduce FSW Flash Formation
- Process Adjustments to Try First
- Tool Shoulder and Material Flow
- Machine Force Control and Fixturing
- What to Confirm Before Accepting Flash in Your Program
- Common Questions About FSW Flash Prevention
- Is flash always a welding defect?
- Does a wide flash band mean the weld is too shallow?
- Why does flash change during a production shift?
- What is the fastest correction to try on an aluminum line?
FSW flash formation is one of the earliest signs that a friction stir weld is running out of balance. A small raised ridge of displaced aluminum along the shoulder edge may look cosmetic, but it signals incorrect heat input, shoulder contact, or axial force. Once flash appears, section thinning, surface defects, and inconsistent penetration often follow. We break down the physical causes, the practical effects on aluminum components, and the prevention measures used in process development at Aerospace Engineering Equipment (Suzhou) Co., Ltd. The aim is to turn flash into a process variable rather than a surface blemish.
The Mechanics of FSW Flash Formation
Flash is displaced plasticized material that escapes from the weld zone at the shoulder edge. In FSW, the rotating shoulder performs two jobs at once: it seals the top of the joint and forges softened material into the weld nugget. When the volume of softened material under the shoulder grows faster than the cavity can contain, aluminum is pushed outward and upward. The result is the fine fin or curled ridge that appears along the weld toe. That ridge is not random. Its location and shape tell us whether the leading edge, trailing edge, or root side is losing containment.

In our process development work, we measure flash width as a routine response variable because a stable flash width usually means stable shoulder engagement. Thin sheet below about 2 mm flashes quickly because there is little parent material to absorb excess heat. Thick plate can also flash when the surface layer becomes much hotter than the material at depth, which is why flash on thick aluminum often appears even when root consolidation is still acceptable. This is one reason we separate surface flash from volumetric defects in quality records.
Primary Causes of FSW Flash
Four conditions create most FSW flash cases in aluminum production: excessive shoulder plunge, high heat input, poor joint fit-up, and worn or unsuitable tool shoulders.
Excessive Shoulder Plunge Depth
Shoulder plunge depth is the first setting we check. For many 6xxx aluminum butt welds, plunge is set between 0.1 mm and 0.3 mm below the original plate surface, depending on thickness and tool geometry. A plunge that is too deep increases forging volume and forces softened metal out at the shoulder rim. The correction is to reduce plunge in small increments of about 0.05 mm and watch flash width shrink. We usually do this before changing rotation speed because it separates containment from heat input.
Heat Input from Rotation Speed and Travel Speed
High rotation speed or low travel speed raises the heat input per unit weld length. The material becomes too fluid, escapes more easily, and leaves a wider flash band. On 6xxx alloys, lowering rotation speed by 100 rpm to 200 rpm, or raising travel speed within the machine’s force envelope, often reduces flash without harming penetration. The tradeoff is that heat input cannot drop so far that root defects appear. We adjust both settings together and verify with a cross-section before transfer to production.
Joint Fit-Up and Fixture Deflection
Gaps, mismatched edges, and weak clamping change the effective plunge and shift material sideways. A butt joint with a 0.2 mm gap may produce flash on one edge while the other edge remains clean. In long aluminum profiles, fixture deflection can create the same pattern in the middle of the weld where local pressure is lowest. The correction is usually not a parameter change; it is fixture reinforcement, better edge preparation, or a backing bar that resists local bending.
Effects of Flash on Weld Quality and Cycle Cost
Flash is not always a rejectable defect by itself. Small continuous flash may be accepted on noncosmetic surfaces. The problem is that flash removes material from the joint cross-section and creates a sharp notch at the weld toe. In thin-wall parts, this notch can reduce fatigue life and act as a stress raiser. In sealed battery trays, cooling plates, and electronic housings, loose flash can become foreign object debris inside a sealed cavity, which is why these applications specify low or zero flash weld surfaces.
| Flash condition | What it indicates | Production consequence | Recommended response |
|---|---|---|---|
| Continuous thin flash on both edges | Moderate heat and stable process | Cosmetic cleanup | Record as acceptance baseline and monitor width |
| One-sided flash | Edge mismatch or local clamp loss | Uneven thinning and possible root shift | Check fit-up, clamping, and backing |
| Wide curled flash | Excess plunge or high heat | Section loss and debris risk | Reduce plunge or rotation speed, or raise travel speed |
| Intermittent flash | Worn tool or inconsistent force | Variable surface profile | Inspect shoulder wear and verify force control |
Flash also adds cost even when it is acceptable. Wire brushing, scraping, and surface dressing increase labor after welding. On high-volume lines, a 0.5 mm flash allowance can turn into an extra secondary operation. That cost is why we treat flash control as a machine-level specification, not an operator cleanup task.
If your program involves long seam welds on thin-wall battery trays, cooling plates, or sealed enclosures, confirm the specified flash width and section thinning before qualifying the tool. Send your alloy, joint thickness, and drawing limits to [email protected], and we will check which shoulder geometry and welding window keep flash inside your specification.
Parameter and Tool Decisions That Reduce FSW Flash Formation
Flash prevention is best handled as a sequence. Set the process to match the tool, then match the tool to the machine’s force control capability, and only then make cosmetic acceptance decisions.
Process Adjustments to Try First
Start with a 0.05 mm plunge reduction and measure the response. If flash remains wide, lower rotation speed or raise travel speed while watching the root side for signs of incomplete penetration. Conventional FSW tools on aluminum are often run with a tilt of about 1.5 to 2.5 degrees, and excessive tilt can create trailing-edge flash. Make small changes because heat input and containment are coupled.
Tool Shoulder and Material Flow
The shoulder design determines how well the softened layer is contained. Concave and scrolled shoulders direct material inward, while a worn or flat shoulder can let it escape. For thin-wall parts and applications that cannot accept flash, a stationary shoulder tool separates pin rotation from shoulder sliding, removing the rotating shoulder edge that produces most flash. Our stationary shoulder series covers aluminum from 1.0 mm to 110.0 mm in thickness and is specified for no flash and no thinning.

Flash control is easiest to lock in while tooling is being specified, not after a process window has already been qualified. <AEE provides customer with optional features FSW tool> covers how optional tool features are matched to a program’s alloy, joint, and thickness before the machine configuration is frozen.
Machine Force Control and Fixturing
The control mode changes how flash develops. In position control, a fixed plunge depth can overload the spindle and increase flash as the plate grows thermally. In force control, the machine holds a set axial force, but the plunge may drift and widen flash if the force envelope is too loose. The practical setup is a controlled force range with a defined plunge limit. Spindle selection matters in the same way: the machine must deliver enough axial force for the section without forcing the shoulder deeper as thermal conditions change.

What to Confirm Before Accepting Flash in Your Program
Accepting flash without a measured limit creates downstream risk because flash width is a leading indicator of heat input and shoulder containment. Before locking a welding procedure, confirm three points on your actual parts: the maximum flash width shown on the drawing, the minimum remaining cross-section after any dressing, and the parameter window that keeps flash repeatable across a full shift. If you are working toward a qualification or ramping a new production line, send your part number, alloy, thickness, and current weld photos to [email protected] or call +86 18325808715. We will check the tool shoulder geometry, machine force control setup, and process window together with you so flash is controlled before the first production lot, not after.
Common Questions About FSW Flash Prevention
Is flash always a welding defect?
Not always. A thin, continuous flash line can be acceptable on noncosmetic surfaces when the remaining cross-section stays inside tolerance and the part does not require a clean weld toe. Flash becomes a defect when it changes section profile, breaks loose as debris, or masks surface cracks. For sealed enclosures, battery trays, and thin-wall aluminum components, even small flash is normally treated as a defect because the notch effect and contamination risk are unacceptable. Acceptance cannot be a visual guess. It should be tied to a measured flash width and a defined location on the joint.
Does a wide flash band mean the weld is too shallow?
A common assumption is that a large flash band means the weld has too little penetration. That is usually backwards on thick aluminum. Good root consolidation and heavy surface flash can appear together because the top layer becomes much softer than the deeper material. The shoulder then expels the overheated surface layer while the pin still stirs the full section. Before changing travel speed or force, cut a cross-section to check nugget shape and root fusion. If the nugget is full and flash is wide, reduce surface heat input instead of forcing the tool deeper.
Why does flash change during a production shift?
The answer depends on how the machine controls the tool. In position control, a fixed plunge depth can overload the spindle and increase flash when the plate grows thermally. In force control, the machine holds a set axial force, but the plunge may drift and widen flash if the force envelope is too loose. The correct setup is a controlled force range with a defined plunge limit. If flash changes during a shift, log force, plunge, and spindle torque together; one value will move first and point to the source.
What is the fastest correction to try on an aluminum line?
In the process development work we run at AEE, the most common correction starts with a 0.05 mm plunge reduction, not a large speed change. Small increments let us separate containment from heat input quickly. If flash remains one-sided, check clamping and backing before changing the recipe because a mechanical problem produces a mechanical-looking flash pattern. If your program has sealed aluminum seams or visible flash limits on the drawing, send your joint design and weld photos to [email protected] and we will confirm which shoulder geometry and process window fit your part.
If you’re interested, check out these related articles:
Refill Friction Stir Spot Welding (RFSSW)
AEE provides customer with optional features FSW tool