01 Overview
Fatigue is the slow accumulation of damage in a part that is loaded and unloaded many times without ever being overloaded once. A frame that survives a single hard landing or a single pothole proves almost nothing about its fatigue life. What matters is the stress range at each cycle, the number of cycles, and the local geometry where cracks start. Steel and aluminum behave very differently under that regime, and the difference is not captured by a single strength number on a data sheet.
Aluminum has no true endurance limit. Its fatigue curve keeps descending as cycle count rises, so a designer must pick a finite life target and design to it. Steel, by contrast, exhibits an endurance limit below which well-made parts can survive an essentially unlimited number of cycles. For a frame that will see millions of small load reversals, that plateau is the whole argument. Steel also has a higher stiffness-to-cost ratio in common structural sections, and it fails gradually rather than suddenly when overloaded.
This article treats the frame as a fatigue structure first and a shape second. The goal is to explain how material choice, joint design, machining practice, and inspection interact, and why a heavier-looking steel frame can be the more durable engineering decision.
02 Designing for the process
Fatigue failures almost always begin at a stress concentration: a weld toe, a sharp corner, a hole edge, or a sudden change in section. The designer's first job is to make those transitions gentle before any metal is cut. In steel frames, that means gussets with generous radii, tube ends that are notched to fit rather than crushed to fit, and weld beads placed away from the highest bending moment. A smooth load path costs a little weight and buys a large increase in cycles to failure.
Process choice then constrains the design. Welding introduces residual stress and a heat-affected zone with altered microstructure. Brazing runs cooler and can be gentler on thin tubing, but it demands tighter joint clearance and cleaner surfaces. Bolted and bonded joints avoid the heat problem but add holes, and every hole is a stress riser. The designer must decide early which joints are welded, which are mechanical, and where post-weld machining will restore a critical interface.
Good practice is to design the fixture at the same time as the frame. A rigid fixture holds tube alignment during thermal distortion, which keeps the frame straight enough that later machining removes a minimal amount of material. Removing less material preserves wall thickness and keeps the fatigue-critical section intact. Safety note: any cutting, notching, or grinding of tubing must be done with the workpiece clamped, with eye and hearing protection, and with guards in place, because thin-wall tube grabs tools aggressively.
03 Material and tolerance decisions
Low-carbon steel is inexpensive, weldable, and forgiving, but its fatigue strength is modest. Chromoly and other low-alloy steels respond to normalizing or stress relief and offer a higher endurance limit for a given section. The trade is that they are less tolerant of poor welding practice and may need preheat or controlled cooling. Aluminum alloys used in frames are typically heat-treatable, and the heat of welding can locally destroy the temper unless the whole assembly is re-aged or the design accepts a reduced strength in the heat-affected zone.
Tolerances should follow function, not habit. Tube-to-tube fit-up around half a millimeter keeps gaps small enough for sound welds without forcing the fitter to distort the frame. Machined bearing bores and pivot faces deserve tighter control, on the order of a quarter millimeter or better, because misalignment there loads bearings unevenly and shortens their life. Over-tightening tolerances everywhere raises cost and scrap without improving fatigue performance.
Wall thickness and surface condition matter as much as alloy. A scratch or a coarse grinding mark on a tension surface can nucleate a crack. Specify a reasonable as-welded finish and reserve fine finishes for machined seats. If weight is critical, it is usually better to reduce section size in low-stress regions than to thin the walls at joints, where fatigue damage concentrates.
04 The machining operation
Frame machining is mostly about preparing joints and restoring interfaces after thermal distortion. Tubes are cut to length, then notched or fish-mouthed so they meet cleanly. A good notch follows the mating tube's curvature closely, leaving a consistent gap for the weld or braze. Milling or abrasive notching both work, but the setup must hold the tube without crushing it, which means soft jaws or custom cradles rather than a bare vise.
After welding, the frame is often fixtured again and the critical faces are machined. Bearing bores are bored or reamed to size, pivot faces are faced square to the bore axis, and dropout slots are cut parallel. These operations remove the distortion that welding introduced, so they must come after any stress-relief step. If the frame is machined first and stress-relieved later, the relief can move the very features that were just brought into tolerance.
Speeds and feeds should favor control over speed. Light cuts with sharp tooling reduce cutting forces and heat, which limits distortion in thin-wall tubing. Coolant or cutting fluid helps control temperature and improves finish. Safety note: never reach into a machine to clear chips while the spindle is turning, always use a brush, and confirm that fixtures are torqued before starting the cycle. A frame that shifts mid-cut can become a projectile.
05 Measurement and quality control
Inspection of a fatigue-critical frame has two halves: geometry and soundness. Geometry checks confirm that bores, faces, and mounting points sit where the design intends. A coordinate measuring approach or a dedicated checking fixture can verify bore position, axis alignment, and face squareness. Simple go and no-go gauges are often enough for production, provided they are checked against a known standard regularly.
Soundness checks look for the defects that become crack starters. Weld toes are inspected visually and, where the consequence of failure is high, with dye penetrant or magnetic particle methods. Magnetic particle inspection works only on ferromagnetic materials, so it suits steel frames and not aluminum ones; aluminum welds are usually examined visually or with penetrant. Any undercut, porosity, or crack indication at a toe should be treated as a reject or a rework item, not blended away casually.
Records matter. Documenting heat numbers, stress-relief cycles, and inspection results lets a later failure be traced to a process change. Calibration of gauges and inspection equipment should be on a schedule, and the person inspecting should not be the person who just finished welding the joint. Independent checks catch the errors that familiarity hides.
06 Troubleshooting bad parts
When a frame cracks in service, the location tells most of the story. Cracks at a weld toe usually mean a stress concentration combined with residual tension, and the fix is better joint geometry, a smoother bead, or post-weld treatment such as stress relief or peening. Cracks starting at a hole or a sharp internal corner point to a design detail that needs a larger radius or a reinforcement. Cracks in the middle of a tube, away from joints, suggest that the section is simply too thin for the load or that the surface was damaged during handling.
Distortion problems show up as parts that will not assemble or bearings that bind. If bores are out of round or faces are not square, the likely causes are welding sequence, insufficient fixturing, or machining before stress relief. Re-machining can recover some parts, but only if enough material remains; otherwise the frame is scrap. It is cheaper to fix the process than to chase individual parts.
Poor weld appearance is not only cosmetic. Undercut reduces the effective section at the toe, and excessive reinforcement creates a sharp transition. Both raise stress. Rework should restore a smooth profile without grinding below the parent metal thickness. Any rework that removes material from a tension surface should be evaluated for fatigue impact, and the part should be re-inspected afterward.
07 Frequently asked questions
Is steel always better than aluminum for fatigue? No. Steel offers an endurance limit and generally tolerates millions of small cycles well, but a well-designed aluminum frame with generous radii, good welds, and a defined life target can be entirely adequate. The right choice depends on the load spectrum, the required life, and the weight budget.
Does a stronger alloy automatically give a longer fatigue life? Not by itself. Fatigue life is governed by stress range and local geometry more than by ultimate strength. A high-strength alloy with a sharp weld toe can fail sooner than a mild steel joint with a smooth transition. Surface condition and residual stress often dominate.
Should I stress-relieve before or after machining? Stress relief should come before final machining of critical interfaces, so that the dimensions set by machining are not disturbed afterward. If the frame is machined first, a later thermal cycle can move bores and faces out of tolerance.
How tight should frame tolerances be? Tight enough to keep bearings and mating parts loaded correctly, and no tighter. Fit-up gaps around half a millimeter and machined bores within a quarter millimeter cover most frames. Tighter tolerances add cost without adding fatigue life.
Can I inspect my own welds? Visual inspection is always worthwhile, but penetrant or magnetic particle testing should be done by someone trained and independent of the welding. Safety note: always wear appropriate eye, hand, and respiratory protection when handling penetrants and developers, and follow the manufacturer's ventilation instructions.