Every material changes size when its temperature changes. This isn't a minor detail — it's the reason bridges have expansion joints, railway tracks buckle in heat waves, and your plumbing uses expansion loops. As a software engineer who may spec hardware enclosures or work with sensor assemblies, understanding thermal expansion prevents costly design failures.

The governing equation is simple:

ΔL = α × L₀ × ΔT

• ΔL = change in length
• α (alpha) = coefficient of linear thermal expansion (units: 1/°C or 1/°F)
• L₀ = original length
• ΔT = change in temperature

Common α values to memorize:

• Steel: ~12 × 10⁻⁶ /°C
• Aluminum: ~23 × 10⁻⁶ /°C (nearly double steel)
• Copper: ~17 × 10⁻⁶ /°C
• Concrete: ~12 × 10⁻⁶ /°C (close to steel — this is why rebar works)
• Invar (nickel-iron alloy): ~1.2 × 10⁻⁶ /°C (engineered for minimal expansion)

Concrete example: You're designing a 2-meter aluminum rail for a server rack in a shipping container that experiences temperatures from -10°C to 50°C (ΔT = 60°C).

ΔL = 23 × 10⁻⁶ × 2.0 × 60 = 0.00276 m ≈ 2.8 mm

That's nearly 3 mm of growth. If both ends are rigidly bolted, the rail will bow, warp, or crack its mounts. The fix: use slotted holes on one end to allow the rail to slide freely.

Why this matters in mixed-material assemblies: When you bolt aluminum to steel, they expand at different rates. A 1-meter joint experiencing a 40°C swing creates a differential of (23 - 12) × 10⁻⁶ × 1.0 × 40 = 0.44 mm of shear at the interface. This is why dissimilar-metal joints use oversize holes, shoulder bolts, or flexible gaskets.

Practical rules of thumb:

• Aluminum moves roughly twice as much as steel for the same temperature change.
• For steel, a quick estimate: ~1 mm per meter per 80°C rise.
• Always check thermal expansion when a part spans more than 0.5 meters and sees temperature swings greater than 30°C.
• Concrete and steel expanding together is not a coincidence — it's the fundamental reason reinforced concrete works without tearing itself apart.

Where software engineers hit this: Outdoor enclosures, rack-mounted hardware, PCB mounting standoffs (boards expand differently than metal chassis), and any IoT deployment exposed to ambient temperature cycles. If your sensor readings drift with temperature, mechanical expansion of the mounting structure may be the cause, not the sensor itself.