Why That Small Notch Just Killed Your Metal: The Hidden Danger of Stress Concentration

Concept Diagram

Concept Diagram

You're designing a flat steel plate. You calculate the nominal stress at 120 MPa—well below yield (250 MPa). You approve it.

Then it cracks. At 120 MPa. In the worst possible place: near a hole.

This wasn't a calculation error. This was stress concentration—the phenomenon where geometry creates local stress peaks that dwarf the nominal stress. And it's the reason why sharp corners kill designs.

The Innocent-Looking Problem

A flat plate under tension looks simple. Apply 100 kN, divide by area, get stress.

σ_nominal = P / A = 100,000 N / (50 mm × 10 mm) = 200 MPa

But drill a small hole in the middle? Now the stress around the hole isn't 200 MPa anymore. It's 600 MPa.

A hole that occupies 2% of the cross-sectional area can triple the local stress.

This is stress concentration. And it's invisible in basic calculations.

---

The Stress Concentration Factor: Kₜ

The relationship is simple but devastating:

$\sigma_{max} = K_t \times \sigma_{nominal}$

Where Kₜ (the stress concentration factor) depends on:

• Geometry (hole size, fillet radius, notch depth)
• Load type (tension, bending, torsion)
• Material (doesn't matter; Kₜ is geometry-only)

Notice: Material strength doesn't appear. A steel plate and an aluminum plate, identical geometry, see the same local stress spike. The steel just happens to tolerate it better.

---

Real-World Example: Stress Concentration in Tensile Bars

Given:

• Flat steel bar, 50 mm wide, 10 mm thick
• Circular hole, 10 mm diameter (20% of width)
• Material: steel (Fy = 250 MPa, E = 200 GPa)
• Applied load: 100 kN

Step 1: Calculate nominal stress $\sigma_{nominal} = \frac{P}{A_{net}} = \frac{100,000}{(50-10) \times 10} = 250 \text{ MPa}$

Wait—that's already at yield! But it's the nominal stress at the weakest point (net section).

Step 2: Calculate stress at the hole's edge (stress concentration)

From published stress concentration curves (Peterson's), for a circular hole in a wide plate: $K_t = 2.4 \text{ (for this geometry)}$

Actual maximum stress at hole edge: $\sigma_{max} = K_t \times \sigma_{nominal} = 2.4 \times 250 = 600 \text{ MPa}$

The trap: You calculated 250 MPa and thought you were safe. The actual local stress is 600 MPa—2.4× higher—and 240% above yield.

The bar will crack.

---

Why Kₜ Matters More Than You Think

Most engineers treat stress concentration as a "detail to worry about later." It's not. It's a first-order design driver.

Why?

1. Stress concentration is local, but failure is permanent. A crack initiates at the stress concentration. Once it starts, it propagates until failure.

2. High-strength materials are more sensitive. A 250 MPa steel has some ductility and can redistribute load locally. A 1000 MPa steel is brittle—it cracks instantly at the stress concentration.

3. Fatigue makes it catastrophic. In cyclic loading, even stresses below yield cause fatigue cracks at stress concentrations. This is why bolts fail, shafts break, and welds crack.

---

Stress Concentration Factors: Where to Find Them

You don't calculate Kₜ; you look it up. Generations of engineers have measured these in labs.

Go-to references:

• Peterson's Stress Concentration Factors — The bible. Every geometry you'll encounter.
• AISC Design Guide 24 — Steel connections and details.
• Machinery's Handbook — Quick reference, surprisingly useful.
• FEA software — Use to verify, not to replace lookup tables.

Common Kₜ values:

• Circular hole in wide plate under tension: 2.4–2.7
• Rectangular notch (sharp, r ≈ 0): 3.0+
• Smooth fillet (r = 0.1d): 1.5–2.0
• Keyway in round shaft: 2.0–3.0

---

The Five Rules of Stress Concentration Design

1. Identify Stress Risers Early

Anything that breaks a smooth flow is a candidate:

• Holes, slots, notches
• Sudden diameter changes
• Fillets and corners
• Welds, rivets, fasteners
• Thread roots

2. Use Smooth Geometry

A fillet's radius is your friend. The larger the fillet radius relative to section depth, the lower Kₜ.

Sharp corner (r ≈ 0)        → Kₜ ≈ 3.0
Small fillet (r = 0.1d)     → Kₜ ≈ 1.8
Large fillet (r = 0.25d)    → Kₜ ≈ 1.4

Adding a fillet can cut stress concentration in half.

3. Apply Stress Concentration Factors to Your Nominal Stress

Don't ignore it in the math:

$\sigma_{design} = K_t \times \sigma_{nominal}$

Compare σ_design to your allowable stress (yield / safety factor), not σ_nominal.

4. Be Extra Careful with High-Strength Materials

A 250 MPa steel has 15–20% ductility and can handle local yielding. A 1200 MPa high-strength steel has <5% ductility and cracks.

For high-strength materials: Kₜ matters even more.

5. In Fatigue, Kₜ Becomes Kf (Fatigue Strength Reduction Factor)

Under cyclic loading, the stress concentration is more severe than static loading predicts:

$K_f = 1 + q(K_t - 1)$

Where q (notch sensitivity) depends on material and notch sharpness. For most steels and notches, Kf ≈ 0.8 × Kₜ.

---

Common Design Mistakes

❌ Mistake 1: Ignoring Concentration at Net Section

"The nominal stress is below yield, so we're safe."

No. The local stress at the concentration can be 2–4× higher.

❌ Mistake 2: Assuming Stress Redistributes

"The stress concentration is small; load will redistribute."

Only true under static loading with ductile material. Under fatigue or brittle conditions, cracks initiate at the concentration and you're done.

❌ Mistake 3: Using Kₜ for Welds and Bolts

Welds and bolts have their own residual stresses. Use design code equations, not raw Kₜ values.

❌ Mistake 4: Forgetting Dynamic Effects

Static analysis gives you Kₜ. Dynamic analysis (impact, vibration, fatigue) often requires higher reduction factors.

---

The Checklist for Stress Concentration

Before finalizing any design with stress risers:

• Identify all stress concentrations (holes, notches, corners, welds)
• Look up Kₜ from Peterson's or your design code
• Calculate σ_max = Kₜ × σ_nominal at each concentration
• Is σ_max below yield (with appropriate safety factor)?
• If fatigue is present, calculate Kf and check infinite-life stress
• Consider optimizing geometry (larger fillets, smoother transitions)
• FEA check: verify stress concentration hotspots match theory

---

The Design Opportunity

Here's where good engineers win: You can't eliminate stress concentration, but you can minimize it.

Add a fillet. Increase the hole spacing. Use a smooth shoulder. Distribute holes across the section.

These small moves cut Kₜ by 30–50%, which translates to 30–50% more load capacity—with zero additional material.

---

The Takeaway

Stress concentration is the reason why your carefully calculated 200 MPa design can fail at 200 MPa. It's the reason bolts snap at threads, shafts crack at keyways, and flat bars break near holes.

Next time you design a component with any stress riser:

1. Identify it. Where will stress concentrate?
2. Look it up. What's the Kₜ?
3. Apply it. σ_max = Kₜ × σ_nominal
4. Optimize. Can you reduce Kₜ with better geometry?
5. Verify. FEA or testing to confirm.

The engineers who master stress concentration design parts that don't fail in unexpected places. The others redesign repeatedly, hoping they've finally found the issue.

Which engineer do you want to be?