6.2 Factor of Safety and Allowable Stress

Factor of Safety and Allowable Stress

Factor of safety and allowable stress are the two main tools engineers use to bridge the gap between theoretical calculations and real-world uncertainty. They help ensure components can handle expected loads plus a margin for the unexpected, while also preventing wasteful over-design.

Defining Factor of Safety and Allowable Stress

The factor of safety ($n$) is the ratio of a material's strength to the maximum stress it experiences under load:

$n = \frac{\text{Strength}}{\text{Maximum Stress}}$

"Strength" here depends on the failure theory you're using. For ductile materials, this is typically the yield strength ($S_y$). For brittle materials, it's the ultimate tensile strength ($S_{ut}$). The choice matters because it changes what failure mode you're guarding against.

A factor of safety of $n = 2$ means the material is twice as strong as the maximum expected stress. Values of $n$ less than 1 mean the design is predicted to fail.

Allowable stress ($\sigma_a$) flips the relationship around. Instead of checking whether a design is safe after the fact, you set a stress ceiling before you start:

$\sigma_a = \frac{\text{Strength}}{n}$

This ceiling accounts for uncertainties in material properties, loading conditions, and manufacturing quality. You then design the component so that the actual stress never exceeds $\sigma_a$.

Design Stress and Working Stress

These two terms show up frequently and are easy to confuse:

• Design stress (or allowable stress) is the maximum stress a component is designed to withstand. It's the limit you set during the design phase using the formula above.
• Working stress is the actual stress the component experiences under normal service loads. It's what you calculate from the real geometry, loads, and boundary conditions.

The fundamental design requirement is straightforward: working stress must be less than or equal to design stress. If it isn't, you need to change the geometry, pick a stronger material, or reduce the load.

Safety Margins and Load Factors

Understanding Safety Margins

While factor of safety is a ratio, the safety margin is a difference:

$\text{Safety Margin} = \text{Strength} - \text{Maximum Stress}$

• A positive safety margin means the component has reserve capacity beyond expected loads.
• A zero or negative safety margin means the component is at or past its limit and the design needs to change (stronger material, larger cross-section, or reduced load).

Some references also define a margin of safety as $n - 1$, which equals zero at the boundary of failure. You'll see this version often in aerospace design. Make sure you know which definition your course uses.

Load Factors and Stress Concentration Factors

Load factors are multipliers applied to expected loads before you calculate stress. They account for the reality that actual loads are uncertain. For example, a building code might require a load factor of 1.6 on live loads and 1.2 on dead loads. The factored load is always larger than the nominal load, building conservatism into the analysis from the start.

Load factors and factors of safety serve related but distinct purposes. Load factors inflate the loads; factors of safety reduce the allowable strength. Some design codes use one, some use the other, and some use both. Don't double-count them unless the code explicitly requires it.

Stress concentration factors ($K_t$) account for localized stress spikes caused by geometric features:

$K_t = \frac{\text{Maximum Local Stress}}{\text{Nominal Stress}}$

Common culprits include holes, sharp notches, keyways, threads, and abrupt changes in shaft diameter. A small hole in a plate under tension can produce $K_t$ values of 2 to 3, meaning the local stress is two to three times the average stress in the cross-section.

When you're checking a design against allowable stress, you need to use the actual peak stress at the concentration, not just the nominal stress. In practice, this means either:

• Multiplying the nominal stress by $K_t$ and comparing to $\sigma_a$
• Or looking up $K_t$ from charts (Peterson's is the standard reference) based on the specific geometry ratios

To reduce stress concentrations in your designs:

• Use generous fillet radii at transitions instead of sharp corners
• Avoid abrupt changes in cross-section
• Where holes or notches are unavoidable, keep them away from regions of highest nominal stress