9.1 Shaft Design and Analysis

Shaft Stresses and Loading

Shafts transfer rotational energy between mechanical elements like gears, pulleys, and couplings. Because they experience complex combinations of twisting and bending forces, designing a shaft means analyzing material properties, geometry, and loading conditions together. This section covers the stress calculations, combined loading criteria, and design considerations you'll need for reliable shaft design.

Torsional and Bending Stresses

Two fundamental stress types act on a rotating shaft: torsional stress from applied torque, and bending stress from transverse loads.

Torsional stress results from the twisting action of torque transmitted through the shaft. It's calculated as:

$\tau = \frac{Tr}{J}$

• $\tau$ = torsional shear stress
• $T$ = applied torque
• $r$ = shaft radius (distance from center to outer surface)
• $J$ = polar moment of inertia (for a solid circular shaft, $J = \frac{\pi d^4}{32}$)

The stress is highest at the outer surface of the shaft (where $r$ is maximum) and zero at the center.

Bending stress occurs when transverse loads or moments act on the shaft. Mounted components like gears and pulleys create these transverse forces. Bending stress is calculated as:

$\sigma = \frac{My}{I}$

• $\sigma$ = bending stress
• $M$ = bending moment at the cross-section of interest
• $y$ = distance from the neutral axis (maximum at the outer surface)
• $I$ = area moment of inertia (for a solid circular shaft, $I = \frac{\pi d^4}{64}$)

Notice that $J = 2I$ for a solid circular cross-section. This relationship is worth remembering because it shows up frequently in shaft problems.

Combined Loading and Stress Concentration Factors

In practice, shafts almost always experience torsional and bending stresses simultaneously. To evaluate whether the shaft can handle this combined state of stress, you need a failure criterion that converts the multiaxial stress state into an equivalent uniaxial value you can compare against material strength.

Two common criteria are used:

• Maximum shear stress theory (Tresca): Conservative and straightforward. The maximum shear stress is $\tau_{max} = \frac{1}{2}\sqrt{\sigma^2 + 4\tau^2}$. This is often used for ductile materials as a quick check.
• Von Mises (distortion energy) criterion: Slightly less conservative and generally more accurate for ductile materials. The equivalent stress is:

$\sigma_{vm} = \sqrt{\sigma^2 + 3\tau^2}$

You compare $\sigma_{vm}$ against the material's yield strength (for static loading) or endurance limit (for fatigue loading), applying an appropriate factor of safety.

Stress concentration factors account for the reality that geometric discontinuities raise local stress levels well above the nominal calculated values. Common discontinuities on shafts include:

• Shoulder fillets (where the diameter changes)
• Keyways
• Grooves for retaining rings
• Transverse holes

The theoretical stress concentration factor $K_t$ is found from published charts (like those in Peterson's Stress Concentration Factors) or from finite element analysis. You apply it to the nominal stress:

$\sigma_{actual} = K_t \cdot \sigma_{nominal}$

For fatigue analysis, a related but distinct factor called the fatigue stress concentration factor $K_f$ is used instead, which accounts for the material's notch sensitivity. The relationship is:

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

where $q$ is the notch sensitivity factor (ranging from 0 to 1, found from material data).

Shaft Materials and Design Standards

Shaft Materials and ASME Code

Material selection depends on the balance of strength, toughness, machinability, and cost required for the application.

• Low-carbon steel (e.g., AISI 1020): Good machinability, lower strength. Used for lightly loaded shafts.
• Medium-carbon steel (e.g., AISI 1045): The most common shaft material. Offers a good balance of strength and cost, and can be heat-treated for higher hardness.
• Alloy steels (e.g., AISI 4140, 4340): Higher strength and toughness. Used when fatigue life or high loads demand better mechanical properties.
• Stainless steels: Selected when corrosion resistance is a primary concern, such as in food processing or marine environments.

The ASME code provides standardized guidelines for shaft design, including allowable stress values, recommended factors of safety, and manufacturing requirements. Following ASME standards ensures that designs meet accepted safety and performance benchmarks across industries. For transmission shafting specifically, the ASME code prescribes maximum allowable shear stress values (typically 30% of yield strength or 18% of ultimate tensile strength, whichever is lower, without keyways).

Fatigue Analysis and Shaft Features

Fatigue failure is the most common mode of shaft failure because shafts experience cyclic loading during rotation. A point on the outer surface of a rotating shaft under a constant bending load sees fully reversed bending stress every revolution, even though the external load doesn't change.

Fatigue analysis steps:

1. Identify the critical cross-section (usually where the bending moment is highest or where a stress concentration exists).
2. Determine the alternating and mean stress components at that section. For a rotating shaft with steady torque and a constant transverse load, the bending stress is fully reversed (alternating) and the torsional stress is steady (mean).
3. Apply fatigue stress concentration factors ($K_f$) to the alternating stress component.
4. Apply endurance limit modifying factors to the material's base endurance limit ($S_e'$) to get the corrected endurance limit ($S_e$). These factors account for surface finish, size, reliability, temperature, and other conditions.
5. Use a fatigue failure criterion. The modified Goodman line is the most common:

$\frac{\sigma_a}{S_e} + \frac{\sigma_m}{S_{ut}} = \frac{1}{n}$

where $\sigma_a$ is the alternating stress, $\sigma_m$ is the mean stress, $S_{ut}$ is the ultimate tensile strength, and $n$ is the factor of safety.

1. Alternatively, use the DE-Goodman or DE-Gerber criteria, which combine bending and torsion directly into the fatigue equation.

S-N curves (stress-life) plot stress amplitude against the number of cycles to failure and are the primary tool for high-cycle fatigue analysis. For steels, the endurance limit (the stress below which fatigue failure theoretically does not occur) is roughly $S_e' \approx 0.5 S_{ut}$ for $S_{ut} \leq 1400$ MPa.

Shaft features for torque transmission:

• Keyways are rectangular slots machined into the shaft and hub. A key sits in the slot and transmits torque through shear and bearing contact. They're simple and widely used, but they create significant stress concentrations ($K_t$ values of 2.0 or higher are typical).
• Splines use multiple teeth machined into the shaft that mesh with corresponding grooves in the hub. They distribute the load over a larger area than a single key, handle higher torques, and allow axial sliding when needed. The tradeoff is higher manufacturing cost.

Shaft Dynamics and Deflection

Critical Speed and Shaft Deflection

Critical speed is the rotational speed at which the shaft's natural frequency of lateral vibration matches the rotation frequency, causing resonance. At resonance, even small imbalances produce large vibrations that can lead to rapid failure.

For a simple shaft with a single concentrated mass, the first critical speed can be estimated using:

$\omega_{cr} = \sqrt{\frac{g}{\delta_{st}}}$

where $g$ is gravitational acceleration and $\delta_{st}$ is the static deflection at the mass location due to its own weight. For shafts with multiple masses, Rayleigh's method or Dunkerley's equation provides approximate critical speeds:

$\frac{1}{\omega_{cr}^2} \approx \frac{1}{\omega_1^2} + \frac{1}{\omega_2^2} + \cdots$

where $\omega_1, \omega_2, \ldots$ are the critical speeds considering each mass individually.

Design practice is to keep the operating speed at least 20-25% away from any critical speed. Shafts that operate below the first critical speed are called rigid shafts; those that operate above it are called flexible shafts and must pass through the critical speed during startup and shutdown.

Shaft deflection refers to lateral bending under applied loads. You calculate it using standard beam deflection methods (superposition, integration, or Castigliano's theorem), treating the shaft as a simply supported or overhanging beam depending on the bearing arrangement.

Excessive deflection causes several problems:

• Gear mesh misalignment, leading to uneven tooth loading and premature wear
• Increased bearing loads and reduced bearing life
• Seal leakage from shaft runout
• Vibration and noise

Typical deflection limits depend on the application. For gear shafts, a common guideline is that the slope at the gear mesh should not exceed about 0.05° (approximately 0.001 rad).

Strategies to reduce deflection:

• Increase shaft diameter (deflection scales with $1/d^4$, so even small diameter increases have a large effect)
• Use a stiffer material (higher modulus of elasticity $E$)
• Move bearings closer to the applied loads to reduce the effective span
• Minimize overhung loads where possible, since cantilever arrangements produce much larger deflections than simply supported ones