Broadband Random Vibration Fatigue

What Is It?

Broadband random vibration fatigue deals with stress processes where the stress PSD has significant energy across a wide frequency range — multiple modes contribute to the stress. In this case, the narrow-band approximation (which assumes a single dominant frequency) is conservative because it overestimates the number of full cycles. Broadband fatigue methods correct for this by accounting for the bandwidth of the stress process.

Why It Matters

Most real engineering structures have multiple modes contributing to the stress at critical locations. The stress PSD is broad-band, not narrow-band. Using the narrow-band approximation for broad-band response produces conservative fatigue life estimates — sometimes by a factor of 2-5 or more. This can lead to over-design, unnecessary weight or overly conservative qualification requirements. Broadband methods provide more accurate estimates.

For broad-band stress response (multiple modes), the narrow-band approximation is conservative by a factor of 2-5 or more. Broadband methods provide more accurate — and less conservative — fatigue estimates.

Wide Frequency Content

A broad-band stress PSD has significant energy across a wide frequency range. Multiple peaks correspond to multiple contributing modes. The stress time history is irregular — peaks of varying amplitude at different frequencies. The cycles are not regular sinusoids; they are a mix of large cycles (from the dominant mode) and smaller, irregular cycles (from other modes and from the interaction between modes).

• Broad-band PSD — multiple peaks from multiple modes
• Irregular stress time history — peaks of varying amplitude and frequency
• Cycles are a mix — large cycles from dominant mode, smaller irregular cycles from others
• Not the regular sinusoidal cycles assumed by the narrow-band method

Multiple Modes

When multiple modes contribute to the stress, the cycle distribution is more complex than the Rayleigh distribution assumed by the narrow-band method. Some peaks are full cycles from the dominant mode; others are partial cycles or secondary peaks from other modes. The narrow-band method counts all peaks as full cycles, overestimating the damage. Broadband methods distinguish between full cycles and partial peaks, producing a more realistic cycle distribution.

Cycle Distributions

The cycle distribution for a broad-band process differs from the Rayleigh distribution. The distribution has fewer large-amplitude cycles and more small-amplitude cycles than the Rayleigh prediction. The exact distribution depends on the bandwidth (irregularity factor) and the shape of the stress PSD. Broadband fatigue methods estimate this distribution using empirical or semi-analytical approaches based on the spectral moments.

Correction Approaches

Several correction approaches have been developed to improve on the narrow-band approximation for broad-band response. These methods use the spectral moments to estimate a more realistic cycle distribution. The most widely used are the Dirlik method, the Wirsching-Light correction, the Tovo-Benasciutti method and the Zhao-Baker method. Each is an empirical or semi-analytical approximation, validated against time-domain rainflow counting for a range of stress PSD shapes.

• Dirlik — most widely used; empirical formula based on spectral moments
• Wirsching-Light — simple correction factor applied to narrow-band damage
• Tovo-Benasciutti — refined version with better theoretical basis
• Zhao-Baker — alternative empirical approach
• All are approximations — validate against time-domain where possible

Limitations

All broadband fatigue methods are approximations. They estimate the cycle distribution from spectral properties (moments) rather than counting actual cycles. Their accuracy depends on how well the method's empirical fit matches the actual stress PSD shape. For unusual PSD shapes — very sharp peaks, unusual bandwidth — the methods may be less accurate. For critical applications, time-domain rainflow counting provides the most defensible estimate.

All broadband methods are approximations. They are validated against time-domain rainflow counting for typical PSD shapes. For unusual shapes or critical applications, validate against time-domain analysis.

Choosing a Method

The choice of broadband method depends on the stress PSD character, the required accuracy and the available tools. The Dirlik method is the most widely used and is generally reliable for a wide range of PSD shapes. The Wirsching-Light correction is simpler but less accurate for some shapes. For critical applications, compare multiple methods and validate against time-domain analysis.

Key Takeaways

• Broadband fatigue deals with stress PSDs where multiple modes contribute — irregular cycles
• Narrow-band approximation is conservative for broadband — overestimates damage by 2-5× or more
• Correction methods (Dirlik, Wirsching-Light, Tovo-Benasciutti) use spectral moments for better estimates
• Dirlik is the most widely used and generally reliable for most PSD shapes
• All broadband methods are approximations — validate against time-domain for critical applications

Engineering judgement — what can change the conclusion

For Broadband Random Vibration Fatigue, the harmonised review should concentrate on the translation from response PSD to cycle-amplitude statistics and damage, including bandwidth, mean stress and modal superposition assumptions. The engineering value comes from identifying the assumptions that can move the governing margin or failure mode, then testing those assumptions deliberately rather than adding complexity indiscriminately. Where simplified and high-fidelity methods coexist, the simpler method should be used as an independent trend or magnitude check so that agreement is based on physics rather than shared modelling assumptions.