How Electrical Signature Analysis Detects Mechanical Faults Through Current Signals Alone

The Core Idea: Mechanical Forces Become Electrically Visible

It sounds counterintuitive at first: how can a sensor reading electrical current tell you anything about a worn bearing or a misaligned shaft? The answer lies in torque. Mechanical faults -- misalignment, imbalance, bearing degradation -- generate small cyclic forces with every rotation, and those forces produce tiny, repeating fluctuations in the torque the motor has to deliver to keep the shaft turning.

A motor's magnetic field responds to torque demand. So when torque fluctuates cyclically due to a developing mechanical fault, that fluctuation shows up as a matching micro-variation in the motor's current and voltage waveform -- faint, but measurable and consistent.

Finding the Signal: Sidebands and the Fourier Transform

On its own, a raw current waveform doesn't reveal much to the eye. The key is spectral analysis -- typically a Fast Fourier Transform (FFT) -- which converts the time-domain signal into a frequency spectrum. Mechanical fault frequencies show up as sidebands clustered around the motor's rotational frequency and its harmonics. The specific spacing and pattern of those sidebands map to specific fault types: a certain pattern for bearing wear, another for imbalance, another for misalignment.

A Concrete Example: Bearing Wear

Bearing failures are one of the most common causes of unplanned downtime in rotating equipment, and they rarely happen in isolation -- degradation often traces back to an earlier issue like misalignment, overload, poor lubrication, or an electrical imbalance nobody was watching for. As a bearing wears, it produces amplitude modulations in the current spectrum clustered around the rotational frequency. Tracked continuously, the growth of that signature over weeks or months shows the defect progressing in real time, well before it would surface as an audible or vibration-detectable symptom.

Why This Matters More Than It Sounds

Because ESA reads electrical and mechanical fault signatures from the same signal simultaneously, it doesn't just flag that something is wrong -- it can often distinguish whether a symptom traces back to a mechanical root cause or an electrical one. That distinction is what turns an alert into an actual diagnosis, and it's a large part of why cross-referencing electrical and mechanical indicators reduces false positives compared to a purely mechanical (vibration-only) view.

Reliability AI's RED sensor applies this same current-signature approach continuously from your Motor Control Center, giving your team a documented trend of mechanical condition, not a single inspection snapshot.