Case Study: Catching a Pulley Misalignment Before It Took Out Another Bearing

The Problem

Note: this example is documented by Insens, the developer of the RED sensor technology Reliability AI deploys -- not a Reliability AI client engagement. We're sharing it because it illustrates exactly the kind of fault our RED sensor is built to catch.

At an aggregate processing site, a vibrating screen driven by an induction motor through a belt-and-pulley transmission kept losing bearings. The maintenance team had chalked it up to normal wear -- the failures kept recurring despite a regular vibration inspection schedule, and no early warning ever showed up in those routine checks.

The Diagnosis

The equipment was fitted with a RED sensor -- an Electrical Signature Analysis device installed at the motor's electrical cabinet. Within a few weeks, the recorded signal showed an unusual, steadily growing amplitude modulation centered on the driving pulley's rotational frequency: a signature consistent with an asymmetric mechanical load somewhere in the transmission.

That pattern pointed toward a progressive pulley misalignment. An on-site inspection confirmed it -- a slightly deformed motor mount had let belt tension go uneven, creating a constant sideways load on the screen's bearings. That load had been accelerating bearing wear well before any vibration symptom was strong enough to trip a standard alarm threshold.

The Outcome

With the root cause identified, the operator corrected the mount geometry and re-tensioned the belts before a failure occurred. The result: the bearing life was extended, the breakdown was avoided entirely, and -- as a secondary benefit of fixing the misalignment -- the system's energy consumption dropped by roughly 6%.

Why This Matters

This is the pattern ESA is built to catch: a fault that produces no meaningful vibration signature until it's already done damage, but shows up clearly in the electrical signal because it's altering the torque the motor has to deliver. A periodic vibration route -- or even a permanently mounted vibration sensor at that specific point -- would likely have missed it at the same stage. Continuous electrical monitoring caught the drift while it was still just a drift, not yet a failure.

This is the same RED sensor technology built into Reliability AI's platform. Contact us to see how continuous Electrical Signature Analysis could catch a similar fault in your own equipment before it becomes a bearing replacement.