Conveyor Idler Maintenance: Hear Bearing Damage Before It Gets Expensive

On a long, fast belt at a transhipment yard, one single idler had stopped turning. Not broken enough to stand out, just stiff enough to seize. Its steel end cap worked its way into the underside of the belt like a pizza cutter while the belt ran over it at several metres per second. By the end, a belt worth as much as a small car was written off, over an idler that costs a few euros to buy.

Why a seized idler gets so expensive

Do the sums and you can see why this story is not a one-off. Large belt installations run several kilometres and carry several thousand idlers. The motor, the drive, the belt and the transformer each get their own monitoring; the idlers usually run with none. They turn until they stop, and nobody is watching.

That is more expensive than it looks. A worn or noisy roller wastes energy, drives up belt wear and pushes the belt off track. In the worst case you end up with a belt tear or even a fire, and at that point it stops being about maintenance and becomes about the safety of the people at the plant.

The economics behind it are simple. An idler costs a few euros to buy. The belt above it is often the single most expensive part of the whole installation. Let the roller seize and it damages itself and works away at the underside of the belt. That costs you twice: once for the new roller, once for the downtime while the belt stands still and nothing moves. Save on the roller and you pay for it on the belt.

The three senses of inspection: listen, look, feel

A sick idler does not report itself with a clear error message. Sometimes it gives itself away through noise, often high up in the ultrasonic range; sometimes through a bearing that warms up; sometimes through vibration. There is no single, always-the-same early signal. That is exactly what makes it tricky: you cannot rely on any one symptom.

That is why the idea of simply bolting a sensor to every roller is less useful in practice than it sounds. Attempts to measure wear through vibration or temperature probes on or inside the roller alone have proven unreliable. The environment that eats the idler eats the sensor along with it. Dust, water and knocks do not stop at the measuring device.

What is left is a person on the walk-round, and a person has more than one channel. Listen for a roller that is out of step with the rest. Look for the one that no longer turns or shows a running surface worn shiny. And feel whether a bearing housing is warmer than its neighbours, but do that with the plant shut down, never with your hand on a running belt. Only the three together give you a picture. You usually hear the damage first, then confirm it with eye and hand.

Making sense of the noises

A shift lead once reported a “noisy roller” to me. My first question was not where, but which noise. A bearing talks; you just have to know the language. What you hear is a fairly good clue to what ails it.

The breakdown that follows comes from general rolling-bearing knowledge, not from any idler-specific measurement. It still carries over well, because what sits inside an idler is, after all, just a rolling bearing.

One detail makes the diagnosis almost clear-cut: temperature. When a bearing runs distinctly hotter than the rollers beside it, the noise gains a second witness. A sharp rise in temperature can come from too much or too little lubricant, from overload or from a fitting error. It can just as well come from the bearing wandering on its seat or the seal rubbing too tight. Keep that one in mind for later, because in the second case a new bearing does nothing at all.

Why idler bearings die

Look at rolling bearings in general and the picture is strikingly one-sided. By one analysis from the bearing literature, around 70 percent of failures trace back to lubrication problems: contamination, the wrong or too-old lubricant, too little of it. A further 15 percent or so are fitting errors. Those are general bearing figures, not a measurement on idlers. But the same pattern holds true out at the belt.

With an idler, it almost always comes down to the seal. As long as it holds, the bearing lives. Let dust and water reach the grease and it is only a matter of time. The seal is the bearing’s front door. Leave it open and in move exactly the guests that ruin the grease. That is why the cheapest measure is also the most effective. Keep it clean, keep it sealed, and the roller lasts far longer than its catalogue price suggests.

You have to prevent the damage where nobody looks: at the seal.

The costly consequence: from a seized roller to belt damage

A seized roller no longer rolls, it grinds. And grinding produces exactly the two things a belt cannot stand: friction and heat. The belt drags across the stationary steel, and depending on the end cap the roller turns from a wear part into a tool. A steel edge cuts where a worn-down plastic cap would only have left a dent. That is precisely what had happened at the transhipment yard from earlier.

But it does not stop at the one roller. A seized or skewed idler brakes on one side and pushes the belt off track. The edge rubs against the steelwork, frays, and in the worst case the belt is torn apart lengthways. At the end of the chain sits the most expensive case of all: a bend pulley that fails catastrophically and takes the plant out of service for a long repair shutdown.

That is why the jammed roller turns up so often in fault-finding. When a belt mistracks, slips or wears faster than it should, a walk-round along the idlers is one of the first things you do. Not because the roller is always to blame, but because it is cheap to check and expensive to ignore.

Repair or replace, and when

The most important rule is already set: change the bearing before the roller stops. A roller that still grinds but turns costs you a bearing. A roller that seizes costs you the roller and a length of belt with it. Between the two there are often only a few days, days in which someone should have been listening.

Whether a bearing swap is even enough depends on the cause. If the raceway shows pitting, if foreign objects have got in or a rolling element is damaged, there is no way around replacing the roller. But if the problem lies in the bearing wandering on its seat or the seal rubbing, a new bearing does nothing. You fit a fresh part onto the same fault, and a few weeks later you are back where you started. Without fixing the cause, the swap is only postponed.

Fixed intervals you will not get from me, because there is no solid figure to give. What there is, is the walk-round. A daily visual inspection that watches for seized rollers, worn parts, anything irregular and the state of the pulley lagging catches most cases before they get expensive. How often and how thoroughly is set by your plant, not by the calendar.

And then there is the most unglamorous maintenance of all: keeping things clean. Material that clings to rollers and bearings has to go. A scraper that takes the carryback off the belt keeps the rollers clean too and extends their service life into the bargain. This is not rocket science, just routine, and it pays for itself.