Chevron Belt Conveyor Problems: Slippage, Profile Damage, Carry-Back

Most ribbed belt complaints reach us as a single sentence. The belt is slipping. Or the ribs are coming off. Or we are shovelling material out of the boot every morning. Three different sentences, occasionally describing the same installation, and the correct fix for one of them quietly makes the other two worse. On a 320 t/h fertiliser incline we watched a crew tighten a belt to stop slippage that was actually caused by a seized return roller. Six weeks later that belt was in the scrap bin with torn rib tips and a stretched carcass. The roller was still seized.

What follows is the order in which we split a chevron belt conveyor problem into symptoms that can be measured, because the three headline complaints share almost no root causes. Slippage is a traction question. Profile damage is usually a geometry and bond-line question. Carry-back is a cleaning and cover question. Tracking sits on top of all three. Treating them as one problem is how a two-day repair becomes a two-month replacement cycle.

Send us the symptom, the slope and the belt size — we will return a diagnosis and a quote

One note on vocabulary before we start. A ribbed belt, a chevron belt and a cleated belt are not interchangeable descriptions on site even though they overlap in catalogues. Chevron patterns are moulded ribs running at an angle to the belt centreline. Cleats are transverse bars, usually taller and square to the direction of travel. Sidewalls are the raised edges that turn a belt into a pocket. A belt can carry all three, and the failure mode of a 25 mm chevron rib is not the failure mode of a 50 mm cleat. If your belt has sidewalls as well as ribs, the extra reading is on the sidewall conveyor belt page, because pocket belts trap material in a way flat ribbed belts never do.

01Split the Complaint Into Symptoms You Can Measure

The first thing we ask a customer is not what the belt is doing but where the evidence sits. Material on the ground under the loading skirt points at one family of causes. Polished rubber on the head pulley points at another. A rib that has lifted at one edge but is still bonded at the other points at a third. As a conveyor belt manufacturer we see the belts after the fact, and the physical evidence on a returned belt is far more reliable than the description that came with it.

So the diagnosis starts with a walk along the line with a tape measure, a torch and a notebook, recording five things at every station. Belt speed against drive running current is the first. Material on the return strand and rib condition on the carrying side come next, then pulley and roller condition wherever the belt changes direction, and finally the free length left at the take-up. That sounds like a survey rather than a repair, and it is. The single most expensive habit in belt maintenance is fixing the first thing that is visible.

Here is the split we use, and it covers the vast majority of ribbed incline problems we are asked about. Longer entries in the last column are deliberate — they describe what actually happens downstream.

That table is a sorting tool, not an answer sheet. The value is that it forces a measurement before a decision.

One more screening step is worth the five minutes it costs. Ask whether the problem appeared suddenly or drifted in. Sudden onset — a belt that ran well on Friday and not on Monday — almost always points at a component. A failed roller bearing is the classic case, and so is a pulley lagging strip that has gone, a splice that has begun to open, or a transfer chute that shifted after a liner change. Gradual onset, spread over months, points at the material, the loading pattern or the belt itself stretching. Those two histories lead to completely different investigations, and mixing them up sends crews down the wrong path roughly half the time. We have seen a shift team spend an entire weekend on belt tension when a single seized return roller was the culprit, and we have seen the opposite as well, a belt replaced as worn out when the real trouble was a chute that had been rebuilt 40 mm lower during a maintenance shutdown.

02Slippage on an Incline: Four Causes That Look Identical at the Head Pulley

When a chevron belt that has run for years starts slipping, the torque demand has changed, the available traction has changed, or both. What makes slippage hard to diagnose on an incline is that all four of the usual causes produce the same noise and the same smell. Rubber squealing at the head pulley, a fine black dust trail under the discharge, and a drive that is drawing more current than its nameplate suggests. Everything downstream of that observation has to come from measurement, not from sound.

The first cause is genuine loss of wrap friction. Lagging wears, becomes glazed or is polished by the very material it is meant to grip. A chevron belt feeding a head pulley at 6 degrees of wrap with worn ceramic lagging will slip at half the tonnage an identical belt with new lagging will carry. We measure this by comparing the belt speed at the head shaft and the belt speed 10 m into the carrying strand. If the head shaft is turning faster than the belt is moving, the belt is slipping on the pulley and nothing else on the line matters until that is fixed. It also matters that the rest of the circuit is not working against the slope. An industrial conveyor belt running flat into the incline, and a plain rubber conveyor belt on the horizontal run downstream of it, belong to the same traction circuit; a polished cover anywhere in that circuit shifts the tension balance the slope depends on. Replacement lagging options and pulley geometry are covered on the conveyor pulleys page, and it is worth reading before ordering a lagging strip, because groove pattern matters more than hardness on a wet incline.

The second cause is the one most often misdiagnosed. A ribbed incline carries material in pockets, and the pockets are full when the belt stops. Restarting from that condition requires the drive to accelerate not just the belt carcass but an entire bed of material sitting on a 20 degree slope, leaning back against the ribs. On a 1,000 mm wide belt with 20 mm ribs at 400 mm pitch and a 250 mm bed depth, the loaded mass per pocket is substantial, and forty pockets do not mind being lifted one at a time but object strongly to being lifted together every start.

Third: the take-up has run out of travel. This is common on short inclined belts with screw take-ups, where the belt has stretched the equivalent of the take-up stroke over two or three years and nobody noticed because the belt still looked tight. Slack side tension collapses, and the drive has lost the wrapping force it needs. A quick check is to measure remaining screw travel. If it is under 25 mm, tension has become a fiction.

Fourth, and the one that surprises people, is a change in the loading. If the feed rate rises 15% or the material becomes wetter and heavier, the belt does not lose traction at all in the friction sense. It simply cannot move the bed. The slip is real from the operator's point of view because the belt appears to hesitate, but the cure is a loading or profile fix rather than a tensioning one.

Distinguishing these four takes an hour. Guessing takes weeks.

There is a fifth possibility that belongs in the same conversation even though it is not strictly slippage: the belt is running on a drive pulley that was sized for a flat belt and never re-checked when the line was converted to a ribbed belt. A ribbed belt is thicker at the rib line, and its minimum pulley diameter is set by the rib height, not the carcass. Running a 25 mm ribbed belt over a pulley that a smooth belt tolerated can crack the base of the ribs even when traction is fine. If you are converting a line, the chevron conveyor belt reference gives the geometry limits we apply to rib height against pulley and roller diameters.

One false lead deserves naming, because it burns whole days. The drive train between motor and head shaft frequently uses wedge belts, and a stretched set there makes a perfectly healthy slope look like a slipping belt. We have chased that twice on the same plant. The conveyor pulley was innocent both times. The real fault was a pair of worn V-belts slipping on the gearbox input, which dropped the head shaft speed and left the belt running slower than the throughput demanded, and the tensioner stayed tight the whole time because nobody was measuring the right shaft. When the speed loss is measurable upstream of the head shaft, the right conversation involves a transmission belt manufacturer and, for the section and cord grade itself, a V-belt manufacturer who can match a replacement to the pulley grooves already on the machine. One tachometer reading at the head shaft would have saved a week of belt work.

The practical point is that slippage on an incline is measured twice before anything is adjusted. Once at the drive, to establish whether the pulley is losing grip. Once at the loading point, to establish whether the belt is being asked to carry more than a pocket can hold. Everything else is a consequence.

03Profile Damage: Peeling Tips, Torn Shoulders and the Bond Line Under the Rib

A chevron rib never fails in the middle of the rib. It fails at its base, at its leading edge, or at a corner where it meets a sidewall, and the reason is always that the rib is being asked to do a job the base belt should be doing. The rib is a mechanical stop. It is not a structural member. Once it starts carrying vertical load, bending load or edge load, the bond line between rib and cover becomes the weakest link in the whole belt, and it will show you exactly where the loads are.

Peel is the most common shape of failure. A rib lifts along one edge while remaining bonded on the other, usually at the upstream-facing side after a period of running. This pattern comes from shear, not from poor adhesion. The rib is being pushed sideways by material sliding back down the slope during stops, or it is being bent repeatedly by a chute lip, or the base belt is flexing over a pulley that is too small for the rib height, and each pass works the base of the rib like a hinge. Re-bonding the rib treats the symptom. If the peel recurs on the same belt within six months and in the same pattern, the cause is geometric, and it will keep recurring until the geometry changes.

Torn tips tell a different story. When the top 3–5 mm of a rib is missing across its full length, something has passed over the rib at speed. Usually it is a large lump riding on the belt from a previous station, or a scraper blade set too deep, or a tramp metal fragment. On a coal incline we traced repeated rib tip loss on the first three ribs behind the join to a misaligned skirt rubber that was standing 8 mm proud of the chute wall, so every join lifted just enough to catch the rubber edge and shear the tips off over a few thousand cycles.

Broken or crushed ribs in a repeating pattern every few metres is a splice or roller problem, not a profile problem. Look upstream of the damage.

There is also a failure that looks like wear but is actually chemical. Ribs that turn chalky, crack in a network of fine lines and lose elasticity long before the cover shows comparable ageing have usually met something they were not selected for. Oils and solvents in the material stream degrade the rubber from the outside in, and the ribs, being the thinnest section with the most surface area relative to volume, go first. If the material carries grease, hydraulic fluid mist or process oil, the compound has to be chosen for it — the oil resistant conveyor belt range exists precisely because standard compounds swell and soften under that exposure. Heat does the same thing faster, and the heat resistant conveyor belt grades trade some cut resistance for retained elasticity at temperature.

The bond line itself deserves a paragraph of its own, because it is where most arguments between site and supplier begin. A moulded rib is formed either by pressing uncured rubber into the belt cover during vulcanisation, so rib and cover cure together as one body, or by bonding a separately cured profile onto a finished belt. Both produce belts that work, but they fail differently. The co-cured version delaminates along the fabric-to-cover interface underneath the rib when the failure is thermal. The bonded version delaminates at the rubber-to-rubber interface when the failure is mechanical. Knowing which you have tells you what to look for, and any reputable conveyor belt supplier will state the manufacturing route in the specification rather than leaving you to infer it from a cut sample.

If you only remember one thing from this section, make it the ratio. A rib is useful for the material it stops and dangerous for the loads it transmits. Keep those separate during design and half of these failures disappear.

04Carry-Back and Material Packing Between Ribs

Carry-back on a ribbed incline is not the same problem as carry-back on a flat belt, and the usual remedies do not transfer between them. On a flat belt the material adheres to a smooth surface and a scraper can shear it off. On a ribbed belt the material sits in discrete pockets, and the pocket geometry changes everything about where it leaves the belt and when.

Material leaves a pocket by gravity as the belt rounds the head pulley, and if the pocket is deep, narrow, or contains damp fines, part of the load reaches the vertical point of the belt path before it starts to move. That fraction either falls straight down onto the return strand or drops into the next pocket on the underside, where it is carried back along the full length of the incline. On a 90 m incline running 240 t/h we measured 1.8 tonnes per shift of carry-back material from exactly this mechanism. Two thirds of it came from pockets deeper than 30 mm that had never been asked to discharge before — because the product had changed from dry screened stone to a washed, damp feed.

The first question in that case is not which scraper to fit. It is whether the material is leaving the belt at all.

Rib pitch matters here more than most specifications admit. The packing combination is wide pitch with a deep rib, carrying a damp feed. A pocket that discharges in 0.4 seconds on a dry feed may need 1.5 seconds on a wet one, and at 2.0 m/s belt speed the belt has already travelled more than a metre by then. Reducing rib height while shortening the pitch — for example from a 35 mm rib at 500 mm pitch to a 20 mm rib at 300 mm pitch — keeps the material retention on the slope while giving each pocket many more discharge opportunities per metre. Traction stays. Discharge improves. The trade is a higher number of smaller ribs, which costs more in moulding and demands a smoother base belt under them. That trade is far easier to manage when an entire incline family comes from one wholesale conveyor belts programme built on a fixed rib pitch ladder, because the plant can then step one rung up or down that ladder without re-machining chute skirts or re-checking pulley diameters against a new profile height.

Where a belt carries both ribs and sidewalls, the packing problem becomes a corner problem. Material lodges in the internal corner where rib meets wall and stays there for the entire circuit. These pockets are visible from the side as a dark wedge that never empties. The sidewall conveyor belt page covers the corner radius and base belt requirements in detail, but the diagnostic sign on site is simple enough: if the material in a corner pocket is wetter than the material in the centre pocket, that corner is not discharging and the belt is carrying dead load.

Cleaning devices need to be chosen for pockets rather than surfaces. A primary scraper blade designed for a flat belt, set to the pressure required to clean a flat cover, will simply be struck by every rib on a ribbed belt. Repeated striking drives the blade back, the blade loses contact between ribs, and the belt begins to carry material underneath it. That is why ribbed belts are usually cleaned with a combination of a counterweighted secondary cleaner set between the ribs and, where the duty allows, a rotary brush under the head pulley. For sticky or abrasive materials where the cleaning load is severe, the compound underneath the ribs needs to survive continuous scouring and the abrasion resistant conveyor belt grades are worth the extra cost in cover thickness alone, because every millimetre of cover is roughly a year of cleaning abrasion on a hard mineral duty.

Measuring carry-back is not complicated and very few plants do it. Lay a tarpaulin under the return strand for one full shift, weigh what lands on it, and divide by the tonnage fed over the same shift. Anything above 0.4% is worth acting on. Anything above 1% is costing you cleaning labour, belt life and, on a long incline, a meaningful amount of drive power. The number also gives you a before-and-after figure to judge whether a profile change actually worked, which is more than most maintenance decisions ever get.

05Tracking, Edge Damage and the Asymmetry Trap

Ribbed belts mistrack for a reason that flat belts do not share: the belt is not symmetric along its length in the way a smooth belt is. Every rib is a transverse stiffener, and any lateral force applied to a rib is transferred to the belt edge much more directly than it would be on a plain cover. A belt that would tolerate 15 mm of roller misalignment running flat will climb a pulley and wear its edge away on the same misalignment once it is ribbed and running up a slope.

We are asked to replace belts for edge damage far more often than the evidence justifies. Two cases in the last year were resolved without a new belt at all. In the first, a single return roller bracket had been re-welded 12 mm out of square after a repair, and the belt was tracking against the frame on one side only, wearing through the edge cover at about 2 mm per month. In the second, the loading chute had been rebuilt with the material landing 60 mm off centre, so the bed sat heavier on one side and pushed the belt across. Both belts were fine. When we are acting as a conveyor belt distributor for a plant, the first request is always for photographs of the training station, the belt edge and the loading point, because roughly half of the edge-wear complaints we see are installation problems travelling under a belt's name.

Edge wear that is genuinely the belt's fault looks different. It is even on both sides, it follows the belt rather than the frame, and it accelerates sharply wherever the belt passes a structure. If the wear pattern matches the spacing of your idlers, the rollers are the cause. If it matches the spacing of your splices, the belt is not square.

Rib height tolerance matters here in a way that surprises most buyers. Because ribs are moulded, a rib that is 2 mm taller on one side of a 1,200 mm wide belt is not unusual in low-cost product, and on an incline that asymmetric rib height makes one edge of the belt ride higher in the trough. The belt then trains towards the low side. This is the single strongest argument for specifying a rib height tolerance with the belt rather than accepting whatever the mould produces, and it is something we check at our own conveyor belt factory before any ribbed belt goes into a container, because a tracking complaint from 8,000 km away costs more to resolve than the tolerance costs to hold.

One measurement settles most tracking arguments: run the belt empty and watch the edge gap at three stations over ten revolutions. If the empty belt tracks straight and the loaded belt does not, the problem is in the loading. If both are off, the problem is mechanical and probably squareness. That two-minute test decides whether you call a fitter or a supplier, and getting that call right is most of the battle.

06Splices and Squaring: Where Ribbed Belts Fail Quietly

A ribbed belt is almost never endless. It is spliced, and the splice has to interrupt a rib pattern that was moulded with a fixed pitch. That interruption is the origin of a large share of the failures that get attributed to the profile itself, and it is usually handled badly the first time.

There are three ways to join a ribbed belt, and each one leaves a different weakness behind. A vulcanised finger splice recovers nearly the full tensile strength of the carcass, but the ribs at the joint are either moulded in a separate press operation or left as a short flat section, and a flat section at a splice is a place where material sits and where the belt flexes differently from everywhere else. A stepped or bias splice spreads the load across a longer length, which is kinder to the carcass, but it produces a joint that is harder to keep square, and on a steep incline an unsquare joint will drag on one side. Mechanical fasteners give a fast repair and a belt that can be shortened on site, but the fastener plates stand proud of the cover, so every rib passing the splice region is locally stressed and every cleaner blade catches the plates.

The rib pattern at the joint deserves more attention than it normally gets. If the splice lands so that the rib pitch is broken by 40 mm, material retention at that point drops and the bed starts to creep during stops. Repeated creep at one location is exactly what peels ribs, which is why the first rib behind a splice appears in so many of the damaged belts we receive.

We keep the specification simple on this point. The rib pitch is continued through the splice, the joint is cut square to within 3 mm across the full belt width, the splice length is at least 600 mm on a ribbed incline regardless of carcass class, and the belt is marked at the splice so the inspection crew can find it. On a 900 mm wide chevron belt carrying damp sand, a splice cut 10 mm out of square was enough to move the belt 25 mm off centre at the head pulley and produce visible edge wear inside four months.

Mechanical fasteners are not forbidden on a ribbed incline, but they come with two conditions. The first is that the belt must be re-tensioned within the first 48 hours, because the plates bed into the carcass and the joint rotates itself loose. The second is that any cleaning device in the line has to clear the fastener height, and on most installations that means the secondary cleaner has to be lifted or counterweighted differently. Where a plant needs the option of a quick repair without a vulcanising press, the sensible arrangement is a mechanical fastener at a single designated service point and vulcanised joints everywhere else, so only one location in the belt has to be treated differently by inspection.

The general rule is that a splice should be the strongest and least interesting part of the belt. If it is the part your operators know about, it is already a problem.

07Rollers, Pulleys and Skirts That Punish Cleated Belts

Support hardware is where a ribbed belt pays a price that a flat belt never pays. Every roller that touches the carrying side of a ribbed belt is in intermittent contact, because the ribs pass over it and the belt surface at that moment is anything from 15 to 40 mm further away from the roller shell than the cover. A carrying idler under a ribbed belt therefore works like a hammer rather than a bearing surface, and a plant that converts a line from flat to ribbed without touching the idlers will find its roller bearings failing within a year.

The functional compromise most plants settle on is to keep the carrying side supported on troughing rolls in the flat sections and let the incline run unsupported between the head and tail, or if support is unavoidable, to use rollers with a larger diameter and heavier bearing class so the impact loading is spread over fewer, stronger units. That is a roller selection decision, and it belongs with the conveyor rollers discussion rather than with the belt specification, but it determines whether a ribbed belt survives longer than three years. The return strand is easier, because the return side is typically plain, and a well-sealed return idler with a labyrinth seal handles a dusty incline without much drama as long as the rollers are correctly aligned.

Where rollers and ribs really fight is at the loading point. A ribbed belt receives material in a series of discrete shots rather than a steady stream, because the ribs interrupt the flow under the chute. That load pattern produces impact loads on the first rollers behind the loading skirt that are noticeably higher than the average tonnage would suggest. The impact idler stations under the chute need their rubber rings checked for flattening rather than grooving, because flattening is what happens when the rings are absorbing rib-frequency impact rather than a continuous bed. On the belt itself, the same pattern is why the compound directly under the ribs has to be cut and impact resistant, and the impact cut resistant conveyor belt construction is usually the right base for a chute-fed incline handling sharp rock or recycled material.

Skirt rubber deserves its own paragraph because it damages more ribbed belts than any other single item on the machine. The rule we give fitters is simple. The skirt must be set to a clearance larger than the maximum rib height at that point, plus a margin for belt sag, and the gap must be measured along the whole length of the loading zone rather than at the two ends. On a chute where the belt sags 12 mm under load, a skirt set to a 20 mm static clearance can be in contact at the mid-point with a 15 mm rib, and every cycle then shears the rib crown. The damage looks like abrasive wear. It is actually contact wear, and no increase in cover grade will stop it. Where the chute geometry cannot be raised, a belt with lower ribs in the loading zone and taller ribs only further up the slope is sometimes the practical answer, and that is a conversation to have with the supplier before the order rather than after the first failure.

Notice what all four rows have in common. The belt is the messenger. If the structure and the loading were rebuilt around the rib pattern rather than around the flat belt that used to be there, most of these failure modes do not develop in the first place.

08Inspection Intervals, Spares and the Repair-or-Replace Decision

Ribbed inclines fail slowly and then quickly. The slow phase produces information that is almost free to collect, and the quick phase starts the moment a rib detaches. A rib that leaves the belt does not politely stop at the discharge; it travels into the chute, the cleaner, or the next transfer point, and on a 1,200 t/h circuit the downstream cost of one loose rib can exceed the price of the belt that shed it. That is the economic argument for a fixed inspection routine rather than reactive maintenance.

The route we recommend takes about twenty minutes on foot for a typical incline of 80 to 150 m. Walk the return strand first, because that is where ribs are visible from underneath without stopping anything, and look for lifted rib edges, dark wedge-shaped patches in corner pockets where sidewalls are fitted, and any fresh material on the ground. Then walk the carrying side from the tail to the head, recording rib height at one station every 10 m with a steel rule. A rib that has lost 5 mm of height is on its way to losing all of it. Then measure the splice, checking pitch continuity either side, plate condition where fasteners are used, and the free length still available at the take-up.

Rib height is the single best indicator of remaining belt life on a ribbed incline, and almost nobody tracks it. If you record rib height at the same station every month, the trend line tells you the replacement date with more confidence than any visual assessment. On a gravel incline we followed, ribs lost 0.6 mm per month consistently for fourteen months, then 2.1 mm in the following two months as the cover underneath began to break up. The change in slope of that curve was the signal, and a plant watching a monthly number would have ordered the replacement eight weeks before the failure that took the line down for two shifts.

The repair-or-replace decision follows from three numbers rather than from opinion.

Spares sit alongside this. A plant with a ribbed incline should hold a spare belt of the correct length, because ribbed belts are not cut and re-spliced on site the way a flat belt can be shortened at will. Not holding one means a 12 to 16 week wait from the point of failure, and the arithmetic rarely favours waiting. Where a line is critical, the sensible structure is one spare belt, one mechanical fastener set, one vulcanising kit appropriate to the belt width, and the roller and pulley spares that the inspection routine has historically flagged. The conveyor components list and the service scope both exist for exactly this reason, and a supplier who cannot tell you the expected lead time for a specific ribbed belt width is telling you something about their stock position. Order documentation, dimensional checks and the incoming inspection that supports all of it are set out under quality assurance, which is where most mismatch problems are caught before they become belt problems.

Two closing observations from the field. First, the plants with the fewest ribbed belt problems are not the ones with the best belts; they are the ones with the most boring maintenance records. Second, the single change with the biggest effect on failure rate is sequencing the shutdown so the belt stops empty. That costs nothing and removes the load case that causes most of the damage described in this article.

Send photos of the belt and the chute — we will identify the failure and quote the replacement

09Frequently Asked Questions

Why does my ribbed belt only slip on the first start after a shutdown?

Because the pockets are full when the drive is asked to move. On a slope, a stopped belt holds an entire bed of material leaning back against the ribs, and the drive has to accelerate that mass from zero in addition to the belt itself. The clue is start-up current sitting 30 to 60% above running current and easing within a few seconds. Sequencing the stop so the belt runs empty before shutdown removes the load case entirely, and a controlled start ramp helps, while extra tension does not.

How do I tell whether a peeled rib is a bonding fault or a geometry fault?

Look at where the lift starts and whether it repeats. A bond defect from manufacture tends to appear on many ribs, in varied positions, early in the belt's life. A geometry fault peels the same ribs in the same place repeatedly, usually immediately behind a splice or right after the loading chute, because a chute lip or a spindle that is too small for the rib height is bending the base of the rib on every pass. Re-bond a single rib once; if the same rib peels again, the geometry has to change.

Can I fit a mechanical fastener to a cleated belt without damaging the ribs?

You can, with two conditions. The plates stand proud of the cover, so every rib passing the joint region is locally stressed, and any cleaning blade in the line has to clear the fastener height or be counterweighted differently. Bolt torques must be re-checked about 48 hours after fitting because the plates bed into the carcass. The usual arrangement is vulcanised joints everywhere except one designated service point fitted with fasteners, so only one location needs special treatment at inspection.

What is an acceptable amount of carry-back on an incline?

Measure it rather than judging it. Lay a tarpaulin under the return strand for one full shift, weigh what lands on it, and divide by the tonnage fed in the same period. Below 0.4% is normal for a well-running belt. Above 1% is worth acting on, because it costs cleaning labour, drive power and belt life at the same time. On a 90 m incline with 240 t/h we measured 1.8 tonnes per shift, and most of it came from pockets deeper than 30 mm that had never been expected to discharge a damp feed.

Do training idlers fix mistracking on a ribbed belt?

They move the symptom, not the cause. Run the belt empty and watch the edge gap at three stations over ten revolutions. If the empty belt tracks straight and only the loaded belt drifts, the problem is off-centre loading at the chute, and training rollers will simply wear the belt into a new shape. If both the loaded and empty belt drift, look for a splice that is not square or a bracket that was re-welded out of position after a repair.

When is it cheaper to replace than to repair?

Three peeled ribs inside a single belt length, rib height down more than 40% of nominal at several stations, or any opening splice, and the arithmetic has turned. Repairs cost less per event but they leave loose fragments to travel into chutes, cleaners and rollers, and the downstream cost of one rib reaching a transfer point can exceed the price of the belt that shed it. Track rib height monthly at the same station; the trend line usually gives eight weeks of warning before the quick failure phase begins.