Bucket Elevator Belts: Construction, Bucket Fixing and Duty Matching

Bucket elevators look simple from the ground. A casing, a boot, a head, buckets bolted on at a fixed pitch, and a belt running up one side and down the other. Almost everything that decides whether the machine lasts, though, ends up in the belt itself, and that is where most premature failures begin. A bucket elevator belt is not a flat conveyor belt that happens to be mounted vertically. It carries the buckets, and the material sitting inside them, as a hanging load. It is pierced several hundred times for bucket bolts. It runs over deliberately small pulleys, sized that way so the buckets can throw their load cleanly at the head. Change any one of those conditions and you are no longer specifying a general-purpose belt.

Builders who make both products say the same thing in different words. A wide flat belt on a troughing line can be a modest multi-ply build and still run for years, because the idlers share the load and the material is supported between them. Put that same construction into an elevator leg lifting wet sand 28 m and it will creep, walk out of the boot pulley, and start shedding buckets long before the cover is worn through. The difference is not cosmetic. It is the ratio between working tension and breaking strength, the amount of stretch the take-up has to absorb, and the way load enters the belt at every bolt hole.

If you are comparing candidate designs and want the range of carcass and cover options in one place, start from the lines we keep for conveyor belt supplier enquiries, then come back here for the elevator-specific selection logic. Everything below assumes the machine is an elevator leg, not a horizontal conveyor. Where the two overlap, and they do overlap more than most buyers expect, the overlap is written down explicitly.

Send us your elevator duty data and we will size the belt build to suit it

01Why an Elevator Belt Is Specified Differently From a Flat Belt

The mechanical argument comes down to load path. On a horizontal conveyor, the belt is a moving floor under the material, and the tension along it is mostly a function of friction between the belt and the load. On an elevator, the belt is a load-bearing member. Every bucket plus its contents hangs from the belt, and the tension grows roughly linearly from the boot to the head, so the section just below the head pulley works hardest all day. A belt that is 100 % utilised at the head is normal on an elevator, whereas a flat belt usually runs with a comfortable margin because it also has to survive impact at the loading point and misalignment on the idlers.

Stretch is the second difference, and it is the one that catches buyers out. A take-up on a flat conveyor has a short travel and only needs to absorb thermal movement and a small amount of running-in stretch. An elevator take-up, whether it is a screw type in the boot or a gravity counterweight in a tower, must absorb the entire elongation of a belt that may be more than 60 m long. Every 0.1 % of extra elongation on a 60 m centre becomes 60 mm of take-up travel that has to exist. If the design elongation is understated, the weight carriage runs out of stroke, tension drops, and the belt slips on the boot pulley under a full bucket load. That sequence is common enough that we now ask for centre distance and available take-up travel before we discuss ply count at all.

Third, the belt is perforated. Flat belting is spliced and left otherwise intact; elevator belting has a disciplined hole pattern down both edges, one row per bucket. Each hole is a stress raiser, and each row of holes reduces the effective width of the belt carrying tension across that station. That is why an elevator belt often needs more plies or a higher-strength carcass than a flat belt of the same width handling the same throughput. On one grain leg we inspected, the belt was correct in width and ply but the buckets had been re-drilled 18 mm inside the original pitch line, and the new holes sat too close to the edge. Twelve months later the belt had a continuous tear line running between the old and new rows.

Fourth, tracking is corrected differently. A flat conveyor steers itself with training idlers and crowned pulleys, and a little edge damage from mistracking is survivable. An elevator leg has almost no room for side movement. The belt has 10 to 20 mm of clearance to the casing on each side, the buckets add inertia, and any lateral drift is contact, abrasion, and sparks. Crowning is limited because the buckets must clear the pulley. So the belt build itself, the splice squareness, and the pulley alignment do the steering instead. Anyone who has wrestled a misaligned leg knows how little patience the machine has for apologies.

A note on scope, because this matters for anyone reading both our elevator material and our flat-belt material. Chevron and cleated tops, which help a flat belt carry material up an incline, are not the same thing as buckets; they lift on a slope and they do not carry a hanging load. Sidewall belts, which stop material spilling off the edges, also work under a completely different tension profile. Those products are covered elsewhere on this site, and the selection rules do not transfer. What does transfer is the habit of matching a belt to a measured duty rather than to a catalogue heading.

The table is a starting point, not a specification. What actually drives the build is the tension the buckets create, and that is where the next section begins. Buyers who want the broader picture of how a belt body is assembled, and how covers and plies are described commercially, will find the general explanation in our belt construction overview, while the notes below stay strictly on the elevator side of the line.

02Carcass Construction, Elongation and the Tension the Belt Must Hold

An elevator belt is a sandwich of load-carrying fabric or cords, a skim layer that keeps those elements together, and two covers that face entirely different conditions. The top cover, the one the buckets sit against, has to resist abrasion from bucket backs and from material trapped between the bucket and the belt. The bottom cover, the one that touches the pulley, has to resist the repeated flexing that comes from running around a small diameter, sometimes a thousand times an hour on a short leg. A build that is generous on one side and thin on the other is an unbalanced design, and the thin side will decide the belt's life.

Textile carcasses, usually polyester warp with polyamide weft, remain the default for most elevator legs because they tolerate a small pulley, they resist impact, and they are easy to splice on site. The trade-off is stretch. A textile belt under working tension will extend noticeably, and the extensions are permanent in part: some of the stretch comes back when tension is released, and some does not, because fibres creep. On a 45 m cement elevator we monitored through commissioning, a four-ply polyester-nylon belt settled at roughly 1.1 % total elongation after eight weeks of running, and roughly two thirds of that was the non-recoverable part the take-up had to absorb permanently. That is why we now ask for the take-up travel figure at the quotation stage, not after the first belt change.

Steel cord construction changes the picture completely. Because the load is carried by steel filaments rather than woven fabric, working elongation drops by roughly a factor of three to five, and the belt holds its length far better over a long service life. On very tall legs, above roughly 60 to 80 m of lift, or on high-capacity legs where the belt width is limited by the casing, steel cord becomes the practical answer. It also lets you use a thinner belt for the same strength, which reduces the bending stress on the small head pulley. Two cautions come with it. The splice is a different operation, requiring controlled curing and careful cord spacing, and a steel cord belt has less tolerance for impact damage because a single broken cord concentrates stress on its neighbours. Buyers evaluating that family should read our steel cord belt page and the steel cord selection guide before comparing it with a textile build for an elevator.

Ply count is the coarse adjustment and carcass grade is the fine one. Adding a ply raises breaking strength by roughly the strength of that ply, but it also stiffens the belt and increases the bending load on the pulleys and shaft. There is a real ceiling here. Past five or six plies, the belt often stops behaving like a belt on a small pulley and starts behaving like a plank, and the splice becomes harder to close properly. When a leg is short but heavily loaded, strength is usually not the binding constraint at all; wear resistance at the buckets and flex fatigue around the boot pulley are. That is a different specification from the one a strength calculation alone produces, and it is a conversation worth having before the drawings are frozen.

Elongation limits are usually written into an elevator purchase as two numbers, and both matter. The first is elongation at a defined reference load, which tells the engineer how much take-up travel to provide. The second is elongation at break, which tells the maintenance team how much warning they get before a belt fails. A belt with 0.4 % working elongation and 3 % elongation at break is a very different animal from one with 2 % working elongation and 3 % at break, even though their breaking strengths may be identical. The first will run a long leg with a compact counterweight tower; the second will need a deep boot and a generous take-up, and it will still wander in length as the load changes.

Covers deserve their own paragraph, because they are where most buyers economise first and regret it later. The pulley-side cover on an elevator sees continuous flexing, so a hard, highly abrasion-resistant compound that cracks under repeated bending is the wrong choice there even though it would be ideal on a flat belt carrying sharp rock. The bucket-side cover faces abrasive sliding from material trapped behind the bucket backs, so it wants abrasion resistance and enough elasticity to close around bolts. Different trades. A single compound that serves both is often the sensible engineering answer, and our colleagues describe the chemistry and the trade-offs on the industrial conveyor belt and rubber conveyor belt pages, which together cover cover compounds and how they are graded commercially.

One more construction detail that gets overlooked. The belt edges of an elevator belt are not decorative. They take the bucket corner bolts in many designs, they rub the casing when tracking drifts, and they are the first place a longitudinal tear becomes visible. A moulded or well-cut edge with the fabric sealed resists water ingress far better than a raw cut edge, and on outdoor legs in wet climates that single detail often decides whether you get four years or eight out of a belt. When we replace an elevator belt that failed from edge delamination while the rest of the body was still sound, the edge finish is usually the first thing we photograph.

03Bucket Fixing: How the Attachment Method Loads the Belt Body

Buckets are attached in one of three ways, and the choice changes the stress pattern in the belt more than most engineers expect. The oldest is a bolted fixing, where each bucket carries two or more bolts through punched or drilled holes in the belt, usually with a plate or a large washer on the pulley side. The second is a riveted or staked fixing, used mostly on lighter duty legs where speed is low and the buckets are small. The third is a clamp-and-bracket arrangement, in which the bucket hangs from a bracket that is itself bolted through the belt, spreading the load over a longer section of a thicker belt. All three work. They fail in different places.

Bolts are the default because they are repairable. A damaged bucket can be unbolted and replaced without touching the splice. But bolts concentrate load. Under a full bucket, the whole weight of bucket and material is transferred to the belt through a handful of holes, and the belt material immediately around each hole carries a stress concentration of roughly two to three times the nominal tension in that region. That is the mechanical reality behind the familiar failure picture: a hole elongating into an oval, then a short crack from the hole edge running with the belt, then a tear line that travels down the leg. Bucket bolts do not cause belt failure by breaking. They cause it by opening holes.

The pitch line position matters as much as the pitch itself. Bucket pitch, the distance from one bucket centre to the next, is fixed by the pulley diameter and the discharge requirement, and it must be punched accurately, because a bucket that sits a few millimetres out of position pulls its neighbours. The pitch line, which is the distance from the belt edge to the centre of the bolt holes, is a separate decision with real consequences. Holes set too close to the edge leave only a narrow strip of belt outboard of the bolt, and that strip tears off first. Holes set too far inboard reduce the width of belt available to carry tension between the two rows, which raises the tension per unit width and hastens fatigue. There is a workable band in between, and it is the belt builder, not the bucket supplier, who should confirm it.

On a fertiliser plant leg we surveyed, the maintenance team had standardised on one bucket size across three different elevators. Two of the legs were fine. The third, the tallest one, had been running for eleven months when the belt developed a tear along the outboard row of holes. The buckets were correct for the duty but the belt was a lighter build than the other two machines, and the same hole pattern that was comfortable on a 30 m leg was too aggressive on a 52 m leg. Nothing was defective. The pattern simply had not been recalculated when the third elevator was specified, and the belt paid for it.

Fastener hardware deserves a mention because it is cheap to fix and expensive to ignore. Bolts that are too long bottom out in the bucket and leave the head proud on the pulley side, where it scrapes the pulley lagging and starts a groove. Nuts that work loose drop into the boot and jam the pulley, and a jammed boot pulley will burn a belt through in minutes at full tension. Countersunk heads, properly sized backing plates, and a torque discipline at every bucket change are the whole of the defence. On high-speed legs we also recommend a thread-locking compound rather than a split washer, because washers deform and lose preload after a few thermal cycles. What the belt asks for, above all, is that the bucket sits flat against it. A bucket that sits on a burr or a weld spatter concentrates load on two bolts and leaves the rest doing nothing.

Spacing is a related variable, and it is where a lot of energy can be saved or wasted. Closer bucket spacing raises the capacity carried by a given belt length but also increases the number of holes, the weight of the bucket chain, and the tension the belt must hold. Wider spacing reduces hole count and chain weight but can leave the belt unsupported across a long span, particularly on the return side where the buckets hang off the belt. A useful check at the design stage is to compare the weight of the empty bucket chain against the weight of the material being lifted. On grain legs we commonly see the empty chain at 15 to 25 % of the material load; on heavy mineral legs with thick buckets it can reach 40 %, and at that point the belt is being sized largely to lift its own hardware.

Whatever method is chosen, the belt builder and the bucket supplier should exchange drawings before either commits. Hole diameter, pitch, pitch line, bolt grade and the flatness of the bucket back are all interface dimensions, and they belong to the pair of suppliers jointly. We would rather see a slightly more expensive bucket with a properly stiffened back than a cheap bucket that deforms and starts working its bolts loose. Buyers assembling a full elevator package, belt, buckets, pulley and casing together, should look at how we describe the interfaces on the conveyor components and conveyor brackets pages, and ask the conveyor belt manufacturer to confirm the punching tolerance at the same time as the strength grade. A punching tolerance that drifts by 5 mm across a belt length is enough to make bucket fitting a fight on the second half of the roll.

One practical note from commissioning work. Fit the first ten buckets, tension the belt, run the leg for an hour empty, then re-check every bolt with a torque wrench before loading. In our experience a third of the fasteners will have relaxed measurably in that first hour as the belt bedding-in stretch settles and the holes seat. Re-torquing after a week, and again after the belt has carried its first full load for a shift, closes the loop. Maintenance crews who skip those two checks tend to find the loose bolts later, in the boot, mixed in with the material.

For a belt supplier or an OEM assembling an elevator package, the same interface discipline applies at the factory level. This is the kind of detail where a conveyor belt factory that punches holes from a controlled template, rather than marking them out by hand, gets a noticeably better result, because hole position tolerance is a manufacturing process, not an inspection activity.

04Damage Mechanisms: Bolt-Hole Tearing and Longitudinal Rip

Almost every elevator belt that fails before its time fails by tearing along its length, not by snapping across its width. That single fact should shape how you inspect a leg. A transverse break means the belt was genuinely overloaded or the splice was defective. A longitudinal tear means something sharp travelled with the belt, or a hole grew, or an edge was rubbing, and the belt then unzipped itself along a line of least resistance. The two failure families have different causes and different prevention, and mixing them up in a maintenance plan wastes effort.

The classic tear starter is a bucket corner. A bucket that has lost a bolt, or was fitted with a bent back plate, tilts slightly on the belt and its corner digs into the casing or into the boot pulley. From that point the corner acts as a blade on the belt surface, and because the belt is under high tension, the cut propagates quickly. The same thing happens when a bucket is fitted with the wrong pitch and touches the pulley before the belt does. We have seen a 950 mm wide wheat belt destroyed in under four hours by a single displaced bucket, and the operator heard nothing over the sound of the leg, because an elevator leg is a noisy machine and a small change in tone does not carry.

Foreign objects are the second starter and they are unavoidable in the industries that use elevators. A length of rebar, a broken bucket lip, a piece of wear plate, a wooden pallet fragment, any of them can lodge between the belt and the boot pulley and start a cut that runs the full height of the leg. The engineering answer is not to keep foreign objects out, which is impossible, but to stop the belt before the damage runs far. Devices that detect a drop in belt tension, a sudden change in pulley speed, or the presence of material in the boot sump will catch most of these events. On a port terminal leg handling clinker, a speed monitor on the boot pulley paid for itself in one incident by tripping the drive four seconds into a jam.

Hole growth is the slow, quiet starter, and it is the one most specific to elevator belting. A bolt hole that is slightly oversized, or a bolt that works loose and hammers its hole, gradually elongates and the belt material around it fatigues. Eventually a crack forms at the widest part of the oval and runs with the belt. Because the crack follows the line of holes, the resulting tear often sits at a constant distance from the edge, which produces a narrow strip of belt that hangs on by a thread and then comes away. In one plant survey we found that four of seven premature elevator belt removals had started at a single bucket station, and in three of those the initial hole oval was visible on the belt before the tear. It was simply never recorded on an inspection sheet, because nobody had been asked to look for it.

Delamination and edge damage are the third route and they look different again. Water gets into a cut edge, the fabric-to-rubber bond softens, and the cover lifts away in patches; once the carcass is exposed, abrasive material works into the plies and the belt thins from the inside. This is not a fast failure, so it rarely causes an unplanned stop. It causes a slow loss of strength that shows up as excessive elongation and a belt that suddenly needs take-up adjustment every month instead of every year. Outdoor legs in humid climates and legs handling wet material are where this pattern dominates, and it is the reason edge quality deserves the attention we give it on the impact and cut resistant belt description and on the plainer abrasion resistant belt page, both of which explain how a tougher cover helps, and why it does not help if the edge is left raw.

Heat and chemical attack run alongside these mechanical causes on certain duties. A leg lifting hot clinker or dried fertiliser at 90 to 120 °C will age the belt from the pulley side first, because that side is flexed and the cover is thin. Acid or alkaline residues left on the buckets will attack the bolts and eventually the belt beneath them. Neither is a tearing mechanism in its own right, but both soften the belt until a tear that would once have arrested itself keeps going. When we are asked to review a leg that has had two belt failures inside three years, the history of temperature and residue is the first thing we ask about, well before we ask about ply count. Buyers who want the wider picture of inspection practice and service intervals will find useful background in our notes on belt inspection and lifespan, and should also look at what wholesale conveyor belts ranges include in the way of repair materials, because a leg that can be repaired in place is worth more than one that cannot.

Notice that five of the six mechanisms in that table are detectable in the field before the belt is lost. What usually fails is not the belt but the inspection routine. An elevator leg is enclosed, so nobody sees it running; the belt is only accessible during a planned stop; and unless somebody is given a specific list of things to look for, the stop is spent changing buckets and greasing bearings. Two photographs of the belt edge at each bucket station, taken through the inspection door every month and filed, will tell you more about belt condition than any single measurement, because the trend is what matters and the trend only exists if you keep the record. Keeping a small stock of repair materials helps as well, and a local conveyor belt distributor can often supply fasteners, patching stock and replacement lagging while a new belt is still on order, which is usually enough to keep a leg running until the planned change arrives.

05Duty Matching: Material, Lift Height and Bucket Speed

Selecting an elevator belt properly means resolving three variables together, because each one changes the others. What the material does to the belt decides the cover. How high the leg lifts and how fast the buckets run decide the tension and the flex duty. How much material must move per hour decides the bucket size, which decides the hole load, which loops back to the belt build. Get the order wrong and you end up with a belt that is generous where it did not need to be and thin where it mattered.

Start with the material. Fine, dry, free-flowing products such as grain, seed and prills are gentle on the belt surface but they also escape through every gap, so boot sealing and belt cleanliness matter more than abrasion resistance. Abrasive, angular products such as clinker, crushed stone and mineral powder wear both the buckets and the belt's bucket-side cover, and they also tend to pack behind the bucket backs where they grind against the belt. Wet or sticky products hang on to the belt after discharge and return down the leg inside the buckets, which doubles the effective load; a leg designed for dry sand that is later fed damp sand will see bucket weights rise by 40 % or more and belt tension with it. Corrosive products attack fasteners first, which then damages the belt at the bolt hole. Each of these pushes the specification in a slightly different direction, and a belt chosen for one duty is often a poor compromise for the next.

Lift height is the variable buyers most often understate, because the belt length is what matters and the belt length is more than the lift height. A head pulley at 40 m above the boot pulley means roughly 80 m of belt in the loop, plus the wrap around the pulleys, plus whatever allowance the take-up requires. Every bit of that length elongates. On a project in a river port we were asked to quote a belt for a leg described as 32 m; the actual belt length, measured off the pulley centres and the path, was 71 m. The difference was almost enough to change the ply count, and it did change the take-up requirement significantly. Always quote belt length and take-up travel, not just centre height, and ask the same of anyone who quotes a belt to you.

Bucket speed is the third variable and it carries a subtle penalty. Elevator buckets discharge by centrifugal action, and the speed needed to throw material clear depends on the pulley diameter and the bucket shape. Run too slowly and the bucket carries material back down the leg, which increases load without increasing output. Run too fast and every bucket starts to fling, the return side of the belt fills with airborne material, and the belt is loaded and unloaded thousands of times an hour with a shock at each bucket. In practice most industrial legs run somewhere between about 1.0 and 2.0 m/s at the bucket line, with grain legs at the higher end and heavy clinker legs lower, and the belt specification should reflect which end you are on. High speed shortens the time available for the belt to flex around the pulley, and flex fatigue moves up the list of design concerns accordingly.

Then there is the drive end, which is a separate belt system on the same machine and one that is often forgotten when an elevator is upgraded. An elevator drive is usually a gearbox with belts between the motor and the reducer or between the reducer and the head shaft. Change the head shaft speed, or increase the motor size for more capacity, and the drive belt selection has to change with it. Fewer, thicker belts running faster are not interchangeable with more, thinner belts running slower, even if both transmit the same power, because they behave differently under shock loads and they wear at different rates on a small pulley. Buyers who are modifying an existing leg should look at what a transmission belt manufacturer offers in matched sets and at the range carried by a V-belt manufacturer before assuming the existing drive arrangement can simply be uprated, and should also check whether a synchronous drive would suit better by reading the timing belt range as a comparison.

Where the site is a specific industry rather than a specific material, it usually helps to read how the duty is described for that sector, because the same belt may be sized quite differently in a quarry than in a food plant. Our sector notes for cement plants, mining and quarrying, port bulk handling, food and packaging, recycling and logistics and warehousing all describe the working conditions that feed into an elevator belt decision, and they are worth reading alongside the material table above. Special cover families such as heat resistant belting, oil resistant belting and chemical resistant belting exist because a single general-purpose compound cannot cover every residue, and it is worth checking whether your material actually requires one of them rather than assuming it does not.

A final word on how these choices interact. You cannot select the cover from the material and the carcass from the height independently and then expect the result to be right. A very abrasive material forces a thicker cover, a thicker cover changes the pulley diameter needed for the same flex stress, and a larger pulley changes the discharge trajectory, which changes the bucket spacing. The chain is connected. When we quote an elevator belt, the material, the belt length, the bucket pitch and the pulley diameters all come into the calculation together, and if any one of them is missing we ask for it rather than assuming a value.

06Tracking, Slippage and Take-Up: Problems Unique to a Vertical Leg

An elevator leg gives you very little room to correct a belt that has decided to misbehave. Clearance to the casing is measured in tens of millimetres, the buckets make the belt stiff in the lateral direction, and there is no training idler to nudge the belt back. That is why most tracking problems on an elevator are solved before the belt is installed, by making the splice square, aligning the head and boot pulleys in both planes, and making sure the bucket holes are punched to a single datum rather than marked from whichever edge was nearest.

When a belt does drift, the direction of the drift tells you where to look. A belt that moves consistently to one side over the whole run, at every position of the splice, is almost always a pulley alignment problem, and the fix is to shim or adjust the pillow blocks rather than to touch the belt. A belt that swings sideways once per revolution, returning to centre in between, has a splice that is not square; the belt is being steered by its own diagonal joint. That is a splice fault and it will not improve with tension. Occasionally a belt wanders only in the region of a few buckets, which points to a distorted bucket or a bucket fitted with the wrong length bolts on that side, and the belt is being deflected rather than steered.

Slippage is different from mistracking and it has different causes. If the belt slips on the boot pulley when the leg is loaded but runs correctly empty, the problem is almost always tension rather than friction. The take-up has run out of stroke, or the belt has elongated more than the design allowed, or the counterweight is too light for the bucket chain that was actually installed. Boost the drive tension without addressing the take-up and you will simply wear the boot pulley lagging faster. On one clinker leg the maintenance crew responded to slip by increasing counterweight mass twice, until the boot shaft failed. The belt had been running beyond its elongation limit for months and the extra tension shortened the shaft's life instead of solving the belt's problem.

Slip that occurs only at start-up, and only in cold weather, has a different explanation and a different answer. Lubricant in the gearbox thickens, the drive cannot accelerate the loaded bucket chain quickly enough, and the belt skids on the boot pulley for a second or two at each start. Repeated cold starts will polish the pulley lagging, and once the lagging is polished the slip reappears in warm weather too. Soft-start arrangements, or a slightly longer acceleration ramp, solve this far better than additional tension, and they cost nothing to run afterwards.

Take-up design deserves more thought than it usually gets. A screw take-up in the boot is simple and cheap, and it works well on short legs where the total elongation is small. It is a poor match for a tall leg, because adjusting it requires stopping the machine and the adjustment is discrete. A gravity take-up with a counterweight keeps tension constant automatically as the belt elongates, but only if there is enough free travel in the frame, and only if the counterweight is calculated for the belt that was actually supplied rather than the one in the original drawing. We have seen gravity take-ups installed with 600 mm of nominal travel, of which 550 mm was consumed by the belt's running-in elongation within the first three months. From that point on the machine was effectively a screw take-up with no screw. Whatever the arrangement, record the operating position of the take-up when the belt is new and measure it monthly. That single number is the most useful indicator of belt condition you can collect on an elevator.

Two components on an elevator deserve specific attention because they interact directly with the belt. The boot pulley must be lagged and the lagging must be kept dry and free of caked material; a smooth, caked boot pulley has a fraction of the grip it needs, and no amount of tension will restore it. The head pulley, meanwhile, sets the flex stress the belt sees at every rotation, so a head pulley that has been re-lagged to a smaller effective diameter, which happens when a worn face is machined back, effectively increases the bending on a belt that was already working hard. Our notes on conveyor pulleys explain diameter selection and lagging choice, and the discussion of self-aligning devices in tracking and downtime is a useful reminder of how much trouble a small alignment error can cause when there is no redundancy in the system.

07Failure Modes and Field Inspection Points Unique to Elevator Belts

An elevator belt has a handful of failure modes that simply do not exist on a horizontal conveyor, and a maintenance plan that only checks the things a flat-belt inspection covers will miss all of them. There is no idler to inspect, no training device to adjust, and no loading skirt to re-line. What there is instead is a bucket chain, a boot that fills with spillage, and a belt that is perforated hundreds of times. The inspection list has to be rewritten around those realities rather than copied from a belt conveyor checklist.

This is also where the boundary with the roller and idler family becomes clear. If you are diagnosing a troughing line, the culprits are usually rollers, idlers, and their bearings, and the pages on conveyor rollers, troughing idlers and the broader roller selection guide are the right place to start. If you are diagnosing an elevator leg, none of those components exist in the machine, and the failure is almost always in the belt, the buckets, or the boot. Borrowing a flat-belt diagnosis for an elevator problem is one of the most common reasons a repair does not hold.

Start the inspection at the boot, because that is where the machine tells you the most. Material depth in the boot sump above the level of the pulley is a serious signal, not a cosmetic one; it means either the discharge is backing up, or the buckets are carrying material back down the return side, or the boot sealing has failed. Any of those conditions increases the load the belt is carrying and puts abrasive material in contact with the belt and pulley. Then check the belt edges through the inspection door, moving the belt a bucket station at a time. What you are looking for is a polished or shiny strip along the edge, which means contact with the casing; a dark line of fretting dust around bolt heads, which means loose fasteners; and any short crack that runs parallel to the belt edge rather than across it. Those last two are the highest-value findings on the whole machine.

At the head, look at the discharge trajectory rather than the belt. If the material is not throwing clear, buckets will carry it back and re-load the return side. Then inspect the pulley side of the belt across the flex zone, at the point where it bends around the head pulley, for fine crazing. On hot duty legs this is where heat ageing shows first, and it appears long before the bucket-side cover looks worn. Finally, count take-up stroke position and compare it with the last record. A change of more than a few millimetres a month on a belt that is several years old is worth investigating, because it usually means the belt is now working at a higher tension than it was designed for.

The buckets themselves matter as much as the belt, and so does the dust they generate. A leg that is producing fine black rubber dust steadily, rather than the coarse dust that comes from material handling, has buckets rubbing the belt or slipping on the boot pulley. A leg that produces no dust at all but whose belt is elongating is running a clean but overloaded duty, and the belt will fail from fatigue rather than abrasion. Reading the dust is not superstition; it is a cheap way of separating two problems that look identical from the outside of the casing.

One thing is worth saying plainly about inspection economics. The checks that matter on an elevator belt take two people under an hour, and they need a torque wrench, a tape, a light and a camera. Nothing on the list requires an instrument that most plants do not already own. The reason these checks get skipped is not cost. It is that the leg is enclosed, it is noisy, and it keeps running, so there is never a felt urgency until the belt comes out in one piece. A short, fixed checklist attached to the machine, with a place to write the take-up figure, converts that situation into something a shift fitter can do without being asked.

08Specifying, Ordering and Changing an Elevator Belt

Every elevator belt enquiry we receive falls into one of two categories. The first arrives with a full set of machine data and can be quoted properly the same day. The second says only the belt width and the word elevator, and it takes three rounds of questions before anyone can size anything. The difference is not how experienced the buyer is. It is whether somebody took the time to write the machine down. The list is short, and the order matters less than having all of it.

What the belt builder needs is the belt width, the belt length measured around the path rather than the centre height, the bucket pitch and the pitch line, the bucket size and mass, the material and its bulk density both dry and wet, the required tonnage, the head and tail pulley diameters, the bucket speed, the available take-up travel, the maximum material temperature, and whether the leg is indoors or exposed. Any of those left blank becomes an assumption, and assumptions on an elevator belt are expensive because the belt is the component that gets changed last and costs the most downtime to reach. When we were asked to quote eleven belts for a group of grain silos, four of the machines turned out to have been re-rated for a higher tonnage years earlier without any change to the belt drawings. The enquiry brought that to light, and two of those belts were undersized for the duty they had been running.

Cost, when we discuss it, is driven by a short list of factors rather than by a single number per length of belt. Ply count and carcass material set the base. Steel cord carcasses cost more than textile ones and change the splicing requirement with them. Cover thickness and compound choice add to the material content, and a heat or chemical resistant compound adds further. Hole punching is a manufacturing step with its own cost, and a non-standard pitch or pitch line is a tooling change. Finally, lead time and order quantity matter, because a single replacement belt cut to an unusual length is not produced at the same rate as a batch of standard builds. Buyers who want to move a project forward quickly should expect the conversation to centre on those five or six variables rather than on a headline figure, and our service team is used to working through them item by item.

Documentation and verification deserve a paragraph of their own, because they are what protect both sides after delivery. A belt for an elevator should arrive with its dimensions checked against the order, the hole pattern inspected against the approved drawing, the splice marked or supplied as an on-site kit with instructions, and a record of the belt's construction and batch. On the receiving side, the sensible checks are belt length, width, thickness at a defined point, hole pitch measured over at least twenty pitches rather than two, hole diameter at several stations, and the edge finish. Measurement over a long run of holes is the one that catches real problems. A belt that is one millimetre long over a single pitch looks acceptable and becomes unusable over three hundred of them. Our quality assurance notes describe how those checks are handled at the producing end, and the same discipline applied at goods-in will save an argument later.

Changing a belt is a planned job, not an emergency, and it is worth planning properly. Measure the old belt before it is removed, so the new one can be made to a verified length rather than to a drawing that may already be wrong. Note the take-up position with the old belt on, so the new belt can be tensioned to the same datum and the difference recorded. Inspect the head and boot pulleys while the belt is off, because a worn pulley is the most common reason a new belt fails quickly. Check the buckets for distortion and replace the damaged ones before they touch a fresh belt. Then tension, run empty for an hour, re-torque, and only then load up. That sequence, followed in that order, is the cheapest insurance available on an elevator leg, and the legs that skip it are the ones we hear from again three months later. Buyers who need the full product range to compare against a proposal can review the products page descriptions before deciding what to order.

Talk to us about your elevator belt duty

09Frequently Asked Questions

How is an elevator belt different from a flat conveyor belt of the same width?

The load path is the fundamental difference. On a troughing line the material rests on the belt and is driven by friction, while on an elevator every bucket and its contents hang from the belt as a tensile load. That means the belt in a leg typically works at 50 to 80 percent of its rated strength near the head. It is also perforated twice per bucket for the full belt length, it runs over much smaller pulleys, and it has almost no room to drift because casing clearance is only tens of millimetres. The same width and the same nominal strength do not make the two interchangeable.

Do I need a steel cord carcass for a tall elevator leg?

Not always, but the case for it strengthens with height. Above roughly 60 to 80 m of lift, or where a high tonnage has to pass through a casing that limits belt width, the very low working elongation of a cord carcass usually justifies itself, because the take-up tower can be shorter and the belt keeps its length over a long life. Below that, a four-ply textile build with a properly engineered splice is often the more practical answer. The trade-off is that cords demand a controlled on-site splice and tolerate impact damage less well than fabric.

Why do the bolt holes in my belt keep getting longer?

Almost always because the bucket is moving. A fastener that has relaxed its preload lets the bucket rock slightly, and the bolt then hammers the hole a little on every revolution. Black fretting dust around the bolt head is the giveaway, and an oval hole is the next stage. The hole itself is not the problem; a hole only grows where the load around it is no longer evenly shared. Re-torquing to a fixed value, fitting plates or clamp bars on the pulley side, and checking that buckets sit flat against the belt will stop the process. Punching larger holes to make assembly easier makes it worse.

My belt slips when the leg is loaded. Should I add more counterweight?

Check the take-up stroke first. Slip under load that disappears when the leg runs empty is normally a tension problem rather than a friction problem, and the usual cause is that the take-up has run out of travel because the belt has elongated more than the design allowance. If that is what has happened, adding mass to the counterweight simply raises the tension on a belt that is already working too hard, and the next failure will be a bearing, a shaft or a pulley rather than a belt. Shortening the belt by re-splicing, or planning a lower-elongation replacement, addresses the actual cause.

How often should the bucket bolts be re-torqued?

Three times at the start of a belt's life and then on a fixed interval. Re-check every fastener after the first hour of empty running, because a significant share will have relaxed as the holes seat and the belt beds in. Check again after a week, and once more after the leg has carried a full load for a shift. After that, a torque check at every planned stop is enough, provided the maintenance record shows the earlier checks were done. Legs where bolts are only looked at when something falls into the boot are the ones that lose buckets.

Can I convert a horizontal conveyor belt into an elevator belt by punching holes in it?

Only in an emergency, and it should never be a planned method. A flat belt is generally built for a lower working tension and a larger minimum pulley diameter than an elevator leg imposes, and punching holes removes width from the section carrying tension between the two rows. Holes drilled by hand also have far rougher edges than a factory punch, and those edges start tears within weeks. If a leg has to be brought back into service quickly, a factory-punched belt made for the duty is the safer route, even if it takes longer to arrive than a length cut from flat stock on the shelf.