What is a belt weigh feeder

 Belt Weigh Feeder: 

How It Works and Why It Goes Wrong

If you've spent any time around a cement plant, you already know the belt weigh feeder isn't a glamorous piece of equipment. Nobody stops to admire it. But ask any process engineer what keeps the kiln stable, the raw mix on-ratio, and the quality lab happy, and the weigh feeder is almost always part of that answer. Get its feed rate wrong by even 2-3%, and you'll see it show up downstream — in free lime, in LSF, in coating buildup, in fuel consumption. This article walks through how these feeders actually work, and then gets into the faults that field engineers deal with most often — the kind you don't always find neatly written up in the OEM manual.

The Basic Idea

A belt weigh feeder does two jobs at once: it moves material, and it weighs it while doing so. That sounds simple, but combining continuous transport with continuous, accurate weighing is where all the engineering complexity lives.

The core components are:

A short conveyor belt running at a controlled, usually variable, speed

A weighing platform or weigh bridge — one or more idlers mounted on load cells, sitting under a defined section of belt

A speed sensor (tachometer or encoder) coupled to the belt or the driven pulley

A load cell that measures the actual weight of material sitting on the weigh span at any instant

An integrator/controller that pulls the weight and speed signals together

The working principle itself is straightforward multiplication:

Mass flow rate = Belt Load (kg/m) × Belt Speed (m/s)

The load cell tells the system how much material (in kg per meter of belt) is sitting on the weigh span. The speed sensor tells it how fast that belt is moving. Multiply the two, integrate over time, and you get an instantaneous flow rate in tons per hour, plus a running totalized value in tons. The controller closes the loop by adjusting belt speed (via a VFD) to hold the setpoint flow rate steady, regardless of how the material bed depth might be shifting upstream.

It's the same principle whether you're running a Schenck, Hasler, K-Tronics, Thayer, or Merrick feeder — the mechanical execution differs, but load × speed = mass flow is the constant underneath all of them.

Why Calibration Matters So Much

A weigh feeder is only as good as its calibration. Two checks matter here, and they get confused with each other more often than they should:

Zero calibration — done with an empty belt running, to cancel out belt weight, idler friction, and mechanical offset

Span/material calibration — done with test weights or chain weights applied to the weigh span, to confirm the load cell output matches a known reference mass

Static test weights confirm the load cell and electronics are reading correctly. But they don't confirm dynamic accuracy — belt tension, tracking, and material distribution only show up once the feeder is actually running. That's why a proper calibration routine includes a material test (drawdown or bin-loss-in-weight comparison) in addition to the static check, not instead of it.

Common Faults — and What's Usually Behind Them

Here's where the real troubleshooting knowledge lives. Most weigh feeder problems fall into a handful of recurring categories.

1. Zero Drift

The feeder reads a false load even with an empty belt. Usually traced to:

Material buildup on idlers, skirt boards, or the belt underside

Belt tension changes as the belt ages or stretches

Temperature effects on the load cell (more common in kiln-area feeders than raw mill feeders)

Moisture or condensation getting into the load cell junction box

A quick housekeeping check — clean belt, clean idlers, dry junction box — resolves more zero drift complaints than people expect.

2. Span/Accuracy Errors

The feeder is consistently reading high or low against a known reference (weighbridge, sampling, or lab check). Common causes:

Load cell calibration has genuinely drifted and needs re-verification

Belt is mistracking, so material isn't centered on the weigh span the way it was during calibration

Idler alignment has shifted, adding friction that the system misreads as load

Speed sensor slippage — if the tach wheel or encoder coupling is slipping, the speed signal is wrong even though the load reading is fine, and your mass flow calculation is off regardless

This last one catches people out because the load cell looks perfectly healthy on diagnostics, yet the totalized output is still wrong. Always check speed feedback integrity, not just the load cell, when accuracy complaints come in.

3. Erratic or Noisy Weight Signal

The reading jumps around instead of tracking smoothly. Usual suspects:

Mechanical vibration from nearby crushers, mills, or fans coupling into the weigh frame

Loose or worn idler bearings on the weigh span

Belt splice passing over the weigh span at a fixed interval — this shows up as a repeating spike pattern, which is a good diagnostic clue in itself

Electrical noise on the load cell signal cable, especially if it's been run alongside VFD power cables without proper shielding

4. Belt Tracking and Mechanical Issues

A feeder that mistracks doesn't just wear the belt edges — it directly corrupts the weight reading, since load distribution across the belt width changes. Tracking problems usually come back to worn idlers, an out-of-square frame, or material spillage building up unevenly on one side.

5. Material Behavior Problems

Not every fault is electrical or mechanical. Sticky, moist, or segregating material can choke the feed inlet, cause surging, or build up on the belt in an uneven profile. This is especially common with raw meal or coal feeders during humid weather, and it often gets misdiagnosed as a "load cell problem" when the real issue is upstream flowability.

6. Communication and Integration Faults

On a modern feeder tied into a PLC/DCS via analog signal or fieldbus, you'll also see:

Signal scaling mismatches between the feeder's local controller and the DCS analog input card

Totalizer resets or communication dropouts during PLC failover

VFD speed reference not matching actual belt speed due to slip or a bad feedback loop on the drive side

A Practical Troubleshooting Sequence

When a weigh feeder starts giving suspect readings, it helps to work through checks in a fixed order rather than jumping straight to recalibration:

Confirm the mechanical basics first — belt tracking, idler condition, buildup, splice condition

Check zero with an empty, clean belt

Verify speed feedback independently (a handheld tach against the displayed speed is a fast sanity check)

Run a static test-weight check on the load cell

Only then move to a full material/dynamic calibration

Recalibrating a feeder that actually has a mechanical or tracking fault just bakes the error into the new calibration — it'll pass the test-weight check and still feed wrong material rates in production.

Closing Thought

The belt weigh feeder is a good reminder that in process instrumentation, accuracy is rarely about one component. It's the load cell, the speed sensor, the mechanical belt system, and the material behavior all agreeing with each other. When one of those falls out of step, the fault usually announces itself as a "weight problem" even when the root cause sits somewhere else entirely. Chasing the right link in that chain — rather than defaulting to a recalibration — is what separates a quick fix from a repeat complaint three weeks later.