Here's the uncomfortable thing about a mechanical sampling system: when it fails, nothing stops. The belt keeps running, the cutter keeps swinging, the sample keeps dropping into the bucket. But the number that comes out of the lab is wrong — and that number is what your ore gets paid on. Solid sampling system maintenance is the difference between an assay you can defend and one you can only hope is right.
Sampling system maintenance that protects the number you get paid on
A worn cutter or a drifting speed doesn't break the plant. It quietly biases every sample — and you mis-value tonnage for months before anyone notices.
Why sampling system maintenance isn't like any other PM on the plant
Most gear tells you when it's unwell. A bearing gets hot, a pump cavitates, a conveyor squeals. A sampling system gives you none of that — it keeps cutting and dropping sample while quietly handing the lab something that no longer represents the stream.
That's because the output isn't motion or flow — it's a number, and a wrong number weighs the same as a right one. A sampler's whole job is representativeness: every particle getting an equal chance of being collected.
Wear and buildup break representativeness quietly, often months before a bias test catches it — book a demo to see cutter drift flagged early.
The sample sets the value. Grade, moisture, and penalty elements are all read off this one small sample. If it's off, the invoice is off.
Failure is invisible. There's no alarm for bias. A narrow cut or a fast cutter produces a clean-looking sample that's simply not representative.
It's cumulative. The error doesn't spike — it drifts shipment after shipment, so the loss is spread thin and hard to trace back to the cutter.
The three ways a sampler lies to you
Sampling bias almost always traces back to one of three mechanical faults. None of them trip an alarm. All of them are catchable on a routine round if you know what to look at — and they're the core of any real sampling system maintenance program.
Rounded, chipped, or corroded cutter lips stop delimiting a clean slice. Material near the edges gets grabbed or shed, so the cut no longer represents the full stream cross-section.
→ Grade biasA cutter that travels too fast, or speeds up and slows down through the stream, over-samples the near side and under-samples the far side. The golden rule is constant travel, with deviation under about ±5%.
→ Size & grade biasFines bridge on the blades or material hangs up in the cutter and doesn't discharge. The next cut carries yesterday's sample, and the discharge chute slowly starves the lab of a clean increment.
→ Contamination biasNotice the pattern: every one is a maintenance condition, not a design flaw. The sampler was correct the day it was commissioned — it drifted. So the fix is a schedule, not a redesign — book a demo to see sampling checks scheduled per cutter.
Stop finding bias in the reconciliation report
Put cutter, speed, and discharge checks on a schedule tied to each sampler — and keep a signed record of every inspection for your next bias audit.
Correct cut vs. biased cut: what actually changes
The whole job of the cutter comes down to one idea — take all of the stream, and only the stream, every single pass. Here's what separates a cut you can defend from one that's quietly costing you.
- Opening is at least three times the top particle size, so no coarse piece is turned away
- Full stream width is captured — edge to edge, every pass
- Cutter travels at a constant speed through the stream
- Everything delimited discharges cleanly; nothing stays in the cutter
Result: a representative sample the lab, the buyer, and the auditor all trust.
- Narrowed or worn opening rejects coarse particles — the sample skews fine
- Edges of the stream get missed, so one side is over-represented
- Variable speed over-weights part of the stream against the rest
- Retained material carries over and contaminates the next increment
Result: a clean-looking sample that misreports grade — and nothing flags it.
The gap between these two columns is almost never visible to the eye. That's why the honest way to run a sampler is to log every part that controls the cut — opening, edge wear, speed, discharge — so sign up free and start tracking cutter condition.
A verification cadence that keeps the sample honest
Good sampling system maintenance runs on two clocks at once: frequent mechanical checks that catch wear early, and periodic statistical checks that prove the sampler is still unbiased. Both belong on a schedule — here's a practical cadence to build from.
Confirm the cutter is parking clean with no retained material, the discharge chute is clear, and nothing is bridging on the blades. Thirty seconds that protects the day's samples.
Inspect the cutter lips for wear, chipping, and corrosion, and verify the opening width against spec. Edge wear is the most common quiet source of grade bias.
Verify cutter travel speed is constant and within tolerance across the stream. On systems with encoders, review the speed/position log for any slowing through heavy material.
Run a formal bias test against a reference method to prove the whole system is still representative. The mechanical rounds keep it honest between these statistical checks.
The trap most plants fall into isn't skipping the bias test — it's having no record of the mechanical checks between them. When the test finally fails, there's nothing to trace it to. Book a walkthrough of cutter and speed-check history and turn guesswork into a timeline.
From a worn cutter to a work order — without the paper
Catching the wear is half the job. The other half is making sure it actually gets fixed, logged, and provable. That's where preventive maintenance scheduling and digital inspections replace the clipboard that usually sits on the sampler platform.
Standardized sampler rounds
Every sampler gets the same digital checklist on your inspection schedule — cutter edge, opening, speed, discharge — so nothing gets skipped and every pass is on the record.
PM by hours and cadence
Weekly edge checks, monthly speed checks, and annual bias tests fire automatically by meter reading or calendar — the two clocks run themselves instead of living in someone's head.
Defects into work orders
A worn cutter flagged on a round becomes a work order with the part, the sampler, and the history attached — and a signed PDF trail ready for the next audit.
The scary part of a sampler is that it never looks broken. We failed a bias test once and had no idea how long we'd been off — the mechanical checks were real, but they lived on a clipboard nobody kept. Could've been three weeks or three months of shipments. Now the cutter edge, the speed check, the discharge, it's all on a schedule with a date stamp. When a bias test comes up, I can show exactly what was checked and when. That's the whole game with sampling — being able to prove the number was honest.
Frequently asked questions about sampling system maintenance
What does sampling system maintenance actually involve?
It's two things running together. First, routine mechanical checks — cutter edge wear, opening width, travel speed, and clean discharge — done daily to monthly to catch drift early. Second, periodic statistical checks, mainly a bias test, that prove the whole system is still taking a representative sample. The mechanical rounds keep the sampler honest between the bias tests, and both belong on a documented schedule rather than being done by feel.
Why does a worn sample cutter cause biased results?
A sample is only representative if the cutter takes the full width of the stream, with every particle having an equal chance of being collected. When the cutter lips wear, round over, or corrode, the opening effectively narrows and the edges of the stream get missed. Coarser particles are turned away first, so the sample skews fine and the grade reads wrong. Nothing stops or alarms — the cut just stops being representative, which is why edge inspection sits near the top of any sampling maintenance list.
How often should a mechanical sampler be inspected?
A practical cadence is: a quick discharge and buildup check every shift or day, a cutter edge and opening inspection weekly, a speed and timing check monthly, and a formal bias or precision test periodically — often annually or per your standard. The exact intervals depend on throughput, abrasiveness, and your sampling standard, but the principle holds: frequent cheap mechanical checks between the less frequent statistical ones.
What is a bias test and why does it matter?
A bias test compares your mechanical sampler against a reference method to statistically confirm it isn't systematically over- or under-reading. It's the proof that the sample is representative, which is what buyers, auditors, and reconciliation depend on. But a bias test is a snapshot — it tells you the sampler was good on test day. The mechanical maintenance record is what lets you trust it on every other day, and what lets you trace a failed test back to when a part drifted.
How does HVI help keep a sampling system unbiased?
HVI puts every sampler on a standardized digital inspection and preventive maintenance schedule — cutter edge, opening, speed, and discharge checks fire automatically by calendar or meter reading. A flagged defect turns straight into a work order with the part and sampler history attached, and every check leaves a signed PDF trail. So instead of hoping the sampler is still representative, you have a documented, auditable record of exactly what was checked and when — ready for your next bias test. Book a demo to see it running on your own samplers.
The number is only as honest as the sampler that took it
Every other asset on the plant tells you when it's hurting. A mechanical sampler hands you a wrong answer that looks exactly like a right one. That's what makes disciplined sampling system maintenance less a reliability task than a financial control.
Worn cutters, drifting speed, and retained material quietly cost you grade until a bias test catches up. So schedule the mechanical checks, keep the bias tests on the calendar, and keep a record tying the two together — then the sampler becomes the thing you can prove.
Make sampling system maintenance something you can prove
Get cutter, speed, and bias-test records on one schedule, with a signed trail for every check — so your next audit starts with evidence, not hope.
