Weighbridge accuracy: calibration, drift and how to check

A maintained 50-tonne weighbridge reads in divisions of about 20kg. On a 14-tonne trailer load that's roughly 0.15% — accurate enough that the bridge should never be the weakest number on the farm. The catch is the word "maintained". Load cells age, junction boxes take on water, muck packs under the deck, and a bridge that's gone off doesn't look any different: it just reads confidently wrong, on every load, all season.

Here's how the measurement works, why it drifts, and how to catch a problem before you've built a harvest's records on it.

How a weighbridge measures

The deck sits on load cells — machined steel elements with strain gauges bonded on, which flex fractionally under load and turn that flex into an electrical signal. Cables carry each cell's signal to junction boxes, the junction boxes feed the indicator, and the indicator sums the lot into the number on the display. (Background on the whole assembly is in what is a weighbridge.)

Every reading passes through those cells, cables and boxes. So does nearly every fault.

Why readings drift

Water in junction boxes. The classic. A perished seal lets damp in, connections corrode, and readings start wandering — often worse in wet weather and frost, which makes the fault maddening to pin down.
Temperature. Cells are temperature-compensated, but the compensation ages. A bridge calibrated in a mild spring can behave differently in an August heatwave or a January freeze.
Debris under the deck. Grain, mud and stones packed between deck and foundation carry part of the load that should be going through the cells. The bridge reads light, and by a different amount every week.
A single failing cell. With six or eight cells sharing the job, one going bad shows up as position sensitivity: the same trailer reads differently depending on where it stops.
Foundation movement. Slower and rarer, but settlement or frost heave changes how the load reaches the cells.

The warning signs

The dangerous faults are the quiet ones. Watch for:

A systematic gap against merchant weights. If every load into the store weighs 60–80kg less on their bridge than yours, all season and always the same way, something is out. Mind you, a small one-way gap isn't automatically your bridge — grain loses moisture in transit, and shrinkage has a direction too. It's the size and consistency of the gap that tells you.
A zero that won't sit still. If the empty bridge shows a changing figure through the day, or needs re-zeroing every few hours, take it seriously.
Position sensitivity. The weight shouldn't depend on where the vehicle stops on the deck.
Poor repeats. The same vehicle weighed twice within minutes should agree within a division or two.

Checks you can do yourself

None of these needs test weights, and together they'll catch most faults:

Zero check, daily. Glance at the indicator with the bridge empty before the first load. Ten seconds.
Repeat test. Weigh a loaded trailer, drive right off, drive back on, weigh again. Do it three times. Readings should sit within 20–40kg of each other.
Position test. Stop the same vehicle at the front, middle and back of the deck. Meaningful disagreement points at a cell or a junction box.
Reference load. A lorry weighed at the merchant's verified bridge and then straight over yours the same hour gives you a genuine cross-check. One load proves little; the pattern across a season proves a lot.
Track the gap. If you record your net weight and the merchant's intake weight per load anyway, a persistent bias beyond about half a percent is worth an engineer's visit.

And sweep under the deck. It's the least glamorous item on this page and fixes more reading problems than anything else.

Annual service and calibration

Once a year, have an engineer test the bridge with certified weights — corner by corner, then combined — and adjust the indicator. Budget a few hundred pounds. Book it for spring or early summer: a fault found in June is an inconvenience, a fault found in the second week of wheat is a season of suspect records.

Keep the calibration certificates. They date any drift you find later, and they're your evidence if a weight dispute ever gets formal.

Calibration is not trade verification

Calibrating a bridge for your own use has no legal paperwork attached — you're doing it because you make decisions on the numbers. Using the bridge "for trade", where the reading determines what someone pays, is a different regime: the installation must be approved and verified under the Weights and Measures Act, with Trading Standards oversight. The line between the two, and which side most farm bridges sit on, is covered in weighbridge for trade rules.

Most farm bridges are own-use. Calibrate them anyway. A bridge you can't trust is a very expensive lump of steel.

Frequently Asked Questions

How accurate is a farm weighbridge?

A maintained 50-tonne bridge reads in divisions of about 20kg, which is roughly 0.15% on a 14-tonne trailer load. In practice, stale tares, mud and part-recorded loads introduce far bigger errors than the bridge itself — provided it's serviced.

How often should a weighbridge be calibrated?

Annually as a minimum, with an engineer and certified test weights, plus after any repair, cell replacement or groundworks near the foundations. Book it in spring so faults surface before harvest rather than during it.

Why does my weighbridge disagree with the merchant's?

First check the size and direction of the gap across many loads. A small, consistent shortfall at their end can be moisture loss in transit rather than a faulty bridge. A gap that's large, growing, or inconsistent between loads points at cells, junction boxes or debris under the deck — time for a service.

How can I check a weighbridge without test weights?

Repeat-weigh the same loaded vehicle three times (results should agree within a division or two), stop it at different positions on the deck, and run a vehicle weighed on a verified bridge straight over yours as a reference. A daily zero check and a clean underdeck catch most of the rest.