Weighing Systems & How They Actually Work

GG50 Weighing Series, Post 1 of 3

Sampling tells you what you've got. Weighing tells you how much of it you've got. We spent the last two posts on the first question. Time to answer the second one.

The Old Problem

Joseph didn't have a scale. As least not one you would be familiar with.

When Pharaoh dreamed about seven fat cows and seven lean ones, and Joseph told him it meant seven years of plenty followed by seven years of famine, the fix wasn’t complicated. Store grain during the good years. Ration it during the bad ones. But somebody had to know how much grain was actually going into those storehouses. Not "a lot." An actual number, tracked year over year, across a kingdom’s worth of harvests. Get that number wrong and people starve. That’s about as high stakes as inventory accuracy ever gets.

A few thousand years later, Byzantine grain traders were moving wheat out of Egypt and off the Black Sea coast into Constantinople by the shipload, packed in jars and sacks, changing hands multiple times before it hit a bakery. Every one of those handoffs was a moment where a buyer had to trust a seller's word on quantity, or check it himself. No trust, no trade.

Different centuries, same problem. Whenever grain moves in volume, off a field, out of a warehouse, onto a ship, somebody has to answer "how much." Everybody downstream has to be able to trust that answer. That’s not a modern problem we solved with load cells and digital indicators. It’s an old problem we’ve gotten better tools for.

From Merchant Balances to Certified Scales

For most of grain trading history, "how much" got answered with a balance scale and a set of reference weights, backed up by the seller’s reputation. That worked fine at the scale of a village market. It stopped working once rail lines and elevators started moving grain by the carload instead of the sackload.

Once volume and distance got big enough that buyer and seller couldn't just eyeball the transaction and shake hands, weighing had to become something more than a private arrangement between two parties. It had to become a system both sides trusted without having to trust each other directly. That's where certified, legal for trade scales came from. Third party verified equipment, subject to inspection, with a paper trail behind every ticket. The scale itself became the referee.

That's the whole point of legal for trade certification, and it's worth understanding even if you never touch the paperwork yourself. It exists because Joseph's problem and the Byzantine trader's problem never went away. We just moved the trust from a person's word to a machine's calibration record.

In the US, the agency most grain people will run into is FGIS, the Federal Grain Inspection Service, part of USDA. FGIS oversees official grain weighing and inspection, particularly for export shipments, where the buyer on the other end of the world has no way to verify anything except the paperwork. Domestically, state weights and measures offices handle scale inspection and certification under standards laid out in NIST Handbook 44. You'll meet other agencies depending on where you're operating and what you're shipping, and we'll get into that fuller picture, who certifies what and how often, in Post 3. For now, just know the requirement isn't bureaucratic box checking. It's the modern version of a very old handshake, with a government inspector standing in for the two parties who used to have to trust each other directly.

The other big leap didn't come from regulation. It came from the scale itself. For most of industrial history, "how much" was answered mechanically. A balance beam, levers, and counterweights doing the work through pure mechanical advantage. That's a fine system, but it has limits. Mechanical scales are slow to read precisely, hard to network into anything larger, and every moving joint is one more place for wear and error to creep in.

The load cell changed that. A load cell is a sensor that converts a mechanical force, the weight pressing down on it, into an electrical signal. It does this with a strain gauge that flexes by a tiny, precise amount under load and changes its electrical resistance accordingly. Feed that signal to a digital indicator and you get a number on a screen instead of a needle settling on a dial. That sounds like a small change. It wasn't. Electronic load cells are what made fast weighing possible: the continuous readings on a belt scale, the rapid cycles on a tower bulk weigher, the instant readout on a truck scale that used to take an operator walking out to read a beam. They're also what made weighing data something a computer could use directly, and that's the foundation everything from inventory software to automated blending systems is built on today. We'll get into what actually keeps a load cell honest, and what quietly wrecks one, in Post 2.

Know Your Terms

If you're new to this industry, here's where people trip. Grain gets weighed, priced, and reported in multiple units, and half the confusion in this business traces straight back to somebody mixing them up.

Gross, tare, and net. Before you worry about tons versus tonnes, get this one down . It's on every single scale ticket you'll ever touch.

● Gross weight is the total weight of the vehicle or hopper and the grain together. Everything on the scale at once.

● Tare weight is the weight of the empty vehicle or hopper by itself. The truck, the trailer, the railcar, whatever's carrying the load, with nothing in it.

● Net weight is gross minus tare. That's the grain. That's the number that gets paid for, tracked, and argued over.

Sounds simple, and it is, right up until the tare weight is wrong. A truck weighed empty with mud caked on the frame. A driver on the scale weighing in and off the scale weighing out. A trailer swapped between weigh-ins. Any of those will throw off your tare, and a wrong tare throws off every net weight calculated from it. Some operations weigh every truck both empty and loaded, every time. Others use a stored tare on file for a known vehicle. Both work fine as long as somebody's paying attention to when a stored tare might not be accurate anymore. This is also where a lot of new hires get tripped up on the language. People will casually say "weight" when they mean net, gross, or tare, and assuming which one they mean is how errors get built into a system before anyone notices.

Ton, ton, and tonne. There are three "tons" out there and they’re not the same weight:

● Short ton (US ton): 2,000 pounds, about 907 kilograms. This is what you'll hear most in US grain and feed operations.

● Long ton (imperial ton): 2,240 pounds, about 1,016 kilograms. Mostly a relic now, but it still shows up in older contracts and some shipping specs, so don't assume every "ton" you see in an old document means the same thing as the one on your scale ticket.

● Metric tonne (t): 1,000 kilograms, about 2,205 pounds. This is the dominant unit in global grain trade, and the only "ton" that's actually part of the metric system.

One metric tonne runs about 1.10 short tons and about 0.98 long tons. Close enough to cause real money problems if somebody assumes "ton" and "tonne" are interchangeable on a contract or a bill of lading. They’re not. Read the letter on the end of that word like it matters, because it does.

Bushels are a trap for the unwary. A bushel is a volume measurement, not a weight. It was never supposed to be a weight. But the industry needed a way to convert bushel tickets into pounds for pricing and accounting, so we assigned each commodity a standard test weight per bushel. 56 pounds for corn, 60 pounds for wheat and soybeans, and so on, based on typical density at a given moisture level. That standard weight is a convention, not a law of physics. Grain that's wetter, lighter, or of different quality than the standard will weigh differently than the bushel math assumes, which is exactly why test weight matters at the scale house as much as it does in the lab.

Outside the US, plenty of the grain world skips bushels entirely and works straight in kilograms and tonnes. Brazil prices a lot of its soybean crop by the saco, a 60 kilogram sack, at the farm level, then reports exports and national statistics in metric tonnes. Same physical grain, two completely different numbers, depending on who's talking and what stage of the chain you're standing in. It's a good reminder that "how much grain is this" only has one right answer once everyone agrees on the unit. Get the unit wrong and you can have two honest people arguing over a number that was never actually in dispute.

Quick reference:

Unit	Equivalent
1 metric tonne	2,204.6 lb / 1.10 short tons / 0.98 long tons
1 short ton	2,000 lb / 907.2 kg
1 long ton	2,240 lb / 1,016.0 kg
1 bu corn (standard)	56 lb
1 bu wheat/soybeans (standard)	60 lb

Pin that table up somewhere if you're new. It'll save you an argument eventually.

Origin and destination weight. These are contract terms, not measurement terms, but they show up on enough tickets and settlement sheets that you need them in this list. A grain contract has to say which end's weight actually governs a transaction, because the shipping-point number and the receiving-point number rarely match to the pound. That single clause decides who gets paid on what and who eats the difference. You'll run into these on truck and rail business as much as ocean freight, domestically and across borders.

● Origin weight is the weight established at the shipping point, wherever the load first got weighed and billed.

● Destination weight is the weight established when the load arrives, at whatever scale or survey the receiving party uses.

● Bill of lading weight is the origin weight specifically on an ocean shipment, set by a load port draft survey and recorded on the shipping documents.

● Outturn weight is the destination weight specifically on an ocean shipment, set by a discharge port draft survey.

● Franchise is the small percentage difference between origin and destination weight that a contract allows without triggering a claim. Anything past it becomes a weight dispute.

Inbound vs. Outbound Weights: Why Both Matter

Every load that crosses your scale gets weighed at least twice, once going in and once going out. The net weight between the two is the number everyone actually cares about.

Inbound weights set your starting point. They establish what came onto your books, what you owe the producer, and what your inventory count is built on. If your inbound weight drifts, your entire operation is wrong before you've done anything else. Every downstream number, storage fees, blending calculations, shrink reports, inherits that error and usually amplifies it.

Outbound weights determine what actually leaves your facility. This is the number your customer checks against their own receiving scale, the number an auditor pulls, and the number a regulator cares about most, because it's the one tied directly to what changed hands.

The net weight difference between inbound and outbound is where shrink lives. Some of it is real, from moisture loss and handling. Some of it is fake, created by a scale that isn't telling the truth. You can't tell the difference between real shrink and phantom shrink if you don't trust both ends of that measurement. That's the whole argument for treating weighing with the same seriousness you treat sampling. Get it wrong at either end and you're not managing your operation. You're guessing at it.

The Scale Types You'll Actually Run Into

Different points in the grain chain call for different kinds of scales, and each one has its own quirks.

Truck scales are the workhorses. Pit type, pitless, or above ground, you'll find them at nearly every elevator, feed mill, and terminal. They're also the most exposed to the elements: temperature swings, moisture, mud, and the occasional truck driver who doesn't quite get the whole vehicle onto the platform before stopping. Approach and exit conditions matter more than people expect. A truck that isn't fully on, level, and still will give you a number you can't trust.

Rail scales do the same job for railcars, but with more moving parts to worry about. Static rail scales weigh a stopped car. Scales that weigh cars in motion weigh them as they roll through, which is faster but adds vibration and dynamic load variables into the mix. Car to car variability and coupling forces make rail weighing inherently trickier than truck weighing, and the calibration and maintenance demands reflect that.

Process scales cover the equipment that weighs grain as it's moving or being handled inside the facility rather than crossing a threshold:

●       Belt scales weigh continuous flow on a conveyor. Useful for totalizing throughput, less useful for a single precise number on a single load.

●       Hopper scales weigh in batches, filling and dumping in cycles.

●       Loss in weight systems track precision by measuring how much weight leaves a vessel over time, which matters a lot in blending and ingredient dosing where small errors compound fast.

●       Impact scales measure flow by reading the force of grain hitting a plate as it falls, usually on a leg or at a feed mill. Cheaper to install than a belt scale and good enough for totalizing throughput. Not something you'd want to see on a ticket that determines who gets paid.

Worth flagging one more thing about process scales: not all of them weigh grain directly. Some infer it. Volumetric metering measures how much space grain takes up, moving through a screw or across a belt at a known rate, and then backs into a weight using an assumed test weight or density. That's the bushel problem all over again, just automated. It works fine as a rough number for a fast-moving process. It's the wrong tool anywhere the actual dollar figure depends on getting weight right, because the assumption baked into the math doesn't know your grain is wetter or lighter than standard today.

Tower bulk weighing systems are the piece people forget about, and they shouldn't. A bulk weigher fills an upper hopper, takes a static weight, dumps the load, and repeats. Cycle after cycle, all day. Those cycle weights often set the actual legal loadout number for a shipment. That means air leaks, gate timing problems, dust buildup, and a zero point that's drifted off true can quietly cost you money on every single cycle without anybody noticing until the totals don't add up at the end of the day. Small errors here don't stay small. They compound.

Draft Weights: When There's No Scale at All

Everything above has been about scales you can walk up to and touch. Once your grain is loaded onto an ocean-going vessel, that stops. Nobody's rolling a barge or a bulk carrier across a truck scale. So the trade falls back on a method that predates every load cell in this post: reading how deep the ship or barge sits in the water.

A draft survey reads the vessel's draft marks, how many feet or meters of hull sit below the waterline, before and after loading. A licensed surveyor combines those readings with the ship's hydrostatic tables, along with corrections for water density, trim, and list, to calculate displacement. The change in displacement between the empty ship and the loaded one is the cargo weight. No load cell touches the grain at any point. The ship itself is the scale.

That’s not a tiny rounding difference; it’s a lot of grain riding on that number. A single Panamax cargo can run 50,000 to 80,000 metric tonnes, and a draft survey is typically accurate to within about half a percent when it's done right. Half a percent of 60,000 tonnes is still 300 tonnes, which is exactly why export contracts spell out who's got a surveyor on the dock and how disputes between the load port figure and the discharge port figure get resolved before the ship ever leaves.

Barges on inland rivers work the same basic principle, just on a smaller scale. They use draft marks and displacement tables instead of a full hydrostatic survey. When trucks dump straight into the barge, you can add up the truck scale tickets to get a running total, and a few barge loaders have bulk weighers just like the ones used for loading ships. But most of the time, the draft‑based number is what you’ve got. It’s not as tight as a certified truck scale, but on the Mississippi or other rivers there usually isn’t another practical option.

This is worth thinking about  for a second, because it's the whole point of this post in one example: the industry didn't wait for a perfect tool before it started trading by weight. It built the best measurement the situation allowed, put a trained professional and a paper trail behind it, and kept trading. That was Joseph's challenge and the Byzantine trader's problem , and the challenge of accurate undisputed weights is a frequent battle across the globe still today.

Origin and destination weight is a contract term, not an ocean-only concept, and it's worth knowing by name because it shows up on truck and rail business too, domestically and across borders. Any grain contract has to specify which end's weight actually governs, the shipping point or the receiving point, because those two numbers rarely match exactly and somebody has to decide up front which one gets paid on and which one absorbs the difference. On a truck or a railcar, both ends usually have a real scale, so the contract language, often as simple as "origin weights govern" or "destination weights govern," is what turns two honest scale tickets into one settled transaction.

At export elevators, the origin weight usually comes from a certified bulk weigher. In the U.S., that means FGIS‑inspected, legally recognized weights. Many ports can also weigh inbound grain whether on a barge or vessel through their bulk weighers. That gives you a clean, audited origin weight without ever touching a draft survey.

But not every country has that capability. In much of the developing world, ports don’t have certified bulk weighers, and the grain arriving at the port, whether it comes in by barge, lighter, or even a full ocean vessel, shows up with no means of verifying the weight with a scale. When there’s no way to weigh inbound grain, the discharge port relies on a draft survey to “challenge” the origin weight, even if the contract says the origin weight governs and the buyer doesn’t have the contractual authority to settle on their survey.

The gap between bill of lading weight and outturn weight is where ocean shrink lives, and it's read exactly the way you'd read the gap on a truck: some of it is real, moisture loss and handling over a multi-week voyage, and some of it isn't, a load port survey done sloppy or a discharge port surveyor with an incentive to find a low number. Contracts handle this with a franchise, typically a small percentage difference that's allowed without triggering a claim, because nobody expects those two draft surveys to land on the identical number. Anything past that franchise becomes a weight dispute, and it gets settled by comparing surveyor credentials, checking who followed proper procedure, and sometimes bringing in a third surveyor to break the tie. The bushels-per-truck version of this argument costs you a phone call. The ocean version can cost real money on a cargo that size, which is exactly why the surveyor's qualifications and the paper trail behind a draft survey matter as much as the number itself.

Operator-Real Takeaway

Weighing isn't a "set it and forget it" function, and it never was. Not for Joseph's storehouses, not for a Byzantine grain trader checking a jar count, and not for you standing at the scale house today. Every scale will lie to you eventually. Environmental drift, mechanical wear, a bad load cell, an operator who's gotten sloppy about zeroing. Something will eventually push a number off true. Your job is to know your terms well enough to catch it, and to treat both ends of every weight, inbound and outbound, like the number that determines whether your operation is telling the truth.

Next up: what can impact accurate weighing with a scale before it ever gets to calibration. Installation quality, mechanical realities, and the environmental factors that quietly wreck accuracy from the ground up. Including the story about the truck scale that got hit by lightning five times in one year.

Grain Guy Fifty invites’ readers to share their thoughts , experiences, knowledge, and insights after exploring the post. Together, we can learn and grow as an industry!

Best, 

Jim Voigt 

“Grain Guy Fifty”