The Tare You Set in March

Truck 12 crosses the scale loaded and the display reads 77,940 pounds. The software subtracts 32,260, prints a net of 22.84 tons, and the gate arm lifts. Call it forty seconds, and the driver never left the cab.

Ask one question about that ticket: where did 32,260 come from? Not from this morning. Truck 12 hasn't crossed the scale empty since March, when somebody ran it over after the winter teardown and keyed the result into the truck file. The gross on the ticket is a measurement, taken today, by a machine. The tare is a memory. Every per-ton ticket in your operation is built from one of each, and only the measurement half ever gets checked.

That should bother you more than it does, because the money on the ticket lives entirely in the difference. Nobody buys gross. The 22.84 tons your invoice will carry is arithmetic between a number weighed at 9:12 this morning and a number weighed in March, and the arithmetic is only as honest as its older half.

What an empty truck weighs depends on the day

An empty truck is not a constant. It just looks like one from the office.

Fuel first, since it's the biggest mover and the easiest to forget. Diesel runs about seven pounds a gallon. A quad-axle dump doesn't carry road-tractor tankage, but most run a tank and a half, call it 120 gallons, and the spread between topped off and running on fumes is around 800 pounds. A truck tared low on fuel and topped off that night starts every morning 800 pounds heavier than its file says.

Then the season. Spring pit mud packs the frame rails and inner fenders of a quad-axle by the hundreds of pounds. Damp fines cake in the corners of the bed and ride back and forth for weeks as carryback. In November the chains go in the box, and somewhere mid-season a new toolbox or a tarp system gets bolted on, and nobody thinks of it as a billing event, because why would they. The truck that tared at 32,260 in March, washed, dry, and low on fuel, crosses the scale in June with an actual empty weight of 33,150. No one did anything wrong, and the file is now wrong by nearly half a ton.

flowchart TD
  A["Gross: weighed at the scale today"] --> N["Net = gross minus stored tare"]
  T["Stored tare: weighed once in March"] --> N
  N --> I["Invoice line: net tons times rate"]
  D["Fuel, mud, added equipment"] --> E["Actual empty weight moves"]
  E -. never re-weighed .-> T

The drift gets corrected in one direction only

Stale tares miss both ways, but the two misses live very different lives, and the difference isn't in your office. It's in who owns a scale.

Some of that 890 pounds is fuel that comes and goes with the day. Call the durable part 700 pounds, the mud and the toolbox, and follow it in both directions.

When the stored tare is too light, meaning the truck is actually heavier empty than the file says, every net prints high and you bill tons that never left the pit. Those tickets have enemies. The ready-mix plant reconciles what your tickets say against what its batch records consumed. The DOT job runs check scales. Sooner or later a customer weighs one of your trucks on their own scale, the call comes in, a credit memo goes out, and truck 12 gets re-tared that week.

When the stored tare is too heavy, tared muddy and full in April, running clean in August, every net prints low. That 700 pounds is a third of a ton and change, real material riding out the gate on every load, weighed, delivered, and never billed. No customer owns a scale that catches your low side. Nobody audits an invoice hoping to owe more.

So the customers with check scales trim your high side all season, and nothing anywhere trims the low side, and a fleet on stored tares finishes the year short. Here's a claim to argue with over coffee: if your pit runs spring mud and nobody has re-tared since the thaw, your certified scale has spent the season making that discount precise.

That last part is the objection to meet head on. The scale company calibrated in January and the state put a seal on the beam, so the tickets must be right. Certification tests the instrument against known weights. It says nothing about your truck file. Nobody from weights and measures reads stored tares, and a perfect scale subtracting a stale number prints an exact, wrong answer forty times a day.

Price one truck's worth

Stay with the 700 pounds: 0.35 tons given away per load. At $13 a ton for base rock, that's $4.55 a load, nine loads a day, a 110-day season: about $4,500 from one truck, all of it hauled at your cost, none of it invoiced. Three trucks drifted like that and you're past $13,500, wearing a certification sticker.

Hired trucks sharpen it. If you pay the subhauler by the load and bill the customer by the ton, his tare is your revenue number, and nobody on your payroll ever set it or checked the date on it.

Give every tare a date and a deadline

A tare isn't a spec of the truck, like axle count. It's a reading, and readings expire. So the truck file has to answer more than how much: 32,260, weighed March 14, clean bed, quarter tanks, good for 30 days or until anything gets bolted on. The re-tare itself is the cheapest control you'll ever run, one empty pass across your own scale on a Monday morning, forty seconds per truck.

The part that has to be structural is what happens when the date lapses, because good intentions don't survive a busy season. This is the slice Termco builds. Tickets get digitized as they're captured, each one carrying the tare it was priced against. A ticket priced against a tare past its window gets flagged as non-compliant that day, instead of surfacing on a customer's check scale in October. The truck list, tares and tare dates included, lives as managed reference data, and invoice totals get assembled only from tickets that cleared. Nothing in that machinery weighs a truck. It just stops a number from March from pretending it's current in June.

Your own test costs one morning. Before the first loads go out, run three trucks across the scale empty, whichever three have the oldest dates in the file, and set each reading against its stored tare. Then multiply each gap by the loads that truck has hauled since the date on its record. That product is the number to read out loud at the Monday meeting, in either direction.

If you'd rather have a second reader on the arithmetic, send 10-20 sample tickets to hi@termco.ai, a few from your oldest-tared trucks especially, and we'll send back what we find within 48 hours.