Skip to content
AxleGuard
Demo

Field notes

Ten loads that break a compliance check

Every product in this category handles the ordinary load. These are the ten that separate them: the ones where the arithmetic is counter-intuitive, the inputs are not what they appear, or the honest answer is that no fix exists.

  • Ten edge cases
  • Every one worked with numbers
  • Use them in an evaluation

How to use this

A test set, not a complaint

These ten are worth keeping as a list, because they are what an evaluation should be built from and what a spreadsheet should be tested against.

Nothing on this page is exotic. Every one of these is a load that a mid-sized fleet meets several times a year, and most of them are cases where the tool returns a confident number rather than failing visibly. That is what makes them worth listing: a check that is silently wrong is worse than no check, because it stops anyone else from looking.

The reference combination is the same one used throughout this section: an 18,000 lb tractor on a 20 ft wheelbase. Its fifth wheel is 17 ft behind the steer axle. It pulls a 14,000 lb trailer with 40 ft from the kingpin to the trailer tandem centre. Limits are the US federal figures for the Interstate System.

A check that is silently wrong is worse than no check, because it stops anyone else from looking.

The ten

Each one, with the arithmetic and what misses it

The last line of each is the useful one: not that the case exists, but what class of tool cannot see it.

01

Legal on gross, and over on the drive tandem

The base case, and the one that accounts for more citations than anything else on this list. Gross weight and axle distribution are independent, and a truck can be thousands of pounds under the federal cap while an axle group is over.

Worked example

Worked example for Legal on gross, and over on the drive tandem.
Gross76,000 lb, 4,000 lb under the cap
Centre of gravity18 ft behind the kingpin instead of 21 ft
Drive tandem36,725 lb against 34,000 lb

What misses it

Any tool whose only weight input is a gross figure. It is structurally incapable of seeing this, because it has never been told where anything sits.

02

Legal on every named group, and over on an interior bridge subset

The bridge formula applies to every consecutive axle group, and the interior ones do not correspond to anything with a name. A load can clear the steer axle, both tandems and the gross and still fail on the four-axle group that spans from the front drive axle to the rear trailer axle.

Worked example

Worked example for Legal on every named group, and over on an interior bridge subset.
Steer, both tandems, grossall legal, with margin
Axles two to five, 38 ft apart67,500 lb limit
Actual load on that group67,700 lb, 200 lb over

What misses it

Any implementation that checks the adjacent pairs and the outer spread. That is three of the ten groups on a five-axle unit, and the seven it skips include the one that usually binds.

03

The tire rating binding below the statutory limit

Three ceilings apply to every axle group and the lowest one governs. On a steer axle fitted with standard tires the tire rating is frequently the lowest, which means the 12,000 lb everybody quotes is not the number that applies.

Worked example

Worked example for The tire rating binding below the statutory limit.
Statutory steer axle limit12,000 lb
Two tires, 8.5 in of tread, at 600 lb per inch10,200 lb
Solved steer axle11,280 lb, legal on statute and over on the tires

What misses it

Every tool whose equipment model records an axle count rather than axles. If there is no field for tread width, the tire rating is not being checked, whatever the marketing says.

04

Spread axles gaining capacity as they separate

Two axles more than 8 ft apart escape the 34,000 lb tandem cap and take the formula result instead. This is why spread-axle trailers exist. It is also a trap in the other direction, because the extra statutory capacity does not come with extra axle rating.

Worked example

Worked example for Spread axles gaining capacity as they separate.
Two axles at 10 ft, formula limit40,000 lb
The same pair treated as a tandem34,000 lb, denying 6,000 lb
Manufacturer rating on the pair39,200 lb, which is what actually binds

What misses it

Tools that hard-code 34,000 lb for any two trailer axles, and tools that grant the formula result without checking whether the axles are rated to carry it. Both errors are common and they point in opposite directions.

05

A lift axle that is raised

A raised lift axle carries nothing and is not part of the group. N appears in both the numerator and the denominator of the formula, so counting an axle that is off the ground does not slightly inflate the limit. It substantially inflates it.

Worked example

Worked example for A lift axle that is raised.
Four axles across 38 ft, lift raised67,500 lb
Five axles across 38 ft, lift counted anyway72,000 lb
Capacity that does not exist4,500 lb

What misses it

Anything that takes the axle count from the equipment record rather than from the axle's current state. The check reports a comfortable margin on a load that is thousands of pounds over.

06

A negative kingpin reaction

Load a trailer far enough aft and the trailer stops pressing down on the fifth wheel and starts trying to lift it. The arithmetic keeps working and returns numbers that look calm, because the steer axle and the drive tandem both read low.

Worked example

Worked example for A negative kingpin reaction.
Cargo centre of gravity46 ft behind the kingpin, past the tandem centre
Kingpin reaction-300 lb, which is upward
What a naive check reportssteer 7,155 lb, drive 10,545 lb, both comfortable

What misses it

Solvers that report the numbers without sanity-checking them. A negative kingpin reaction is not a tight load, it is a load that cannot exist as described, and the correct output is an error rather than a verdict.

07

Two jurisdictions on one leg

A leg that crosses a state line is governed by two rule sets, and the limits, the bridge table and sometimes the units all change at the boundary. There is no moment at which the truck is subject to only one of them.

Worked example

Worked example for Two jurisdictions on one leg.
Drive tandem, as planned33,800 lb
Governing limit for the first 260 miles34,000 lb, legal
Governing limit for the last 40 miles33,500 lb, over by 300 lb

What misses it

Tools where jurisdiction is a global setting rather than a property of the load and the leg. They give one confident answer for a run that has two correct ones.

08

Fuel burn, and where the weigh station is

Diesel comes off the tractor as it burns, roughly evenly across the steer axle and the drive tandem. A load that is over at the pump gets lighter every mile, so whether it passes depends on which scale it meets and when.

Worked example

Worked example for Fuel burn, and where the weigh station is.
Drive tandem at the pump34,200 lb, 200 lb over
After 120 miles, 137 lb of fuel burned34,131 lb, still over
After 400 miles, 458 lb burned33,971 lb, legal

What misses it

Anything that computes one verdict at the origin. The load has a different answer at every point on the route, and the only points that matter are the ones with a scale on them.

09

A load that is legal at no slider position at all

There is a payload ceiling implied by the axle limits, and it is lower than the one implied by the gross limit. Above it, no arrangement of freight and no slider setting produces a legal truck. The right output is not a correction, it is a refusal.

Worked example

Worked example for A load that is legal at no slider position at all.
Payload from the gross limit48,000 lb
Payload the axle limits actually allow47,294 lb
A 47,500 lb order206 lb beyond what this combination can carry

What misses it

Tools that always return a suggested fix. Offering to move the tandems on a load that cannot be made legal wastes an hour on the dock and ends where it started. The honest answer is a different trailer, two trucks, or a permit if the freight is non-divisible.

10

Lateral imbalance and centre-of-gravity height

Neither of these appears anywhere in the axle arithmetic. Move a thousand pounds from the left side of the trailer to the right and every axle group reads exactly the same, because the group totals have not changed. The truck is a different vehicle in a curve.

Worked example

Worked example for Lateral imbalance and centre-of-gravity height.
Effect on axle group weightsnone at all
Effect on the rollover thresholdmaterial, and it grows with stack height
Where it is caughtwheel-by-wheel weighing, or a curve

What misses it

Almost everything, because axle compliance and stability are treated as separate subjects. A load model that records lateral offset and centre-of-gravity height can at least warn; one that records weight and position along the deck cannot.

On some of these the honest output is not a correction. It is a refusal.

How far over, in the six cases that end in a number

lb
  • Legal on gross, over on the drive tandem2,725 lb
  • The tire rating binding below the statute1,080 lb
  • A second jurisdiction on the last forty miles300 lb
  • An order beyond the payload ceiling206 lb
  • Over on an interior bridge subset200 lb
  • Over at the pump, before any fuel burns off200 lb

Six of the ten produce an overage measured in pounds, and four of those six are under 400 lb. That is the shape of the problem: a check that is quietly wrong by a few hundred pounds on a truck nobody suspects. The other four fail differently — a limit invented by counting a raised axle, capacity denied to a spread pair, a kingpin reaction that goes negative, and a lateral imbalance that changes no axle group at all.

The pattern

Three reasons a check goes quiet

The ten cases above fail for three reasons, and knowing which one you are looking at tells you whether a tool can be fixed or has to be replaced.

The model has nowhere to put the information

Cases three, five and ten fail because the equipment or cargo model has no field for the thing that matters: tread width, the axle's deployed state, lateral offset. This is not a bug and it cannot be configured around. A product that stores an axle count cannot check a tire rating, and no amount of rules on top of it will change that.

The calculation is incomplete

Cases two, four and six fail because the arithmetic is partial: some groups are skipped, a cap is applied where it should not be, or the result is not sanity-checked. These are fixable, and a vendor who shows you their worksheet is a vendor you can have this conversation with.

The answer was computed once for something that changes

Cases seven, eight and nine fail because the verdict is a snapshot of a run that has many states. Freight boards, jurisdictions change under the wheels, fuel burns, and the load that was checked at 05:00 is not the load at the scale at 14:00. The fix is not a better formula, it is computing the verdict at every point where the inputs change.

The load that was checked at five in the morning is not the load at the scale at two in the afternoon.

Dated, and honest about it

Assessed from publicly published documentation as of August 2026. Limits are the US federal figures on the Interstate System. Every weight, distance and spacing here is a worked example on a representative combination, chosen to make the arithmetic visible.

Next step

Put the ten edge cases against one real load.

Bring the route, configuration or scale ticket that made the ten edge cases matter. We will work through it in plain English and show where AxleGuard changes the decision.