Safe Stacking Load: What a Compression Figure Actually Allows

A spec sheet says 900 pounds. The warehouse question is how much of that 900 a stack may use.

Nobody stacks to 900. The usable share is some fraction, and that fraction carries every condition the test did not reproduce. Weeks instead of a minute, damp air instead of a conditioned room, a pallet that is not quite square.

Most published guidance names those conditions and then stops. It rarely gives the arithmetic, almost never gives a source, and never mentions that the published factors disagree with each other.

This reference covers the parts of the conversion, and the standard that defines the divisor. It also covers what humidity actually costs, the method that measures time under load, and why the starting figure is an estimate too.

What the Conversion Actually Is

The arithmetic is simple, and that is the problem: take a measured compression figure, divide by a factor, and the result is an allowable load.

The first term comes from a test. A box compression test crushes a sample at a fixed rate and records the peak, which is a real number taken under stated conditions.

The second term is a judgment. It stands in for storage time, humidity, pallet overhang, stack alignment and handling, none of which the crushing test reproduced.

So the quality of the answer rests almost entirely on the second term. A 900 pound box divided by three allows 300 pounds; divided by seven it allows 129.

Those are both defensible divisors in published practice, and they differ by a factor of more than two. Picking one without stating why is the step most guidance skips.

The Divisor Has a Standard

Most guidance treats the factor as shop wisdom, but a current practice defines it. ASTM D5639 names it: a safety factor, F, “calculated from the expected environmental hazards” of storage and shipping.

Two words there carry weight: calculated, not chosen, and anticipated, which puts the burden on whoever knows the lane.

The same practice draws a second line that matters. Rail and motor freight classifications have established minimum requirements for certain attributes of corrugated packaging.

Then it qualifies them. Those minimums may or may not be appropriate for application in the complete distribution system, which makes a carrier rule a floor rather than a design basis.

That is the same distinction behind the stamp on a box flap. Meeting a classification minimum says the board qualified to move as freight, not that a stack of it will hold for a quarter in a warehouse.

Humidity Is the Best-Measured Factor

Of the four conditions, humidity is the one with real measurements behind it. A university thesis on the subject conditioned every specimen to meet or exceed TAPPI T402 before testing.

It found strength falling with water in the board, roughly linearly. The reported rate is about 7 percent of compression strength lost for each 1 percent of moisture gained.

Moisture itself tracks the air. The same work puts board moisture near 6.7 percent at 30 percent relative humidity, rising to about 13.7 percent at 90 percent.

An older result, cited in the same thesis, states the loss per point of humidity directly. Fellers and Bränge put it at about 8 percent for every 1 percent rise in relative humidity above 50 percent.

Set that against what circulates. One guide offers potential strength losses of 40 to 60 percent in humid conditions, and a 30 to 50 percent allowance for long-term storage. Neither range carries a humidity value, a method or a source.

Which direction the board arrived from also counts. ASTM D685 notes that properties at 50 percent humidity depend on whether the sample reached 50 percent from a higher or a lower humidity.

So a figure carries a history as well as a condition: board that dried down to the test room reads differently from board that damped up to it.

Three Published Answers, One Box

The disagreement is not a rounding matter. The thesis reproduces a prediction table for a single box rated 1,000 pounds at 50 percent relative humidity, asked what it holds at 80 percent.

Source	Predicted at 80% RH	Implied loss
Marcondes	550 lb	45%
Fiber Box Association	680 lb	32%
Maltenfort	760 lb	24%

Same box, same air, a 210 pound spread. Choosing the optimistic source over the cautious one nearly doubles the load a planner would sign off on.

Where the Load Actually Sits

The alignment factor hides the largest effect of the four. A box does not carry load evenly, so where the box above lands decides what the box below can hold.

The split is lopsided. One packaging supplier puts two-thirds of the stacking strength at the corners, and describes strength as greatest at the corners and least at the side of the box.

Column stacking keeps corner over corner. Interlocking turns alternate layers, which lands an upper corner on a lower wall, and the wall is the weak part.

The cost is measurable. A unit load study reports that interlocking boxes between layers significantly reduces resistance to compression, with estimates between 35 and 55 percent. The supplier above puts the reduction at up to 50 percent.

Two independent figures agreeing is worth noting in an article about factors that disagree, and on this one point the numbers line up.

But interlocking is not simply worse, and here the trade-off turns. The same study finds interlocking cuts pallet deflection by up to 53 percent against column stacking, through a stress redistribution the field calls load bridging.

So the pattern protects one component at the other’s expense. Interlocked layers spare the pallet and load the box walls, while column stacks spare the boxes and work the pallet harder.

Racking changes the question again, since a pallet in a rack carries only its own load. A floor stack puts every pallet above onto the bottom one, so the bottom boxes see the whole column.

The same study ran its supports both ways, racked and floor stacked, because the two load the pallet differently.

Overhang belongs in the same section. The footprint comes from standard pallet sizes, and hanging boxes past that edge removes support from exactly the corners that carry the load. The same supplier puts that penalty at 32 percent.

Time Under Load Is a Different Test

Humidity at least draws an argument. Duration usually gets a round number and no method, which is odd, because a method exists.

ASTM D4577 determines the resistance of a shipping container to a vertically applied constant load, for either a specified time or to failure. It holds a weight instead of raising one.

The scope is broader than a single box. It subjects a container, empty or filled, to a predetermined static load. The method also covers palletized or unitized load configurations.

That last clause matters for a warehouse. The thing that fails is a unit load, not a box, so a method that accepts the whole pallet answers a question a single-box test cannot.

The practical consequence is narrow. Where a duration factor is a guess, that guess stands in for a test, and the test could have run on the actual load for the actual time.

The Starting Figure Is an Estimate Too

All of this assumes the first term is solid. Often nobody measures it at all, and it comes instead from board properties through the McKee formula. Its board inputs are the ones an ECT grade chart tabulates.

That prediction has been checked, and the result is not reassuring. A verification study reports the formula can be significantly inaccurate, with a 50.48 percent overestimate for one box size and a 69.36 percent underestimate for another.

The study also notes what the formula leaves out, which includes transportation conditions and humidity. So a predicted figure already omits two of the conditions the safety factor is meant to cover.

That changes what the divisor is doing. It is not purely a margin against hazards; where the input was predicted rather than tested, it is also absorbing error in the prediction.

Which is why a divisor cannot be read as conservatism. Applied to an overestimated starting figure it may not be conservative at all, and applied to an underestimated one it wastes board.

Common Questions