Box Compression Service Derating

This page has no default values, and that is the most important thing on it.

Why there are no defaults. The humidity, storage-time, overhang and stacking-pattern factors that circulate on vendor websites, in course notes and in half the packaging spreadsheets in existence are unattributed restatements of tables published in the Fibre Box Handbook, which is copyrighted by the Fibre Box Association. Reproducing them here would be laundering someone else's intellectual property, and this site will not do it. Bring your own factors: from your own testing, from your board supplier's data for your board, or from a licensed copy of the source. What this page can honestly do is tell you what the OPEN literature has measured, and warn you where the numbers everybody quotes have been shown to be wrong.

Moisture is the biggest lever, and it is physics rather than a fudge. Paper is hygroscopic; it equilibrates with the air around it, and its compression strength falls steeply as it does. A box conditioned and tested at 50% relative humidity in a laboratory is not the box sitting in an unheated warehouse in a wet spring, in a cold chain, or under shrink wrap that has trapped the product's own moisture. This factor is also the one most often left out entirely, because the laboratory certificate never mentions it.

Creep is about time, and a compression test takes about thirty seconds. A box under a constant load deforms slowly and continuously and will eventually collapse under a load far below the one it survived in the press. That is a genuine long-term failure mechanism, not a safety margin — the box does not "have" the laboratory strength for six months and then lose it. The loss is steepest in the first days and continues for months, so the storage duration you design around is a real decision. Worse, creep and moisture interact: board under CYCLING humidity creeps dramatically faster than board held steadily wet or steadily dry, an effect called accelerated or mechanosorptive creep. An unheated warehouse with a day-night swing is harder on a stack than a consistently damp one.

Two open findings you may cite freely, and both contradict the common tables. First, in the box's favour: an open-access 2023 study found that the industry-standard 20 to 40 percent overhang derate significantly OVER-derates single-side overhang. A box hanging off one edge of a pallet is being penalised as though it hung off two, which it is not. Second, sharply against the box: an FBA-commissioned study at the Rochester Institute of Technology measured misaligned column stacking at 0.56, where the circulating tables give 0.85 to 0.90 for the same condition. That is more than twice the loss the tables predict, and misalignment is not something you opt into — it is what a real palletiser and a real forklift produce. If your pattern is nominally column but the corners do not actually line up, you are in the 0.56 case.

And one factor you will be offered that has no honest source at all. Scorelines — the creased folds — plainly affect compression, and vendors will quote you a number for it. The USDA Forest Products Laboratory wrote in 1985 that the effect of scorelines is "not fully quantified nor understood", and no sourced score derate has appeared in the open literature since. There is therefore nothing legitimate to publish, and a calculator that offers you one has invented it.

Now the part nobody says out loud, and it is the reason this page exists. This multiplier chain and the trade rule "divide the laboratory BCT by three to seven" are COMPETING METHODS OF DOING THE SAME JOB. They are not complementary. The three-to-seven band is a lumped stand-in for exactly the humidity, creep, overhang and pattern losses the chain models explicitly. Work through a realistic chain — 90% relative humidity, 180 days of storage, an interlocked pattern — and you land near 0.12 all by itself, a divisor of about 8.5. Apply a safety factor of 5 on top of that and you have divided the laboratory strength by more than forty, specified a board nobody makes, and paid for fibre that is doing nothing. Choose one method, and write on the drawing which one you chose. If you derated factor by factor, the remaining safety factor's only job is to cover what the factors did not model — handling damage, board variability, the pallet the forklift caught — and 1.2 to 1.5 is honest for that.