Two different numbers, one label
Ask whether a crate is strong enough to stack three high and the honest answer is another question: for how long, and in what conditions?
A plywood box that carries a load for an hour and the same box carrying that load for six weeks are not the same problem. Wood and wood-based panels deform gradually under sustained load — the technical name is creep, and the practical consequence is that the load a crate passes in a short test can be a load it slowly loses the ability to carry. A stack that is stable on the day it is built can be leaning at week four.
Stacking strength therefore has two values worth knowing: the short-term load the box survives, and the long-term load it can hold for the duration of the voyage with a margin. The second is the one that matters, and it is always the smaller of the two.
The load travels through the edges, not the panels
The single most useful thing to understand about a stacked plywood crate is where the load goes.
In a stack, each box passes the weight above it down to the box below. That transfer happens at the contact between them, which is the top and bottom edges of the walls — not the middle of a panel. A panel is a membrane: it is stiff in its own plane and flexible across it. The edge is what turns the panel into something that behaves like a column.
This is why two boxes made from the same panel and the same thickness can carry very different stack loads. The difference is not the panel; it is how the panel is finished at its boundary and how the walls meet the base.
What limits a stack
Four failure modes account for most stack collapses, and they arrive in a rough order.
- Edge crush. The top or bottom edge of a wall crushes locally, the wall shortens, and the stack settles to one side. A raw plywood edge is vulnerable because it has no member resisting this; a formed edge distributes the load along the boundary instead of letting it concentrate.
- Joint slip. The wall-to-wall joint slides or opens under vertical load, which lets the box rack out of square. Once a box is out of square its stack load drops sharply, because the load is no longer travelling straight down.
- Base deflection. The floor of the box bows between its supports, so the goods sit in a dish and the walls are pulled inwards.
- Panel buckling. The wall bows outwards when the load is not perfectly axial, which happens whenever the stack is not level or the boxes are not the same height.
Why the joint is usually the weak link
Walls carry the stack, and walls are joined to each other. If that joint can move, the wall cannot act as a column.
Two things decide whether it moves: how much load the fastening can take in shear, and whether the fastenings are where the design assumed. A joint fastened by eye has a variable number of fastenings at variable spacings, so its capacity varies from box to box in a stack where every box is carrying the same load. The box that happens to be weakest becomes the one that fails, and it looks like a random failure.
Programmed fastening positions remove that variable. On a buckled joint the fastenings are placed at set intervals along the joint line by the machine, so every box in the stack has the same joint capacity, and the design calculation applies to the whole stack rather than to the best box in it.
Moisture and time both reduce capacity
Two environmental effects reduce stacking strength after the box leaves the workshop, and both are predictable.
Moisture. A panel that takes up moisture loses stiffness, and stiffness is what resists buckling. A crate that is dry when stacked and damp three weeks later has less capacity than its specification, which is the reason a moisture plan and a stack plan belong in the same conversation.
Time. Sustained load causes creep in the panels and in the joints. The longer the stack stands, the more the box deforms, and deformation itself reduces capacity — a bowed wall carries less than a straight one.
The practical consequence is that the design load for a stack should be set against the worst condition during the voyage, not the condition on the loading dock.
How to specify and verify it
Three things make a stack specification usable.
1. State the load per box, not the number of boxes. A stack height only means something with the weight of each unit and the contact footprint. Two stacks of three can be very different loads. 2. State the duration and the environment. Warehouse storage for a month under cover and a six-week sea voyage are different specifications, and quoting the first for the second is the most common error. 3. Verify by load test on a first-off box, not on a drawing. A compression test to the intended load, held for a representative period, tells you what the design actually does. It also gives you a number to reuse for every identical box.
Where a customer has no test capability, the honest alternative is to design with a margin and to keep the stack height in the handling instructions, so the box is not asked to do something it was never specified for.
The design levers, in order of effect
When a stack load has to go up, these levers work, roughly in this order.
- Stiffen the edge. The boundary is where the load enters the wall. A formed steel edge on the panel edge turns a vulnerable raw edge into a load-bearing member and is the most direct change available.
- Make the joint consistent. Identical joints in every box mean the weakest box is the designed box.
- Change thickness last. Thickness adds weight to every box in every shipment, and it is a blunt instrument compared with edge stiffness. It is the right answer when the panel itself is being loaded, which is a different problem.
- Fix the base. A base that distributes load to the skids keeps the walls vertical, and vertical walls are what the whole calculation assumes.
- Reduce the stack. Sometimes the correct engineering answer is a lower stack and a second container position, and it is cheaper than a heavier box across the whole order.
What to send us
Send the load per box, the number of boxes in the stack, the contact footprint, and how long the stack will stand and where. We can tell you which failure mode will limit your stack, and which change — edge, joint, base or thickness — moves the number for the least weight.