FIG. 1 · NAVAL ARCHITECTURE · BOX BARGE, END ON
On 10 August 1628 the Swedish warship Vasa set sail for the first time. She had gone roughly 1,300 metres when a gust of wind leaned her over, water poured in through the open gunports and she sank. It was a calm day. Nothing was wrong with her planks. What was wrong was where her weight sat.
A floating ship is a balancing act between two forces. Her weight pulls down through one point, her centre of gravity, G. The water pushes up through another, the centre of buoyancy, B, which is the centre of the shape of water she has pushed aside. Float level and the two points line up and nothing happens. The interesting part is what happens when she leans.
Tilt the hull and the underwater shape changes. One side sinks deeper, the other lifts. The pushed-aside water is now lopsided, so its centre B slides towards the side that went down. G, fixed to the ship's own load, hardly moves. Now the upward push and the downward weight are in different places, and they form a pair of forces that twists the ship. The sideways gap between them is the righting arm, called GZ.
If B has slid far enough past G, the twist rolls her back upright. If not, the twist rolls her further over.
For small tilts there is a neat shortcut. Draw the vertical line up from B and see where it crosses the ship's centreline. That crossing is the metacentre, M, an idea worked out by the French scientist Pierre Bouguer in the 1700s. For small tilts the ship behaves as though she were hanging from a hinge at M. If G is below M, she hangs and swings back. If G is above M, she is balanced on a pin and falls.
For our box-shaped barge, which is wide W with draft d, the maths is exact and short: M sits above B by W² / (12 d). Notice the square. Double the width and the metacentre rises four times as far. That is why wide, flat hulls are so stubbornly stable, and why a tall narrow one is easy to tip.
Below is the barge in cross-section. Drag sideways on the water to heel her and let go, or hit a gust button. Then slide the cargo up from the hold to a tall stack on deck. The cargo never changes in weight. Only its height does. Watch G (amber) climb towards M (blue-white ring), and watch the curve underneath: it plots the righting arm at every angle of heel.
With the cargo low in the hold, G sits well under M. Every heel is met with a strong push back, the curve stays high and she survives a heavy shove. Raise the cargo and G climbs. GM shrinks, the curve flattens, and the angle at which the push back disappears creeps in. Push the stack high enough and GM goes negative: she no longer wants to be upright at all. She leans over to some angle and sits there, or rolls right over.
Real ships have the same problem in a more complicated form. Cargo and fuel are shifted around, tanks are filled with ballast water to lower G, and the loading plan for every voyage is checked against stability calculations. One famous trap is liquid sloshing in a half-empty tank: it slides to the low side as she leans and effectively lifts G, shrinking GM.
The Vasa's hull above the waterline was too tall and too heavily built, so her centre of gravity was too high and she had very little initial stability. The guns alone were not enough to doom her. The first gust leaned her over and she righted slowly. The second pushed her down until the lowest gunports dipped under. That is the same story as your slider: not a broken ship, just a G that rose too close to M. She was raised from the seabed in 1961 and now sits in a museum in Stockholm.