Vasa was a Swedish warship that took three years to build and sank on its maiden voyage after traveling less than a mile.
It wasn’t until the 20th century, however, when the ship was recovered by the Swedish government that the cause of the sinking was discovered.
Vasa sank because it had very little initial stability, which can be thought of as resistance to heeling over under the force of wind or waves acting on the hull. The reason for this is that the distribution of mass in the hull structure and the ballast, guns, provisions, and other objects loaded on board puts too much weight too high in the ship. The centre of gravity is too high, and so it takes very little force to make the ship heel over, and there is not enough righting moment, force trying to make the ship return to an upright position. . . . The problem is in the hull construction itself. The part of the hull above the waterline is too high and too heavily built in relation to the amount of hull in the water.
The use of different measuring systems on either side of the vessel caused its mass to be distributed asymmetrically, heavier to port. During construction both Swedish feet and Amsterdam feet were in use by different teams. Archaeologists have found four rulers used by the workmen who built the ship. Two were calibrated in Swedish feet, which had 12 inches, while the other two measured Amsterdam feet, which had 11 inches.
A sort of 17th century version of the Mars Climate Orbiter which failed because different contractors used different units of measurement,
The primary cause of this discrepancy was that one piece of ground software supplied by Lockheed Martin produced results in a United States customary unit, contrary to its Software Interface Specification (SIS), while a second system, supplied by NASA, expected those results to be in SI units, in accordance with the SIS. Specifically, software that calculated the total impulse produced by thruster firings produced results in pound-force seconds. The trajectory calculation software then used these results – expected to be in newton seconds (incorrect by a factor of 4.45)[16] – to update the predicted position of the spacecraft.