When Did the Titanic Sink? The Forensic Timeline and Physics of a 2-Hour 40-Minute Failure

The Titanic sank in the early hours of April 15, 1912, striking an iceberg at 11:40 p.m. ship's time on April 14 and disappearing beneath the North Atlantic at about 2:20 a.m. on April 15. The sinking took roughly 2 hours and 40 minutes from collision to final plunge. She went down about 400 nautical miles south-southeast of Newfoundland, taking more than 1,500 lives with her.

The question sounds like a single fact, but the honest answer is a timeline with estimated boundaries, disputed clock conventions, and physical processes that engineers still model today. This explainer goes past the date and into the mechanics: how six small breaches broke a ship advertised as unsinkable, why she stayed afloat as long as she did, and why the death toll was governed less by the damage than by decisions made in the first hour.

The collision and the breach geometry

Titanic's hull was divided into 16 watertight compartments by 15 transverse bulkheads. These bulkheads were not sealed at the top. They rose only partway up each deck, so water could spill over the top from one flooded compartment into the next, like a slow-motion waterfall moving aft.

At 11:40 p.m. on April 14, the ship grazed an iceberg along roughly 250 to 300 feet of her starboard side. The glancing contact opened the hull across six forward compartments. The openings were small — the total area of the damage is often estimated at around 12 square feet — but they were distributed along a long stretch of hull, which mattered far more than their combined size.

The flooding arithmetic was brutal. With four compartments flooded, the ship could stay afloat. With five, buoyancy was marginal. With six, the bow would settle low enough that the forwardmost bulkheads would submerge, water would pour over their tops into the seventh compartment, and the process would cascade aft with no stopping point short of the ship's loss. The famous "unsinkable" design assumed a worst case of four compartments breached. The iceberg delivered six.

The counterintuitive detail: a single large hole would likely have been survivable. It was the lengthwise spread of small breaches, not their area, that defeated the bulkhead system. Ship designers learned from this that compartment subdivision must account for damage extent, not just damage volume.

The two-hour-forty-minute physics of foundering

Why did a 52,000-ton ship take 160 minutes to sink? Because flooding is throttled by the openings themselves, and the openings were small. Water entering through narrow rents flows at a rate set by their area and by the pressure difference across the hull, which grows as the ship settles deeper.

As the bow dipped, water advanced through the forward compartments, over successive bulkhead tops, and aft. Each stage added weight and changed the ship's trim. The stern rose, lifting the rudder and propellers clear of the water. Around 2:18 a.m., the hull's structure could no longer bridge the unsupported length between the flooded bow and the rising stern. The ship broke apart near the surface between the third and fourth funnels. The bow section went under first; the stern, still containing trapped air, upended and sank a couple of minutes later.

The physics answer to "how long does a ship take to sink" is that it is not a single event. Foundering is a feedback process: flooding changes trim, trim changes which openings are submerged, and submerged openings change the flooding rate. Titanic's 2-hour-40-minute span reflects a slow, sequential cascade rather than a sudden failure.

Timeline: the sinking in ship's time

All times below are Titanic's clocks as commonly reconstructed, with noted uncertainty of several minutes because the ship was scheduled to adjust time overnight.

The 40-minute gap between the last boat and the final plunge is the deadliest interval in the timeline. Thousands remained aboard with no craft left to carry them.

The lifeboat arithmetic and the death toll

Titanic carried 20 lifeboats with a combined rated capacity of 1,178 people. Aboard were about 2,208 people. The boats could hold only about 53 percent of those on board — roughly 1,178 seats for 2,208 souls.

The British inquiry found that about 1,503 people died; the American inquiry put the figure near 1,517. Modern estimates cluster around 1,496 to 1,503 lost, with about 705 to 712 survivors. The exact number remains uncertain because passenger and crew lists were not reconciled, and some travelers were listed under multiple names or not at all.

The boat capacity was legal under the merchant-shipping rules of the era, which set lifeboat requirements by gross tonnage rather than by the number of people aboard. That rule was written for a different era of ship design and was overtaken by the size of ships like Titanic. After the disaster, the first International Convention for the Safety of Life at Sea, adopted in 1914, required lifeboat capacity for all persons on board, among other measures.

Red-flag warnings that the lifeboat arithmetic exposes, and that apply to any emergency planning today: Capacity rules that scale with the vehicle or facility, not the number of people present. Evacuation equipment that is inspected but never drilled under realistic load. Launch procedures that assume calm conditions and full crews. Communication systems that cannot reach help within the survival window. One-way information flow: passengers told to stay put while crew improvise.

Misconceptions that distort the timeline

Several popular claims are either wrong or unsupported, and each one changes how the timeline should be read.

The ship did not sink in one piece. The break between the third and fourth funnels was not confirmed until the wreck was found in 1985. Eyewitnesses described the stern remaining afloat briefly, but the mechanism was not understood until the debris field was mapped.

The iceberg did not slice a 300-foot gash. The damage was a series of small openings along the shell plating, some from rivet failures and plate seams, consistent with a glancing impact rather than a clean cut.

The band did not necessarily play to the very end. Survivors disagreed on the last tune and on when the music stopped; the popular image of "Nearer, My God, to Thee" as the final number is a cultural tradition, not a verified record.

The ship was not traveling at reckless speed for the conditions. She was making about 22 knots in an ice field reported by other ships. The practice of maintaining speed through known ice was common, and the bridge had received ice warnings but not all of them reached the officer on watch. The failure was systemic, not a single act of folly.

The wreck is not intact and never will be. The bow section sits upright in about 12,500 feet of water; the stern is crushed and the hull is being consumed by iron-eating bacteria. The site is a grave and is protected by international agreement.

What the timeline teaches about decision-making under uncertainty

The most transferable lesson is about the first hour, not the last two. The collision happened at 11:40 p.m. The first lifeboat left at about 12:45 a.m. — over an hour after impact. During that hour, officers and passengers largely did not believe the ship would sink, and boats were launched partly filled because people preferred the apparent safety of the deck.

The wireless operators sent distress calls and received replies from nearby ships, but the nearest, Carpathia, was about four hours away at her best speed. The survival window was shorter than the rescue window, and no one on the bridge appears to have fully internalized that arithmetic in the first minutes.

For anyone planning safety systems today, the thresholds that matter are concrete. Time-to-failure must be compared against time-to-rescue, not assumed to be generous. Evacuation capacity must be compared against the number of people present, not the rated maximum. And the first decision window — the minutes when uncertainty is highest — is where outcomes are set.

A practical checklist derived from the timeline: Calculate time-to-failure explicitly for the worst credible damage case, not the design case. Confirm rescue time against that failure time before departure, not after an incident. Staff and drill for full-capacity evacuation, not partial loads. Verify that warning information reaches the person with authority to act. Treat "unsinkable" claims as marketing until independently validated.

Limits of the reconstruction

Every time in the timeline carries uncertainty of several minutes because Titanic's clocks were being adjusted overnight and different accounts use different conventions. The death toll is a range, not a single verified number. The damage geometry is inferred from survivor testimony, the wreck, and engineering models rather than from direct observation at the moment of impact.

What is not in dispute is the sequence: collision on the night of April 14, foundering in the early hours of April 15, roughly 2 hours and 40 minutes from impact to disappearance, and a death toll above 1,500. The mechanisms — compartment flooding over unsealed bulkhead tops, progressive loss of buoyancy, structural failure between the third and fourth funnels — are well supported by the wreck and by the inquiries.

The deeper point is that the answer to "when did the Titanic sink" is a clock, and the clock was set by physics and by human decisions made in the dark. The ship's fate was sealed in the first hour; the next 100 minutes were the slow playing out of that arithmetic. Understanding the timeline is less about memorizing a date than about recognizing the same pattern — slow cascade, underestimated window, capacity shortfall — wherever it appears today.