The Gimli Glider, the Metric Conversion That Silenced Two Jet Engines, and the Slip That Saved Sixty-Nine Lives

On July 23, 1983, Air Canada Flight 143 - a Boeing 767 with 69 people aboard - ran out of fuel at 41,000 feet over Manitoba because of a single unit conversion error made during Canada’s transition to the metric system. Both engines flamed out. The crew glided the aircraft to a forced landing on an abandoned military runway being used as a racing circuit. Everyone survived.

How a Metric Conversion Error Grounded a Boeing 767

Canada’s national metrication program had been underway for years. For most Canadians, it meant adjusting to Celsius and kilometers. For the aviation industry, it meant overlapping measurement systems, different procedure manuals for different aircraft types, and conversion factors still being reconciled in the field.

Air Canada had taken delivery of its first Boeing 767 just seven months before the incident. The aircraft assigned to Flight 143 that day had a known problem: its fuel quantity processors were not reading correctly. Maintenance reviewed the situation before departure from Montreal and determined that operating with one of the two processors inoperative was permissible under the aircraft’s Minimum Equipment List (MEL). The aircraft was signed off and released.

With the automated gauging system degraded, the crew fell back on a manual measurement - a drip stick, a physical probe lowered into the fuel tank to measure depth, converted first to volume in liters, then to mass in kilograms.

The Number That Looked Correct

Jet fuel has a density of approximately 0.803 kilograms per liter. That is the conversion factor required to translate liters into kilograms. Someone on the ground used a factor tied to pounds instead.

The worksheet looked internally consistent. The documentation was clean. It showed 22,300 kilograms of fuel on board.

The actual fuel aboard was approximately 22,300 pounds - roughly 10,100 kilograms. Less than half what the paperwork indicated.

The full trip from Montreal to Edmonton required approximately 22,000 kilograms of fuel. Flight 143 had less than half of that.

A refueling stop in Ottawa added more fuel, but used the same flawed methodology. Nobody caught the error - not in Montreal, not in Ottawa, not during preflight. Flight 143 pushed back from Ottawa with 61 passengers and 8 crew and climbed toward cruise altitude.

The Crew Aboard Flight 143

Captain Bob Pearson was experienced and methodical. He also carried something unusual in his logbook: a commercial glider pilot rating. He had flown unpowered aircraft. He understood energy management without thrust in a way most airline captains simply don’t.

First Officer Maurice Quintal had flown for the Royal Canadian Air Force and, years earlier, had been stationed at a base called Gimli in Manitoba. He had landed fighters there. He could picture the runway.

Neither fact seemed relevant as they climbed through 35,000 feet on a clear summer afternoon.

Both Engines Flame Out at 41,000 Feet

The first warning came as they approached 41,000 feet: a fuel pressure alert on the left side. Pearson called for the checklist. They ran the procedures. The warning didn’t clear.

The left engine flamed out.

A jet engine that runs out of fuel simply stops - no mechanical protest, just silence, a yaw toward the dead side, and the absence of thrust. Pearson declared Mayday. His voice was level. Winnipeg was ahead. Quintal pulled approach plates and worked the radio.

Then the right engine flamed out.

The primary electrical system - driven by engine-mounted generators - went with it. The ram air turbine (RAT) deployed automatically into the slipstream, generating just enough power for emergency hydraulics, basic navigation, and limited radio.

No thrust. No fuel for a restart. 69 people at 41,000 feet over northern Ontario.

Choosing Gimli

Pearson understood immediately that he was flying a glider. The 767 has a serviceable glide ratio in clean configuration - significant altitude translates to significant distance - but that margin evaporates quickly at the weight and speed of a loaded widebody.

Quintal ran the numbers for Winnipeg. As he watched the altimeter unwind, the math wasn’t coming out cleanly. Then he thought of Gimli. He had been stationed there. Long military runway. Closer than Winnipeg.

They turned for Gimli.

What no one could tell them through their degraded radio - what no one knew to say - was that the eastern runway at Gimli had been decommissioned and converted to a racing circuit. The Winnipeg Sports Car Club used it on weekends. On that particular Tuesday afternoon, a race was in progress. Cars were on the pavement. Spectators lined the shoulders. Families had set up chairs. Kids were on bikes.

Air Canada Flight 143 was descending silently toward a car race.

The Slip That Saved 69 Lives

On final approach, Pearson saw another problem: they were high. Too much altitude, not enough distance, and no way to burn it off conventionally. Without engines, there is no go-around. A dead-stick approach gets one attempt.

He put the 767 into a forward slip.

A forward slip - cross the controls, rudder one direction and opposite aileron - presents more of the airframe to the airflow. It builds drag without building airspeed, trading altitude at an accelerated rate while preserving forward progress. Flight instructors teach it in Cessnas and Pipers for high finals and crosswind approaches.

No procedure manual covers slipping a Boeing 767. No simulator scenario had ever been written for it. But Pearson had flown gliders. He understood the physics in a way that went beyond academic knowledge. He knew what he needed.

He slipped the airplane.

The car racers saw the aircraft before they understood what they were looking at: a massive airliner descending steeply, nearly silent, gear down, holding a bank angle no airliner holds on final. Drivers abandoned their cars and ran. Spectators sprinted from the pavement. A track official reportedly waved a checkered flag - the only thing he had to wave - at an 87-ton jet on final approach.

The Landing on the Racetrack

Pearson touched down at approximately 175 knots - well above normal approach speed, but survivable for an aircraft of that size. The nose gear collapsed on contact. The aircraft’s nose came down on the pavement and it slid forward on its belly, trailing sparks, and stopped well short of the runway end.

All 69 people walked away. Ten were treated for minor injuries, most sustained on the evacuation slides. The collapsed nose gear actually helped: it lowered the rear of the aircraft, keeping the aft slides at a manageable angle.

Spectators who had scattered came back. Some helped passengers away from the aircraft. Two groups stood on a Manitoba racetrack in the summer afternoon looking at each other - the people who had been watching a car race, and the people who had just survived an emergency with no textbook resolution.

What the Investigation Found

The Transportation Safety Board of Canada spent months on the investigation. The conclusion was direct: the fuel had been calculated using the wrong conversion factor during Canada’s metric transition, compounded by the decision to depart with a degraded fuel quantity indication system under MEL - removing the automated cross-check that might have caught the discrepancy.

The industry response was substantive. Fuel verification procedures were revised. MEL policy around fuel measurement systems was tightened. The specific calculation error behind Flight 143 became a case study in airline safety training worldwide.

The aircraft was repaired and returned to service. It flew for another 25 years before Air Canada retired it in 2008 - the last 767 of its generation in the fleet. It was scrapped in 2013, more than 30 years after it became famous for landing on a racetrack without engines.

Why This Story Still Matters to Pilots

Bob Pearson, when he spoke about the flight in later years, was characteristically matter-of-fact: he had an airplane, he had a runway, he put them together.

That understatement obscures something important. The slip that saved Flight 143 wasn’t instinct - it was a skill developed flying unpowered aircraft in a discipline that has no obvious connection to airline flying. Pearson knew what a slip did to an energy state because he had done it hundreds of times in gliders.

The cross-country endorsement you keep deferring. The tailwheel checkout. The mountain flying course. The glider rating that sits on a someday list. These skills accumulate in a logbook that gets opened in moments nobody planned for.

The other lesson is less dramatic and equally important: check your units. The Gimli Glider didn’t happen because someone was careless. It happened because two measurement frameworks were in simultaneous use, and a number that looked correct was catastrophically wrong. Every fuel calculation you’ve ever done - every gallons-per-hour estimate, every conversion, every preflight check - exists because that kind of failure is always possible.

The fuel is either there or it isn’t. Get the units right.

Key Takeaways

Air Canada Flight 143 ran out of fuel at 41,000 feet on July 23, 1983, due to a unit conversion error: pounds used where kilograms were required, leaving roughly 10,100 kg aboard against a requirement of approximately 22,000 kg
The error was compounded by departing with a degraded fuel quantity indication system under MEL, removing the automated cross-check that should have caught the discrepancy
Captain Bob Pearson’s glider rating was the decisive factor: he slipped a 767 on final approach - a maneuver with no procedure manual entry - to bleed excess altitude on a powerless approach
First Officer Maurice Quintal identified Gimli as a closer alternate from memory of his RCAF service days, likely saving everyone aboard
All 69 people survived; the aircraft was repaired, flew another 25 years, and was retired in 2008