The Spacecraft Lost Between Pound-Force Seconds and Newton-Seconds
NASA launched the Mars Climate Orbiter in December 1998.
The spacecraft was intended to study the Martian climate and act
as a communications relay for another mission. After travelling
for more than nine months, it reached Mars in September 1999—but
contact was lost during its attempt to enter orbit.
The investigation found a mismatch between two units of impulse.
A ground-software component produced thruster data in
pound-force seconds, while the navigation
software expected newton-seconds.
The two units are not interchangeable:
1 pound-force second ≈ 4.44822 newton-seconds
Because the expected conversion was not performed, the navigation
system underestimated the effects of small spacecraft manoeuvres.
The errors accumulated during the journey, and the orbiter
approached Mars on a trajectory substantially lower than planned.
NASA’s investigation identified the missing unit conversion as
the root cause of the loss. It also found broader weaknesses in
communication, software-interface compliance, verification and
the processes intended to detect navigation discrepancies.
The lesson is larger than “remember to convert the units.”
Software interfaces must state which units they accept, systems
should validate incoming data, and independent checks should flag
results that do not agree with the expected physical behaviour.
Sources:
NASA’s Mars Climate Orbiter mission history
and
NASA Jet Propulsion Laboratory’s investigation summary
.
The Passenger Jet That Became the “Gimli Glider”
On 23 July 1983, Air Canada Flight 143 was travelling from
Montreal towards Edmonton when its Boeing 767 exhausted its fuel
in flight. Both engines lost power, leaving the pilots to glide
the aircraft towards a former air-force base at Gimli, Manitoba.
The aircraft’s fuel-quantity indication system was not working
normally, so the amount of fuel had to be determined manually.
This required converting the measured volume of fuel in liters
into its mass in kilograms.
Air Canada’s other aircraft had traditionally used pounds for
fuel mass, while the new Boeing 767 used kilograms. During the
manual calculation, a density factor expressed in pounds per liter
was used as though it represented kilograms per liter.
The aircraft required a total fuel load of approximately
22,300 kilograms. The incorrect calculation
resulted in a much smaller amount being carried—roughly the mass
that 22,300 would represent if the number were interpreted as
pounds rather than kilograms:
22,300 lb × 0.45359237 ≈ 10,115 kg
This was less than half the intended fuel mass. The discrepancy
was not detected before departure, and the aircraft eventually
ran out of fuel at cruising altitude.
The crew successfully glided the unpowered aircraft to Gimli and
completed an emergency landing. The flight subsequently became
known as the “Gimli Glider.”
The official inquiry described a chain of contributing problems,
including the unavailable fuel indication system, unfamiliar
metric procedures, incorrect conversion factors and unclear
responsibility for verifying the fuel load. It was not simply a
case of one person confusing pounds with kilograms.
The incident demonstrates why safety-critical calculations need
clearly labelled units, standard procedures and an independent
reasonableness check. A fuel quantity should be verified as both
a volume and a mass before an aircraft departs.
Source:
Final report of the inquiry into Air Canada Flight 143,
hosted by the Federal Aviation Administration
.