Unit Conversion Errors: What NIST's Cases Actually Show
Unit Conversion Errors: What NIST's Cases Actually Show
By James · May 28, 2026
When I was a child, I loved playing with a flashlight in the dark. I would shine it at the school, the church, my neighbor's house, the trees, and even straight up into the night sky, wondering whether that beam was one continuous stream or a flow of tiny particles.
In college, I came across an idea that puzzled me even more: the official definition of one meter is based on the speed of light. The meter was first tied to a fraction of Earth's meridian, later embodied in a metal standard bar, and since 1983 has been defined as the distance light travels in a vacuum during 1/299,792,458 of a second.
Years later, while traveling abroad, I kept running into a practical problem. I was used to miles and feet — the everyday units most Americans grow up with — and suddenly I was hearing things like, "The airport is about 18 kilometers away." That meant stopping, pulling out my phone, and converting kilometers into miles just to get a basic sense of distance.
Eventually, a simple question came into focus: if nearly the entire world uses the metric system, what do Americans actually gain by doing things differently? NIST argues that running two systems in parallel creates preventable unit conversion errors, added complexity, and real financial cost.
What a Meter Really Measures
The Mars Climate Orbiter (1999): How NASA Lost $327.6 Million
The Boeing 767 That Flew Without Fuel
When the Costs Are Smaller — and When They Aren't
Frequently Asked Questions
What a Meter Really Measures
The meter's current definition is anchored to a physical constant — the speed of light in a vacuum. No object, no reference bar, and no geographic measurement can match that for stability.
The original definition went the other way. In the 1790s, French scientists defined one meter as one ten-millionth of the distance from the North Pole to the equator, measured along the Paris meridian. The result was a platinum reference bar, later refined and distributed to member nations as an international prototype. Countries sent their bars to Paris for periodic comparison.
In 1983, the International Bureau of Weights and Measures redefined the meter entirely. According to BIPM, it is now the distance light travels in a vacuum in exactly 1/299,792,458 of a second.
The original definition required surveying a planet. The current one is 1/299,792,458 of a second.Unit conversion errors happen when two teams working on the same system use different measurement units — and neither catches the discrepancy before it matters. That kind of mismatch has caused spacecraft losses, flight emergencies, and avoidable financial cost — and four such cases are laid out in NIST's own published record.
The Mars Climate Orbiter (1999): How NASA Lost $327.6 Million to a Unit Mismatch
NASA launched the Mars Climate Orbiter in December 1998. The spacecraft was designed to study the Martian climate and atmosphere, and to serve as the communications relay for the Mars Polar Lander, which was due to touch down on the Martian surface within months — the kind of infrastructure investment that takes years to build and a single navigation error to lose.
The navigation software ran without visible problems for nearly ten months. Then, on September 23, 1999, the orbiter attempted to enter Mars orbit. It came in at the wrong angle, passed too close to the Martian atmosphere, and burned up. Contact was never reestablished.
The navigation software performed as implemented — but the unit mismatch had never been caught in testing or validation. One engineering contractor had been delivering thruster performance data in pound-force-seconds; NASA's navigation team was reading those values as newton-seconds. Working-level navigators did notice the spacecraft's trajectory readings drifting from predictions during the cruise to Mars, but those concerns were raised only informally and never resolved through the project's formal anomaly-tracking process — so the root cause itself went unaddressed for the entire flight.
One number. Two units. $327.6 million gone.According to NASA and NIST, the mission loss is documented at $327.6 million in 1998 dollars. The spacecraft's final navigation data made the unit discrepancy plain — but only after the loss.
The Boeing 767 That Flew Without Fuel
Air Canada Flight 143 departed Montréal on July 23, 1983. The Boeing 767 was newly in commercial service, with fuel systems calibrated to the metric units Canada had recently adopted.
During the fueling stop in Montréal, ground crew and the flight crew together calculated the required fuel load using 1.77 pounds per liter instead of the correct metric factor of approximately 0.8 kilograms per liter. The aircraft received about 10,000 kg of fuel. It needed 22,300 kg. The shortfall was not caught before departure.
The Boeing 767 that became known as the Gimli Glider left Montréal carrying less than half the fuel it required.Near Red Lake, Ontario, both engines flamed out. The aircraft — carrying passengers and crew on a scheduled flight — glided to an emergency landing at a former Royal Canadian Air Force base in Gimli, more than 200 kilometers away. CBC's archival record of the event identifies the conversion factor error explicitly. NIST lists it among the documented cases of metrication mishap. No fatalities resulted, and the Boeing 767 was repaired and returned to service.
When the Costs Are Smaller — and When They Aren't
Not every unit mistake ends in wreckage.
In 1999, the crew of Korean Air Cargo Flight 6316 was cleared to climb to 1,500 meters after takeoff from Shanghai. According to China's Civil Aviation Administration (CAAC) 2001 investigation, the crew interpreted the clearance as 1,500 feet — an incident that also appears in NIST's own incident record. The confusion between meters and feet contributed to a loss of control shortly after departure. The aircraft crashed, killing all three crew members and five people on the ground.
In 2007, the US House of Representatives approved the purchase of 24,000 short tons of carbon offsets. NIST reports that 30,000 metric tons were purchased instead — a unit mismatch that produced an extra cost of approximately $24,447.
The dollar figures across these cases — $327.6 million and $24,447 — exist on entirely different scales. That gap is precisely the point.
To be fair: the House overpurchase was caught, documented, and reported. That is exactly how NIST knows about it. Cases that surface in audits tend to be the recoverable ones. The fatal ones, like Korean Air 6316, were not — which is the strongest argument for standardization.
| Incident | Year | Unit Error | Documented Cost |
|---|---|---|---|
| Mars Climate Orbiter | 1999 | Pound-force-seconds vs. newton-seconds | $327.6 million (1998 USD) |
| Air Canada Flight 143 (Gimli Glider) | 1983 | Pounds per liter vs. kg per liter | Aircraft damaged; emergency landing; no fatalities |
| Korean Air Cargo Flight 6316 | 1999 | Feet vs. meters (altitude clearance) | Fatal crash; 8 killed |
| US House carbon offset purchase | 2007 | Short tons vs. metric tons | ~$24,447 extra |
The deeper issue is not sentiment or habit. It is cost. Every time miles, feet, pounds, short tons, meters, and kilograms are used side by side without a clear conversion, another opening for error appears — and NIST's own case record shows that those errors can waste money, destroy equipment, and contribute to deaths. (NIST)
The uncomfortable truth is this: maintaining two measurement systems is not preserving tradition; it is preserving inefficiency. Aviation, engineering, science, and international trade all work better when everyone speaks the same numerical language — and the United States keeps paying the price for refusing to fully join them.
Frequently Asked Questions
What caused the Mars Climate Orbiter to fail in 1999?
NASA's Mars Climate Orbiter was lost because one engineering contractor delivered thruster performance data in pound-force-seconds while NASA's navigation team read those values as newton-seconds. The unit mismatch went undetected for nearly ten months of flight. The spacecraft burned up in the Martian atmosphere on September 23, 1999. According to NASA and NIST, the mission loss totaled $327.6 million in 1998 dollars.
How did Air Canada Flight 143 end up running out of fuel?
Air Canada Flight 143 ran out of fuel because ground crew and the flight crew calculated the required load using 1.77 pounds per liter instead of the correct metric factor of approximately 0.8 kilograms per liter. NIST and CBC's archival record show the Boeing 767 received about 10,000 kg instead of the required 22,300 kg. Both engines flamed out near Red Lake, Ontario, and the aircraft glided to an emergency landing at a former airfield in Gimli, Manitoba — earning the name "Gimli Glider."
Why does the US still use miles and feet instead of the metric system?
The United States continues to use customary units — miles, feet, and pounds — primarily because of historical adoption, entrenched infrastructure, and the political and economic cost of conversion. NIST advocates for metrication, noting that running two systems side by side creates preventable unit conversion errors and added complexity in aviation, science, and international trade.
How much did the Mars Climate Orbiter cost NASA?
The Mars Climate Orbiter mission cost approximately $327.6 million in 1998 dollars, according to NIST's documentation of metrication mishaps. That figure covers the full cost of building, launching, and operating the spacecraft, which was lost in September 1999 after a navigation error sent it too close to Mars.
What is the official definition of one meter today?
Since 1983, one meter is defined as the distance light travels in a vacuum in exactly 1/299,792,458 of a second. The International Bureau of Weights and Measures (BIPM) established this definition to replace the earlier standard based on a fraction of Earth's meridian. The current definition ties the meter to a universal physical constant rather than any geographic measurement.
Can unit conversion errors cause plane crashes?
Yes. Korean Air Cargo Flight 6316 crashed in 1999 after the crew confused meters and feet during an altitude clearance. According to China's CAAC 2001 investigation, the crew was cleared to climb to 1,500 meters after takeoff from Shanghai but interpreted the instruction as 1,500 feet — an incident that also appears in NIST's own record of metrication mishaps. The altitude confusion contributed to a loss of control shortly after departure, killing all three crew members and five people on the ground.
Sources & References
- NASA — Mars Climate Orbiter mission: science.nasa.gov
- NIST — Metrication Errors and Mishaps: nist.gov
- NIST — US Metrication: nist.gov
- CBC Archives — When a Metric Mix-Up Led to the Gimli Glider Emergency: cbc.ca
- CAAC — Korean Air Cargo Flight 6316 Investigation Report (2001), via: en.wikipedia.org
- Britannica — Metre (measurement): britannica.com
- BIPM — History of the SI, the metre: bipm.org
Comments
Post a Comment