Quick question: If it’s 30 degrees outside, should you wear a coat?
The only technically correct response to this is, 30 degrees of what, Celsius or Fahrenheit?
You might as well assume the units depending on what’s commonly used around your place, in any case, 30°C is beach weather, but 30°F (-1.12° C) is below freezing.
You might make a similar mistake with units while working out a physics problem, but you know what, who cares?
What’s the worst case scenario?
You would lose a mark or simply get no marks at all? That is all, right? Well, NASA has something to tell you.
Someone actually messed up with units and it cost NASA a 600 million dollar space mission.
Story time.
NASA’s Mars Climate Orbiter Mission Failure (1999)
As a part of NASA’s Mars Surveyor Program, they launched two separate missions, the first one being the Mars Climate Orbiter (MCO) mission (11 December 1998), and the other was the Mars Polar Lander (January 1999).
The Mars Climate Orbiter was primarily supposed to be deployed into an orbit around Mars and act as a communication relay for the upcoming Mars Surveyor ‘98 Lander or the Mars Polar Lander and its Deep Space Probes.
From 11th of December 1998 to about nine and a half months later on 23rd of September the MCO arrived at its destination, and was supposed to execute aerobraking to reduce its velocity and move to a near circular orbit around Mars.
However “discrepancies” in the on ground-based systems on Earth, caused an error resulting in incorrect calculations of the trajectory that sent the spacecraft too close to Mars, about 170 km from the surface of the planet and that eventually led to a crash of the spacecraft onto the Martian surface.
Due to these subsequently failed missions to Mars, JPL and NASA had commissioned two failure review boards, followed by the establishment of the Mars Climate Orbiter (MCO) Mishap Investigation Board (MIB).
What Went Wrong with the Mars Climate Orbiter
Like many spacecraft, the Mars Climate Orbiter (MCO) relied on thrusters to manage its translational motion, while reaction wheels were employed to control its attitude i.e., its orientation in space.
However, as reaction wheels accumulate angular momentum over time due to external torques (like solar radiation pressure or other minor asymmetries), they must periodically be desaturated.
This process, known as Angular Momentum Desaturation (AMD), involves offloading the built-up momentum by firing the spacecraft’s thrusters in a carefully calculated counter-direction.
So very simply what happens is that when any object gets close to the planet, the planet’s gravity pulls the spacecraft inside, making the spacecraft fall onto the surface.
To prevent that from happening, the thrusters on the spacecraft are used to slow down the velocity, and move it in a specific direction, so that it doesn’t get too close to the planet, and starts stably orbiting the planet.
And these AMD events take place in batches, so each time an AMD event would occur, approximately once every 17 hours, the spacecraft’s onboard “Small Forces” software kept a count of the number of thruster pulses performed, and its duration, then the thruster’s alignment and spacecraft’s altitude was used to was used to determine the inertial direction of the imparted change in velocity.
This data was then transmitted back to the ground-based Spacecraft Performance Analysis System (SPAS) which would use the data to refine estimates of the spacecraft’s current position and velocity. The trajectory team used these updated outputs to track the MCO’s flight path and also to direct future AMD events.
The mission plan anticipated a brief loss of signal approximately 5 minutes after the start of the Mars orbit insertion burn which was basically an expected blackout due to the spacecraft passing behind Mars.
However, contact was lost 49 seconds before that expected point, and communication was never re-established.
Later calculations, made using corrected data, revealed that the spacecraft had entered the Martian atmosphere at an altitude of roughly 57 kilometers which was far below, and that led to a crash.
The primary contributor to the failure was a unit mismatch between two collaborating teams. Lockheed Martin, which developed both the spacecraft and parts of the ground-based software, operated using British Imperial units (specifically, pounds-force for force measurements).
The software onboard on the spacecraft did calculate the predicted ∆v from each pulse in the correct units, the ground based SPAS did not use that data, and instead performed their own calculations..
Meanwhile, NASA’s navigation team assumed that the software outputs were in metric units (Newtons), because SI units are the agency standard.
This misunderstanding meant that values computed in pounds-force were incorrectly interpreted as Newtons by the ground navigation software.
Since 1 lbf ≈ 4.448 N, the thruster impulses calculated during AMD events were underestimated by a factor of about 4.45.
As a result, each AMD event introduced a larger change in velocity than expected by increasing the thruster force. This trajectory deviation pushed the spacecraft significantly off-course, ultimately dooming the mission.
Total Loss Incurred by NASA Due to MCO
The Mars Surveyor ‘98 program spacecraft development cost 193.1 million dollars. Launch costs are estimated at 91.7 million dollars and mission operations at 42.8 million dollars, which totals to 327.6 million dollars.
Accounting for inflation from 1998 to 2025, it’s about $646 million, which *ahem*… is quite a heavy price to pay for such a trivial human error.
Except for that they also had to call off the subsequent mission the Mars Polar Lander, because the MCO was supposed to be the bridge between the lander and the Earth.
Closing thoughts
“People make errors. The problem here was not the error. It was the failure of us to look at it end-to-end and find it. It’s unfair to rely on any one person.”
Tom Gavin, NASA Jet Propulsion Laboratory
This serves a good lesson to anyone who underestimates the importance of units in any routines of science. This should also absolutely be used in pedagogical settings to make students really feel the importance of units.
And this one from NASA is only one of other such events that went wrong because of unit mismatch, for instance, In 1983, a Boeing 767 flight from Montreal to Winnipeg had to do an emergency landing because the plane was not adequately fueled because the Air Canada staff had recently started using the metric system instead of the customary system.
There are more such events. We recommend checking out this article: “Metrication Errors and Mishaps” by the National Institute of Standards and Technology, U.S.
References
