Recently, I singed up for a course on Coursera, titled : Re-Engineering Your Science Curriculum. Not that I’m a science teacher, but I do love to expand my knowledge and skills. I find taking these courses to be the perfect way of leisure studying and more important, to learning things I might have not ever learned about.
This week, the theme of the study is Engineering Failures. The teaching stuff shares two cases of engineering failures, famous and quite costly! I’ve decided to share these with you, as I found them very fascinating! Hope you will enjoy reading about them, as well as learing something new.
Case Study #1: The Mars Climate Orbiter
What Happened: On December 11, 1998, NASA launched the Mars Climate Orbiter (formerly the Mars Surveyor ’98 Orbiter), a 338-kg (750 lb.) robotic space probe. Its mission was to study the climate, atmosphere, and surface changes on Mars, and also to relay communications to the Mars Polar Lander. However, nine months into the mission, on September 23, 1999, NASA lost all communication with the spacecraft as it attempted to enter into a stable orbit around Mars. Communication was never re-established.
What Caused the Failure: NASA’s crash report attributed the failure to errors in the computer navigation system, caused by an embarrassing conversion mistake between the contractor, Lockheed Martin, and NASA’s flight controllers. Lockheed Martin’s computer calculations were based on English measurements, non-SI units of pound-seconds (lbs), common in the U.S. launch industry, rather than the metric units of newton-seconds (Ns) specified by NASA. The result is that the spacecraft’s trajectory around Mars brought it too close to the planet, causing it to enter the upper atmosphere and disintegrate. The cost of the mission was $327.6 million total for the orbiter and lander: $193.1 million for spacecraft development, $91.7 million for launching it, and $42.8 million for mission operations.
Takeaway Lesson: Not using and specifying the correct scientific units caused a very expensive loss of spacecraft. Always write down the units!
Case Study #2: The Hubble telescope mirror
What Happened: The Hubble Space Telescope (HST), named after famed astronomer Edwin Powell Hubble, was carried into orbit by the space shuttle Discovery in 1990. It is a 2.4-meter (7.9 ft.) aperture telescope, with five instruments that can observe ultraviolet, visible, and near infrared spectra. Although the Hubble telescope remains in operation today and has become an important science observatory in space, the discovery of flaws in its primary mirror soon after launch was a huge embarrassment for NASA.
Problems arose almost immediately after the telescope was placed into orbit. Images transmitted from the telescope were blurrier than expected, indicating a serious problem with the optical system. Eventually, computer analysis of the flawed images showed that fault lay with the primary mirror, which had a “spherical aberration.” The mirror had been ground to the wrong shape, making it impossible to achieve proper focus.
What Caused the Failure: The optics company commissioned to build the mirror, Perkin-Elmer, failed to confirm that it was the correct shape, relying solely on a single instrument test to verify its accuracy and ignoring other test results that suggested errors.
“During the initial grinding and polishing of the mirror, Perkin-Elmer analyzed its surface with two conventional null correctors. However, for the final manufacturing step (figuring), they switched to a custom-built null corrector, designed explicitly to meet very strict tolerances. Ironically, this device was assembled incorrectly, resulting in an extremely precise (but wrong) shape for the mirror. There was one final opportunity to catch the error, since a few of the final tests needed to use conventional null correctors for various technical reasons. These tests correctly indicated spherical aberration. However, the company ignored these results, as it believed they were less accurate than the primary device which reported that the mirror was perfectly figured.” (Dunar, 1999)
The investigating commission pointed to numerous managerial failures that contributed to the technical error, both at NASA and at Perkin-Elmer. The contractor failed to supervise the final mirror construction adequately and NASA was criticized for not recognizing the problems with quality control. Relations between NASA and the optics company had also become increasingly hostile during construction due to cost overruns and scheduling delays, resulting in an environment that discouraged speaking out even in the face of obvious errors.
Takeaway Lesson: Don’t rely on a single test to tell you what’s working. Examine your environment to ensure a project is being managed appropriately and doesn’t discourage investigation of problems.
Dunar, A. J.; S. P. Waring (1999). Power to Explore: History of Marshall Space Flight Center 1960–1990. U.S. Government Printing Office. ISBN 0-16-058992-4. Chapter 12, The Hubble Space Telescope PDF (260 KB).