The Third Field

Cases from the record · 1998 – 1999

Mars Climate Orbiter

The discrepancy was noticed. It stayed informal.

A spacecraft was lost at Mars because one file used the wrong units. The investigation board looked past the arithmetic to how the warning signs were handled.

Process lens Follow through → Review

4.45the factor by which navigation underestimated the thrusters’ effect (Mishap Investigation Board)
49 show much earlier than predicted the signal was lost
9 monthsof flight between launch and arrival at Mars

What happened

NASA’s Mars Climate Orbiter launched in December 1998 and was lost on 23 September 1999 as it fired its engine to enter orbit around Mars. It passed far closer to the planet than planned.

The Mishap Investigation Board traced the root cause to a ground software file that reported thruster data in pound-force seconds when the agreed specification required newton-seconds. Its reports then examined why the error was not caught during nine months of flight.

  1. Launch from Cape Canaveral.

  2. Thruster data files are not used in navigation for about four months because of format errors.

  3. Navigation data repeatedly suggest a closer path to Mars. The discrepancies are noted but not resolved.

  4. An additional course correction is discussed verbally. It is not performed.

  5. The orbiter begins its engine burn behind Mars. The signal is lost 49 seconds early.

  6. The Mishap Investigation Board releases its first report.

What the record shows

  1. 01

    The interface was specified, and not followed.

    The board traced the loss to “the failure to use metric units in the coding of a ground software file, ‘Small Forces,’ used in trajectory models.” The specification required metric units; the software on the spacecraft itself was correct.

  2. 02

    The symptom was seen and reported informally.

    The board found the navigation discrepancies were “noted but only informally reported,” often by email rather than through the formal process for raising anomalies, and that “these discrepancies were not resolved.”

  3. 03

    Concerns did not reach the people deciding.

    The board found that the operations navigation team “did not communicate their trajectory concerns effectively” to the spacecraft operations team or project management, and that concerns were not elevated to the highest levels within the contractor.

  4. 04

    Ownership was unclear.

    The board’s later report records that “Who’s in charge?” was a recurring question during its investigation.

Our reading · Follow through → Review

The condition that mattered was written down: use metric units. What failed was the path between noticing that something did not fit and making that observation part of the decision. A concern raised in an email is easy to lose; a concern attached to the decision it affects is harder to overlook.

Context and limits

The error was in units of impulse, not in inches and feet, and the board named no individual. NASA gives program-level costs for the two Mars ’98 spacecraft rather than a figure for the orbiter alone, so none is printed here.

From the record to a design question

When the work contradicts an assumption, where does that signal go—and who must respond?

What this asks of the process

Connect evidence from the work to the assumption or condition it affects. Identify a reviewer, preserve the unresolved issue and bring the relevant decision back into view when circumstances change.

What we do not conclude

A connected record supports coordination; it does not replace engineering assurance, interface testing or formal anomaly resolution. No claim is made that Sikizana would have prevented the loss.

This is our design inference from the case. The case does not test Sikizana. See the research that informs the sequence

Sources

  1. Mars Climate Orbiter Mishap Investigation Board, Phase I Report (10 November 1999)
  2. Mars Climate Orbiter Mishap Investigation Board, Report on Project Management in NASA (13 March 2000)
  3. NASA news release 99-134 (10 November 1999)
  4. NASA Space Science Data Coordinated Archive, Mars Climate Orbiter (1998-073A)

Based on the linked official record, participants’ accounts and original research. Retrospective accounts and investigations have different limits; they are not controlled tests of a decision method. No organization named here is presented as a customer, partner or endorser of The Third Field.