Confidence with Doubt
A Followup To the UPI vs NASA Saga
by Steve Russell

     

    I am 100% sure that I think I am right!

    The reports on all of the failed Mars missions including the loss of the Mars Polar Lander (MPL) are now out for the public to digest and discuss. The review boards have confidently determined a probable cause despite previous doubts that a cause would be found. Within the pages of the report are countless degradations to NASA's publicly perceived ability to explore Mars.

    UPI allegations of tampering with test conditions on the catalyst bed ignition processes have not been disproved or confirmed within the reports. However, such actions could certainly have transpired, given one's close examination of the report findings relating to work conditions and the knowledge of personnel.

    1. Probable Cause
    2. According to the reports, the most probable cause for the loss of the MPL is a premature shutdown of the engines. As the spacecraft descended, the landing legs extended triggering sensors as they opened. The sensors are used to indicate a successful touchdown and tell the computer to shutdown the engines. The landing legs triggered the sensors at an altitude of 40 meters causing the MPL to hit the surface at 22 meters/second.

      What makes this scenario more interesting is the claim that the problem went unnoticed. Either an amazing sequence of coincidental bad luck plagued the software engineering team, or NASA's "Faster, Better, Cheaper" mentality caused yet another serious blow to the space program.

    3. The Failure Process
  1. An essential requirement was somehow forgotten and left out of the Software Requirements Specification document. The requirement specified that the touchdown sensor data would not be used until 12 meters above the surface.
  2. Meetings were held to walk through and examine the logic of the software. The attendance record was perfect with the exception of one person, the only one who could have discovered the problem.
  3. Unit test cases did not include a test to find such a problem.
  4. Software integration tests did not find the problem.
  5. A test was performed on 4 June 1998 using the faulty software. The results indicated that the sensors were not wired correctly and so they were not being monitored. The wiring was fixed but the tests were not performed again.
  6. The problem was eventually discovered, not by the MPL team, but from tests on the 2001 Lander. The problem was discovered accidentally when a test engineer mistakenly pushed a button which was for indicating an early landing. This resulted in a premature shutdown of the engines and the problem was found.
The decision to make this scenario the most probable was made despite a convenient total absence of corroborating flight data. The JPL Special Review Board (SRB) appears to make confident conclusions and then raise doubts about those conclusions.

The Confidence:

The Doubt: How can there be compelling evidence with a lack of data?

UPI Allegations

UPI has accused NASA of covering up evidence of tampering, carried out by someone in middle management. Information was given to UPI by sources close to a panel that was investigating the loss of the MPL. Their sources stated that the catalyst beds were tested at the predicted temperatures but failed. Someone then increased the temperatures in the tests so that the catalyst beds would successfully pass.

Data provided by Lockheed Martin Astronautics predicted the temperature of the catalyst beds to be approximately -30 degrees C, well below the freezing point of the hydrazine fuel, which is 1.5 degrees C! Was this data available and understood during the testing phase of the catalyst beds? It would explain why someone had to increase the temperature of the tests to get the catalyst beds above freezing point and approved.

In a normal situation where two teams are working together, it would be either difficult or impossible for one person to cover up something like this. Inside NASA anything and everything appears to be possible. The two teams involved here are the propulsion and thermal (P&T) groups.

The SRB report stated that the MPL development staff worked 60 hours/week with some extended periods of 80 hours/week for some workers. The P&T groups were also over worked leaving no time for analysis or collaboration on critical issues. The report also criticised the poor communication between the two groups.

The SRB report also highlighted the inadequacies of personnel knowledge regarding the data they were looking at. The temperature requirements for the propulsion units and the safety margins involved were not fully understood by the thermal team. The test data used was never fully evaluated or understood by either team! The temperature models used throughout testing were never even verified due to inadequate instrumentation.

Despite the problems, the SRB has again made a confident conclusion followed by serious doubts.

The Confidence:

The Doubt:

"·the Propulsion Subsystem and thermal designs did contain four potentially serious, if not catastrophic, weaknesses."5

A system cannot be reliable if it has catastrophic weaknesses.

Corrective Action

On 29 March 2000 at the Jet Propulsion Laboratory, NASA Administrator Dan Goldin said, "I salute [the Mars 98] team's courage and conviction. And make no mistake: they need not apologize to anyone. They did not fail alone. As the head of NASA, I accept the responsibility. If anything, the system failed them."

Does the system need changing? Does NASA need improved management, or perhaps a new administrator? We the people need a manned mission to Mars. The question is, can Dan Goldin and his current management staff achieve this task?

Conclusion

The majority of the SRB report contains a very detailed and accurate analysis with many findings and recommendations. However, the findings in the two most important areas of the report do not appear to be as precise or confidently concluded. The sections describing the probable cause along with the propulsion and thermal sections contain confident conclusions followed by doubts leaving no real definite answers.

The information given to UPI was either disinformation, or truthful facts that the sources knew would be removed from the final report after being reviewed by NASA and the White House. The report does provide enough information to confirm that the circumstances were favourable for the allegations to be true. However, if more people do not come forward with information, the truth will never be known.

1 JPL Special Review Board, Report on the Loss of the Mars Polar Lander and Deep Space 2 Missions, 22 March 2000, Page xi

2 JPL Special Review Board, Report on the Loss of the Mars Polar Lander and Deep Space 2 Missions, 22 March 2000, Page 20

3 JPL Special Review Board, Report on the Loss of the Mars Polar Lander and Deep Space 2 Missions, 22 March 2000, Page xi

4 JPL Special Review Board, Report on the Loss of the Mars Polar Lander and Deep Space 2 Missions, 22 March 2000, Page 81

5 JPL Special Review Board, Report on the Loss of the Mars Polar Lander and Deep Space 2 Missions, 22 March 2000, Page 81


All material is copyrighted by THE MILLENNIUM GROUP and may not be used without their express written permission.

RETURN TO THE REPOSITORY