View Single Post
  #10   Spotlight this post!  
Unread 07-11-2002, 02:39
dlavery's Avatar
dlavery dlavery is offline
Curmudgeon
FRC #0116 (Epsilon Delta)
 
Join Date: Jun 2001
Rookie Year: 1996
Location: Herndon, VA
Posts: 3,176
dlavery has a reputation beyond reputedlavery has a reputation beyond reputedlavery has a reputation beyond reputedlavery has a reputation beyond reputedlavery has a reputation beyond reputedlavery has a reputation beyond reputedlavery has a reputation beyond reputedlavery has a reputation beyond reputedlavery has a reputation beyond reputedlavery has a reputation beyond reputedlavery has a reputation beyond repute
Re: Re: Programming errors

Quote:
Originally posted by Wetzel
Guess you missed Dave's post about that....

And yes, it was Dave's fault.
Yeah, I know, it is always my fault - just ask anyone on 116!

-dave

p.s. Even though Wetzel makes me take the blame, the reality is that it wasn't exactly a NASA "bonehead engineer" that made the mistake. A very young engineer, working for the contractor that actually built the spacecraft, was relatively new to the job and had not been fully trained on "how the system worked." His job was to calculate very small propulsion forces used to make minute corrections in the spacecraft navigation system. The other projects he had worked on used English units for their calculations (as required by those other contracts). But he did not know that all NASA contracts specify the use of Metric units for all force (and other) calculations.

The forces he calculated were sent to JPL and then uplinked to the spacecraft. The rest is history. The simple-minded answer is to say that the engineer made a mistake that caused the loss of the spacecraft. A more insightful analysis leads to the understanding that the real problem is that the system was set up with insufficient review and validation of the navigation calculations being performed, and that insufficient checks and balances allowed a small mistake to make it's way all the way through to the spacecraft without being caught. The lesson learned is that systems and procedures need to be designed in anticipation of people making "human mistakes" and be robust enough to recognize and correct them.