Failures/Galileo (Jupiter orbiter)

Galileo high-gain antenna failed to fully open

galileo-hga-1991

What happened

When commanded open in April 1991, the umbrella-like main antenna opened only partly. JPL analysis concluded that 4 of its 18 ribs were stuck.

Why

High friction between a few rib pins and their sockets. The bonded MoS2 dry lubricant on those pins was most likely lost before launch, and the bare titanium pins galled (seized) against the Inconel sockets. The deployment motor then bent the pins during the first attempt.

Explanations in the sourcesSource
Lubricant shaken or worn off during pre-launch transport (cross-country truck trips).Galileo spacecraft description (NASA Planetary Data System catalog file); The Galileo trials (Spaceflight Now, 2003)
Lubricant degraded during the long storage caused by launch delays.Unconfirmed; from a secondary source.Galileo's antenna problem (The Logbook)

Evidence

MeasurementValueSource
friction coefficient, lubricated Ti-6Al-4V on Inconel 718, in vacuum0.04Properties Data for Opening the Galileo's Partially Unfurled Main Antenna
friction coefficient, lubricated Ti-6Al-4V on Inconel 718, in air0.13Properties Data for Opening the Galileo's Partially Unfurled Main Antenna
dry-film lubricant endurance life, vacuum vs airabout 1000 times longer in vacuumProperties Data for Opening the Galileo's Partially Unfurled Main Antenna
friction coefficient, unlubricated Ti-6Al-4V on Inconel 718, in vacuum (galling)1.2 (1.45 when wear debris clogged the interface)Properties Data for Opening the Galileo's Partially Unfurled Main Antenna
friction coefficient, lubricated pin run first in humid air then in vacuum (galling)0.32 in humid air, 1.4 in vacuumProperties Data for Opening the Galileo's Partially Unfurled Main Antenna

Consequence

The mission ran on the small low-gain antenna. Without changes it would have sent only 8 to 16 bits per second from Jupiter. Software changes on the spacecraft and on the ground raised this to as much as 160 bits per second, and lossy compression shrank some image and plasma-wave data to as little as 1/80 of its original size.

Lessons

Space Mushroom's reading of the sources, not quotes from them.

  • MoS2 dry-film lubricants wear far faster in air than in vacuum, so handling, transport and ground tests in air can use up lubricant life before launch.
  • Once a dry-film lubricant is gone, titanium sliding on nickel alloy in vacuum can gall and seize, with friction above 1.
  • Check lubricant condition after every transport and before final integration.