Spirit Condition Serious

Since Wednesday, the Spirit rover’s previously flawless performance has generated anomalies that an engineering team has been analyzing, along with preparing for Saturday’s arrival of the second rover, Opportunity. Peter Theisinger, the Mars Exploration Rover Project Manager, said "An anomaly team has been setup, and their activities will be severed from the Opportunity landing team." Any change in Spirit’s health "could be days, could be weeks," said Theisinger. "Don’t expect a big sea-change in the next several days."

The Mini-Thermal Emission Spectrometer (Mini-TES), which was undergoing a motor movement when the first fault was detected Wednesday.
Credit: NASA/JPL

The Spirit rover’s software currently reports intermittently that its high-gain antenna is in ‘X-band fault condition’, which may or may not show a relation to earlier movements of this rapid data-transfer broadcaster. "The rover is not sleeping at night", said Theisinger, referring to the scheduled low-power phase of the martian day, when its solar panels cannot charge the rover’s batteries. "Power and thermal states are okay, meaning the rover is in a safe, stable mode, but after prolonged activity without shutting down, the batteries will run down. A command was tried to go to low-power, but was initially ignored. …The batteries can go to zero-charge and recharge, their chemistry allows that."

Corrective actions depend on first identifying ‘the root cause’, said Theisinger, noting that software may ‘not be behaving normally [but] we don’t know if this is related to a hardware event that is impressing itself onto software". On Wednesday, when the fault was first observed, the rover tried to reset itself over sixty times–a step built-in to the rover’s architecture comparable to rebooting a home computer. " When the rover reboots, things don’t get better. We need a starting place, because the rover’s software package is built in modules and this may be a software contention issue. These states can be very time-intensive to reproduce. All we know is when the fault occurred, the rover was doing a motor operation that didn’t go to completion. That may indicate the root cause was hardware, not purely a software thing."

The motor movement that did not go to completion was related to the infrared camera mirror for the mini-TES instrument, which images the temperature difference and deduces many rock properties such as density, thermal inertia during cool and hot portions of the martian day. The possible choices for such hardware anomalies include a "sensor failure, electronics, cabling–a mechanical anomaly not visible" to the rover’s camera itself. "This problem can be triggered outside software; it occurred first when running the mini-TES mirror’s motor, which didn’t run to completion." Because of the software’s capability to diagnose faults, if the root cause turns out to be hardware-related, contradictory readings may "impress [itself] onto software." One workaround proposed was to isolate a possible hardware fault, to get software to ignore contradictory data.

The rover’s Pancam Mast Assembly with camera bar in stowed position.
Credit: NASA/JPL/Cornell

Current data feedback from the rover comes intermittently in small amounts, ranging from 10 bits per second (bps) to 120, about a thousand times less than the 128 megabits per second transmitted from the high-gain antenna.

When asked to speculate on Spirit’s prospects, Theisinger suggested future "functionality. The chances for a return to perfect are not good, the chances for it not working are low. We’re probably at a mid-point with unknown effects on rover science…To me, this looks like a one-off, not systemic," he said, referring to whether changes would be needed or possible with Spirit’s twin rover Opportunity.

In analogy to dialogue adopted from the movie, Apollo 13, in which flight controllers were polled to find out "what we’ve got that’s good", Theisinger concluded "we don’t know ‘what we’ve got that’s good’. "We don’t know what’s broken, or what started this chain of events. And we don’t know the consequences."

Mars Exploration Rover with main instruments indicated by location on the unfurled instrument after stand-up.

Chief Engineer Wayne Lee described some of the interesting observations from Spirit’s landing, the "first test of this system, end-to-end." Lee noted that while many subsystems can be tested prior to launch, martian conditions cannot be simulated such as gravity and some integrated systems working together. Since the other rover Opportunity will be the "highest altitude landing NASA has ever tried," Lee stressed that only one non-cosmetic change was made from the lessons learned during the Spirit landing. "Opportunity will deploy its parachute 4500 feet higher than Spirit for three reasons: this is the highest altitude landing tried, we want to guard against the martian atmosphere changing, so we will buy more time to fire the retro-rockets; and finally, to buy insurance against descent rate limiters."

Because of a Christmas dust storm on Mars, the entire equator showed slightly hotter than predicted temperatures. The dust traps incoming solar heat in these cases, and the hotter atmosphere puffs up or becomes thinner–a problem for the supersonic parachute design which depends on drag.

One unpredictable part of Spirit’s descent was the local wind conditions near Gusev crater. "Two gusts of wind hit Spirit," said Lee. "One at 40,000 feet, when the spacecraft was traveling at Mach 6 (or six times the speed of sound). This is comparable to clear-air turbulence one gets when crossing over the Rocky Mountains, towards the Denver airport. The effect is the capsule began to wobble. The other gust hit Spirit near the ground, which in the absence of corrective firing would have driven the airbag and capsule into the ground at 50 miles per hour."

The airbag and parachute system for the Mars Exploration Rovers (MER) "is based on the Pathfinder design which has sensitivity to horizontal winds," said Miguel San Martin, the Guidance and Control Chief. "Airbags don’t like horizontal velocity," said San Miguel, referring to the possibility of snagging the bladder fabric on sharp rocks as distinct from a true vertical bounce. The MER "compensates for this weakness with sensors and a last second steering rocket", said San Miguel. Opportunity’s descent will rely on the same performance, although from a slightly higher landing destination and thus less atmosphere for slowing its descent.

Click for larger view. Annotated image from Mars Orbital Camera showing the backshell, parachute, and rover base petal. Approximately 300 meters separated the parachute from it eventual resting place. After 28 bounces, include a fourth bounce on the lip of a crater, the lander came to a stop with its base petal down. The rover itself is not far enough from the Columbia Memorial Station to be visible in the reflective glare until it is able to move further.

The descent team has completed their engineering re-enactment of what Spirit actually went through when bouncing nearly 300 meters –about three football lengths–from a height of 28 meters, or about the roof distance of a five to six story building. What makes this reconstruction possible is the real-time health and status reports broadcast from the MER back to orbitting relay satellites, the Mars Global Surveyor and Mars Odyssey. The team reports ‘folding this knowledge into Opportunity’s deployment to get a descent rate limit that is comfortable."

Dr. Mike Malin, who is principal investigator for the Mars Orbital Camera on the orbiting Surveyor, presented one of the more remarkable pieces of remote sensing data yet obtained: an actual picture of the Spirit rover from a 400 kilometer altitude. The picture shows the base station’s orientation, and positive limits to its actual location within one-half meter. This high-resolution was possible using a new imaging technique. If the spacecraft’s error ellipse for location is small, this survey method previously has imaged both Viking and Pathfinder sites, but Malin notes that martian dust makes those two spacecrafts resemble rocks after so long. The high contrast possible because of illumination and lack of dust at Gusev, Malin said, may enable their camera eventually to picture the rover itself, once it drives out of the glare from its base petal.

The Spirit/Columbia Memorial Station is clearly seen as a bright feature in the image, as are the parachute and backshell from which Spirit was detached during the landing on 4 January 2004. Also evident is a dark scar on the rim of a crater to the northeast of the lander; this dark marking was not present prior to landing, and is believed to have been caused by the impact of Spirit’s heatshield. The lander is white because the data received from Mars were saturated at this location–that is, the lander was so much brighter than the surrounding terrain that the camera saw it as a white object.

Related Web Pages

Water Signs
Microscopic Imager
Gusev Crater
Pancam– Surveying the Martian Scene
Mössbauer spectrometer
Alpha Proton X-ray Spectrometer
Mars Rover: The Owner’s Manual
Reverse Robotic Origami