Mars Polar Lander: What Went Wrong?

Welcome to Mars; a world where water once flowed freely and pooled into oceans; a world once very much like Earth, a world where life may have evolved; a world now in the grip of a billion-year-old ice age; a world that we will learn to take care of and make bloom again; but until then, a world unforgiving of mistakes.

Ten days ago, the Mars Polar Lander spacecraft entered the Martian atmosphere and was never heard from again. Ten weeks before that, its sister spacecraft, the Mars Climate Orbiter, was destroyed as is entered the Martian atmosphere.

Are the two failures related? Probably not. The Mars Climate Orbiter was not designed to encounter the Martian atmosphere, and only did so because of a navigational error. However, the Mars Polar Lander, along with the two Deep Space 2 microprobes, should have passed through the atmosphere unscathed to perform their missions on the surface of Mars.

What went wrong? We may never know for certain what happened; however, with the few facts that we do have, we can logically speculate on the fate of the mission.

All times are for December 3, 1999 and are in Earth Receive Time (ERT). The spacecraft altitude and velocity relative to Mars are provided in parentheses.

In the mission timeline, we see that the spacecraft oriented itself for atmospheric entry twelve minutes prior to the scheduled landing, and that Loss of Signal was planned to occur at this time, as the spacecraft’s antenna alignment with Earth was broken. Two minutes later, the Cruise Stage and the two Deep Space 2 microprobes called Amundsen and Scott were to have separated from the aeroshell containing the lander.

The Cruise Stage contained propulsion, guidance, communications, and power systems needed during the eleven-month cruise from Earth to Mars, as well as the DS2 microprobes. It was critical that this separation take place so that the entry body would be unencumbered, and would be able to aerodynamically point itself during entry like an arrow in flight.

According to the timeline, the lander should have deployed its solar panels five minutes after landing, pointed its medium gain antenna at Earth eight minutes after landing, and begun transmitting to Earth 24 minutes after landing. About seven hours later, the DS2 microprobes should have begun transmitting as well.

The key fact is that not only was contact with the Mars Polar Lander never established after landing, but that neither of the two Deep Space 2 microprobes were ever heard from. Three separate mission-critical failures, one for each mission element, is an extremely unlikely scenario.

In failure analysis, one seeks the simplest explanation that fits the facts. What single failure event can account for the loss of all three mission elements?

Logically, once the mission elements separated from each other, three separate failures would have been required to take all of them out. For instance, if the Mars Polar Lander landed on the edge of a large rock and fell over, this would not have affected the Amundsen and Scott probes, which were designed to impact on Mars several kilometers away. One would have to speculate on two additional failure modes, caused by encountering the surface of Mars, that would have affected these probes.

Thus, the simplest explanation is that a single failure occurred while these three mission elements were still attached. According to the mission timeline, the only two events planned during the two minutes between Loss of Signal and the separation of the Cruise Stage and the Deep Space 2 microprobes were the reorientation maneuver and the separation sequence.

If the reorientation maneuver failed, it would have had to fail spectacularly. The spacecraft would have had to have gone into a high rate of tumble, otherwise the microprobes would have been able to separate cleanly, survive atmospheric entry, impact on the surface, and perform their mission.

What might have happened if the separation had failed?

From a hardware standpoint, it is unlikely that all of the pyrotechnic separation devices failed. The most likely scenario is that there was a partial separation, and that the Cruise Stage remained attached to the backshell by one unexploded bolt. Would Amunden and Scott have deployed properly under this condition, or is the more unlikely scenario necessary? If the separation sequence flat out did not come off, this would point to a software failure, in which case an investigation of the software will reveal the "bug".

Interestingly, I have it from a Cal Tech student that "rumor had it that the pyrotechnics were messed up and that JPL was trying to solve the problem in the weeks before arrival." Obviously, no one could go up there and fix a hardware problem, so what could they have been trying to solve but a software problem?

In any case, let us speculate that both the Cruise Stage and the Deep Space 2 microprobes were still attached to the lander’s backshell as it encountered the atmosphere five minutes after a failed separation sequence. The first item to be blown off the vehicle would have been one of the Cruise Stage’s solar panels. We would then have an aerodynamically asymmetric configuration. Have you every faced the palm of your hand into a strong wind, such as when you are driving down the freeway? Imagine that effect magnified many-fold. The spacecraft would have immediately gone into a violent tumble.

As the back end of the vehicle, including the Cruise Stage, the DS2 microprobes, and the backshell, repeatedly rotated forward to be exposed to the extreme aerodynamic and aerothermal forces of high-speed entry, the Cruise Stage would have fallen away in chunks, possibly damaging the backshell. Exposing the backshell to the full brunt of entry forces would have eventually caused it to structurally fail. The lander itself, certainly already damaged by the breakup of the backshell, would now have been exposed, and the hot, high-speed stream of gasses would have blown the vehicle apart like a bug on a windshield.

As for the two microprobes, Amundsen and Scott, they would have blown off the Cruise Stage at some point, but could have been damaged during the breakup and/or spun off from the vehicle in a condition that precluded their surviving atmospheric entry.

We have had a bad streak this year with the failure of the Mars Climate Orbiter, the Mars Polar Lander, and the Amundsen and Scott microprobes. We will have successes in the future on Mars, but we must expect occasional failures as well, and some day failures on missions to Mars will cost us human lives. This is not a pleasant prospect to contemplate, but frontier life has always entailed greater risks in the search for greater opportunities, and the New Frontier is certainly no exception. It is not safe out there, and it is not for the faint at heart. But taking the reasonable risks that are necessary to turn the Unknown into the Known, and to make our home there, is nothing new in human experience. Our species came out of Africa just a few tens of thousands of years ago; now, in our time, we are in the process of coming out of Earth. What we have seen this year are a few missteps in the great migration that is to come.