The Design and Engineering of Curiosity

by Emily Lakdawalla

  4.3.2.3 Flight software version R11.0.5 (sol 772)

  In October 2014, engineers patched R11.0.4 to R11.0.5 to permanently resolve the problem with the A-side computer that had existed since the sol 200 anomaly. The patch, applied only to the A-side computer, instructed it not to read its bad memory cells, and gave the rover a fully functional backup computer again.

  4.3.2.4 Flight software version R12.0.3 (sol 879)

  In January 2015, they upgraded to version R12. Its improvements were more modest than R11’s:Ability to use inertial measurement units to sense slipping during drilling operations on slopes.

  Other improvements to driving that made the guarded mode easier to sequence.

  Added “hooks” into the flight software that made it easier to add a new ability in the future without a complete flight software upgrade. The hooks were for software added in 2017 that permitted traction control, in which wheels can be driven at different speeds when certain conditions are met, in order to allow wheels going over obstacles to travel faster than other wheels.

  4.4 THERMAL CONTROL

  Maintaining the temperature of a spacecraft’s components within allowable ranges presents challenges for any mission. Space is cold; the Sun is hot; a vacuum doesn’t conduct heat. Mars has a thin atmosphere that mitigates the temperature swings that would occur in a vacuum, but not as well as Earth’s does.

  Curiosity’s external parts can handle temperatures as high as 50°C and as low as –128°C, and can cope with wide temperature swings from day to night every sol. In fact, the rover was designed to handle more extreme weather than it needs to, because the landing site hadn’t been chosen when the rover design was finalized. Located close to the equator and at low elevation, Gale’s weather is relatively benign. The REMS instrument has measured overnight lows in the air just above the ground averaging around –75°C (ranging from –85°C to –65°C), and daytime highs averaging around –10°C (ranging from –30°C to +5°C) (see section 8.​4.​3). Ground temperatures have been as warm as 15°C and as cold as –100°C.17

  Different parts of the rover have different thermal requirements. The warm electronics boxes inside the rover’s body and head keep their interiors warmer than –40°C and cooler than 50°C, although their contents won’t fail as long as temperatures stay between –55°C and +70°C. 18 But some instruments, particularly ChemCam’s body unit, don’t like running hot. Other instruments, like SAM, have components that generate a lot of heat. These instruments have integrated coolers to keep their electronics at safe temperatures.

  4.4.1 Rover avionics mounting panel

  All the temperature-sensitive electronics and systems are bolted to a rover avionics mounting panel (RAMP). The panel is, in turn, attached to the top deck of the rover with titanium structures designed not to conduct heat from the interior to the exterior of the rover. Martian atmosphere occupies the small amount of space inside the rover that is not filled with electronics. The mostly carbon dioxide gas inhibits the transfer of heat between the internal hardware and the rover’s sides and belly panel.

  4.4.2 Sensors and survival heaters

  A total of 221 temperature sensors monitor conditions all over the rover, though only about half are in use at any given time, since there are redundant sensors for the rover’s A- and B- side electronics. A few critical components (such as the batteries) have their own survival and warm-up heaters controlled by a total of 8 mechanical thermostats.

  Curiosity’s 17 cameras and 32 motors can survive all expected ambient temperatures outside the rover’s body, but have minimum operating temperatures between –55°C and –40°C. When they are too cold (as they always are overnight and early in the morning, and can be during the day, depending on the season), they have to be warmed before use. The rover’s main computer switches these warm-up heaters on and off as commanded.

  4.4.3 Rover heat rejection system

  The rover heat rejection system (RHRS) pumps Freon (trichlorofluoromethane, CFC-11) through tubing that loops near the MMRTG to pick up waste heat and then into the rover body to warm the electronics (Figure 4.6). The heat rejection system contains a total of 60 meters of aluminum and stainless-steel tubing. The rover integrated pump assembly (also visible in Figure 4.3) acts like the rover’s heart, pumping Freon near all the parts of the rover that need to be warmed and cooled. It contains a large accumulator, or tank, that gives the Freon room to expand when it warms. Peak temperatures in the system have never risen above 72°C; the system can handle Freon temperatures as high as 90°C.19

  Figure 4.6. Rover heat rejection system. (a) Layout of the tubing. (b) Schematic diagram of the system, color coded to match (a). (c) Pumps, valves, filters, and manifolds that make up the interior of the pump assembly. Emily Lakdawalla after Novak et al ( 2013 ).

  There are two heat exchangers on the back of the rover, one on each side of the MMRTG (Figure 4.7). The heat exchangers have tubing bonded to both sides. There is a hot plate on the inward-facing side of each heat exchanger, where the fluid picks up waste heat from the MMRTG and returns to the pump. On the outward-facing side of each heat exchanger is a cold plate, where fluid flowing through the tubing radiates heat away. Aerogel fills the honeycomb core of the heat exchangers, thermally separating the hot inner face from the cold outer face. If the interior of the rover needs to be heated, the pump sends fluid warmed by the MMRTG through the tubing connected to the rover avionics mounting plate. When the rover runs hot, the pump can send fluid from the rover avionics mounting plate to the cold plates on the outside of the heat exchangers and just underneath the rover’s top deck.

  Figure 4.7. External parts of the rover’s heat rejection system. Cruise and ground heat rejection system tubing in direct contact with the MMRTG is not used on Mars; it was for cooling the MMRTG on Earth and during cruise (see section 2.2.1 ). The base image is the Mastcam self-portrait taken on sol 1197. NASA/JPL-Caltech/MSSS/Emily Lakdawalla.

  The MMRTG is exposed to the Martian elements, including wind. During the coldest winter months, high winds could rob the rover of heat necessary to survive. The heat exchangers and body of the rover shield the MMRTG from winds blowing from the front or sides of the rover, but the back is unprotected. A fabric windbreaker (Figure 4.7) bridges the cold plates on the back of the rover, dramatically reducing the wind’s chilling effect.

  Because the heat rejection system is absolutely essential to rover health, there are two redundant pumps and two redundant mixer valves and splitter valves (Figure 4.6c). The mixer and splitter valves allow the heat rejection system to selectively heat or cool the Freon as needed. For rover safety, they work independently of any computer, operating passively in response to the temperature of the fluid flowing through them.

  The mixer valve controls the amount of flow across the rover’s hot plates. If the mixer valve falls below a temperature of –10°C, it opens all the way, sending 97% of the fluid through the hot plates. If the mixer valve measures a temperature of 10°C, it closes to its minimum setting of 55%, which runs just enough fluid in the hot plates to keep the fluid temperatures below 90°C. The splitter valve controls the flow to the cold plates and top deck. When its temperature rises above 35°C, it opens all the way to 96%; it closes to its minimum setting of 4% at 15°C.20

  4.4.4 Heater tables

  All of Curiosity’s components have minimum allowable operating temperatures. Some components would spend too much time at those temperatures without assistance, so have built-in heaters. The most heat-demanding components are the motors. Curiosity’s motors do not operate well at temperatures below –55°C because the wet lubricant inside the gearbox is highly viscous at that temperature. All the motors have heaters to permit their operation when ambient temperatures are lower than that. Heating requires both power and time, two limited resources, so during tactical planning it is imperative to predict how long and how much power it will take to prepare motors for use. Time of day, season, wind speeds, and rover orientation (potentially causing shad
owing) all have strong effects on the start temperatures of rover hardware. With so many variables involved, engineers can’t predict exactly how much heating will be necessary for a given motor on a given sol to be operated at a given time. Instead, they budget enough power to heat the motors as much as necessary for a predicted worst-case scenario.

  Prior to landing, thermal engineers prepared two “heater tables” that laid out the energy requirements for motor heating for the worst-case environments for every motor for each hour of the day for two representative days in the Martian year: landing day, which was on Ls 151 (approaching the southern hemisphere vernal equinox), and the coldest day, winter solstice, Ls 90. The landing-day table would allow conservative budgeting of energy expenditures for heating throughout the rest of spring and most of summer until fall started to bring cooler temperatures, when they would have to switch to the winter table.

  The heater tables stipulate, for each heater, for each hour of the day: Warmup period. This varies with the mass of the motor. The largest motors that drive and steer the wheels weigh 6 kilograms and can take up to 2 hours to preheat when they are coldest.

  Target temperature of the temperature sensor. The sensors and heaters are on the outside, not inside, the motors. But it’s the deep interior of the gearbox that has to be brought to –55°C. This unfortunate arrangement results from the late switch to wet-lubricated motors (see section 1.​5.​2). To drive in the morning, the outside of the motor and its temperature sensor may be heated as high as –5°C in order to generate a big enough pulse of heat to bring the interior to temperature within a reasonable amount of time.

  High and low set points for temperature cycling during the maintenance heating period. Once the motor preheats, the exterior temperature is allowed to fall to a lower set point before the heater turns on again and continues to oscillate within a small temperature range. For some motors, after preheating, the heater turns on when needed to maintain the exterior temperature of the motor between –49°C and –44°C.

  Duty cycle prediction. The amount of time during the maintenance heating period that the heater will be turned on.

  Timeout period. How long to wait for the target temperature to be reached before giving up, shutting down the heater, and aborting the day’s plan. If this happens, the rover will continue performing remote sensing activities, but will not proceed with any drives or arm work.

  Energy (in watt-hours) associated with warmup and maintenance heating of the motor. This is taken out of the power budget in the day’s plan.

  Fortunately, it isn’t always necessary to preheat motors. Even on the coldest winter days, Gale crater heats up to about –25°C, well above the motors’ minimum operating temperature. The biggest motors take the longest to heat, and are the ones that enable Curiosity to drive. During the winter, there is a 3-hour period when the rover can drive without spending time or energy preheating, from about 14:40 to 17:50 local true solar time each day. (Heater tables are a case where it is necessary to employ true rather than mean solar time; see section 3.​2.​2.) During the spring and summer this is a 6-hour period, but still in the afternoon, from about 12:30 to 18:30 each day.

  Therefore, waiting for the motors to preheat naturally requires waiting until the afternoon to drive. However, the mission would often prefer to move the rover in the morning in order to allow sufficient time for driving and post-drive imaging to complete before the afternoon orbiter relay. Rover planners compromise by usually starting drives between 11:00 and 12:30, which means motors usually need to be preheated, but for a relatively short time.

  4.4.5 Performance on Mars

  Curiosity’s thermal control systems have operated flawlessly.21 The temperature of the interior of the electronics box has varied within acceptable limits, ranging from lows near 5°C to highs of 17°C in winter and 37°C in summer. The rover’s temperature profile has been reliably the same, day after day, making it easy for rover planners to decide when to operate the instruments that need cool ambient temperatures. The battery survival heaters have never been turned on and likely never will be.

  There were several surprises after the spacecraft landed on Mars. While REMS measured ground temperatures that were in good agreement with predictions, it found air temperatures to be much higher than predicted: 25°C warmer than predicted during the day, and 10°C warmer at night. The team now uses current REMS data on atmosphere temperatures to help them predict rover temperatures (see section 8.​4.​3).

  The rover operated using the Ls 151 or spring-summer heater table until sol 434, when some of the wheel motors got cold enough that the thermal team switched to the Ls 90 (winter) heater table. The sudden switch to preheating for winter solstice temperatures dramatically reduced available power and drive time, and was, of course, overly conservative. It also added complexity, because sometimes heating had to happen in one sol’s plan for the subsequent sol’s activities. The worst impact was on drive time, because increased preheat time imposed a limitation on drive time at a point in the mission when they were attempting to extend drives to cover more distance.

  The team briefly tried a hybrid approach (using the spring-summer heater table for some systems and winter table for others), but this was operationally complex, and couldn’t last long in any case because of rapidly cooling temperatures. They switched all of the mobility system completely to the winter table on sol 456, and all remaining subsystems to the winter table on sol 463. Because the stepwise switch to winter heating requirements dramatically affected rover activities, the thermal team began the process of developing an intermediate set of tables, optimized for Ls 130, covering early spring and late fall seasons.

  4.5 TELECOMMUNICATION

  Curiosity receives commands directly from Earth, but returns more than 99% of its data through an orbital relay. Telecommunications bandwidth is one of the primary limitations on the science return from Curiosity (or any other deep-space mission). Curiosity typically returns about 500 megabits per sol. Actual volumes in any given transmission depend on many factors, especially the geometry of an orbiter’s communications pass (range, elevation, and duration of the pass).

  Curiosity’s operational schedule is dictated by communication opportunities, scheduled months in advance. The sol begins at about 10:00 a.m. local time, when Curiosity usually receives the day’s command sequence directly from Earth via an X-band transmission between a Deep Space Network dish and Curiosity’s high-gain antenna. Curiosity warms up and performs the commands – driving, arm operations, and/or remote sensing – and typically wraps up work in time for the afternoon communications passes. Overnight, Curiosity usually rouses from sleep to return more data. Whichever communications pass is the last one before the next sol’s tactical planning shift begins is called the “decisional” data pass (see section 3.​4 for more about tactical planning).

  4.5.1 The Deep Space Network

  For more than 50 years, Earth has listened to faint signals from distant spacecraft with the giant radio antennas of NASA’s Deep Space Network (DSN). The DSN consists of three ground stations positioned approximately 120° of longitude apart from each other, so that at least one station can “see” a spacecraft at all times. The three stations are Goldstone, located near Barstow, in California; Madrid, in Spain; and Canberra, Australia. Each station has multiple dishes, including one 70-meter dish and several 34-meter dishes (Figure 4.8). The 34-meter dishes can be arrayed to create a single aperture comparable to a single 70-meter dish. For Mars, which is nearby, this usually isn’t necessary. In fact, a single antenna can receive signals from multiple Mars spacecraft simultaneously. The DSN provides support for European and Indian Mars missions as well as NASA ones.

  Figure 4.8. Dishes of the Canberra Deep Space Network pointed at Mars. In this photo, taken on November 18, 2013, two 34-meter dishes (DSS-34 at center and DSS-45 at right) were listening to signals from the MAVEN orbiter as it arrived at Mars. At the same time, at left, the 70-meter DSS-43 simultaneously received d
ata from Mars Odyssey and Mars Reconnaissance Orbiter. Photo courtesy Glen Nagle, Canberra Deep Space Communication Complex.

  4.5.2 Curiosity hardware

  Figure 4.9 details external parts of Curiosity’s telecommunications hardware; refer to Figure 4.3 for locations of internal parts.22 Curiosity has three antennas. Two X-band antennas can communicate directly with Earth. A UHF antenna links Curiosity with orbiters. There is one high-gain and one low-gain X-band antenna. X-band communications happen through one of two redundant Rover Small Deep Space Transponders (RSDST). The transponders are an improved version of the design used for the Mars Exploration Rovers.

  Figure 4.9. External parts of Curiosity’s telecommunications hardware as seen in the sol 1197 Mastcam self-portrait. NASA/JPL-Caltech/MSSS/Emily Lakdawalla.

  4.5.2.1 High-gain antenna

  The high-gain antenna is hexagon-shaped, 28 centimeters in diameter, and is steerable in both azimuth and elevation. It can both receive commands from and transmit telemetry to Earth, but it has to be aimed properly. It has a 5° pointing accuracy, limited in part by the accuracy of the rover’s knowledge of its own orientation. It can provide a downlink of 160 bits per second to a 34-meter Deep Space Network radio antenna, or 800 bits per second to a 70-meter antenna. Running in the other direction, the high-gain antenna can receive uplinked commands at a rate of 1 or 2 kilobits per second. A typical command load is about 225 kilobits. Planners schedule 15 minutes for communication sessions to allow sufficient margin.