I know that many artificial satellites are tide-locked with Earth, which allows them to always face a directional antenna towards their ground station. But some vehicles, such as space telescopes, and orbiters observing other bodies are required to rotate to point at other objects in space. What are the common ways used by such satellites to maintain communication links with Earth?
What are the common ways used by such satellites to maintain communication links with Earth?
In general, the strategy is to
- Have an antenna that can be pointed toward the Earth.
- Have multiple antennae that can be pointed toward the Earth as a backup plan in case something goes wrong.
- Have a backup backup plan in case something goes very, very wrong.
The need to send data to and receive commands from the Earth can strongly influence overall vehicle design. The mechanisms by which a vehicle will achieve these goals is made very early on in the planning and design of the vehicle and of the mission. While these are the overarching design goals, there is no one single answer as to how to achieve these goals.
I'll look at three spacecraft to illustrate, The New Horizons mission on its way to Pluto, and the Wilkinson Microwave Anisotropy Probe that mapped the cosmic microwave background radiation from a pseudo-orbit about the Sun-Earth L2 point, and the Mars Reconnaissance Orbiter that is currently orbiting Mars.
The New Horizons is currently en route to Pluto. It's cruise phase operations are fairly simple. It is currently operating as a spin-stabilized vehicle in hibernation mode. The vehicle rotates about the axis of the communications system with the axis pointed more or less earthward. This is the stable axis of rotation, by design. The vehicle gets once a year checkups from the Earth. The communications system comprises a stack of three antennae, plus one other antenna elsewhere. One is a high gain but low beam angle antenna. This will be the primary antenna for sending data to the Earth. The second is a medium gain antenna with a 4 degree beam angle. The third is a low gain antenna with a hemispherical beam width. (The fourth antenna is identical to the third, but aimed in the opposite direction.)
The low gain antennae are no longer of any use. The vehicle is too far from the Earth. They were the backup to the backup plan during the very first part of the cruise phase. If something had gone very, very wrong early on, controllers on the Earth could have sent low bandwidth signals to the vehicle to try to urge it back to health. There's only one backup now, the medium gain antenna.
Sometime in the next year, the vehicle will switch from it's current spin-stabilized mode to 3-axis mode to perform its primary mission. The vehicle will use some of its precious hydrazine to de-spin itself and then use more of that precious hydrazine for active attitude control. To keep things simple, the antenna on New Horizons are not steerable. The vehicle will temporarily lose comm with the Earth during its closest approach with Pluto. The main job there is to record as much data as possible. Once the approach is over, the vehicle will reacquire comm by turning the vehicle as a whole so antenna axis once again points Earthward. It will then transmit that recorded data toward the Earth.
What if that doesn't work? The vehicle has two emergency modes, Earth acquisition and Sun acquisition. The first resort on failing to reacquire comm is to enter Earth acquisition mode. If that first backup plan doesn't work, the vehicle will switch to Sun acquisition mode and aim the comm axis toward the Sun. Since the medium gain antenna has a 4 degree beam width, the Earth should be in view at distance of 40 AU. If that backup to the backup doesn't work, it's goodbye mission.
Wilkinson Microwave Anisotropy Probe
The WMAP was also a spin-stabilized vehicle, but with a twist. It's rotation axis was not around one of the principal axes of the vehicle. It intentionally precessed by a bit. It was also slowly reoriented throughout its pseudo orbit so the science package pointed away from the Sun and Earth, shielded from radio noise from the sun and Earth by a large sun shield. The communications equipment and solar panels were on the sunward side of the sun shield. This orientation obviated the need for gimbalable communications antennae.
The WMAP communications system comprised dual redundant medium gain antennae and two low gain omnidirectional antennae. The backup plan was to switch from one medium gain antenna to the other. The backup plan to the backup plan was to use the omni antennae to find the Earth.
Mars Reconnaissance Orbiter
The MRO also has multiple levels of redundancy. It has but a single large antenna dish. That the MRO dish needs to be gimbal able means that the dish is a single point of failure. However, failures are much more likely to happen in the electronics rather than in the dish and its controls. The vehicle has two X-band transponders that use this single dish. A third Ka band transponder uses the same dish (but this is a demo project rather than a redundancy). In addition to these transponders, the vehicle also has two omni antenna as independent backups to the backup. Mars is close enough to Earth so that those omnis provide an alternate means of contacting the vehicle (at a low bit rate) even if the vehicle is lost in space / lost in time or if some failure has occurred in the main communications system.
I omitted something in the above: How does the spacecraft know where to point the antenna? Once again there is no single answer.
All of the above spacecraft, along with almost all other modern spacecraft, autonomously keep track of their orientation in space. Now all the spacecraft needs to know is how to point toward the Earth. One way for a spacecraft to know this pointing information is via command from ground controllers, typically in the form of a function of time.
Another approach is for the spacecraft to keep track of where it is in space. This can be done autonomously, via command from the ground, or a combination of autonomy and ground update. Now it's a simple matter of calculating where the Earth and Sun are as a function of time. That's the job of an ephemeris model. The ephemerides are calculated on the ground. The ephemeris model is just a bunch of Chebychev polynomial coefficients, where the polynomial is a function of time.