Mission Planning Updates, Part 4: Thermal

This post is part of a series of updates on our mission planning process, which began with Mission Planning Updates, Part 1: Trajectory. We now turn to considering the effects of thermal radiation on our spacecraft, and the analyses we have performed and will repeat to ensure mission success.

We discussed solar radiation in our previous post, but only visual light for photovoltaics powering our spacecraft. As everyone is familiar with, sunlight also imparts heat.

Thermal Radiation:

Is space hot or cold? The answer is a confusing one, because the vacuum of space is an excellent insulator. Convection and conduction are nonexistent (outside of the spacecraft itself) with almost nothing to carry heat to or from the spacecraft. The only significant form of heat transfer to and from the spacecraft is radiation.

Radiators on the International Space Station. Image: NASA

Keeping the spacecraft at a safe and stable temperature requires balancing the incident thermal radiation (mainly from the Sun) with the emitted thermal radiation from the spacecraft, as well as heat generated onboard. Larger, more complex spacecraft have dedicated radiators they can deploy for this purpose, but our CubeSats do not. In our case, balancing the radiation on the spacecraft means reorienting it to adjust the amount of surface area that is exposed to the Sun, and carefully managing the heat generated onboard.

Thermal Requirements:

Our spacecraft have two thermal goals to keep in mind. It is important that the water in the propellant tank remain liquid, because even though there is enough room in the tank for it to expand and freeze, it cannot be used as propellant until it thaws! Additionally, a frozen propellant tank would eliminate the slosh damping effect, harming our attitude control strategy.

In addition, it is vital that the spacecraft electronics, particularly the batteries, are keep cool to prevent potentially destructive overheating. This goal may seem to be add odds with the need to keep the water heated, but the two are actually complementary. By heat-sinking the spacecraft electronics to the propellant tank, we are able to keep the water warm at the same time as we remove waste heat to keep the electronics from overheating. This allows for a symbiotic relationship between subsystems, a design philosophy we have made use of in our architecture wherever possible.

However, in space, that heat needs to go somewhere eventually, and this is where the spacecraft orientation comes in. In the previous post, we discussed how the spacecraft naturally receives the most solar radiation when its spin axis is oriented towards the Sun, exposing the largest of its surface area.

Passive Thermal Analysis:

In our analysis we mainly consider two extremes:

  1. Can the spacecraft keep its propellant tank from freezing indefinitely in its minimum solar incidence orientation? Also, can this be done while maintaining a positive power balance (still being able to charge the batteries, instead of needing to spend more power heating the spacecraft than it receives from solar panels in this orientation)?
  2. Can the spacecraft keep its electronics (primarily, the batteries) from overheating in its maximum solar incidence orientation? Also, does being in the maximum solar incidence orientation retain all available mission operations, or will some be precluded by fear of overheating?

In our design process, we have considered these two questions repeatedly using finite-element analysis, iterating on the structural design of the spacecraft, the locaion of the electronics within it, and the thermal coating applied to the surface, until the answer to both questions was “yes.”

Our modeling included both steady state and transient analyses, with the latter including cases such as the spacecraft operating its radio amplifier or firing its thruster (producing lots of heat) in the maximum solar orientation, or crashing and losing its ability to heat itself temporarily during the minimum solar orientation.

The end result of all of this is that we were able to refine our thermal design until we finalized it last year, with:

  1. A margin of 10 C above freezing while maintaining net power production, and a margin of 5 C above freezing while running at minimum power, in minimum solar orientation.
  2. A margin of 15 C below dangerous temperatures while activating the spacecraft radio in the maximum solar incidence orientation. We also found that the thruster pulses do not take place often enough to significantly heat the spacecraft.

Thermal Mission Planning:

The steady-state thermal analysis above might give the impression we do not need to factor thermal concerns into our mission planning. Thanks to our careful structural and coating design, the spacecraft will passively remain at acceptable temperatures even during maximum power output in maximum solar incidence, or during minimum power output in minimum solar incidence.

However, there are still a few considerations to be made. Most notably are eclipses, when the spacecraft position is such that the Sun is obscured by the Earth or Moon. This does not happen too often or too long in our trajectory, but can be dangerous. The spacecraft receives no power and very little heating in this situation, meaning it must rely on cranking up onboard heaters to survive. But without solar power, those heaters are sustained by the batteries, giving the spacecraft a finite time until it can no longer keep the water liquid.

The spacecraft are able to survive an eclipse of any duration that might be encountered in our planned mission, but we must be careful not to let them be caught by surprise. A spacecraft that has just finished massively discharging its batteries to perform some sequence of maneuvers may not be prepared to keep itself heated for as long as might be necessary.

This is part of the motivation for the power and trajectory analysis in the previous posts: ensuring that the spacecraft enters any eclipse in a state of maximum charge to be best prepared to survive until it can see the Sun again.


Mission Planning Updates, Part 3: Power Budget

I’m giving it all she’s got, Captain!

–Montgomery “Scotty” Scott, Star Trek

This post is part of a series of updates on our mission planning process, which began with Mission Planning Updates, Part 1: Trajectory.


Solar energy is abundant in the inner Solar System. Therefore, solar panels are a ubiquitous choice for powering spacecraft. For shorter missions, such as Space Shuttle flights or launch vehicle upper stages like Centaur, energy storage such as fuel cells or batteries may be all that is needed. But for lengthy missions, spacecraft need to generate their own power to sustain mission operations.

Apart from those few that use radioisotope thermoelectric generators (mostly deep space probes such as Voyager and Pioneer), that is done with photovoltaics. Our own spacecraft use surface-mounted Solaero ZTJM triple-junction cells, together with a GomSpace P31u for power management, including a lithium-ion battery for energy storage.

Our spacecraft, like many, will not produce enough power to continuously run all of its components at once. Their surface-mounted solar panels produce at most 10 W for each spacecraft, while the electrolyzers alone consume about 6 W, for example. High performance lithium-ion batteries mean we will be able to run a power deficit for many hours, but not forever. Therefore, we will need to pick and choose carefully how to match the available power to the needs to the mission at all times.

Planning a power budget for the spacecraft cannot be totally decoupled from other mission planning elements, such as the trajectory previously discussed, and the thermal concerns that will be the subject of the next post. We have already developed detailed power budgets and analysis for the spacecraft through their entire trajectory, but the scripts we made to do this will need to be re-run on the new trajectory data after it is finalized.

In the rest of this post, I want to explain a bit about why the spacecraft trajectory and attitude are linked to the power budget, and then discuss the results of our existing power analysis and how we expect it to change this semester.


As our spacecraft traverse their planned path as described in the first post of this series, one might expect the solar power available to it to change. For example, the solar irradiance at Mars is about 44% that at Earth. Although the distance from the Sun does affect the solar irradiance, it has an inverse-square relationship with distance from our star:

Inverse square law. Image: NASA

This is because the light emitted from the Sun spreads out like an expanding sphere as it moves away, and the surface area of a sphere increases proportionally to the square of its radius. However, this is not a very significant effect in our case.

Our spacecraft will be deployed close to the Earth and remain in its orbit (or the Moon’s) for the entire mission duration, meaning they will never be more than about 1 million km closer or father from the Sun than the Earth itself is.

Because the Earth is about 150 milion km from the Sun already, that means the solar irradiance on the spacecraft will vary on the order of 1%. And our spacecraft will spend most of their time closer to the Moon’s distance or in its orbit, which is even closer to the Earth, and so solar irradiance will vary even less there.

There is one important effect the trajectory has on our power generation, however: Occasionally, our spacecraft will enter an eclipse where the Sun is blocked by the Earth or Moon. In these cases, there will not be any power available from the solar panels at all. The spacecraft will be dependent on its battery reserves. Eclipses also pose a danger to the water in the propellant tank, which is susceptible to freezing, but that is a topic for another post.


Our spacecraft use surface-mounted solar panels, which have the advantage of being mechanically simpler than deployable solar arrays such as those on the ISS, but have a limited surface area in comparison, and also no ability to orient the solar panels towards the sun independently of the spacecraft.

ISS in 2007. Note the way the solar panels can reorient themselves. Photo: NASA


Because of this, the power generated by the spacecraft is dependent on orientation. All orientations produce some power, but the amount varies around an average of about 7 W. Our spacecraft have an excess propellant budget for their attitude control cold gas thrusters, so it may be advantageous to reorient the spacecraft to face the sun when power is needed for long durations. At other times, it may not be worth the propellant cost to do so, if the spacecraft will not need to perform high power operations, or if it does but there is time to wait for the batteries to recharge in between them.

Power Budget:


This slideshow requires JavaScript.

The above images are the results of the previous power analysis. Our next run will show improved performance, but more on that later. The first script takes as an input the spacecraft solar panel layout to determine power production as a function of spin axis angle relative to the Sun.

The next uses the result of the first, plus the trajectory and orientation of the spacecraft as inputs, as well as the ephimerides of the Sun, Earth, and Moon. With all of this information, we can compute the power production of the spacecraft across the planned mission.

The final step is to add in the planned power consumption across the trajectory. This means determining how much power each component uses, when it will be needed, and at what duty cycles. By subtracting production from consumption over the entire trajectory, we can then compute the net power production of the spacecraft across the whole mission.

We can then assess the times when net power production is negative against the total energy capacity of the cells in the P31-u power unit. Even in this outdated analysis, we found that the spacecraft would never go below 40% depth of discharge (60% battery capacity) in the nominal mission.

It is possible to alter the mission planning to improve this, by for example, setting a DOD limit and halting electrolysis at this limit to allow the battery time to recharge. In our case, the final design of the spacecraft will improve the results as well.


We expect the outlook to be more favorable in our upcoming power analysis this semester, for two main reasons. First, the final spacecraft design has a greater solar panel surface area, especially on the spinning edge sides. Therefore, the spacecraft produces more power at all orientations than it did previously. The improved power production means the spacecraft will nearly always produce more power than the electrolyzers for the main thruster consume, which was previously not the case. This means the spacecraft will be able to continuously thrust in most orientations, instead of needing to take breaks every few hours to recharge. This eliminates several of the negative net power cases in the existing analysis.

Second, with our improved ground station and other changes to the communications architecture, we were able to reduce the required power draw of the spacecraft transmitter. This was previously the highest power draw item on the spacecraft (that honor now belongs to the cold gas thruster solenoid valve) and will be needed frequently during one of the most critical phases of the mission, because the spacecraft will be downlinking as much as several minutes every hour when working to verify lunar orbit. Therefore, the spacecraft consumes less power than it did previously.

Because the spacecraft now consumes less power and produces more power than it did previously, the expected depth of discharge will be decreased (improved) and the spacecraft will spend more time with its batteries at a high state of charge. This will ensure it is prepared for emergencies and unexpected net power drains, for example, if a solar panel were to fail and the spacecraft needed to reorient to begin producing power again.

Mission Planning, Part 2: Micrometeoroids

This post is part of a series of updates on our mission planning process, which began with Mission Planning Updates, Part 1: Trajectory.

Micrometeoroids and orbital debris (MMOD) are small objects in space, the possibility of collision with which poses a hazard to spacecraft. Relative velocities between objects in space can be in excess of 20 km/s. Just about anything can be dangerous at such incredible speeds; a speck 1 mg in mass has about as much energy at 30 km/s (450 J) as a 9 mm handgun bullet fired on Earth.

MMOD damage is a concern for all spacecraft, not only for their own sake, but also because a spacecraft lost to MMOD damage can become a derilect or debris, creating a new hazard to avoid for future spacecraft.

The relevant standards we need to comply with are found in NASA-STD 8719.14 and NPR 8715.6A, and we have written an Orbital Debris Assessment Report as required by them. We have also completed an End of Mission Plan (EOMP) and Planetary Protection Plan (PPP), but those are topics for another post.

The MMOD environment depends heavily on the region of space one considers–for example, human-made debris such as paint chips are a much greater concern in Low Earth Orbit, where they are distressingly common, than they are near the Moon.

Not-to-scale representation of Earth’s orbital debris. Image: NASA

Because our deployment options begin past the geostationary orbit distance, and we never get closer to the Earth than that, we mainly have to worry about micrometeoroids and not human-made debris.

Efforts have been made to quantify the micrometeoroid environment in the inner solar system, resulting in tools such as NASA’s Micrometeoroid Engineering Model (MEM), which is what we have used for our analysis. Software for this purpose can be obtained on request from NASA’s Meteoroid Environment Office.

The expected flux of micrometeoroids on a spacecraft at a given moment depends mainly on the surface area of the spacecraft (a larger one such as the ISS will naturally be struck more often), the environment where the spacecraft is orbiting, and the threshhold of size and velocity for what one considers a potentially disabling meteoroid. Fortunately for us, our spacecraft are very small targets, which improves our odds.

We have ran MMOD analysis for multiple different size thresholds and are confident that our spacecraft will survive in the cislunar MMOD environment for the duration of the mission and beyond. Example: For the entire mission, we computed odds of 0.04% chance of being struck by a micrometeoroid of 1 mg mass or larger moving 20 km/s relative to the spacecraft.

The odds are increased to as much as 0.8% across the entire mission for a 0.1 mg g or larger micrometeoroid, but the smaller objects included in this chance are much less dangerous. Any new trajectories we develop will not travel through any different regions of space, so we do not expect these results to change in future runs.


Mission Planning Updates, Part 1: Trajectory


Last semester, the Cislunar Explorers team completed our latest and most successful trajectory plan. We have done work on this before, but the planned launch date given to us by SLS has changed several times, necessitating a re-run. Because the launch date of SLS is nearly two years away and still subject to change, it is possible that we may need to regenerate our trajectory again based on a new launch date in the future. Each time we need to do so, we take the opportunity to improve the new trajectory based on lessons learned from working the previous launch date.

There is more to mission planning than just a “bus schedule” of where the spacecraft will be and when. Within the next week, we will have several follow up posts about some different examples, including orbital debris assessment, power budgeting, and thermal analysis. But first, an examination of some of the challenges facing our trajectory plans, and a description of the current trajectory.

Trajectory Challenges:

In the past, we have designed our trajectories semi-manually, using STK software with the Astrogator module and its built-in targeter for individual maneuvers, after manually planning out what we want from each maneuver and inputting initial guesses for optimization by the software. Now, we have improved on this by writing monte carlo scripts for STK using its MATLAB interface, allowing us to automatically generate arbitrary numbers of slightly perturbed trajectories. In this way, we can see how robust our trajectory is with regards to e.g. deployment by the Interim Cryogenic Propulsion Stage (ICPS) of the SLS in random directions.

Another challenge in designing our trajectory is that our electrolysis propulsion thruster does not behave exactly like either a typical chemical or chemical thruster. Our spacecraft produce many small thruster pulses over time, instead the constant steady thrust of an electric thruster or the large, nearly impulsive burns of a chemical thruster. In practice, the most realistic way to design trajectories using this thruster has been to average out its thrust over a long period of time and treat it like an electric thruster.

Finite burns are more challenging to calculate than impulsive burns, and more difficult to target in the STK software. The best way we have found to do this is to start by implementing an impulsive burn that does what we need, then having the software “seed” a finite burn based on that. We also need to avoid any maneuvers that require an actual impulsive burn, such as orbit injections with a sudden and large DeltaV, because our thruster is not capable of doing so. To avoid this and still achieve lunar orbit, we need to set up a ballistic capture using only lunar flybys and low-thrust course correction maneuvers over the days or weeks in between lunar encounters.

Current Trajectory:

The current trajectory begins with deployment from the SLS ICPS at the first available “bus stop,” mere hours after launch. The reason for this is that our spacecraft need time to perform their separation from each other and reach their desired stable 6 rad/s spin before the mission can begin. With this deployment, we ride out the inner Van Allen belt while powered off inside the ICPS, but the spacecraft must survive the outer Van Allen belt by itself. Fortunately, the Raspberry Pi flight computer is quite tough when it comes to radiation, and we expect it to survive easily.

When our spacecraft our deployed from the ICPS, they are on course for an unavoidable lunar flyby, between 5-6 days after deployment depending on when exactly deployment occurs. These few days include the most time-sensitive maneuvers of the mission, because we need to tweak this lunar flyby to prevent the resulting gravity slingshot from ejecting our spacecraft from the Earth-Moon system. This is obviously key to mission success–for spacecraft called the “Cislunar Explorers,” being hurled out of cislunar space entirely would be embarassing!

Once that is avoided, the spacecraft will be on a trajectory that takes them to about one million kilometers away from Earth, several times farther than the Moon is, but staying within Earth’s gravitational influence the whole time. Course corrections over about a month will guide the spacecraft to a second lunar encounter, one which we will have more control over.

We are working on being able to capture into lunar orbit directly from that second encounter, but currently, we require using it as a second flyby and gravity assist, to set up a third and final encounter. At this encounter, and several months after launch, our spacecraft will arrive in a highly elliptical orbit around the Moon.

Over the following months, the spacecraft will spiral its orbit down closer to the Moon for the dual purpose of claiming the CubeQuest prize by getting close enough to it, and preparing for the eventual disposal of the spacecraft. For the end of the mission (up to one year after the launch), the spacecraft will continue to lower its periapsis until it intersects with the far side of the Moon, ensuring an impact far from any historic landing sites.

Upcoming Work:

Our current trajectory is sufficient, and better than past efforts, but not yet optimal. Every gram of propellant we can save from needing to achieve lunar orbit is another gram that can be used for stationkeeping to increase the longevity of the spacecraft, as well as margin in case things don’t go as expected. Therefore, we are working on optimizing the trajectory this semester.

We have already tweaked our approach for the first flyby to setup a more favorable approach for the second lunar encounter. This semester, we will work on attempting to capture into lunar orbit on that second encounter, eliminating the need for a third, and the propellant cost that comes with it.

We will also re-run our existing mission analyses based on that more optimal trajectory. We have performed a number of mission-related simulations, and the code used for most can take arbitrary trajectory inputs , so they can be re-run at will. Expect posts coming in the next few days specific to:

  • Micrometeoroids and orbital debris (MMOD) damage assessment, including the expected odds of a mission failure from MMOD impact (<<1%).
  • Power generation, consumption, and budgeting for the mission, including the maximum battery depth of discharge expected during eclipses.
  • Thermal analysis, coating, and heating strategies to keep the spacecraft electronics cool and the water liquid.


CubeQuest GT-4 Presentation

As part of Ground Tournament 4, we and the other teams gave presentations about our projects. Although the presentations were recorded, we were not notified of where (or even that) they had been posted. However, we recently came across the presentations video when looking up CubeQuest information.

The video includes our GT4 presentation, as well as the presentations of the other GT-4 competitors, along with remarks by CubeQuest Challenge Administrator Jim Cockrell.

Note that the screen capture recording of the slides seems to be slightly out of sync with the video recording of the speakers.

Fall Semester Accomplishments: Optical Navigation

This fall semester saw improvements to multiple subsystems as we prepare to build the flight units in the spring. Our optical navigation system is one such area. We have published multiple papers about our research into this navigation method in the past. The most significant improvements this semester were in the area of image processing and recognition. In the past, our system has been capable of recognizing and distinguishing between the Sun, Earth, and Moon, but only when these bodies were visible as full circles.

As you can see in the image slideshow below, that is no longer the case. Our algorithm is now capable of recognizing partial and even crescent bodies, and can even detect and distinguish the Earth and Moon from each other when they overlap. Coupled with the already completed Kalman filters for estimation of attitude and position from this data, the optical navigation algorithm is now complete.

This slideshow requires JavaScript.

Several tasks remain for the spring semester. First, we want to make improvements to the hardware interface with the Raspberry Pi cameras. We already havey implemented a camera multiplexer for the Raspberry Pi, and captured images of the Sun and Moon with these cameras in the field, but the rate of image capture and camera calibration can be improved.

Second, we need to begin testing the combined hardware and software. The Kalman filters have been tested with representative data, and the image processing has been tested with images taken in the field, but the Kalman filters need to be tested using data computed by processing actual images of the Sun, Earth, and Moon. The main obstacle to doing this is finding a stream of images of the Sun, Earth, and Moon from the same spacecraft along a cislunar trajectory. Individual images of any of these bodies are readily available, but we need:

  • Images of the Sun, Earth, and Moon from roughly the same location.
  • Knowledge of the angular separation between the camera facing when each image was taken.
  • Many times at different locations in cislunar space.

One way to obtain such images is to simulate them. Mission planning software such as STK or GMAT, or planetarium software such as Stellarium, is capable of doing this. Last year, we published a short video showing the spacecraft point of view for part of a simulated Cislunar Explorers trajectory using STK. Simulated images of the Sun, Earth, and Moon could be created in a similar way. This could provide us with arbitrarily many images of the three bodies, and can be easily repeated for different trajectories.

We will do this next semester, but it only tests the software, because the images are simulated and not captured with the spacecraft cameras.

We can also repeat field tests of the Sun and Moon to test our improved image capture rate and camera calibration, but there is an obvious obstacle to capturing images of the Earth from its surface, so it is not possible for us to collect fully representative data from here on Earth. So, this mainly tests the hardware.

In order to test the hardware and software together, we need to create a representative environment for the spacecraft to spin in and take images of a fake Sun, Earth, and Moon. Fortunately,  we already have a spinning air bearing test rig for our slosh damping measurements. Our existing CubeSat EDU structure can rest on it and spin exactly as the spacecraft will in orbit. We will create a sort of darkroom/planetarium around the model, for the cameras to capture images of and feed to the navigation algorithm. Similar techniques have been used by other researchers to test star trackers in the past, with projected starfields on the walls of a darkroom.

We look forward to confronting this and other challenges in the spring semester, as we move towards final testing and integration of flight hardware.

GT4: First place, selected to fly on EM-1!

This morning, the NASA Centennial Challenges program announced the results of the fourth and final Ground Tournament in the CubeQuest.

The Cislunar Explorers team is proud to say that we won first place, earning a $20,000 prize! More importantly, we are officially one of three CubeQuest selections to fly as a secondary payload on EM-1 in 2019. This will allow us to compete in the Lunar Derby and become among the first CubeSats to depart Earth orbit. We are grateful for the opportunity to demonstrate our new technologies in orbit. Success will prove the viability of water electrolysis propulsion and interplanetary optical navigation, both of which will contribute to expanding the capabilities of the CubeSat platform.

Congratulations also to our neighbors on EM-1, CU-E3 and Team Miles, both of whom will demonstrate their own incredible technologies in the Deep Space Derby. We are looking forward to completing our spacecraft and flying into space alongside them, the other secondary payloads, and the Orion space capsule.

Above all, we are excited to continue with our multi-year journey from whiteboard scribbles, to a completed design, to a manifested payload… and hopefully, to the moon!