Dave Riepe and Jeanine Fritz have turned the Boulder Outdoor Cinema into a local institution.  Every summer on Friday and Saturday nights, they show great current and classic movies on the back of a downtown building.  You bring your own chair (or couch or blanket), and they supply the movie, popcorn, opening acts, the works.  A number of years ago, Dave asked me if it were possible to show trailers of the coming attractions before the feature, just like a “real” movie theater.  It sounded like a fun project, and I was happy to give it a shot.

Like many problems, this one sounds simple at first, but becomes increasingly complicated the deeper you get into it.  Dave wanted to show trailers for the next three movies before the main feature.  If you think about this for a minute, you realize that you’ll be repeating each trailers a number of times – the first week you’d show trailers 1, 2 and 3.  Next week you’d show 2, 3 and 4, the next week 3, 4 and 5, etc.  You could just generate individual sequences with all the trailers you need, but that’s (A) a lot of editing work, and (B) all that repeated data takes up a lot of space, requiring multiple DVDs to get through the season.  Another problem is that some movies are shown in widescreen, and some are shown in fullscreen.  The projector is framed differently for each type of movie, and reframing the projector between the coming attractions and the feature wouldn’t be professional.  Add a number of other technical, programmatic and legal issues, and you can see how this became a very interesting project.

After a few years of refinement (and thanks to some very clever software), all of these problems were neatly solved.  The DVD specification allows for a limited amount of primitive programming, so to solve the multiple-instance problem, mini-playlists were developed.  These allow one instance of each trailer to be placed on the DVD, but each trailer can be accessed as many times as necessary.  With further work, this also allowed the coming attractions to be shown in widescreen or fullscreen as the main feature requires.

The current iteration of the Coming Attractions DVD is highly automated, to make the projectionist’s life as easy as possible.  It allows the operator to set up the night’s show with one click, automatically pauses on black, starts the show with another click, automatically plays the 3 trailers, plays a short informational animation about the outdoor cinema, pauses on black again while the operator switches to the main show, and later on plays a longer animation for the intermission.  All from one DVD that is used all summer long.

A few words about tools:  for general video capture and editing, I use the Video Toaster from Newtek.  It’s billed as the fastest editing system in the world (which it usually lives up to), and it’s iphone-easy to use.  My DVD authoring software of choice is DVD-lab PRO, from a small company called Mediachance.  Unlike most drag-and-drop DVD software, which typically limit you to the few templates they’ve provided, DVD-lab exposes every level of the DVD specification to the user, while still being intuitive to use (which is quite a trick).  It’s a bit more expensive than the cheapest ones, but for a few hundred dollars its feature set is competitive with high-end packages costing ten thousand or more.  It’s that good, and highly recommended for serious DVD authoring.

Another interesting aspect of this project was obtaining the trailers themselves.  Most DVDs will throw the theatrical trailer on as a bonus feature.  But some don’t, which often happens when they’re not sold as a “special edition” disc.  What’s worse is that a DVD will sometimes include trailers for other movies, but not bother including the trailer for the film on that disc, which often leads to a (usually) fruitless hunt through other discs trying to find the trailer for this one.  And before the days of YouTube, Apple and other trailer sites on the internet, there weren’t a lot of options if you couldn’t put your hands on a specific trailer.  Since I have a strong dose of “the show must go on”, there were a number of times in the early days when given no other option I sat down and handmade a trailer for the movie in front of me.  It was a lot of work, but also a lot of fun, and I’m especially proud of those editing jobs.

See you at the movies!

• Boulder Outdoor Cinema
• Newtek (video editing systems)
• Mediachance (DVD authoring software)

SPARTAN was an R&D project from Sierra-Nevada Corporation for a tactical military UAV system.  I assisted with several aspects of this project, including design of the onboard power system, flight termination, and problem solving.

• Power system

A versatile power system was developed all the way from requirements to I&T, which could run the aircraft off either the ultra-fast rotational output of the ducted-fan engine, or in the event of engine failure, on backup batteries.  The system provided regulation and on-off control for payload and critical flight systems.  A bank of status LEDs indicated which power source was in use, and whether all voltages were within acceptable limits.   This system was even adaptable to a customer request to use the generator as a motor to start the ducted-fan engine.

• Flight termination

Per requirement, flight termination features were built into the power system.  Several channels of servo control signals generated by the power system microcontroller were used to cut off fuel flow to the engine, and release a recovery parachute.  Termination could be initiated by the main avionics system (via either autonomous or direct command), or by a completely separate radio system using a coded signal (the iconic “big red button”).  Logic was built into the system to reduce the possibility of inadvertent activation, while ensuring that the intended signals were promptly applied.

• Problem solving
  

In addition to designing an effective RF and DC power-grounding architecture for the aircraft, several black boxes were built to solve issues that were preventing effective flight testing.  These boxes allowed standard radio-control systems to interface with various parts of the aircraft and test fixtures.  At the core of these boxes were microcontrollers programmed to input, process, and regenerate servo actuator signals in real-time.  These signal translators solved several difficult problems in an inexpensive and elegant way.

As these were flight-critical systems, considerable care was given to circuit and code correctness, reliability and speed.  Particularly gratifying was the fact that the onboard power system, based on a four-layer PCB and fine-pitch surface-mounted parts, worked perfectly the first time it was assembled.

• Sierra-Nevada Corporation website

Individual final project, Interplanetary Mission Design (ASEN 5519)

As a final project for the Interplanetary Mission Design course taught at the University of Colorado, I designed a mission to Jupiter.  This entailed choosing an appropriate gravity-assist course by evaluating planetary alignments over the mission time window for lowest (energy) cost and highest (velocity) return.  Once the transfer course was selected, a useful science orbit around Jupiter was chosen and targeted.  Finally, for extra credit, the transfer course was evaluated for possible asteroid encounters along the way.  Using a JPL database, the spacecraft was found to pass near four asteroids during its transfer to Jupiter, though additional maneuvering would be necessary to create close flyby encounters.

This project was particularly rewarding.  Before entering the course I had been fascinated by astrodynamics but nervous about the mathematics involved.  However, I was extremely gratified by the talent of the instructor (George Born) and his veritable army of TAs.  This project also got me deeper into Matlab and Satellite Tool Kit (STK) than I had ever been before.  STK turned out to be very useful in not only verifying Matlab results, but also in producing visualization which made this work much more understandable.

• Final report (800KB, PDF)
• STK animation (3 minutes, 9.7MB, MPEG4)

HOMER (the High-Altitude Ozone-Measuring Educational Rocket) was the third sounding rocket flown by COSGC.  It successfully launched out of NASA Wallops Flight Facility in Virginia in August 1996.  It carried three photometers and one spectrometer to perform remote measurements of ozone and related gases.  It was designed and built on a 2-year schedule by a dozen undergraduate and graduate engineering students.  The skin, nosecone, and specialized mating sections were donated by NASA and other organizations. Almost everything else was designed and built from scratch by students, with a budget of $20,000.

I started out as a Command and Data Handling (CDH) team member, but soon became CDH lead when the former lead (the talented Wes Brandley) graduated.  HOMER’s onboard computer consisted of an off-the-shelf 68HC16 industrial control board coupled with house-designed PCBs for I/O.  The system was tasked with real-time data collection and transmission (interfacing to a high-speed radio link; there was no onboard data storage), and general spacecraft housekeeping including turning subsystems on and off at specific times during the flight.  A lot of thought went into the build quality and failsafe operations, such as how to recover from an unexpected reset during the flight with maximum safety to the payload and minimum science data loss.  The system performed flawlessly during the flight with no resets even during the 20G launch accelerations.

I also modified a commercial off-the-shelf video camcorder for this flight after the responsible student graduated.  The camcorder operated successfully, and even with the limited field of view through the tiny porthole, the resulting footage became a (locally) famous example of what students could create.

HOMER was my first major space mission, and it completely changed the direction of my life.  I had never before had the experience of working on a small team to build something extraordinary (the rocket reached an altitude of 100km, which is “officially” in space).  The team was also extraordinary, and has gone on to accomplish many great things.

Additional information

• Homer fact sheet (50KB, PDF)
• Construction montage video (2 minutes, 10MB, MPEG4)
• Physical layout animation (1 minute, 5MB, MPEG4)
• “HOMER: A Rocket Odyssey”
  documentary made for HOMER’s 10 year reunion
  (35 minutes, 150MB, MPEG4)

DATA-CHASER was a student-designed, student-built, and student-operated space experiment which flew on space shuttle Discovery mission STS-85 in August of 1997.  It was the third space shuttle payload built and operated by the Colorado Space Grant Consortium, a group of colleges and universities in Colorado that support student-run space experimentation.

DATA-CHASER consisted of two separate but linked experiments. The first, DATA (Distributed Automation Technology Advancement), tested remote control and autonomy technologies.  During the mission, the hardware onboard the space shuttle was monitored and controlled in real-time from the University of Colorado via the Internet.  At that time, this was a significant advancement over previous shuttle experiments which had to be controlled from NASA centers.  Onboard autonomy software, developed in partnership with JPL, was also tested.  This allowed the payload to make its own decisions when it wasn’t in contact with the ground (although it was monitored in the background for testing purposes).  DATA helped pave the way to simpler, less expensive, and more effective spacecraft operations.

DATA controlled the second experiment, CHASER (Colorado Hitchhiker And Student Experiment of solar Radiation).  Chaser consisted of three instruments: LASIT, SXEE (pronounced “sexy”), and FARUS, which measured the full-disk solar ultraviolet and soft x-ray radiance of the sun.  These wavelengths drive ozone creation and depletion in the upper atmosphere, which in turn affects the amount of ultraviolet radiation reaching the Earth’s surface.  The data from this experiment helped benefit atmospheric, climate and environmental research, as well as giving DATA a set of science instruments to exercise.

My role on this mission consisted of coming in late in the project to troubleshoot outstanding onboard firmware issues. This involved ferreting out two power-up reset bugs (whose symptoms were masking each other) and coming up with workarounds. I followed the payload from Colorado to Goddard Space Flight Center in MD, which involved extensive work on the payload bay bridge in a cleanroom environment, and was also lucky enough to tag along on the integration trip to Kennedy Space Center FL, where we got to see space shuttle Discovery closer than I could have imagined.

Additional information:

• DATA-CHASER web pages on archive.org’s wayback machine
• Promotional video (coming soon)

Citizen Explorer was Colorado Space Grant Consortium’s do-it-yourself educational science satellite.  It carried two ozone-measuring instruments for low-resolution global measurements, and was designed to transmit that information directly to participating K-12 schools worldwide. The satellite was first conceived in 1996, received a launch vehicle slot in 1997, and came very close to launching in November 2000.

I served as both systems engineer and Command and Data Handling (CDH) lead on this mission. The systems engineering position was particularly challenging;  the extremely tight mass, volume, power, and especially budget restrictions, along with issues unique to a project run by undergraduate engineering students, required innovative technological and programmatic solutions.

Citizen Explorer’s Flight Computing system was designed for extremely low cost, ruggedness, and rapid development in an educational environment.  It consisted of an off-the-shelf 80386SX PC-compatible industrial control computer with minor but important modifications for spaceflight, networked via RS-485 to four 8051-based microcontrollers for low-level subsystem I/O.   The subsystem controllers were programmed in C and Assembly.  The main computer ran VxWorks, Spacecraft Command Language (SCL), and autonomy software developed in partnership with JPL.  The architecture required careful attention to reliability and recovery concerns, and required the identification and execution of many design tradeoffs.

With an original schedule of 14 months from paper to launch, CX-1 would have been an ambitious project for any team. Due to a series of launch vehicle slips, this schedule eventually stretched out to 3 years, but this was still not enough time to get everything working, and ultimately the Delta-II launched without CX-1 aboard. Several subsequent generations of students worked on CX-1 and eventually created a functioning spacecraft, but a second launch was not forthcoming.  CX-1 is now stored in COSGC’s cleanroom in the hope that it could still be manifested on a future flight.

Additional information:

• Mechanical drawings (PDF, 240KB)
• Wiring harness (PDF, 1MB)
• Architecture lecture series: Critical Decoder (PDF, 181KB)
• Qualification flowchart (PDF, 61KB)
• Flight computer fact sheet (PDF, 36KB)
• Microcontroller network (“Micronet”) fact sheet (PDF, 66KB)
• “Make your own CX-1″ (PDF, 38KB)

Experience is one of the main factors in creating a successful space (or any other) project. However, this is one thing that students, by definition, lack.  An effective way around this issue is to invite corporate and academic experts into a project to advise and assist the student engineers.

Mentorship has grown to become a core value at the Colorado Space Grant Consortium.  But it took some time to get that way.  One issue is that a mentorship relationship has to be carefully managed to get the most out of it.  Another is that some of the best mentors at COSGC are former students; the longer a student organization like COSGC exists, the more alumni resources it will produce.

Bruce Davis (DANDE structural lead / I&T manager) and I noticed that other organizations weren’t taking advantage of this resource as well as COSGC was, so we wrote a paper for the 22nd AIAA/USU Conference on Small Satellites.  The paper illustrates the need in the industry for student-mentor relationships, describes some of CSGC’s best practices for creating and maintaining mentorship ties, and provides some success stories.

“The Creation and Impact of Corporate Mentorship on Student-Led Satellite Projects”

• Paper (PDF, 1.2MB)
• Presentation (PDF, 630KB)
• AIAA/USU Small Satellite Conference web site

NEMO science orbit animation 2 minutes, STK 6 Click image to play NEMO Functional Block Diagram (FBD) Visio Click image to enlarge Multilobe high-gain rigid antenna concept

NEMO, the Neptune Exploration Mission of Opportunity, was a team project for ASEN 5148 Spacecraft Design.  This project was carried out in an industry-standard fashion, including answering an Announcement of Opportunity, formulating requirements, performing trade studies, and generating a final phase-A report and presentation.  Ten students were on this team; I served as C&DH lead, with responsibilities including investigating hardware and software options, and performing radiation analysis over the mission lifetime.  I also produced Functional Block Diagrams (FBDs) of the spacecraft systems, and performed STK modeling to visualize and refine the mission’s science imaging plans and concept of operations.  This project was very interesting due to the extreme distance and time scales involved.  It’s nontrivial to design a spacecraft that has to last 20 years before it even reaches its destination.

The Satellite Tool Kit (STK) visualization here was challenging because STK does not include Neptune’s moons by default.  I retrieved the orbital elements from JPL’s database, and entered them manually.  The blue box in the main window (and the blue cone in the small window) is the main science instrument’s field of view.  The red oval in the small window is NEMO’s science orbit; every dot is one day apart.  The primary science goal of the AO was to perform long-term observations of Neptune’s auroral ring, so the hundred-day orbit was not a hindrance.  Neptune’s largest moon Triton is a fascinating object because it is in a retrograde orbit – it orbits in the opposite direction to nearly every other body in the solar system.

Three teams each produced a different design to meet the same AO.  Our team was commended on sticking closely to the requirements and producing a straightforward design, which was both (relatively) inexpensive, and (relatively) low-risk.

Selected project documents:

• Final report (PDF, 2.2MB)
• Spacecraft Functional Block Diagram (PDF, 350KB)
• CDH Functional Block Diagram (PNG, 350KB)
• Spacecraft Electrical Interconnect Diagram (PNG, 650KB)
• STK animation (h264, 8MB)

• C-SMARTS Opening Animation (2004)
  60 seconds
  Lightwave, Video Toaster
  Click image to play

C-SMARTS stands for Colorado Students and Mentors Applying Research and Technology in Space. We took a college course called “Gateway to Space”, designed and taught by Chris Koehler at the University of Colorado, and put it onto 42 DVDs for distribution to smaller and less-advantaged schools throughout Colorado. Chris and I shot and edited the course over a very busy summer in 2004. The course includes lectures on the history of space exploration, the basics of rocket propulsion and orbital mechanics, advice for working on projects and teams, and a number of guest scientists and engineers lecturing on a variety of topics.

Click to continue reading…

DANDE, the Drag and Atmospheric Neutral Density Explorer, was CoSGC’s entry in the Air Force Research Laboratory’s University Nanosat 5 (UN5) competition. DANDE went from a proposal in late 2006, to flight hardware and a fully-realized mission concept in 2009, and won first place at the competition’s conclusion, earning it a government-sponsored flight sometime in the next few years.

DANDE’s mission is to perform in-situ studies of the neutral thermosphere, a region of the tenuous upper atmosphere which is too high for aircraft to study, and too low for most spacecraft to enter. This region is highly variable, and coupled to both solar activity and lower weather patterns in ways which are not clearly understood. The winds, weather, and variability of this region have serious effects on spacecraft maneuvering and lifetimes, and is an important area of study.

DANDE will study this area using two unique instruments. The first is an accelerometer suite (ACC) which will be used to measure spacecraft acceleration and deceleration due to local variations in density and in-track winds. The accelerometer suite uses six radially-mounted commercial-grade accelerometers, and a unique data-processing scheme which takes advantage of the spacecraft’s spin, to measure nano-g accelerations using relatively inexpensive micro-g components.

The other on-board instrument is the Neutral Mass Spectrometer (NMS), a miniature instrument which can measure the atomic composition of the local neutral atmosphere. This instrument includes a unique “imaging” capability which can measure this composition across a 16-pixel fan, allowing the determination of not only the composition of the local atmosphere but also the presence of any cross-track winds. These two instruments will allow DANDE to produce an accurate profile of the composition and weather effects of the neutral thermosphere it will pass through on its brief 100-day mission, hopefully leading to improved atmospheric drag models in the future.

I served as lead systems engineer on the DANDE project.  This unique mission required a unique architecture.  To produce an accurate drag profile, the spacecraft needed to be as spherical as possible. This requirement created numerous design challenges, including conforming photovoltaic panels and flush-mounted antennas to the spherical structure.  Solutions were found to all of these issues, and on many occasions weaknesses were turned into strengths. The judges at the UN5 competition commended the DANDE team on their requirements management, well-integrated design, professional build quality, and strong science mission relevance, especially for a student project.

Additional information:

• DANDE web site (http://dande.colorado.edu)
• DANDE photo gallery (local)