a16z Podcast | New Year, New Horizons -- Pluto!

Mission Context and Personnel

• The transcript details the design and execution of the Radio Experiment (REX) for the New Horizons probe, hosted by Frank Chen (a16z) and Ivan Linscott (Stanford senior scientist).
• The REX experiment was designed to perform radio occultation, radiometry, and bistatic radar measurements of Pluto and Charon, utilizing a novel signal acquisition architecture where signals are transmitted from Earth and received on the spacecraft.
• The team collaborated with Southwest Research Institute (led by Alan Stern) and the Johns Hopkins Applied Physics Lab (APL).

Technical Challenges: Power and Propulsion

• Pluto's distance (approx. 40 AU) renders solar power insufficient; the mission utilized a Radioisotope Thermoelectric Generator (RTG) converting plutonium-238 decay heat into electricity.
• The initial power budget assumed ~300 watts, but supply constraints forced a reduction to a maximum of ~250 watts at launch, degrading to ~245 watts by arrival.
• Power constraints required strict scheduling; only 4 of 5 payload instruments could operate simultaneously due to limited thermal and electrical capacity.
• Plutonium supply faced a critical bottleneck when the U.S. Department of Energy's processing facility shut down indefinitely due to a radioactive leak discovered midway through production.

Critical Problem Solving and Acquisition

• Glenn Fountain, the program manager, negotiated a deal with the U.S. State Department to purchase half the required plutonium from Russian stockpiles (surplus from the Soviet era) to meet the narrow launch window.
• The acquisition was expedited to under six months, a drastic reduction from the typical three-to-four-year licensing cycle for such missions.
• Remaining fuel shortages were addressed via a trade of older, depleted plutonium pellets for fresh ones, albeit with reduced initial output.

Signal Processing and FPGA Implementation

• The mission originally planned to use two radiation-hardened Field-Programmable Gate Arrays (FPGAs) with 25% unused gate margin; budget cuts forced a migration to a single FPGA with near-zero margin.
• A 10,000:1 bandwidth reduction (filtering a 1 kHz signal within a 4.5 MHz bandwidth) was required using Finite Impulse Response (FIR) filters, a task deemed unimplementable on existing technology.
• Student Kamakshi Krishnan adapted a 1950s radar filtering technique by Glenn Houghtonmeier to create a computationally efficient solution.
• A "SWAT team" led by Mark Johnson (Southwest Research Institute) intervened after a year of failed prototypes, enforcing strict coding discipline to eliminate "clever" algorithmic optimizations that caused routing failures.
• The final design fit on the single FPGA with only five gates remaining (less than 0.1% unused), requiring precise "herding" of signal nets during routing.

Communication Hardware and Ultra-Stable Oscillators (USO)

• The system relied on an Ultra-Stable Oscillator (USO) based on quartz, maintaining frequency stability to one part in 10^13, derived from Cold War-era submarine surveillance technology.
• Commercial suppliers for USOs had gone bankrupt; APL transitioned internal high-frequency stability capabilities to a startup, which delivered five units with two meeting specifications just six months before launch.

Operational Incident During Encounter

• Nine hours before the critical encounter sequence, the spacecraft entered "safe mode" due to a computer overload caused by a race condition.
• The fault occurred because the onboard JPEG compression algorithm was delayed by unexpected star patterns in the image, causing the system to miss a command deadline to update flash memory.
• This scenario had not been triggered during rehearsals because the background sky was dark (few stars) during testing, whereas Pluto's background was crowded.
• The operations team successfully rebooted the system and restored the timeline within minutes, preserving the scientific data collection sequence.

The Stanford Dish: History and Revival

• The Stanford Dish, originally built in the 1960s with U.S. Navy funding for over-the-horizon communications, utilized surplus anti-aircraft hydraulic motors from battleships for its altitude-azimuth drive.
• The dish was decommissioned in the 1980s due to the risk of high-pressure hydraulic failure and a lack of qualified technicians to maintain the system.
• Linscott and SRI engineer Mike Cousins revived the facility by replacing the hydraulic system with high-torque electric stepper motors derived from the disk drive industry.
• The revived dish is now used for SETI research, solar wind characterization, and tracking "dead" or malfunctioning spacecraft by detecting leaked radio signals (as low as 1 microwatt) from orbit.

Scientific Outcomes and Observations

• The REX experiment successfully captured Pluto's temperature and pressure profiles via radio occultation.
• Radiometric measurements revealed the surprising discovery that Pluto's night side is significantly warmer than its day side.
• A successful bistatic radar signal was bounced off Pluto's surface, a first for such a distant radar experiment.
• The signal-to-noise ratio at Pluto was approximately 1,000,000:1 (amplitude ratio of 1,000:1), despite a total power of ~20,000 watts transmitted from Earth and a signal strength at Pluto reduced by distance and solar wind interference.