ARTEMIS II

Current velocity (LEO): 7.8 km/s Required for trans-lunar injection: 10.8 km/s ───────────────────────────────────────────────── Delta-v needed: ~3.1 km/s Comparison ladder: ├── A bullet from a rifle: 0.9 km/s (you need 3.4 bullets) ├── Speed of sound at sea level: 0.34 km/s (you need Mach 9) ├── ISS orbital speed: 7.7 km/s (you need 40% more on top) └── Apollo TLI delta-v: 3.05 km/s (Artemis is nearly identical)The Moon is not far in the sense of "more distance." It is far in the sense of "more velocity." Distance in space is measured in km/s, not km. You don't drive to the Moon -- you accelerate to it.

Delta-v = v_e x ln(m_initial / m_final) ICPS + Orion stack: ├── m_initial (ICPS full + Orion): ~30,700 kg └── m_final (ICPS empty + Orion): ~15,200 kg Mass ratio: 30,700 / 15,200 = 2.02 Delta-v = 4,532 x ln(2.02) = 4,532 x 0.703 = 3,186 m/s = 3.19 km/s Required: ~3.1 km/s. It fits. Barely. The mass ratio is 2.02 -- meaning the ICPS burns away slightly more than half its loaded weight in propellant. If Orion were 500 kg heavier, the mass ratio drops to 1.99, and delta-v drops to 3,110 m/s. Margin: 80 m/s. That's it.This is the tyranny of the rocket equation from Rocket. Every kilogram of crew supplies, every experiment rack, every bolt -- it all comes out of that mass ratio. The Moon doesn't negotiate.

Direct Injection Parking Orbit ────────────────────────────────────────────────────────────── Abort window: None after T+8 min Up to ~2 orbits (3 hrs) Launch window: Seconds-wide Minutes-wide Position error: Accumulates from T+0 Measured, corrected in LEO Extra hardware: None ICPS must restart in orbit Risk if engine fails in TLI: Stuck in random orbit Same -- but you confirmed it works first Every crewed lunar mission in history used a parking orbit. Apollo did it. Artemis does it. Nobody bets lives on direct injection.The parking orbit adds a stage, adds complexity, adds mass. But it converts an irreversible gamble into a verifiable sequence. The same engineering philosophy behind fly-by-wire: check before you commit.

Moon's orbital velocity: 1.022 km/s Moon's orbital period: 27.3 days Moon's orbit circumference: 2 x pi x 384,400 = 2,415,000 km In 4 days of transit: Distance moved = 1.022 km/s x 4 x 86,400 s = 353,200 km The Moon moves 353,200 km during your 4-day coast. Comparison ladder: ├── Earth-Moon distance: 384,400 km ├── Moon movement in transit: 353,200 km (92% of the total distance!) ├── Earth's diameter: 12,742 km (Moon moves 28 Earths sideways) └── If you aimed at where the Moon IS at TLI, you'd miss by 353,200 km. That's like shooting a bullet at New York from London and aiming at where New York was yesterday.Orbital mechanics is not ballistics. You don't point and shoot. You solve a differential equation where the target, the shooter, and the gravitational field are all moving simultaneously.

Moon's orbit . . . . . . . . . . MOON . . .-'''-. . . / ● \ . . \ / . . `-...-' . TLI . .''' '''. . burn ●---->. / ... \ . at . | ●crew . | . Earth . | passes . | . \ behind . / . \ Moon ./ . `-..-' . . . . . . . The trajectory forms a figure-8: 1. Depart Earth at 10.9 km/s 2. Coast for 4 days, slowing as you climb out of Earth's gravity well 3. Moon's gravity captures you into a hyperbolic pass 4. Swing behind the Moon at ~130 km altitude 5. Moon's gravity redirects you back toward Earth 6. Coast 4 days home, accelerating as you fall back into Earth's well 7. Re-enter at ~11 km/s Total engine burns required for return: ZERO. If every engine on Orion fails after TLI, you still come home.Apollo 13 used exactly this trajectory to survive. After the oxygen tank explosion, the crew had almost no propulsion. The free-return trajectory -- already built into the flight plan -- brought them home. The math saved their lives.

Range measurement: ├── Send radio pulse from Goldstone ├── Pulse hits Orion, Orion sends it back ├── Round-trip time: t ├── Distance = c x t / 2 └── At 200,000 km: t = 2 x 200,000 / 300,000 = 1.33 seconds Accuracy: ~1 meter (from timing precision of ~3 nanoseconds) Doppler measurement: ├── Transmitted frequency: f_0 = 8.4 GHz (X-band) ├── Received frequency shifts by delta-f = f_0 x v/c ├── If spacecraft moves at 1 km/s away: delta-f = 8.4e9 x 1000/3e8 └── delta-f = 28,000 Hz -- easily measurable Velocity accuracy: ~0.1 mm/s Comparison ladder: ├── GPS accuracy on Earth: ~1 m (24 satellites overhead) ├── DSN range accuracy: ~1 m (3 stations, one spacecraft) ├── DSN velocity accuracy: ~0.1 mm/s (Doppler is extraordinary) └── Your car's speedometer: +/- 2 km/h = +/- 560 mm/s DSN is 5,600x more precise than your speedometer.The Deep Space Network is the same system that tracked Voyager 1 at 24 billion km. At that distance, the signal takes 22 hours each way and arrives with the power of a refrigerator light bulb spread over a sphere 48 billion km wide. And they still measure the velocity to 0.1 mm/s.

Typical navigation-grade IMU drift: ~0.001 degrees/hour Over 4 days of lunar transit: Angular drift = 0.001 x 96 hours = 0.096 degrees At 384,400 km range: Position error = 384,400 x tan(0.096 deg) = 384,400 x 0.001676 = 644 km Comparison ladder: ├── Moon's radius: 1,737 km ├── IMU drift after 4 days: 644 km (37% of the Moon's radius!) ├── Target approach altitude: 130 km └── You'd miss by 5x your target altitude. Without corrections, you'd either crash into the Moon or fly past it into deep space.This is why Apollo relied on mid-course corrections -- small burns during the coast to nudge the trajectory back on target. Artemis II will execute at least 3-4 mid-course correction burns using DSN tracking data uplinked to the crew. Same physics as inertial navigation in a fighter jet, but over 1,000x longer timeframes.

Per person per day: ├── Oxygen consumed: 0.84 kg (metabolic breathing) ├── CO2 produced: 1.00 kg (exhaled waste gas) ├── Water consumed: 2.50 kg (drinking + food prep) ├── Food consumed: 1.80 kg (2,500 kcal/day) └── Heat produced: 100 W (basal metabolism) Total for 4 crew x 10 days: ├── Oxygen: 4 x 0.84 x 10 = 33.6 kg ├── CO2: 4 x 1.00 x 10 = 40.0 kg (must be removed) ├── Water: 4 x 2.50 x 10 = 100.0 kg (if not recycled) ├── Food: 4 x 1.80 x 10 = 72.0 kg └── Heat: 4 x 100 W = 400 W (continuous) Comparison ladder: ├── A scuba tank: ~2.3 kg of O2, lasts 1 person ~1 hour at depth ├── Artemis II O2 supply: 33.6 kg = 15 scuba tanks ├── ISS resupply per crew member per year: ~3,400 kg └── Artemis II total consumables: ~206 kg for 10 days That's 7,500 kg/year rate -- 2.2x more than ISS per person. WHY? Because ISS recycles water. Orion mostly doesn't.Every gram of this was launched from Earth's surface on SLS. At roughly $60,000 per kilogram to the Moon, that 206 kg of consumables cost ~$12 million just in launch mass. The physics of life support is also the economics of life support.

Reaction: 2 LiOH + CO2 --> Li2CO3 + H2O Molar masses: ├── LiOH: 23.95 g/mol ├── CO2: 44.01 g/mol └── Li2CO3: 73.89 g/mol 2 moles LiOH absorb 1 mole CO2: 2 x 23.95 = 47.9 g LiOH per 44.0 g CO2 LiOH needed per kg of CO2: 47.9 / 44.0 x 1000 = 1,089 g = 1.09 kg LiOH per kg CO2 Total CO2 to remove: 40.0 kg (over 10 days) Total LiOH needed: 40.0 x 1.09 = 43.6 kg Canister size: each canister holds ~2.5 kg LiOH Number of canisters: 43.6 / 2.5 = ~18 canisters Comparison ladder: ├── If using Ca(OH)2 instead: need 1.68 kg per kg CO2 = 67.2 kg total ├── LiOH saves: 67.2 - 43.6 = 23.6 kg (34% lighter) ├── At $60,000/kg to TLI: that's $1.4 million saved just from chemistry └── A single canister lasts about 13 hours for 4 crew members Swap one canister every half-day. Miss a swap, CO2 rises in 2 hours.Apollo used the same LiOH chemistry. Apollo 13 nearly died because the Command Module's square canisters wouldn't fit the Lunar Module's round receptacles. Same chemistry, different packaging -- and three astronauts had to build an adapter from cardboard, plastic bags, and duct tape to survive. Artemis II uses standardized canisters.

P = epsilon x sigma x A x T^4 Where: ├── P = power radiated = 2,400 W ├── epsilon = emissivity of radiator surface (~0.9 for white paint) ├── sigma = 5.67 x 10^-8 W/m^2/K^4 ├── A = radiator area └── T = surface temperature (what we solve for) Orion's radiators: ~14 m^2 of effective area T^4 = P / (epsilon x sigma x A) = 2,400 / (0.9 x 5.67e-8 x 14) = 2,400 / 7.14e-7 = 3.36 x 10^9 T = (3.36 x 10^9)^(1/4) = 241 K = -32 C The radiator surfaces must run at -32 C to dump 2,400 W. But wait: the sun heats one side of Orion to ~120 C. And the dark side cools to ~-150 C. Temperature gradient across the hull: 270 C. Comparison ladder: ├── Inside your house on a cold day: 20 C inside, -10 C outside = 30 C gradient ├── Inside a pizza oven vs. room: 300 C gradient ├── Orion hull gradient in space: 270 C -- between pizza oven and your house └── The crew cabin must stay at 21 C while the hull ranges from -150 to +120 This requires active fluid loops pumping coolant between hot and cold sides.Orion uses a two-loop thermal system. An inner water loop collects heat from equipment and crew. An outer Freon loop transfers it to the external radiator panels, which face away from the Sun. The same basic design as your car's radiator -- except there's no air flowing over it. The heat must RADIATE into the void. Same physics as a reactor cooling system.

Outer belt . . . . . . . . (electrons, . . 1-6 MeV) . . . . Inner belt . . .-----------.. . . / (protons, \ . . | 10-100 MeV) | . . | EARTH | . . | [*] | . . \ / . . `-----------' . . . . Inner: 1,000-12,000 km altitude . Outer: 13,000-60,000 km altitude . . . . . . . . Dose rate in inner belt: ~100 mSv/hour (peak) Time to transit inner belt: ~30 minutes each way Dose per transit: ~50 mSv Artemis II transits: ├── Outbound through inner belt: ~50 mSv ├── Outbound through outer belt: ~20 mSv ├── Return through outer belt: ~20 mSv ├── Return through inner belt: ~50 mSv └── Deep space coast (6-8 days): ~20 mSv (galactic cosmic rays) ──────────────────────────────────────────────── Total mission dose: ~160-200 mSv Comparison ladder: ├── Chest X-ray: 0.1 mSv ├── Annual background on Earth: 2.4 mSv ├── CT scan of abdomen: 8 mSv ├── ISS per year: 150 mSv ├── Artemis II (10 days): ~200 mSv = 83 years of background in 10 days ├── NASA career limit: 600 mSv └── Lethal acute dose: 4,000 mSv200 mSv in 10 days won't kill you. It increases your lifetime cancer risk by roughly 1% -- from the baseline ~25% to ~26%. But a solar particle event during the mission could change everything.

A 100 MeV proton's range (how far it penetrates before stopping): Material Density Range of 100 MeV proton ─────────────────────────────────────────────────────── Aluminum 2.7 g/cm3 7.5 cm (Orion hull: ~3 cm -- NOT enough) Water 1.0 g/cm3 7.7 cm (same stopping power, 2.7x less dense) Polyethylene 0.95 g/cm3 8.1 cm (hydrogen-rich, excellent) WHY is water as good as aluminum despite being 2.7x less dense? Because what stops protons is HYDROGEN -- the lightest nucleus. When a proton hits a hydrogen nucleus, it transfers up to 100% of its energy in a single collision (equal mass billiard balls). When it hits aluminum (mass 27), it transfers at most 4/27 = 15% per collision. More collisions needed. Same physics as a nuclear reactor moderator -- hydrogen slows neutrons fastest because it's the closest mass match. Orion's radiation shelter strategy: ├── Crew moves to the aft bay (most shielded location) ├── Water tanks (~200 kg) positioned between crew and Sun-facing wall ├── Equipment racks provide additional mass shielding ├── Total effective shielding: ~20-30 g/cm2 of material └── Reduces SPE dose by roughly 80% A 1,000 mSv unshielded SPE becomes ~200 mSv behind the shelter. Survivable. Barely. But survivable.There's no way to shield against everything. The highest-energy galactic cosmic rays (heavy ions at GeV energies) punch through ANY practical amount of shielding. You can reduce the dose but never eliminate it. This is the fundamental limit of human deep-space exploration -- and why radiation damage to DNA is a career-limiting factor for astronauts.

Probability Total Mission Dose Effect ───────────────────────────────────────────────────── 50% 1,000 mSv Acute radiation sickness possible Apollo missions faced the same gamble. Apollo 14 (Jan 1971): 11.4 mSv -- quiet Sun. If Apollo 14 had flown in Aug 1972 instead: A massive SPE erupted. Estimated unshielded dose: 3,840 mSv. Even inside the CM: ~1,000 mSv. It would have been lethal. The astronauts were lucky. Artemis can't rely on luck.This is why Artemis carries radiation monitors on every crew member, has real-time solar activity feeds from NOAA, and can abort to the shelter in under 10 minutes. It's also why the mission is 10 days, not 30 -- every extra day in deep space is another roll of the solar dice.

Moon's radius: R = 1,737 km Altitude above surface: h = 130 km Distance from center: d = R + h = 1,867 km Angular half-width = arctan(R / d) = arctan(1737 / 1867) = arctan(0.930) = 42.9 degrees Full angular diameter = 2 x 42.9 = 85.8 degrees But that's the geometric disc. The limb curves away, and your peripheral vision sees the surface extending in all directions. Effective visual coverage: roughly 120-160 degrees of your field of view. Comparison ladder: ├── Moon from Earth's surface: 0.5 degrees (a fingernail at arm's length) ├── Sun from Earth's surface: 0.5 degrees ├── Moon from ISS (420 km): 3.8 degrees ├── Moon from 130 km altitude: 86 degrees -- 172x larger than from Earth └── It fills your vision like a wall. Not a disc in the sky -- a WORLD beneath you.Every Apollo astronaut who saw this reported the same thing: the Moon stops being an object and becomes a place. The craters have depth. The mountains cast shadows. The horizon curves. For the first time, you understand viscerally that it's a real world, not a light in the sky.

Earth * ~384,400 km MOON .---------. / xxxxxxx \ x = radio shadow | xxxxxxxxxx | | xxOrionxpath | Orion passes through | xxxxxxxxxx | the radio shadow zone \ xxxxxxx / behind the Moon. '---------' Duration of LOS: Orbital velocity at closest v = ~2,400 m/s Shadow width at 130 km alt: ~3,200 km (chord length) Time in shadow: 3,200,000 / 2,400 = ~22 minutes Comparison ladder: ├── ISS comm blackout (typical): ~0 minutes (TDRS relay satellites cover it) ├── Apollo LOS behind Moon: ~25 minutes ├── Artemis II LOS: ~22 minutes ├── Mars comm delay (one way): 4-24 minutes (always delayed, never lost) └── Voyager 1 signal delay: 22 hours (but continuous contact) The Moon creates a HARD blackout -- not a delay, but a total cutoff. Nobody in human history has been this isolated since Apollo 17.Apollo 8 astronaut Bill Anders described the far side: "The backside of the Moon is all beat up... it's all just craters on top of craters." No atmosphere has ever smoothed the surface. Every impact from 4 billion years of bombardment is preserved.

Lunar escape velocity at 130 km altitude: v_esc = sqrt(2 x G x M_moon / r) = sqrt(2 x 6.674e-11 x 7.342e22 / 1,867,000) = sqrt(5,260,000) = 2,293 m/s Your approach velocity: ~2,400 m/s 2,400 > 2,293. You're above escape velocity. The Moon can bend your path but cannot capture you. This is a hyperbolic flyby. The trajectory is a hyperbola around the Moon, not an ellipse. Hyperbola = you leave. Ellipse = you're captured. Circle = you're in orbit. The deflection angle depends on how close you pass and how fast: delta = 2 x arcsin(1 / (1 + r_closest x v_inf^2 / (G x M_moon))) Where v_inf = hyperbolic excess velocity = sqrt(v^2 - v_esc^2) = sqrt(2400^2 - 2293^2) = sqrt(509,151) = 714 m/s The Moon bends your trajectory by roughly 140-160 degrees, sending you back toward Earth on the free-return path.If you were going 2,200 m/s instead of 2,400 -- below escape velocity -- the Moon would capture you into an orbit. You'd circle the Moon until you ran out of supplies. This is why TLI velocity must be precise to within ~1 m/s.

System Uncrewed Test Crewed Test ───────────────────────────────────────────────────────────── Life support at full No CO2 produced 4 humans exhaling 4 kg/day thermal load Manual piloting No pilot Hand controller, visual ref Star tracker navigation Partial Crew verifies, takes over in deep space if automated system fails Comm at lunar distance Data only Voice, video, human latency Thermal control with Electric heaters 4 moving heat sources, human heat sources simulate poorly variable metabolic rates Waste management N/A Toilet, hygiene, storage Crew psychology N/A Confinement, sleep, stress Emergency procedures Simulated Real humans, real time pressureThe ISS taught NASA that human systems fail in ways no simulation predicts. The first time astronauts used the ISS toilet, the suction was wrong and the result was... documented in engineering reports that no one discusses at dinner.

Distance: 384,400 km = 3.844 x 10^8 m Speed of light: c = 3.0 x 10^8 m/s One-way delay: d / c = 3.844e8 / 3.0e8 = 1.28 seconds Round-trip delay: 2.56 seconds You say "Houston, we have a problem." Houston hears it 1.3 seconds later. Houston says "Roger, Orion." You hear it 1.3 seconds after that. Total: 2.6 seconds between your words and their response. That sounds trivial. It isn't. Comparison ladder: ├── Normal conversation gap: 0.2 seconds ├── Phone call, same country: ~0.05 seconds ├── Satellite phone call: ~0.6 seconds (feels laggy) ├── ISS to Houston: ~0.5 seconds round-trip ├── Moon to Houston: ~2.6 seconds round-trip ├── Mars to Houston (closest): 8 minutes round-trip └── Mars to Houston (farthest): 48 minutes round-trip At 2.6 seconds, conversation is still possible but strained. At 8+ minutes (Mars), real-time conversation is impossible. Artemis II is the last mission where ground control can talk to the crew. After this, crews must be autonomous.The Apollo astronauts noticed the delay. Gene Cernan (Apollo 17) described how you had to consciously wait before speaking, or you'd talk over Houston's response. For Artemis crews heading to Mars, this delay will stretch to minutes -- and the crew must make life-or-death decisions alone.

Proximity operations with ICPS: ├── Fly to within 20 m of spent stage ├── Station-keep (hold position) using hand controller ├── Fly around the stage (360-degree inspection) ├── Test reaction control thrusters in all 6 degrees of freedom: │ ├── Translation: forward/back, left/right, up/down │ └── Rotation: pitch, yaw, roll └── Verify displays show correct relative motion WHY this matters for Artemis III: ├── Starship HLS is 50 m tall and 9 m wide -- enormous ├── Orion must dock with a port on Starship's nose ├── Closing speed must be < 0.1 m/s at contact ├── Misalignment tolerance: < 5 degrees └── If automated docking fails in lunar orbit, manual backup is the difference between mission success and four astronauts stuck in lunar orbit with no lander.The Apollo Lunar Module pilots practiced this in Earth orbit before attempting it at the Moon. Artemis II serves the same role: validate the human interface before betting lives on it during the actual landing mission.

Kinetic energy: KE = 1/2 x m x v^2 For Orion (m = ~9,000 kg at re-entry): LEO return (v = 7.8 km/s): KE = 0.5 x 9000 x 7800^2 = 2.74 x 10^11 J = 274 GJ Lunar return (v = 11 km/s): KE = 0.5 x 9000 x 11000^2 = 5.45 x 10^11 J = 545 GJ Ratio: 545 / 274 = 1.99 40% more speed = 99% more energy. Nearly DOUBLE. Comparison ladder: ├── 545 GJ = energy of 130 tonnes of TNT ├── = 151,000 kilowatt-hours (average US home uses 30 kWh/day) ├── = enough to power 5,000 homes for a day ├── = 15,000 liters of gasoline └── All of this energy must be converted to HEAT in about 10 minutes. That's a sustained power dissipation of ~900 MW. Nearly a gigawatt. The output of a nuclear reactor.A nuclear reactor runs at ~1 GW thermal for years behind meters of concrete and steel shielding. Orion must absorb the same power level through a heat shield 5 meters wide and 5 cm thick for 10 minutes. The engineering margins are astonishing -- and paper-thin.

DIRECT ENTRY at 11 km/s: ├── Peak deceleration: ~7-8 g for ~2-3 minutes ├── The crew weighs 8x normal. A 70 kg person feels 560 kg. ├── Breathing is difficult. Vision grays at the edges. ├── Arms weigh 40 kg each. Reaching for controls is exhausting. └── Survivable -- but barely. And no margin for error. SKIP ENTRY at 11 km/s: ├── First dip into atmosphere at ~60 km altitude ├── Decelerate from 11 km/s to ~8 km/s: peak ~4 g for ~1 minute ├── Capsule "skips" back up to ~90 km altitude (like a stone on water) ├── Coasts for ~2 minutes (crew feels ~0 g briefly) ├── Re-enters a second time at 8 km/s: peak ~4 g for ~2 minutes └── Total peak: ~4 g. Half the direct-entry load. Direct entry profile: 8g ─────╱╲──────────── ╱ ╲ 4g ────╱ ╲────────── ╱ ╲ 0g ───╱ ╲─────── | 3 min | Skip entry profile: 4g ──╱╲──────╱╲────── ╱ ╲ ╱ ╲ 2g ─╱ ╲──╱ ╲─── ╱ ╲╱╱ ╲ 0g ─╱ skip ╲─ | 1 | 2 | 2 | min Same total energy dissipated. Half the peak load. The physics is time: spread the deceleration over more minutes.Apollo used direct entry because its guidance computer couldn't reliably execute a skip. The computer had 74 KB of memory and ran at 0.043 MHz. Orion's computers have 2.5 GB and run at 200 MHz. The processing power to track a skip trajectory in real-time didn't exist in 1969. It does now.

Nominal entry angle: -5.86 degrees (below horizontal) (negative = diving into atmosphere) If entry is too SHALLOW (-5.0 degrees): ├── Capsule doesn't decelerate enough in first dip ├── Skips too high, travels too far downrange ├── Misses recovery zone by hundreds of km ├── At -3.5 degrees: capsule skips off atmosphere entirely └── Back into space. No second chance. Crew dies in orbit. If entry is too STEEP (-7.0 degrees): ├── Too much deceleration too fast ├── G-loads exceed 10g -- crew injuries likely ├── Heat shield may exceed design temperature └── At -8.5 degrees: heat shield fails. Crew dies. Safe corridor: roughly -5.0 to -7.0 degrees = 2 degrees wide At 384,400 km range, 0.1 degree of angular error equals: 384,400 x tan(0.1 deg) = 384,400 x 0.001745 = 671 km To land within 10 km: you need angular accuracy of: arctan(10 / 384,400) = 0.0015 degrees = 5.4 arcseconds Comparison ladder: ├── Width of a human hair at arm's length: ~60 arcseconds ├── Required pointing accuracy: 5.4 arcseconds ├── That's like threading a needle from 1/10th of a hair's width ├── In practice, mid-course corrections during the return coast keep adjusting the entry angle. The final burn, hours before entry, trims the angle to within 0.01 degrees.Artemis I demonstrated this in December 2022. The uncrewed capsule landed 2.4 km from the target -- within the recovery zone. But the heat shield showed unexpected erosion patterns. Material charred unevenly and some pieces broke off. For Artemis II, the shield was modified. This is exactly WHY test flights exist.

Parameter Saturn V (1967) SLS Block 1 (2022) ────────────────────────────────────────────────────────────────── Total thrust (liftoff): 34.0 MN (7.65 Mlbf) 39.1 MN (8.8 Mlbf) Height: 110.6 m 98.1 m Launch mass: 2,970 tonnes 2,603 tonnes Payload to LEO: 130 tonnes 95 tonnes Payload to TLI: 48.6 tonnes 27 tonnes Stage 1 engines: 5x F-1 (kerosene) 4x RS-25 (hydrogen) + 2x SRBs Engine reusability: Expendable Expendable (RS-25s were designed reusable for Shuttle) Cost per launch (2024$): ~$1.5 billion ~$4.1 billion Saturn V sent more payload to the Moon for 1/3 the cost. SLS has more thrust, but less payload. WHY? The RS-25 runs on hydrogen -- higher Isp (452s vs 263s for F-1) but hydrogen is absurdly low-density (71 kg/m3 vs kerosene 810 kg/m3). Hydrogen tanks are ENORMOUS -- the SLS core is mostly empty space holding a gas that weighs nothing. Saturn V's kerosene was dense and compact. Thrust-to-weight was better.This is the propellant tradeoff from Rocket Phase 4. Hydrogen wins on Isp. Kerosene wins on density. For a first stage fighting through atmosphere where drag penalizes size, kerosene's compactness mattered more than hydrogen's efficiency. SLS uses hydrogen everywhere. Saturn V used kerosene for the first stage and hydrogen only for upper stages. The engineering choice was different -- not necessarily better.

Parameter Apollo AGC (1966) Orion (2022) ────────────────────────────────────────────────────────────── Memory (RAM): 2 KB 2.5 GB Memory (ROM): 72 KB included in 2.5 GB Clock speed: 0.043 MHz 200 MHz Weight: 32 kg ~10 kg per unit Power consumption: 55 W ~50 W per unit Redundancy: 1 computer 3 flight computers (triple) Display: DSKY (numeric only) Glass cockpit, 3 screens Memory ratio: 2,500,000,000 / 74,000 = 33,784x more memory Speed ratio: 200,000,000 / 43,000 = 4,651x faster What this buys: ├── Apollo: pre-computed trajectory tables uploaded before launch. │ Mid-course corrections calculated on the ground, radioed up. ├── Orion: full trajectory simulation running in real-time onboard. │ Can recompute free-return path autonomously if ground contact lost. ├── Apollo AGC: skip entry impossible (couldn't compute fast enough) └── Orion: skip entry with real-time guidance, continuous correction Comparison ladder: ├── Apollo AGC memory: 74 KB (a single small JPEG image) ├── Your phone: 6+ GB (80,000x more than AGC) ├── Orion flight computer: 2.5 GB (34,000x more than AGC) └── Your phone has more computing power than Orion. WHY? Because Orion uses radiation-hardened chips that run 10-20 years behind consumer hardware. Your phone would crash in the Van Allen belts. Orion's computer won't.The AGC was hand-wired by women at MIT's Instrumentation Lab. Copper wire threaded through magnetic cores -- literally weaving software into hardware. A single bug required physically rewiring the machine. Orion's computers are reprogrammable in flight via software upload from Houston.

System Bandwidth What it can carry ────────────────────────────────────────────────────────────── Apollo S-band: 2.4 kbps Voice + telemetry only ISS Ku-band: 300 Mbps HD video, science data Orion Ka-band: 20+ Mbps HD video, high-rate telemetry Orion DSOC (demo): 267 Mbps 4K video (optical laser test) Ratio: 20,000 / 2.4 = 8,333x more bandwidth Comparison ladder: ├── Apollo: audio-only comm + 10 fps low-res TV = newspaper photo quality ├── Artemis Ka-band: 1080p video + voice + full telemetry = Netflix quality ├── Artemis DSOC demo: 4K video from lunar distance = YouTube 4K └── Apollo couldn't show the world what the Moon looked like in real time. Artemis can stream the experience live in high definition.DSOC (Deep Space Optical Communications) uses a laser instead of radio. Light can be focused into a tighter beam, so more power reaches the receiver. The demonstration on Artemis II will test whether optical communication works reliably at lunar distance -- critical for future Mars missions where radio bandwidth drops to ~2 Mbps.

Apollo program total (1961-1972, 2024 dollars): ├── Total cost: ~$280 billion ├── Missions to lunar orbit or surface: 9 (Apollo 8-17, excluding 9) ├── Successful landings: 6 └── Cost per lunar mission: ~$31 billion Artemis program (2011-present, through Artemis II): ├── Total cost so far: ~$93 billion ├── Missions completed: 1 (uncrewed Artemis I) ├── Landings: 0 └── Cost per mission so far: ~$93 billion WHY is Artemis so much more expensive per mission? ├── Apollo built 15 Saturn V rockets (production line economics) │ Artemis builds SLS one at a time (bespoke manufacturing) ├── Apollo: 400,000 workers, wartime urgency, 4.4% of federal budget │ Artemis: ~30,000 workers, peacetime pace, 0.5% of federal budget ├── Apollo accepted higher risk (3 astronauts died in Apollo 1 fire) │ Artemis demands higher safety margins (post-Challenger, post-Columbia) └── Apollo's supply chain existed and was scaling UP Artemis's supply chain had to be rebuilt from NOTHING Comparison: SpaceX Starship development cost so far: ~$5 billion Starship: fully reusable, 150 tonnes to LEO (vs SLS: 95 tonnes, expendable) The economics of space access are being rewritten by reusability.SLS uses RS-25 engines that were built for the Space Shuttle -- designed to fly 55 times each. Artemis throws them into the ocean after one use. Each RS-25 costs ~$146 million. Four per flight = $584 million in engines alone, dumped in the Atlantic. The fastest path to the Moon used existing hardware. The cheapest path would have waited for new hardware. Artemis chose speed.

Company / Facility Apollo Role Status by 2000 ────────────────────────────────────────────────────────────────────── North American Aviation Command Module builder Absorbed into Boeing Grumman Lunar Module builder Absorbed into Northrop Rocketdyne F-1 engine manufacturer Sold, resold, merged Michoud Assembly Facility Saturn V first stage Repurposed for Shuttle ET Kennedy Space Center LC-39 Saturn V launch pad Repurposed for Shuttle MIT Instrumentation Lab Guidance computer Became Draper Lab, moved on Thousands of subcontractors Components, materials Closed, pivoted, gone Engineers: ├── Average age of Apollo engineer in 1969: ~28 years old ├── Average age in 2000: 59 years old ├── Average age in 2020: dead └── The knowledge wasn't in documents. It was in PEOPLE. "We have the blueprints" is meaningless without the workforce that knew which tolerances actually mattered and which were over-specified, which welds needed extra inspection, which vendors delivered quality parts.When NASA tried to build the Ares V rocket in 2005, they discovered they couldn't even source the same aluminum alloy used in Saturn V's tanks. The specific alloy (2219-T87) was still manufactured, but the rolling mill that made it in the right thickness had closed. They had to qualify a new mill, a new process, and new inspections -- adding years and billions to the program.

RS-25 specifications: ├── Thrust: 2,279 kN (sea level) ├── Isp: 452 s (vacuum) ├── Propellant: liquid hydrogen + liquid oxygen ├── Design life: 55 flights (7.5 hours cumulative firing) ├── Actual use on SLS: 1 flight (~8 minutes) └── Cost per engine: ~$146 million SLS uses 4 RS-25s per flight: 4 x $146M = $584 million in engines The Shuttle flew each engine ~20 times: Cost per flight per engine: $146M / 20 = $7.3M per use SLS uses each engine once: Cost per flight per engine: $146M per use Ratio: 20x more expensive per engine-use on SLS than Shuttle. WHY not build cheaper expendable engines? ├── Developing a new engine: 7-10 years, $3-5 billion ├── Using existing RS-25: available now, human-rated now ├── Political reality: Congress mandated SLS use Shuttle hardware │ to preserve jobs in existing NASA centers and contractor facilities └── The "Senate Launch System" nickname exists for a reasonCompare: SpaceX's Raptor engine costs ~$1 million each. Fully reusable. Higher thrust (2,750 kN vs 2,279 kN). Starship uses 33 Raptors on the booster and 6 on the ship -- $39 million total in engines. Even if expendable, that's 1/15th the cost of SLS's engines alone. The economics are brutal.

Vehicle Payload to LEO Cost per kg to LEO Reusable? ────────────────────────────────────────────────────────────────────── Saturn V (1969): 130 tonnes ~$11,500/kg No Space Shuttle: 27 tonnes ~$54,000/kg Partial SLS (2022): 95 tonnes ~$43,000/kg No Falcon 9 (reused): 17 tonnes ~$2,700/kg Yes (booster) Starship (target): 150 tonnes ~$100-200/kg Yes (full) SLS costs 16x more per kg than Falcon 9. Starship targets 200-400x cheaper than SLS. The comparison isn't fair -- SLS exists and flies. Starship is still in development. But the trajectory is clear: expendable rockets are a dead end for sustained exploration. If Starship achieves even $1,000/kg to LEO: ├── 100 kg of life support consumables: $100,000 instead of $6,000,000 ├── Lunar Gateway resupply: $15M instead of $645M per mission ├── Building a Moon base becomes economically feasible └── Mars transit: the cost barrier drops from "national GDP" to "large company"SLS is the last of the expendable heavy-lift rockets. It exists because it was the fastest way to return to the Moon using proven technology. But the future of space access is reusable. The question is not whether reusable rockets replace expendable ones -- it's when.