How a Spacecraft Gets to Space: From Materials to Launch, Explained

A spacecraft gets to space on a rocket that carries its own fuel and oxidizer, climbs above the dense atmosphere, then accelerates sideways until it moves fast enough to keep falling around the Earth instead of onto it. Crossing 100 km counts as reaching space; staying there takes about 7.8 km/s of horizontal speed in low Earth orbit, which means a launcher that can deliver roughly 9 km/s or more of total velocity change once losses are added. Everything else in a space program exists to make that climb survivable and repeatable.

This guide follows the order a real program works in: requirements, physics thresholds, materials, design, build and test, regulation, launch and orbit. Figures come from NASA, ESA, the FAA, the FCC, the FAI and the Code of Federal Regulations; where we derived a number, we show the arithmetic. Propulsion stays conceptual: this is a systems explainer, not a manual for engines or propellants.

Key takeaways

• Space and orbit are different goals. The FAI uses 100 km as the boundary of space; the FAA recognises commercial crew who pass 50 statute miles (about 80.5 km). Neither height keeps you up. Orbit is about horizontal speed.
• Gravity sets the speeds. NASA Glenn gives 17,478 mph (about 7.81 km/s) for a circular orbit 100 miles up. NASA's Earth fact sheet gives escape velocity as 11.186 km/s.
• The rocket equation sets the price. Velocity grows with the logarithm of the mass ratio, so each extra km/s costs a multiplying amount of propellant. That is why rockets stage.
• Materials are chosen for strength per kilogram, temperature range and how well they can be joined and inspected. Examples: aluminium lithium 2195 on the Shuttle's Super Lightweight External Tank, welded aluminium on the SLS core stage, NASA's PICA ablator on heat shields.
• Testing reproduces launch on the ground. NASA's GEVS standard and ESA's test centre shake, blast with noise (up to 156 dB) and vacuum bake hardware before it flies.
• A US launch needs several permissions: an FAA Part 450 licence, FCC spectrum authority, a NOAA licence for imaging satellites, insurance sized by the FAA, and a plan to leave low Earth orbit within five years of mission end.
• Flight software has a supply chain. NASA's core Flight System is open source under Apache 2.0, and 14 CFR 450.141 requires configuration management, reuse policies and third party product policies for safety critical launch software.

The whole journey at a glance

Real programs loop back constantly (a test failure sends you back to design, a licence condition changes the trajectory), but the dependencies run roughly like this.

What does "space" actually mean?

The atmosphere thins gradually, so any boundary is a convention. Three numbers matter.

The Karman line: 100 km (FAI)

The Fédération Aéronautique Internationale (FAI) describes the Karman line as "the 100km altitude used by FAI and many other organisations" to mark the boundary of space. In a statement dated 30 November 2018 it noted published analyses making a scientific case for lowering the line to 80 km and proposed a workshop with the International Astronautical Federation to examine it. In that statement, 100 km remained the line the FAI uses.

The US convention: 50 statute miles (FAA)

FAA Order 8800.2, effective 20 July 2021, set eligibility for the FAA Commercial Space Astronaut Wings Program: meet the crew training rules in 14 CFR part 460, fly beyond 50 statute miles (about 80.5 km) on an FAA licensed or permitted vehicle, and perform activities essential to public safety or that contributed to human spaceflight safety. The FAA's human spaceflight page now says anyone on an FAA licensed vehicle who reaches 50 statute miles is listed on its Commercial Human Space Flight Recognition page.

A licensing threshold: 150 km

The FAA's licensing page says Part 450 covers launch operations that exceed 150 km in altitude, exceed 200,000 lb-sec (the page calls it thrust; it is a total impulse unit), or launch a payload for hire. That decides which rulebook applies, not where space begins. Separately, the FCC's 2022 debris order describes 2,000 km as the upper limit of low Earth orbit (LEO).

Reaching space versus staying in space

Going up 100 km is the easy part. Staying up is the hard part.

NASA Glenn describes orbital flight as "a combination of altitude and horizontal velocity", and gives the example of a privately built suborbital spaceplane that reached the altitude to go into space but lacked the horizontal velocity to go into orbit. An orbiting spacecraft is still in Earth's gravity and falling all the time; it simply moves sideways so fast that the ground curves away as fast as it falls. NASA describes the International Space Station as traveling at five miles per second (about 8 km/s) and orbiting about every 90 minutes, 16 orbits in 24 hours.

• Suborbital: crosses into space and falls back on a ballistic arc.
• Orbital: reaches circular or elliptical orbital speed for its altitude.
• Escape: reaches escape velocity (11.186 km/s from Earth's surface, per NASA) and leaves for the Moon or beyond.

The thresholds: altitude, velocity and delta-v

Delta-v is the total velocity change a vehicle's propulsion can deliver, the currency mission designers budget in.

Why delta-v to orbit exceeds orbital speed

Three losses eat into the climb:

• Gravity loss: effort spent holding the vehicle up against gravity rather than adding horizontal speed. NASA Glenn's rocket equation derivation shows this weight term before setting it aside for the ideal case.
• Drag loss: air resistance in the lower atmosphere.
• Steering loss: thrust pointed away from the direction of travel during turns.

Launching east near the equator gives a head start of about 0.465 km/s from Earth's rotation. Losses depend on the vehicle and trajectory, and figures around 9.3 to 9.5 km/s are widely quoted for LEO, but we could not find a single NASA primary source stating one, so the table shows our derived estimate with its assumption spelled out.

Phase 1: requirements come before rockets

Engineers start with a payload and a destination, not a rocket. A satellite's requirements answer:

1. Orbit: altitude, inclination (tilt relative to the equator) and shape. Geostationary orbit costs far more delta-v than LEO.
2. Mass at separation, including the satellite's own propellant.
3. Lifetime, which drives radiation tolerance, propellant reserves and battery cycling.
4. Power and thermal: how much power it needs and how much heat it must reject.
5. Communications: data volume, ground stations and radio frequencies, which later becomes a licensing task.
6. End of life: how it will leave orbit, now a US regulatory requirement for LEO.

Each becomes a budget (mass, power, delta-v, radio link) with margins, because early estimates are almost always too light.

Phase 2: the physics of getting there

Thrust: a rocket pushes on its own exhaust

A rocket does not push against air or ground. NASA Glenn explains that it accelerates gas backward and is pushed forward by the reaction (Newton's third law). Because the oxidizer is on board, "rockets can generate thrust in a vacuum", which is why rockets work in space and jet engines, which need atmospheric air, do not. Glenn's simplified thrust equation says thrust is mass flow rate times exit velocity, plus a nozzle exit pressure term. In short: thrust depends on how much mass you throw out per second and how fast.

Specific impulse: how efficiently propellant is used

Specific impulse (Isp, in seconds) is the standard efficiency measure. NASA Glenn says the engine with higher Isp "produces more thrust for the same amount of propellant". Multiply Isp by standard gravity (Glenn uses 9.8 m/s²) to get effective exhaust velocity, the number in the rocket equation.

These NASA published figures are conceptual reference points, not design data:

The 1981 abstract says "455 vs 300" without stating vacuum or sea level, and Isp changes with outside pressure, so treat those two as a contrast. Solid motors are a major booster family, but we found no NASA figure we could verify for this post, so they are left out rather than guessed.

The trade is visible in the table. Hydrogen is efficient but very low density and extremely cold, so tanks get large. Kerosene is denser and easier to store but less efficient. Electric propulsion is very efficient but its thrust is tiny, so it can never lift a vehicle off the ground.

The rocket equation

NASA Glenn's "Ideal Rocket Equation" page derives the velocity a rocket gains as it burns propellant and gets lighter. Ignoring gravity and drag:

delta_v = Isp * g0 * ln(m_initial / m_final)

  delta_v    change in velocity (m/s)
  Isp        specific impulse (s)
  g0         standard gravity, 9.80665 m/s^2
  ln         natural logarithm
  m_initial  mass at ignition: structure + engines + propellant + payload
  m_final    mass at burnout: everything except the propellant just burned

The logarithm is the whole story. Doubling the mass ratio adds a fixed amount of speed; to go from 4 km/s to 8 km/s at the same Isp you must square the mass ratio.

Worked example: one stage versus two

An illustrative two stage vehicle, with invented round masses and the NASA reference Isp values above (300 s first stage, 455 s second stage):

import math

G0 = 9.80665  # standard gravity, m/s^2

def delta_v(isp_s, m_initial_t, m_final_t):
    """Ideal rocket equation (no gravity or drag losses)."""
    return isp_s * G0 * math.log(m_initial_t / m_final_t)

# Illustrative masses in tonnes
payload        = 20
stage2_prop    = 90
stage2_dry     = 10    # tanks, engine, avionics
stage1_prop    = 360
stage1_dry     = 40

liftoff = payload + stage2_prop + stage2_dry + stage1_prop + stage1_dry  # 520 t

# Stage 1 burns its propellant while carrying everything
s1 = delta_v(300, liftoff, liftoff - stage1_prop)            # 520 t to 160 t

# Stage 1 dry mass is dropped; stage 2 starts lighter
stage2_start = payload + stage2_prop + stage2_dry              # 120 t
s2 = delta_v(455, stage2_start, stage2_start - stage2_prop)    # 120 t to 30 t

print(round(s1), round(s2), round(s1 + s2))
# 3468 6186 9653   (m/s)

# Same total propellant and dry mass, but one stage, nothing dropped
single_300 = delta_v(300, liftoff, liftoff - stage1_prop - stage2_prop)
single_455 = delta_v(455, liftoff, liftoff - stage1_prop - stage2_prop)
print(round(single_300), round(single_455))
# 5900 8948   (m/s)

What the numbers say:

• Two stages deliver about 9.65 km/s of ideal delta-v, a small margin over our illustrative 9.3 km/s for LEO.
• The same hardware as one stage reaches about 5.9 km/s at 300 s, or about 8.9 km/s even at 455 s throughout. Still short, and optimistic, because a hydrogen first stage needs much bigger tanks.
• The difference is the 40 tonnes of empty first stage. Dropping it lets the second stage start at 120 tonnes instead of 160.

To reach 9.3 km/s in one stage, the mass ratio would need to be about 8.0 at 455 s or about 23.6 at 300 s: 87.6 or 95.8 percent of liftoff mass as propellant, leaving almost nothing for structure, engines and payload. That is why nearly every orbital launcher stages.

The ascent trajectory and Max Q

NASA Glenn's Flight to Orbit page describes the sequence. Thrust exceeds weight and the vehicle rises vertically, steered by a guidance system that gimbals (swivels) the engine. It leaves the dense atmosphere "as quickly as possible", pitches toward horizontal, drops the first stage several minutes in, and the upper stage keeps accelerating until it cuts off at a precise altitude and speed, in orbit.

On the way it passes Max Q, maximum dynamic pressure. Glenn explains that dynamic pressure depends on air density and speed: zero at liftoff, rising with speed, then falling as the air thins faster than speed grows. Glenn calls Max Q "a design constraint on full scale rockets", and many vehicles throttle down through it.

Phase 3: from matter to materials

Material choice is driven by a few properties:

• Specific strength and stiffness: every kilogram of structure is a kilogram less payload.
• Temperature range: cryogenic tanks, hot engines and reentry surfaces, and the swing between sunlight and shadow in orbit.
• Fracture behaviour and weldability: tanks are pressure vessels, and welds are where they fail.
• Behaviour in vacuum and radiation: some materials release gas or degrade under radiation and ultraviolet light.
• Manufacturability and cost: a material you cannot form, join or inspect reliably is not useful.

Aluminium alloys and aluminium lithium

NASA's Small Spacecraft Technology State of the Art report says conventional machined aluminium alloy assemblies "remain the dominant approach for primary structures" for larger small satellites, and that many standard structures are machined from alloy 6061 or 7075.

Adding lithium lowers density and raises stiffness. The Shuttle's Super Lightweight External Tank used aluminium lithium alloy 2195. A NASA Marshall presentation on the Technical Reports Server notes that the tank was welded with 4043 filler wire, and documents six years of work on a new filler wire (B218) to improve weld and repair weld properties, tested at room and cryogenic temperatures. A new alloy is only as good as the process for joining it.

NASA's SLS core stage shows welded aluminium at scale: about 212 feet tall and 27.6 feet across, holding 537,000 gallons of liquid hydrogen and 196,000 gallons of liquid oxygen. Each barrel section is eight aluminium panels, welded into five major components: hydrogen tank, oxygen tank, engine section, intertank and forward skirt.

Composites

Carbon fibre composites put strength where loads run. NASA's report calls composites "inherently inhomogeneous and anisotropic", letting properties be tailored to load paths, and describes a graphite epoxy sandwich panel with structural mass under 3 kg per square metre. NASA Langley's Deployable Composite Booms, flown on the Advanced Composite Solar Sail System launched in April 2024, show composites unfolding in orbit. The trade off: internal damage is harder to inspect, and joints where metal fittings attach need care.

Stainless steel and titanium

Stainless steels keep strength from cryogenic to hot temperatures and suit large builds, at higher density than aluminium. Titanium alloys offer high strength per kilogram and heat tolerance for fittings, fasteners and pressure vessels, but are harder and costlier to machine. We found no NASA or ESA page with specific figures we could cite for these families, so we keep to general terms.

Thermal protection systems

A 2019 NASA Ames presentation describes PICA (Phenolic Impregnated Carbon Ablator), used on the Stardust capsule's forebody heat shield (about 0.8 m across), the Mars Science Laboratory heat shield (4.5 m) and OSIRIS-REx. Ablators char and erode in a controlled way, carrying heat off with the shed material. Reusable systems such as the Shuttle's tiles instead insulate and survive many flights.

Phase 4: design, where subsystems meet

A spacecraft is a set of subsystems sharing one mass and power budget:

• Structure and mechanisms: carry launch loads, hold alignment, deploy arrays and antennas.
• Propulsion: usually small thrusters for attitude and orbit changes.
• Power: solar arrays, batteries, power conditioning.
• Thermal control: radiators, blankets, heaters, heat pipes.
• Command and data handling (C&DH): onboard computer, memory, data buses.
• Guidance, navigation and control: star trackers, gyroscopes, reaction wheels or thrusters.
• Communications: radios and antennas on licensed frequencies.
• Flight software: schedules tasks, handles faults, processes commands, ties it all together.

Avionics and radiation hardening

NASA's small spacecraft avionics report calls space radiation "a major factor to consider" and sorts its effects in two:

• Cumulative effects: total ionizing dose (TID, in krad), which slowly degrades parts, and displacement damage dose (DDD), non ionizing damage from particle collisions.
• Single event effects (SEEs): one energetic particle flips a bit or triggers a fault.

The report says modern C&DH avionics use radiation hardened by design techniques, fault tolerant architectures and error mitigation in software, hardware and circuits. In practice: tolerant parts, error correcting memory, watchdog timers, redundant computers, and software that detects and recovers from upsets.

On the launcher, NASA Glenn notes that full scale vehicles rely on "a sophisticated guidance system" that gimbals engine thrust. Guidance software compares the vehicle's state with the planned trajectory and steers within the structural limits set at Max Q.

Phase 5: build and test, launching on the ground first

A spacecraft gets one launch, so programs reproduce launch and space on the ground first. NASA Goddard's General Environmental Verification Standard (GEVS, GSFC-STD-7000) is a baseline for showing, by test or analysis, that hardware performs in its mission environments and meets minimum workmanship standards, with guideline test levels and methods.

Qualification versus acceptance

ESA's description of its ESTEC vibration facilities draws the key distinction. Qualification tests run on a structural model during development to show the design survives launch with margin and to check the mathematical model's resonant frequencies. Acceptance tests run on the flight model to verify workmanship and confirm it works in its final configuration.

Vibration and acoustic

ESA lists three vibration modes: sine (a progressive sweep of frequencies and amplitudes), random, and shock (a sudden input simulating stage separation and engine firing). Its shakers include an 80 kN unit for small items and two 160 kN units for subsystems.

Launch noise is itself a load, especially on large light panels. ESA's Large European Acoustic Facility is 11 m wide, 9 m deep and 16.4 m high with half metre reinforced concrete walls. Horns with cut off frequencies of 25, 35, 80 and 160 Hz plus three high frequency generators reach up to 156 dB, with provisions for 158.5 dB. The chamber, commissioned in 1990, rests on springs.

Thermal vacuum

In orbit there is no air to carry heat. ESA's Large Space Simulator, Europe's largest vacuum chamber, is 15 m high and 10 m wide. ESA says it reaches a vacuum "a billion times lower" than sea level, its nitrogen cooled shrouds run between 353 K and 100 K, and its sun simulator produces a 6 m beam: 12 of 19 xenon lamp modules give one solar constant (1,380 W per square metre in ESA's figure), and all lamps exceed 2,700 W per square metre. Tests can run for weeks.

Engines and software

NASA says the RS-25 engine used on SLS has had more than 3,000 starts and more than 1 million seconds of ground test and flight firing time, and was tested up to 113 percent thrust to allow safety margins. Software is tested alongside hardware: unit tests, hardware in the loop rigs and end to end tests from a ground station. For safety critical launch software, the FAA requires testing by a team independent of the developers.

Phase 6: the regulatory path

This section covers the US, whose rules are published openly in the Code of Federal Regulations. Other countries have their own regimes.

The FAA launch licence (14 CFR part 450)

The FAA's Office of Commercial Space Transportation regulates commercial launches and reentries under the Commercial Space Launch Act of 1984, codified at 51 U.S.C. 50901 to 50923 (as FAA Order 8800.2 summarises). A vehicle operator licence may authorise launch, reentry or both, including pre and post flight operations. The FAA encourages pre-application consultation and publishes accepted means of compliance.

Part 450 sets safety outcomes rather than prescribing designs. Section 450.101 includes:

• Collective risk to the public (excluding people in aircraft and neighbouring operations personnel) of no more than an expected 1 × 10^-4 casualties.
• Individual risk to any member of the public of no more than 1 × 10^-6 per launch.
• Individual risk to neighbouring operations personnel of no more than 1 × 10^-5 per launch.
• Aircraft hazard areas that keep the probability of a casualty causing debris impact on aircraft at or below 1 × 10^-6.

Operators meet these through hazard control strategies such as flight abort with a flight safety system (450.108), physical containment (450.110) or wind weighting (450.111), backed by trajectory, debris, population exposure and failure probability analyses (450.115 to 450.135).

Software is regulated too: 14 CFR 450.141

Section 450.141, "Computing systems", requires operators to identify any software or data that could present a public hazard by working, failing or working unexpectedly, rate its criticality, and develop and verify safety requirements. Safety critical items need verification including testing by an independent team. The development process must define, among other things:

• review "such that no person approves that person's own work";
• traceability from requirements to verification evidence;
• configuration management specifying the content of each released version;
• reuse policies for reused software;
• third party product use policies for any third party product.

In plain words, a federal requirement to manage the software supply chain of launch safety software.

Insurance (14 CFR part 440)

The FAA determines a maximum probable loss (MPL) for each licensed activity, within 90 days of receiving complete information. Under 440.9, required third party liability insurance is based on the MPL but capped at the lesser of $500 million or the maximum available on the world market at reasonable cost; coverage for US government property is capped at the lesser of $100 million or the maximum available. Above that, the rule describes a federal payment mechanism for covered third party claims up to $1.5 billion, adjusted for inflation since 1 January 1989. The FAA also notes the legislative moratorium on human occupant safety rules, set in 2004, now expires on 1 January 2028.

Spectrum, remote sensing and debris

Spectrum. A satellite that cannot talk to the ground is debris. In the US the FCC's Space Bureau leads policy and licensing for satellite communications, and satellite frequencies are also coordinated internationally through the International Telecommunication Union, with US filings made through the FCC. Frequency choices shape radio and antenna design, so this starts early.

Remote sensing. Imaging satellites need a licence from the Commerce Department's Commercial Remote Sensing Regulatory Affairs division, issued under 15 C.F.R. 960 pursuant to 51 U.S.C. 60101 and following. It encourages consultation before applying.

Debris. On 29 September 2022 the FCC adopted (and on 30 September released) a rule requiring LEO satellites to be disposed of within five years of mission end, replacing the 25 year guideline. After a two year grandfathering period, satellites launched after 29 September 2024 must comply, with waivers considered case by case. Disposal is therefore a day one design requirement: a low enough orbit to decay in time, or propellant and working systems to deorbit.

What launch costs

A 2018 paper by Harry W. Jones of NASA Ames, "The Recent Large Reduction in Space Launch Cost", divided cost per flight by maximum payload to LEO:

• The Space Shuttle: about $1.5 billion for 27,500 kg, or $54,500 per kg.
• 1970 to 2000 average: $18,500 per kg, typical range $10,000 to $32,000.
• Commercial launch from 2010 on cut the cost to LEO by about a factor of 20 versus the Shuttle, by the paper's figures, with one reusable first stage vehicle advertised at about $2,720 per kg.

These are one NASA author's calculations from 2018 published prices, not an agency price list. We cite them for scale, not to compare providers.

Phase 7: launch day

The spacecraft ships to the launch site, is fuelled if it carries propellant, encapsulated in the fairing and mated to the launcher.

Range safety. The operator controls hazard areas on land, sea and air (450.161), holds agreements with the relevant authorities (450.147) and tracks the vehicle throughout flight (450.167). Part 450 requires the data needed to evaluate flight abort rules to be available to the flight safety system "under all reasonably foreseeable conditions during normal and malfunctioning flight".

Go or no go. Flight commit criteria (450.165) cover weather including lightning (450.163), vehicle health and range status; a readiness review (450.155) and crew rest rules (450.151) apply too.

Ascent:

1. Liftoff: thrust exceeds weight.
2. Vertical climb, then pitch over downrange.
3. Max Q, often with engines throttled down.
4. First stage cutoff and separation, a few minutes in.
5. Upper stage ignition and acceleration toward orbital speed.
6. Fairing separation once the air is thin enough.
7. Upper stage cutoff at the planned altitude and speed.
8. Payload separation.

Phase 8: in orbit and at the end of life

Separation starts the mission. Commissioning comes first: the spacecraft stabilises, deploys arrays and antennas, contacts ground stations and checks each subsystem. Then operations: data collection, orbit adjustments, software updates, anomaly handling. In low orbits, residual atmosphere drags satellites down, so stations and low satellites periodically reboost. Glenn notes that higher orbits need less speed and take longer per lap. Finally, under the FCC rule, a LEO satellite must leave orbit within five years of mission end, closing the loop that began with requirements.

The software inside the spacecraft has a supply chain

Software runs through every phase: ground systems, launcher guidance, flight software, test tools and simulators. Much of it is reused.

NASA's core Flight System (cFS) is the clearest example. NASA describes it as a reusable flight software framework built from a platform support package (PSP), an operating system abstraction layer (OSAL) and the core Flight Executive (cFE), used on 40 or more NASA missions including the Roman Space Telescope. Its GitHub repository says the open source version is released under Apache 2.0 and covers the framework plus a set of common apps, libraries and tools; related elements carry their own licences.

Reuse brings the questions every software team faces:

• Which versions of which components are in the build that flies?
• If a component vulnerability is disclosed, which missions and ground systems carry it?
• Do component licences fit the mission's distribution plans?
• Can you prove what was in a released version months later?

NASA's Software Assurance and Software Safety Standard (NASA-STD-8739.8) defines software assurance, software safety and independent verification and validation requirements for software created, acquired, provided or maintained by or for NASA, from conception to retirement. With the FAA's 450.141, both point to one practice: know what is in your software, where it came from and how it was verified.

More on this: software supply chain controls for space systems, the emerging space industry software supply chain, why embedded and space code still uses C and aerospace and defence software supply chains.

To be clear about scope: Safeguard does not design, test, certify or license spacecraft or launch vehicles. Its fit is the software supply chain: inventorying open source components in the code and container images you build (flight software builds, ground systems, test tooling), matching them against vulnerability data, and helping move vulnerable components to fixed versions.

Common mistakes and misconceptions

• "Getting to space means going up." Altitude gets you across the line; about 7.8 km/s of sideways speed keeps you there.
• "There is no gravity in orbit." Gravity at station altitude is still strong. Astronauts float because they and the station fall together.
• "Rockets push against the air." NASA Glenn is explicit that rockets carry their oxidizer and generate thrust in a vacuum.
• "Orbit needs escape velocity." Escape (11.186 km/s) is for leaving Earth. Low orbit needs about 7.8 km/s plus losses.
• "Just add more fuel." The rocket equation is logarithmic; more propellant means heavier tanks and structure. Staging and higher Isp are the levers that work.
• "Specific impulse is thrust." Isp is efficiency. A high Isp electric thruster has tiny thrust.
• "The Karman line is a border everyone agrees on." It is a widely used convention; the FAI uses 100 km and reviewed an 80 km proposal, while US practice recognises 50 statute miles.
• "Testing is one big shake at the end." Programs run qualification on dedicated models and acceptance on flight hardware, across vibration, acoustic, shock and thermal vacuum, with software tested throughout.
• "The FAA licence is the only permission." Spectrum, remote sensing, insurance and debris disposal are separate requirements.
• "Flight software is custom, so supply chain risk does not apply." Frameworks, operating systems, libraries and build tools are reused widely, and the FAA requires third party product policies for safety critical launch software.

FAQ

How high do you have to go to reach space?

The FAI uses 100 km (about 62 miles). The FAA recognises commercial crew who pass 50 statute miles, about 80.5 km. Both are conventions.

How fast does a rocket need to go to get into orbit?

NASA Glenn gives 17,478 mph, about 7.81 km/s, for a circular orbit 100 miles up. With gravity, drag and steering losses, the launcher must deliver more, roughly 9 km/s or more depending on trajectory.

What is the difference between suborbital and orbital flight?

A suborbital flight reaches space and falls back on a ballistic arc. An orbital flight also reaches the horizontal speed to keep falling around the Earth.

What is Earth's escape velocity?

NASA's Earth fact sheet gives 11.186 km/s, the speed needed without further thrust to leave Earth's gravity.

What is the rocket equation in simple terms?

Velocity gained equals effective exhaust velocity times the natural logarithm of starting mass divided by burnout mass. Each extra unit of speed costs a multiplying amount of propellant.

Why do rockets have stages?

Dropping empty tanks and engines means later stages do not accelerate dead weight. In our example the same hardware reached about 9.65 km/s as two stages and about 5.9 km/s as one.

What materials are spacecraft made of?

Mostly aluminium alloys (including aluminium lithium for tanks), carbon fibre composites, titanium and stainless steel, plus thermal protection such as NASA's PICA ablator.

What tests does a spacecraft go through before launch?

Vibration (sine, random, shock), acoustic, thermal vacuum and functional tests, first at qualification levels on test models, then at acceptance levels on the flight model, as described in NASA's GEVS and by ESA's test centre.

Do you need a licence to launch a rocket in the US?

Within the scope of 14 CFR part 450, yes: an FAA vehicle operator licence. Satellites also need FCC spectrum authority, a NOAA licence if they image the Earth, FAA set insurance, and a plan to leave LEO within five years of mission end.

How much does it cost to launch a kilogram to orbit?

It varies by vehicle and over time. A 2018 NASA Ames paper calculated about $54,500 per kg for the Shuttle and an average of $18,500 per kg for 1970 to 2000 systems, with newer commercial systems about 20 times cheaper than the Shuttle by its figures.

Fix it without waiting for upstream

Spacecraft physics is not a software problem, but modern space programs run on reused software: frameworks like NASA's cFS, operating systems, libraries, ground services and container images, and the FAA requires third party product policies for launch safety software. The steps below apply only to the open source dependencies and images those teams build and ship, not to hardware, propulsion or licensing, and fork and patch is in early access.

Whatever the vulnerability, most of the risk sits in two gaps. The first is the time between a fix being published and that fix actually running in your production builds. The second, and worse, is the time before any fix exists at all, while a maintainer works out a patch and a release schedule. Waiting through both gaps is a habit, not a requirement. Safeguard gives teams three ways to close them.

1. Autonomous remediation: pull requests, not tickets

When a fixed version exists, Safeguard's Griffin models open a tested, policy gated pull request that moves the affected dependency to the safe version. You can run it on one repository or in bulk across every project that carries the vulnerable version. Remediation covers eight categories, including dependencies, licenses, security posture and findings. Reachability analysis ranks the vulnerable code paths your application actually calls first, so the first pull requests you merge are the ones that matter most. Autonomous remediation ships switched off, with every category set to Manual, so your team decides how much runs on its own and what still needs a human approval.

2. Gold Open Source: the safe version in seconds, free

gold.safeguard.sh is free and needs no login. Search any CVE or package to see affected and fixed versions, CVSS, EPSS, CISA KEV status and a recommended safe version, or use the "Am I affected?" checker. The same data is available through a free read API, README status badges and RSS threat feeds. The safeguard-sh/gold-scan-action GitHub Action can fail a build on a known exploited vulnerability, on any known vulnerability, or never, so a version you have just removed cannot quietly come back in next week's merge.

3. Gold Registry: start from components that are already fixed

registry.safeguard.sh provides hardened, drop in replacements for upstream packages and container images. Every Gold artifact is held continuously to zero exploitable critical or high CVEs, zero malware and a completed license check, with SLSA L3 or higher provenance, a Sigstore signature and a public evidence page you can hand to an auditor. An artifact loses its Gold badge if it regresses. Teams can point their own package directory at the registry so hardened equivalents resolve without changing how developers install anything.

When upstream has no fix yet, or will not ship one, Safeguard first works to get the fix accepted upstream. If the issue is severe or upstream will not move, Safeguard builds a fork, fixes the vulnerability there and publishes a hardened Gold build, so you are not stuck on a maintainer's release schedule. You can also request a hardened, compatibility tested build of a specific package. This fork and patch flow is in early access today, so ask which packages it covers before you plan around it.

One honest limit: when the vulnerable code lives in a vendor appliance, a hosted service or an operating system runtime you do not build yourself, the fix still has to come from that vendor or distribution. The steps above apply to the open source dependencies and container images you build and ship, which is where most teams carry the largest and least tracked share of their exposure.

See how Safeguard would help

These demos show the software side only; none touches spacecraft hardware, launch operations or licensing. You can download Safeguard or read the CLI documentation.

The first demo is a real CLI session: safeguard opens the interactive terminal, is asked to "fix the lodash CVE in this repo", reads the finding and advisory, asks before each command, edits package.json to lodash 4.17.21, runs npm install and npm test with approval, prints a summary and pushes nothing. The same ask before acting flow suits a ground software or tooling repository where every change needs review.

Click through it at your own pace, or open it full screen.

The second demo sets up the runner on your own machine, approves it with its device code, then scans from the terminal. Teams whose flight software builds must stay on their own hardware can use the same pattern to scan locally.

Click through it at your own pace, or open it full screen.

The third demo connects Claude to the Safeguard MCP server with OAuth and lets you choose which tools Claude can use. An engineer can then ask which projects carry a given component version, such as a specific framework release, while controlling what the assistant can reach.

Click through it at your own pace, or open it full screen.

Quick check

Sources

• FAI: Statement about the Karman Line, 30 November 2018 (archived copy)
• FAA Order 8800.2: Commercial Space Astronaut Wings Program, effective 20 July 2021 (PDF)
• FAA: Human spaceflight
• FAA: Vehicle operator licensing under Part 450
• eCFR: 14 CFR part 450, Launch and Reentry License Requirements
• eCFR: 14 CFR part 440, Financial Responsibility
• FCC: FCC adopts new 5-year rule for deorbiting satellites
• FCC: Second Report and Order, FCC 22-74, adopted 29 September 2022 (PDF)
• FCC: Space Bureau
• Office of Space Commerce: Commercial Remote Sensing Regulatory Affairs
• NASA: Earth fact sheet
• NASA Glenn: Ideal rocket equation
• NASA Glenn: Specific impulse
• NASA Glenn: Flight to orbit
• NASA Glenn: Dynamic pressure and Max Q
• NASA Glenn: Liquid rocket engine
• NASA: Space station facts and figures
• NASA: Space Launch System RS-25 core stage engine
• NASA: Space Launch System core stage
• NASA NTRS 19810057776: Space Shuttle main engine progress through the first flight
• NASA NTRS 20000112954: Aluminum Lithium Alloy 2195 Fusion Welding Improvements with New Filler Wire
• NASA NTRS 20190028929: Sustaining Phenolic Impregnated Carbon Ablator (PICA) for future NASA missions (PDF)
• NASA NTRS 20200001093: Jones, The Recent Large Reduction in Space Launch Cost, ICES 2018 (PDF)
• NASA: Small Spacecraft Technology State of the Art, Small Spacecraft Avionics
• NASA: Small Spacecraft Technology State of the Art, Structures, Materials and Mechanisms
• NASA: Small Spacecraft Technology State of the Art, In Space Propulsion
• NASA Technical Standards: GSFC-STD-7000, General Environmental Verification Standard
• NASA Technical Standards: NASA-STD-8739.8, Software Assurance and Software Safety Standard
• NASA: core Flight System
• NASA core Flight System on GitHub (Apache 2.0)
• ESA: Electrodynamic shakers, ESTEC test centre
• ESA: Large European Acoustic Facility (LEAF)
• ESA: Large Space Simulator (LSS)

Last reviewed: 23 October 2026.