OSP-RA3-001 · Programme Architecture · Rev 0.3
The Open
Spaceplane Project
Spaceplane Project
Unified Programme Architecture · Supernova Labs · Manchester
Build · Race · Space
MCR UK · SN-001
Home to the dreamers
The Open Spaceplane Project exists to answer a specific question: can a small, open-source team build a reusable single-stage-to-orbit spaceplane, starting from a hackspace, without the resources of a nation state or a billionaire?
We believe Open source can move faster, and crosses all moats. Governments and primes have been working on spaceplanes for sixty years; the business case has existed just as long. What hasn't existed is a community-driven, openly published programme that anyone can contribute to, verify, and build on. That is what this project is. Every design file, test result, and specification is published under CC BY-SA 4.0.
The structure of the programme is three words: Build. Race. Space. Build world-class engines and airframe demonstrators. Get them into competition — a racing league that generates revenue, proves reliability, and keeps the engineering honest. Then go to orbit. Each phase funds the next.
The central technical decomposition is this: airframe and propulsion are separate concerns. The airframe can be built incrementally in a community workshop, validated at subscale, and scaled methodically. Propulsion cannot; the physics of SSTO demand a combined-cycle engine that does not yet exist as a commercial product. This programme solves that by treating propulsion as a drop-in interface — exactly as Formula Student treats the engine.
The programme does not attempt to prove the MAV engine and the orbital airframe simultaneously. The airframe programme validates everything else — structures, TPS, avionics, re-entry, landing — using whatever propulsion is available at each phase. The MAV engine is a parallel R&D track that plugs in at Phase 3.
Riveted Al-6061 with waterjet-cut frames and TIG-welded 4130 steel fittings. Repeatable, inspectable, repairable in the field, and executable with tools available in any well-equipped hackspace. The same construction logic that built early jet-age experimental vehicles applies here.
The Tsiolkovsky rocket equation is non-negotiable. A hybrid cluster at Isp 260s cannot reach orbit from a single stage regardless of vehicle size. The end-state vehicle requires a combined-cycle propulsion system delivering effective Isp well above 1,000s across the full trajectory.
All intermediate vehicles are derived from this specification by removing capability. Design choices at every earlier phase must be compatible with this envelope.
| Parameter | Value | Basis |
|---|---|---|
| Overall length | 15.0 m | Cryogenic tank volume + MAV engine bay |
| Wingspan | 10.0 m | Re-entry L/D ≥ 2.5, blended wing-body |
| GLOW | 25,000 kg | Rocket equation: mass ratio 5 at avg Isp 1,500s |
| Dry mass | 5,000 kg | CF + CMC structure, TPS, systems |
| Payload | 1,000 kg | 4 passengers or cargo equivalent |
| Target orbit | 400 km LEO | ISS-compatible inclination |
| Propulsion | 1× MAV combined-cycle | Air-breathing + MPD rocket |
| TPS | C/C leading edges, TUFI tiles, blankets | Peak re-entry 1,200°C |
| Landing | Tricycle gear, RTLS glide | HTOL |
| Reusability | Target 50 flights | Airline operations model |
All programme vehicles share a common engine bay definition. Teams may fly any propulsion unit that conforms to EIS-1. The MAV engine is the intended final occupant; earlier vehicles use whatever is physically appropriate.
| Interface Parameter | RA3 Full Scale | RA2 Subscale | RA1 Demonstrator |
|---|---|---|---|
| Mount pattern | EIS-1-FULL 800mm, 8×M24 | EIS-1-SUB 400mm, 6×M16 | EIS-1-DEMO 200mm, 4×M12 |
| Max dry engine mass | 800 kg | 150 kg | 40 kg |
| Electrical power in | 28V DC, 400A | 28V DC, 100A | 12V DC, 30A |
| CAN bus | CAN 2.0B, OSP v1 | CAN 2.0B, OSP v1 | CAN 2.0B, OSP v1 |
| TVC | ±8° gimbal | ±8° gimbal | ±5° fixed or gimballed |
| Emergency shutdown | Hardware, <50ms | Hardware, <50ms | Hardware |
Four vehicles, each a strict superset of the previous. Every phase ends with something that flies. Intermediate vehicles are prototypes and are the revenue-generating competition vehicles that fund the next phase.
3D-printed and carbon-fibre blended wing-body drone, 3-metre span, flying autonomously. Aerodynamic proof of concept — delta planform, autonomous flare and runway landing. The OSP flight software stack is validated here before it goes near a crewed vehicle. Also the fundraising tool.
6-metre riveted aluminium airframe, hackspace-built. Target altitude 50km. Re-entry heating demonstration. Autonomous runway landing. Proves the structural and thermal concept. Propulsion: Class A hybrid cluster or Class C COTS turbojets.
10-metre CFRP airframe, scaled 1.7× from RA1. Enters the spaceplane racing league — competing for prize money, generating media, building the brand. Racing league revenue provides the runway to Series B. MAV engine air-breathing mode tested in this phase.
15m, 25t GLOW, MAV combined-cycle engine, LH₂/LOX propellant. Takes off from a runway, reaches 400km LEO, returns, lands. Unrefuelled. Unstaged. Reusable to 50 flights. The Holy Grail of Human Spaceflight and the dawn of s new era..
The following sequence describes how to build RA1 from raw materials in a well-equipped hackspace or light industrial workshop.
Waterjet-cut 2mm Al-6061 frames at 500mm spacing, formed on the brake. Four 50×25×3mm rectangular-tube longerons, 5.5m span. Cleco-assembled to verify fit before any riveting.
Four 3mm plate bulkheads at engine mount, main spar, nose gear, and aft pressure bulkhead. EIS-1-DEMO bolt pattern machined to ±0.1mm. All structural loads route through here.
1.5mm Al-6061 sheet, waterjet blanked, brake-formed over the frames. 3M Scotch-Weld DP460 bond line before riveting. Pneumatic rivet gun, AN470 rivets throughout. Leave engine bay panels removable.
Two 4mm plate spars, CNC machined with lightening holes. Sixteen 2mm waterjet-cut ribs. Main spar root fitting is 4130 steel, TIG-welded, bolts to the fuselage main spar bulkhead.
Bottom surface: Insulfrax ceramic blanket, 50mm, stainless wire ties to standoffs. Top surface: 25mm blanket + Cerakote spray finish. Leading edges (C/C panels) bolted on last.
Aluminium tray on captive rails. Primary flight computer (CubeSat-class ARM), triple-redundant single-board computer backup cluster with TMR voting — Linux-based OS running the OSP flight stack. VN-300 IMU, u-blox RTK GPS, RM3100 magnetometer. All connectors MIL-spec.
Welded 4130 chromoly tube truss, oleo-pneumatic strut, 15-inch GA tyre. All retract hydraulically into flush bays. Trunnion mounts CNC-machined 4130 billet.
Any EIS-1-DEMO conformant engine bolts to the aft bulkhead. CAN bus harness plugs in. If the engine is swapped between flights, only the propellant feed lines and harness change — the airframe is unmodified.
| Parameter | RA1 | RA2 | RA3 |
|---|---|---|---|
| Scale factor k | 1.0× | 1.7× | 2.5× |
| Length | ~6 m | ~10 m | 15 m |
| Wing area | ~12 m² | ~34 m² | ~75 m² |
| Skin | 1.5mm Al | 2.0mm CFRP | CFRP laminate |
| Avionics bus | OSP-BUS v1 CAN 2.0B | OSP-BUS v1 | OSP-BUS v2 (redundant) |
| EIS standard | EIS-1-DEMO | EIS-1-SUB | EIS-1-FULL |
| Propulsion class | A, B, or C | C or D | MAV |
Because the frame spacing, avionics bus standard, and EIS bolt pattern are fixed across all phases, a hackspace team that builds RA1 is not starting over for RA2. They are scaling known-good solutions. Every decision at Phase 0 is made with RA3 in mind.
The MAV engine is not on the critical path for Phases 0–2 and is a separate programme that develops concurrently to plug into EIS-1-FULL when ready.
| Milestone | Timeline | Airframe Dependency |
|---|---|---|
| 100kW plasma-augmented RDE ground test | Year 1–2 | None |
| 1MW scaled demonstrator, Westcott | Year 2–3 | None |
| Air-breathing mode flight test in RA1 | Year 3 | EIS-1-DEMO functional |
| Full combined-cycle ground qualification | Year 4–5 | None |
| MAV engine flight test in RA2 | Year 5–6 | RA2 flying |
| MAV engine integrated in RA3 | Year 7–8 | RA3 airframe complete |
| Phase | Amount | Sources | Unlock |
|---|---|---|---|
| Pre-seed (RA0) | £15–30k | Founder capital, hackspace in-kind, crowdfund | RA0 flying |
| Seed (RA1) | £300–500k | Innovate UK, UK Space Agency SBRI, angels, Kickstarter | RA1 suborbital flight |
| Series A | £2–5M | Seraphim Capital, Space Capital, racing league sponsorship | RA2 racing launch |
| Series B (RA3) | £30–60M | Project finance, government contracts, strategic investor | MAV qualified, RA3 begins |
This programme requires no leaps of faith — every milestone is a real, physical thing that either works or doesn't.
Spaceplane development is accelerating globally. Germany, China, India, the US, and New Zealand all have active programmes. The window to establish an open, independent, non-state-aligned programme in this space is not unlimited.
A 3-metre blended wing-body flying autonomously and landing itself on a runway. Watch for: stable delta-wing flight, autonomous flare and touchdown, open-sourced flight logs.
The EIS-1 specification released publicly. If you build an engine that fits the spec, it can fly on an OSP airframe. Watch for: bolt pattern drawings, CAN protocol definition, propellant class rules.
A 6-metre vehicle reaching 50km, surviving re-entry heating, landing on a runway. Watch for: TPS post-flight results, peak heating telemetry, landing footage.
The combined-cycle engine completing a full-duration test — air-breathing modes transitioning to rocket mode. Watch for: published Isp measurements, mode transition data, test stand footage from Westcott.
A 15-metre, 25-tonne vehicle taking off from a runway, reaching 400km orbit, and returning to land — unrefuelled, unstaged, reusable.
Design files, flight software, test data, and the EIS specification will be published under CC BY-SA 4.0 as they are produced. If you are an engineer, a maker, a pilot, or just someone who thinks this should exist — follow the work and get involved.
The Open Spaceplane Project until mow has been a one-person founding effort, twelve years in the making. RA0 — the 1-metre blended wing-body drone testbed. The avionics stack is real. The airframe architecture is being built now.
Not render porn. It's a programme built on physics, open collaboration, and the conviction that the tools for space should belong to everyone.
Supernova Labs is based in Manchester. Open to collaborators, propulsion teams, engineers, and anyone who wants a future where everyone can go to space.