Cross-section (to scale)
The numbered bubbles are standard gas-turbine stations: 0 ambient β 2 compressor face β 3 compressor exit β 4 turbine inlet (your TIT) β 5 turbine exit (βEGT) β 8 nozzle exit. Numbers 6β7 are reserved for the afterburner ducts big engines have β hobby jets jump straight to 8. Hover a bubble for its gas state. Fuel enters at the rear (turbine end): a manifold ring outside the case feeds needles through the rear cover into the vaporizer sticks, which run forward and hook 180Β° so the vapor discharges rearward into the primary zone β that's the yellow feed line.
Design box β constraint checks
Flight performance (your aircraft from the sidebar β first-order, parasite drag only)
The design box β a map (drag the dot β it moves your Οc / TIT sliders)
Every combination of pressure ratio and turbine temperature, graded by the constraint checks: teal = inside the design box, amber = check limits, red = over a hard limit, dark = the cycle doesn't close at all (the turbine can't feed the compressor). Brightness = specific thrust β note how the brightest cells sit right next to the red cliff: that's the whole tension of jet design. Your other sliders (wheel, efficienciesβ¦) reshape this map live.
Brayton cycle β T-s diagram
The cycle's fingerprint: temperature vs entropy through the engine. Up-and-right legs cost you (compression heat, combustion), the down leg 4β5β8 pays you back. The rightward lean of the compression line is inefficiency (ideal compression would be vertical); the long orange leg is fuel heat; whatever temperature survives to leg 5β8 becomes jet velocity. A taller, wider loop = more work per kg of air.
Station conditions
The gas state at each numbered station (see the bubbles on the Design cross-section). Practical anchors: 3 is what your combustor and its o-rings must survive, 4 is the turbine material limit, 5 is what your EGT probe reads, and 8β0 pressure tells you whether the nozzle chokes.
| Station | Where | T [K] | T [Β°C] | P [kPa] |
|---|
Pressure-ratio sweep (at your TIT β dot = your design)
The fundamental trade study: raising Οc buys specific thrust (blue, less air needed per newton) and fuel economy (orange TSFC falling). So why not max it out? Because compressor work rises with Οc, and work = tip speedΒ² β the RPM and wheel stress climb until the aluminum gives up. The curve also flattens: past ~3.5 you pay a lot of stress for little thrust.
TIT sweep (at your pressure ratio)
Hotter gas = more energy left after the turbine = more thrust per kg of air (blue, nearly linear). But EGT (red) tracks TIT almost 1:1 β and uncooled Inconel quits around 1050 K TIT / ~750 Β°C EGT. This chart is why turbine material, not aerodynamics, is the real thrust limit of a micro engine.
Compressor map β the operating line
The chart every real engine lives on. Gray lines are constant-spool-speed lines of the (scaled, generic) compressor map; the red dashed line is surge β flow reversal, bang, possible blade contact. The teal operating line is where the turbine and the fixed nozzle force the compressor to run at each speed: your engine has no choice about it once built. The gap between the operating line and the surge line is your surge margin. The amber ring is the throttle point from the card beside this one.
Operate the engine (steady state, matched on the map)
Drag the throttle. Everything below comes from re-matching compressor, turbine and nozzle at that spool speed β not from scaling the design point. Note the very real micro-turbine quirk: EGT rises as you throttle back, because component efficiencies collapse while the fixed nozzle still needs a temperature push to pass the flow. This is why ECU manuals obsess about idle EGT.
Throttle curve (matched on the compressor map)
Thrust and EGT along the whole operating line. Thrust falls steeply β at 70% RPM you have well under half thrust, which is why turbine models need momentum management on landing. EGT is highest at the bottom of the range: a fixed-geometry engine runs hot at part speed, and the range simply stops where the model (and roughly the real engine) can no longer sustain itself.
Sensitivity β which knob matters most?
Each parameter nudged Β±3% while the rest hold still; bars show the resulting change in specific thrust. Long bars = high leverage β this is your build-effort priority list. Note how much the "boring" loss terms (intake recovery, nozzle Cv) matter: polishing a bellmouth is free thrust, while the faint mirror-image bars confirm the effect works both ways.
Energy accounting
Where the fuel's chemical power actually goes. Only a few percent becomes jet kinetic energy β the rest leaves as hot exhaust. This is normal at micro scale (low Οc, small parts, big relative clearances); compare the TSFC line against a full-size engine and appreciate why real aircraft engines are big.
3D preview (drag to orbit Β· scroll to zoom Β· double-click to reset β cutaway shows the gas path)
The STL is the full (non-cutaway) engine at your exact dimensions in mm β open it in any CAD package, slicer, or Windows 3D Viewer. It's a visualization mesh of the major components, not a machining model. The plume, combustion glow and heat-tint are viewer-only effects and never end up in the STL.
Build notes / bill of materials
| Part | Specification | ~ cost |
|---|
Rough 2026 hobby prices for bought parts and raw stock β excludes tools, fuel, and the inevitable second turbine wheel.
Manufacturing sensitivity β what to machine carefully
Each row re-matches the whole engine at 100% spool speed with one realistic build error introduced. The bars rank them: long bar = an error that actually shows up on the test stand. This is your machining priority list β sweat the parts at the top, relax about the bottom. (Deltas are at constant spool speed; a real ECU trims fuel to EGT, which shifts the numbers but not the ranking.)
Rules of thumb behind the efficiency rows: ~0.4% compressor efficiency per 1% of exducer blade height in tip clearance, ~1.5% turbine efficiency per 1% blade height. A row marked SURGE means the engine will not reach full speed with that error β the nozzle or NGV chokes it into surge first.
Engineering drawing β general arrangement
An A4-landscape GA sheet: dimensioned section, front view and the full computed-dimensions table. The sheet is NTS β the axial layout is illustrative, so work from the printed dimension values, never by measuring the sheet. First-order sizing, not a stress-checked machining drawing.
Export
The share link encodes your entire design in the URL β send it to a friend and they'll open JetStudio with your exact parameters loaded.
Saved designs (stored in this browser)
Designs live in this browser's storage β use Link to copy a share URL that moves one to another machine. Tick up to three designs to compare them below.
Side-by-side comparison
Your current design (the sidebar) is always the first column β drag a slider and watch the columns fight. Green marks the best value in rows where "best" is unambiguous across engine sizes: TSFC, thermal efficiency, thrust:weight and the flight numbers.
How a micro turbojet works
A single-spool turbojet is the Brayton cycle cast in metal: squeeze air (centrifugal compressor, pressure ratio ~1.8β4), burn kerosene in it at constant pressure (up to the turbine inlet temperature, TIT), expand it through a turbine that extracts exactly the work the compressor needs, then let the leftover pressure accelerate the gas out of a converging nozzle. Thrust = mass flow Γ exit velocity (plus a pressure term if the nozzle chokes).
Micro engines are thermodynamically bad on purpose: low pressure ratio and small size give ~5β10% thermal efficiency and roughly 10Γ the specific fuel consumption of a full-size engine. They trade efficiency for simplicity, cost, and robustness β the right trade at this scale.
The five core components
1 Β· Compressor β a turbocharger wheel, spun by you
The hobby standard is an automotive turbocharger compressor wheel (KKK/BorgWarner, Garrett), 54β76 mm
tip diameter, cast or billet aluminum. Work scales with tip speed squared (w β ΞΌΒ·UΒ², ΞΌ β 0.75β0.80):
pressure ratio 2.2 needs ~360β400 m/s of tip speed β ~110,000 rpm on a 66 mm wheel.
Behind the wheel a diffuser (wedge vanes or a drilled ring) converts swirl into pressure β the
#1 home-build efficiency killer. Expect only 70β78% stage efficiency.
2 Β· Combustor β annular, with vaporizer sticks
An annular flame tube of 0.5 mm stainless with rows of carefully sized holes: primary zone (recirculating, near-stoichiometric flame), secondary (burnout), dilution (mix down to a TIT the turbine survives). Hobby engines don't atomize fuel β they boil it inside hooked vaporizer tubes sitting in the flame, fed by needles that enter through the rear cover (the fuel manifold sits outside the case at the turbine end); each tube runs forward and its hook turns the vapor to discharge rearward into the primary zone. Rules of thumb: reference velocity 15β25 m/s, residence time 4β6 ms (sets length ~80β100 mm), pressure loss 5β8%.
3 Β· Turbine β the part that wants to die
A ring of nozzle guide vanes accelerates the gas onto a single axial rotor of cast Inconel 713C (usable to ~900β950 Β°C metal temperature). There is no blade cooling at this scale, so TIT above ~1050 K means creep: blades stretch until they rub. Buy the cast wheel β it is one of two parts you should never improvise (the other is the compressor wheel).
4 Β· Shaft & bearings
Two angular-contact hybrid bearings (ceramic SiβNβ balls) with light spring preload, lubricated by
~5% turbine oil mixed into the fuel and misted through the shaft tunnel. Speed metric:
DN = bore(mm) Γ rpm β keep under ~2.0β2.5 million. The shaft must stay below its first
bending critical speed: short and stiff wins.
5 Β· Nozzle
A simple converging cone, often over a center bullet. At hobby pressure ratios it is just barely choked or unchoked (exit 300β450 m/s). Exit area is a live tuning knob: too big β cold and weak; too small β hot and surgy.
The design box β what limits you
| Constraint | Hobby limit | If exceeded |
|---|---|---|
| Compressor tip speed | ~450 m/s cast Β· ~520 billet | wheel burst |
| TIT (uncooled Inconel 713) | ~1000β1050 K | blade creep, rub, burst |
| EGT continuous | 600β750 Β°C | ECU cuts fuel; nozzle suffers |
| Bearing DN (fuel-mist lube) | ~2.0β2.5 M mmΒ·rpm | bearing cooks, shaft whip |
| Οc, one centrifugal stage | 2β3 realistic, 4.5 max | tip speed the alloy can't take |
| Combustor loading | 15β25 m/s, β₯4 ms | blowout, flame in the turbine |
| Shaft speed | < 1st bending mode | destructive whirl |
The design loop this app automates: pick thrust β cycle gives mass flow & compressor work β wheel diameter sets rpm β check every limit β iterate.
Reference engines
| Engine | Thrust | RPM | Οc | Mass flow | Claim to fame |
|---|---|---|---|---|---|
| Schreckling FD3/67 | ~25 N | ~85k | ~1.9 | ~0.10 kg/s | 1992 original; plywood compressor |
| KJ66 | 75β92 N | 110β128k | ~2.2 | ~0.23 kg/s | THE homebuild reference, plans available |
| P100 class | ~100 N | ~150k | ~2.6 | ~0.26 kg/s | modern commercial sport engine |
| P160/P180 class | 160β180 N | ~125k | ~3+ | ~0.4 kg/s | 2 m sport jets |
Fuel, starting, control
- Fuel: kerosene / Jet-A1 (43.1 MJ/kg, ~0.80 kg/L) + 4β5% turbine oil (Mobil Jet II, AeroShell 500). Budget 150β400 ml/min at full thrust for KJ66-class.
- Starting: electric starter spins the rotor β ignite propane/butane pre-heat gas (or a kerostart glow vaporizer) β ramp to self-sustain (~45β60 krpm idle) β transition to kerosene. Failed starts that pool fuel cause the classic tailpipe fireball β purge before retrying.
- ECU/FADEC: monitors rpm + EGT, drives the fuel pump, enforces limits (max EGT, max rpm, flameout detection, auto-cooldown). Use an off-the-shelf ECU (Xicoy, JetCat-style, Orbit); do not run open-loop needle-valve control.
Safety β non-negotiable
- A turbine wheel failure at 120,000 rpm releases several kilojoules of shrapnel (see your design's rotor energy on the Build Sheet). First runs: engine bolted to a test stand, bystanders out of the rotor plane, barrier between you and the engine.
- Kerosene pools burn. COβ or dry-powder extinguisher within arm's reach; never fuel near the hot section.
- Hearing protection β >120 dB close up. The exhaust is ~700 Β°C for a meter behind the nozzle.
- Follow your national model-flying body's turbine code of practice (e.g. the GTBA's). Check insurance.
A sensible build path
- See one run first. Attend a jet meet or find a local turbine flyer.
- Build from proven plans β KJ66 or a GTBA-published design. Buy the compressor wheel and the cast Inconel turbine wheel; machine the housings, shaft, diffuser, and combustor yourself.
- Instrument from day one: EGT + rpm minimum, ECU strongly recommended.
- Only after a proven runner, iterate your own aero: diffuser, combustor holes, NGV angles, nozzle area. Change one thing at a time; log everything.
Sources and further reading are in RESEARCH.md next to this app β including Kurt Schreckling's
book (the classic), GTBA design pages, KJ66 documentation, and papers on wheel tip speeds, Inconel 713
limits, and kerosene-lubricated hybrid bearings.