Tools › Orbital Mechanics › Interstellar Travel › Relativistic Trip Planner › Method
Relativistic Trip Planner: Method
Open the Relativistic Trip Planner tool
Relativistic Trip Planner: the Constantly Accelerating Ship
What the tool does
You pick a destination (or a distance, or a time on board), a steady acceleration and a trip profile. The tool returns the time that passes on the ship's clock and on Earth's clocks, the top speed as a fraction of the speed of light, Newton's answer for comparison and the propellant a rocket would need, by special relativity.
The idea
Push on a ship with a constant force and Newton says its speed grows without limit. Nature says otherwise: nothing with mass reaches the speed of light c. A crew that feels a steady push, say 1 g, the same as standing on Earth, keeps gaining speed as measured on board, but an observer on Earth sees each further year of thrust add less and less speed. The ship's path, plotted as distance against Earth time, is a hyperbola that bends toward the path of a light beam and never touches it.
The reward is in the clocks. A fast-moving clock runs slow by the Lorentz factor γ = 1 / √(1 − v²/c²). At 1 g, γ doubles about every 0.67 years of ship time once the ship is fast, so after a few years on board each ship-day covers centuries of Earth time and light-years of distance. That is why the crew can cross 2.5 million light-years in under 29 years of their own lives, and why 2.5 million years have gone by at home when they arrive.
The dots mark equal intervals on the ship's clock. Early on they are evenly spaced; later each tick spans more Earth time and more distance than the one before. The ship never catches the light beam that left with it, but it ends up trailing by a fixed lead: about c/a of Earth time, a little under one year at 1 g.
The equations
Let a be the acceleration felt on board (the proper acceleration), τ the time on the ship's clock (the proper time), t the time on Earth's clocks and x the distance measured by Earth. Starting from rest:
| Quantity | From ship time τ | From distance x |
|---|---|---|
| Lorentz factor | γ = cosh(aτ/c) | γ = 1 + a x / c² |
| Speed | v = c tanh(aτ/c) | v = c √(1 − 1/γ²) |
| Earth time | t = (c/a) sinh(aτ/c) | t = (c/a) √(γ² − 1) |
| Distance | x = (c²/a)(cosh(aτ/c) − 1) | given |
| Ship time | given | τ = (c/a) acosh(γ) = (c/a) ln(γ + √(γ² − 1)) |
The quantity φ = aτ/c is called the rapidity. Unlike speed it simply adds up: every ship-year at 1 g adds 1.0323 to φ, and v = c tanh φ turns it back into a speed that stays below c.
Units that make it easy. Measure time in years and distance in light-years. Then c = 1 light-year per year, and 1 g = 9.80665 m/s² × 31,557,600 s / 299,792,458 m/s = 1.0323 light-years per year². The fact that 1 g is so close to 1 in these units is a coincidence, and a convenient one.
Speed up halfway, then brake. The braking half is the speeding-up half run backward, so use the formulas for half the distance and double the times: γ = 1 + a(d/2)/c², τ = 2(c/a) acosh(γ), t = 2(c/a)√(γ² − 1).
Fly-by. Use the formulas once for the full distance. The ship passes its destination at top speed.
Speed up, coast, brake. With a chosen coasting speed v, γ = 1/√(1 − v²/c²). Each powered leg covers x = (c²/a)(γ − 1) in ship time (c/a) acosh(γ) and Earth time (c/a)γv/c. The coasting leg covers the rest of the distance at constant speed: Earth time xc/v, ship time shorter by the factor γ. If the two powered legs alone would overshoot the distance, the ship never reaches the coasting speed and the trip is a plain speed-up and brake.
Very large and very small numbers. For Andromeda at 1 g, γ is about 1.29 million and the speed is 0.999 999 999 999 7 c. Computing 1 − v/c by subtraction would lose every digit, so the tool uses 1 − v/c = 1 / (γ(γ + √(γ² − 1))) and, for slow trips, works with γ − 1 instead of γ. The result is accurate from a trip to the Moon to the edge of the form's range.
Newton's answer and why it fails
With no speed limit, a ship that speeds up halfway and brakes takes t = 2√(d/a) and peaks at v = √(a d). For 4.25 light-years at 1 g this gives a peak of 2.09 c, twice the speed of light, which cannot happen. For short trips the two theories agree: to Mars at 1 g the peak speed is under 0.3 % of c and Newton is right to better than one part in a million. Relativity is not a different law for fast ships; it is the one law, and Newton's is what it looks like when v is much less than c.
The rocket
A rocket moves by throwing exhaust backward at speed ve. The relativistic rocket equation (Ackeret, 1946) says the rapidity gained is (ve/c) ln(m0/m1), where m0 is the mass at the start of a burn and m1 the mass at its end. Turned around:
m0 / m1 = eΔφ c / ve, with Δφ the total rapidity change of all powered legs.
Braking costs as much rapidity as speeding up, so a trip that stops has twice the Δφ of a fly-by to the same top speed, and its mass ratio is the square. The best conceivable engine is a photon rocket whose exhaust is light (ve = c); then m0/m1 = eΔφ, which for one powered leg equals γ(1 + v/c). Any slower exhaust raises the ratio to the power c/ve. This is the real obstacle: the clocks allow the trip, the propellant does not.
What it assumes
- The acceleration felt on board is constant during each powered leg, and the engine can deliver it for the whole trip.
- Start and destination are at rest relative to each other and the distance between them is fixed. Earth's and the stars' own motions (tens to hundreds of km/s) are ignored.
- Space is flat and empty: no gravity, no expansion of the universe, no gas, dust or radiation in the way.
- The year is the Julian year of 365.25 days, the one in the definition of the light-year. c = 299,792,458 m/s and 1 g = 9.80665 m/s² are exact by definition.
- The rocket section assumes a single stage, exhaust at one constant speed, and no mass other than payload and propellant.
When it misleads
- Beyond roughly 100 million light-years the expansion of the universe changes the answer and general relativity is needed. The tool computes such trips but flags them.
- The trip is not survivable as described. At γ in the thousands, the thin gas between the stars arrives as intense radiation and starlight ahead is shifted to hard X-rays. Nothing here accounts for shielding.
- No known engine comes close. The mass ratios are lower bounds for an ideal engine; tanks, structure and the means of making the energy are left out.
- Distances are round figures. The center of the Milky Way is given as 25,000 to 27,000 light-years on different NASA pages; ship time barely changes (it depends on the logarithm of distance), Earth time changes in proportion.
- "Time on Earth" is the time between departure and arrival in Earth's frame. People on Earth learn of the arrival only when a signal returns, a further d/c later.
Common mistakes
- Using v = a t. That is Newton's formula. With a t/c = 5 it gives five times light speed; the right speed is c tanh(aτ/c) in ship time, or a t / √(1 + (a t/c)²) in Earth time.
- Mixing up the two times. τ (ship) goes with cosh, sinh and tanh of aτ/c; t (Earth) goes with √(1 + (a t/c)²). Putting Earth time into cosh gives nonsense.
- Forgetting to halve the distance for a trip that stops. The turnaround is at d/2; γ there is 1 + a d/(2c²), not 1 + a d/c². Then remember to double the times.
- Treating a as the acceleration seen from Earth. The constant quantity is the push felt on board. Seen from Earth the acceleration is a/γ³ and fades toward zero.
- Applying the time-dilation formula τ = t/γ with the peak γ. γ changes all the way; τ is the integral of dt/γ, which is where the logarithm comes from. Using the peak γ for Andromeda would give 1.9 years instead of 28.6.
- Computing 1 − v/c on a calculator by subtraction. With v/c = 0.999 999 999 999 7 the calculator shows 0. Use 1/(2γ²) for large γ.
- Mixed units. a in m/s² with time in years and c in km/s is the usual source of answers wrong by powers of ten. Convert everything to SI, or use years and light-years with c = 1 and g = 1.0323.
- Doubling a one-way fuel figure for a trip that stops. The mass ratio is squared, not doubled.
- Reading the result as "faster than light". Covering 2.5 million light-years in 28.6 ship-years is not a speed of 87,000 c: the distance is Earth's measure and the time is the ship's. In either frame alone nothing outruns light; in the ship's frame the distance itself is contracted by γ.
How the tool is checked
The closed forms are compared with the same quantities in 60-digit decimal arithmetic for 134 trips, with a numerical integration of the equations of motion that uses no closed form, and with the printed tables of the Usenet Physics FAQ article "The Relativistic Rocket" (ship time, Earth time, distance, speed, Lorentz factor and photon-rocket fuel), which they reproduce to the printed digits when 1 g is rounded to 1.03 light-years per year² as that article does. The Einstein, Misner–Thorne–Wheeler and Ackeret references are cited from memory and were not read for this page.