Why light speed wouldn't matter

[mjl1966]

So, when we think of going to some place like Alpha Centauri, we think "4 light years away".  Implementations such as the Valkyrie from Avatar where the ship spends half the trip accelerating to near light speed and the other half decelerating start to sound feasible.

But wouldn't we be facing the same problem we face with interplanetary travel?  If we want to go to Mars, we can't just wait for it to come by for a close pass and then hop over to it.  We *must* use a transfer orbit.  There's no way around it.  A transfer orbit is still going to be about half a full orbit around the sun.  We *may* be able to shorten that up by intercepting at some point along the transfer orbit than the aphelion, but that's going to require a monster burn for orbital insertion.  no free lunch.

Well, aren't both sol and Alpha Centauri orbiting the center of the galaxy?  Doesn't that mean you have to get there using a transfer orbit?  Doesn't that mean that  you're going to have to complete a large portion of an orbit around the center of the galaxy to get there?  If you use the short orbit method, then you have to do a really monster burn for orbital insertion.  I don't know what the numbers are, but I imagine the dV simply is not feasible.  (I'm guessing you would have to use about the same amount of energy as would be provided by the potential energy of a proper transfer orbit.  Math dudes, what say you?)

And here's the real middle finger from the universe: the speed at which you travel along that transfer orbit is not limited to the speed of light.  It is limited to the speeds dictated by the mechanics of the orbit itself, which are going to be much much slower than the speed of light.

The orbital period of the sun is 225x10(6) years.  This does not bode well for interstellar travel. 

Is there a way around this?

[Dfthu]

Hmmm. I never thought of that. Maybe there is so much gravity it doesn't matter?

FYI: I know nothing about this stuff, just my idea.

[jwenting]

you've clearly never heard of a torch drive, or don't comprehend the concept.

An FTL drive is effectively a torch drive.

[Gaarst]

The speed of light is so huge compared to galactic orbital speeds (230km/s for the Sun) and the distance to Alpha Centauri is so small compared to the distance to the galactic core that you don't really care about orbiting anything.

You're not going to use a Hohmann transfer orbit for anything past Jupiter. You'll just go as fast as you can (using propulsion or gravity assists) to get there; it is the same for interstellar travel.

Edited January 23, 2016 by Gaarst

[Shpaget]

If you can achieve near light speed, then you just don't worry about using the most efficient transfer, you already have all the dV required.

Hohmann transfer is not mandatory. In the case of interstellar travel, the time it takes to use Hohmann would be prohibitive. Launch windows happen very rarely. Since its birth, the Sun has made only 18 or so rounds around the galactic core.

[Frybert]

21 minutes ago, mjl1966 said:

We *must* use a transfer orbit.

I'm going to stop you right there. We don't HAVE to use a transfer orbit. If we wanted to get there really quick we could burn half the way there, flip around, and slow down the other half and get there really quickly.  It would just use up a tremendous amount of fuel which is why we do the orbit method. It isn't required, but is for our purposes more convenient. 

[A35K]

The key thing is having something that has A LOT of dV, so not only can you reach ludicrous speeds, but also just burn until your trajectory is a straight line pointing straight at your destination. This can be done without stuff like Alcubierre drives, just by having really efficient engines. KSP Interstellar simulates this pretty well, if you have over around 40k dV you can burn in a straight line to pretty much anywhere in game.

[mjl1966]

39 minutes ago, Gaarst said:

The speed of light is so huge compared to galactic orbital speeds (230km/s for the Sun) and the distance to Alpha Centauri is so small compared to the distance to the galactic core that you don't really care about orbiting anything.

You're not going to use a Hohmann transfer orbit for anything past Jupiter. You'll just go as fast as you can (using propulsion or gravity assists) to get there; it is the same for interstellar travel.

So when we send probes to Saturn and beyond, it's just point and burn?  Is this how they do it or is it just theoretically feasible?  I'm very curious about when it becomes practical to abandon a transfer orbit because that solves a lot of problems.

[mjl1966]

27 minutes ago, Frybert said:

I'm going to stop you right there. We don't HAVE to use a transfer orbit. If we wanted to get there really quick we could burn half the way there, flip around, and slow down the other half and get there really quickly.  It would just use up a tremendous amount of fuel which is why we do the orbit method. It isn't required, but is for our purposes more convenient. 

Any travel between orbiting bodies is going to be orbital.  It doesn't matter if the intercept with another body requires a short piece of orbit or a long piece of orbit.  It's still an orbit.  However, if we are able to use an exceedingly small segment of an orbit around the galactic core, then it could be considered negligible.  That's where the meat lies.  Can we use a super small segment or do we have to use a longer segment?  More importantly, what is the difference in dV?  This is what determines the realistic feasibility.  I don't know the numbers, but I imagine if we could have point and burned to the Moon, we would have.  Why didn't we?  Why would it be any different from Sol to another star? 

Here's a different perspective on the same problem.  At the beginning of our trip to Alpha Centauri, or any star for that matter, we are in orbit around the galactic core.  Just because we leave the SOI of Sol does not change this fact.  Any maneuvers are changes in that orbit, just as maneuvers between Earth and Mars are changes in the initial orbit of the Earth around Sol.  That orbit is baggage you can't get rid of.  You can just shift the weight around a little bit.

So the question becomes, can we change the initial orbit sufficiently to make the trip to AC substantially shorter in time than it would be with a regular Hohmann transfer orbit?  What kind of dV are we talking about?  Is it feasible?

Edited January 23, 2016 by mjl1966

[Gaarst]

7 minutes ago, mjl1966 said:

So when we send probes to Saturn and beyond, it's just point and burn?  Is this how they do it or is it just theoretically feasible?  I'm very curious about when it becomes practical to abandon a transfer orbit because that solves a lot of problems.

For Saturn and beyond, we usually flyby Jupiter to get a gravity assist. But even then, to get to Jupiter in the first place you don't use a Hohmann transfer orbit.

For missions to Jupiter itself, we either swing by Mars, just burn in its general direction or use Hohmann transfers. That is because a Hohmann to Jupiter takes about 3 years to get there which is reasonable.

The flyby method is better because it spares a lot of delta-V and is a lot quicker for outer planets (10 years to Pluto instead of 60 years with a Hohmann transfer); but its downside is that you usually arrive at the target body with a lot of velocity and you can't get to orbit it unless you spend tremendous amounts of fuel circularising: you just swing by it and end up in a solar escape orbit.

You can take a look at real missions to the outer solar system and you will see that all of them used single of multiple gravity assits at Jupiter (or Mars sometimes). The Voyager probes basically flew by everything: Jupiter, Saturn, Uranus and Neptune.

Edited January 23, 2016 by Gaarst

[mjl1966]

49 minutes ago, jwenting said:

you've clearly never heard of a torch drive, or don't comprehend the concept.

An FTL drive is effectively a torch drive.

FTL not allowed.  No wormholes or Tesseracts either.  For the sake of argument, we're using high impulse, low thrust propulsion.  We can thrust half way if we want.  How much time will that actually save?

[problemecium]

Indeed. We don't need to use transfer windows, but with our current state of affairs it's preferable to other options. Our solar system is small enough that Earth orbits the Sun every year, and thus we get roughly one transfer window for every planet every year. If things were different and Earth took 500 years to orbit the Sun, we'd probably get impatient and just design (more expensive) spacecraft to go there on our schedule.
Also, the galaxy is much denser as far as the number of gravity wells go than our solar system is. A transfer trajectory that would move us from one star to another in half a galactic orbit would be grossly perturbed by all the stellar gravity wells nearby (or on the way, depending) and thus require near-constant correction thrust, to the degree where it wouldn't be any more practical. That, combined with the fact that on a galactic scale, going to Proxima Centauri is like going from Earth to the ISS (just distance-wise of course), means it's about the same trip no matter when we decide to go. And of course if things were different and our solar system orbited the galactic core every ten years for instance, things would be different and we'd look into using transfer windows.

[Frybert]

1 minute ago, mjl1966 said:

.  I don't know the numbers, but I imagine if we could have point and burned to the Moon, we would have.  Why didn't we?  Why would it be any different from Sol to another star? 

Except we can burn straight to the moon. We didn't because it uses up a lot more fuel. For an orbital transfer you need to change your speed only to get into orbit and then again to transfer, and again to put yourself in orbit at the destination. It takes longer, but we only have to accelerate to a lower velocity.

[Vanamonde]

One uses a hohman transfer to travel between bodies which will have moved a considerable proportion of their orbit during the flight time. But in the time it would take a spacecraft to travel between Sol and Alpha Centauri, the two stars will barely have moved, on galactic rotation time scales. Also, the distance between the two stars is so small in relation to their distance to the galactic center-of-mass that their orbital paths are nearly parallel to each other anyway. There's just no need to do anything fancier than fly straight there. 

[Frybert]

You can actually simulate this (though on a much smaller scale) in KSP. Put two objects into a medium to high orbit over kerbin about 1 km apart. Now get from one to the other.

Edited January 23, 2016 by Frybert

[fredinno]

2 hours ago, Gaarst said:

For missions to Jupiter itself, we either swing by Mars, just burn in its general direction or use Hohmann transfers. That is because a Hohmann to Jupiter takes about 3 years to get there which is reasonable.

No, you use Earth and Venus Gravity Assists. or a direct trajectory, which saves a lot of time.

[Shpaget]

To answer the question you are repeating and not getting the answer... It depends on how long you want the trip to take.
For the purposes of this post, disengage brain activity related to relativistic effects.

If your goal is to get to Alpha Centauri in 10 or so years, your average speed should be around 0,5 c, so to accelerate to it, and then come to a stop once you reach AC, you need 2 x 0,5 = 1 c of dV.

If you can accelerate at a comfortable 1 G, your acceleration and deceleration burns will take (relativistic effects put aside) about half a year to reach the speed of 0,5 c and again that much to stop, so they won't affect the travel time a lot.

If you can manage 20 years in space, your dV requirements drop by half.

[YNM]

The fact that Alpha Centauri pretty much stays there from our sky in the tens of thousand years range implies something very different - you can actually just propel straight there (or if possible, propel to the supposed intersect / arrival site). Mars, on the other hand, moves quite fast compared to near-stationary distant star - going all around in ~2 years - so you need a proper transfer orbit. An analogue between Sun and Alpha Centauri is quite like Earth to some similar-orbit NEOs (say, Cruithne) - you don't hohman transfer to it, you just propel yourself there.

[K^2]

Oh, you could totally take the transfer orbit around galactic center. It'd take over 100 million years, but you could do it with modern rockets. Fact that no human will ever learn if mission was a success puts a damper on things, though.

[Shpaget]

@K^2, you believe that there will be no humans in 100 million years? As in an extinction, or that our descendants will not call themselves humans any more?

[Bill Phil]

But you could just accelerate to a few hundred kps in the solar system to get to planets, and 50-70 % C to go to nearby stars.'

And if you could build a Heinlein style Torchship, it would hardly matter.

[magnemoe]

12 hours ago, jwenting said:

you've clearly never heard of a torch drive, or don't comprehend the concept.

An FTL drive is effectively a torch drive.

No an faster than light drive cheats using warp or other methods to arrive faster than light. 
An torch drive is an drive who can keep accelerating for an long time, dV is almost infinite. 

And many plans for an manned Mars mission is to not use an lowest cost transfer orbit but go faster. Benefit is less radiation and less time for something to go wrong during the trip.
Manned missions to places outside Mars, main asteroid belt or the outer planets can not use an minimum cost transfer, 
 

[KSK]

I always understood the Hohmann transfer to be a low energy but long time solution for getting somewhere else in space. If we're seriously contemplating a mission to Alpha Centauri, then we want a short time solution and are probably less concerned about low energy.

[K^2]

1 hour ago, Shpaget said:

@K^2, you believe that there will be no humans in 100 million years? As in an extinction, or that our descendants will not call themselves humans any more?

Given that humans have been humans for less than a million years, and 100 million years ago we were tree-climbing rodents, even if we don't go extinct, expecting anything like humanity to exist 100 million years from now is silly.

[Shpaget]

There are living fossils such as Queensland lungfish, horseshoe crabs, nautilus, sharks... which haven't changed much during the last 100 or more million years.

While the evolution of humans will most likely go on and we can't predict what the descendants of humans will look like in a million or more years, it is possible that they will retain fundamental features we observe today. Presently, there is a significant lack of predators that concern humans, so natural selection from that perspective is not happening. I'm not saying that humans have stopped evolving, but the main mechanism of evolution - the ability to procreate is practically universal among the population.