Hacker News new | past | comments | ask | show | jobs | submit login
Scientists to unveil new Earth-like planet (phys.org)
65 points by vinnyglennon on Aug 13, 2016 | hide | past | web | favorite | 60 comments

Project Longshot estimates 100 years to reach Alpha Centauri B. And that's assuming we launch tomorrow. Add 4 years to receive the telemmetry.


Starshot estimates 20-30 years for a flyby probe (or, more likely, a swarm of them)... starting a few decades from now: https://en.wikipedia.org/wiki/Breakthrough_Starshot

That's ~1000 times the velocity of our fastest probe so far when you exclude gravity assist. Stopping at the other end doubles travel time.

> Stopping at the other end doubles travel time.

Wow, I knew the physics but never realized the implication: where on earth hitting the brakes is easy and costs next to zero energy, in space breaking takes just as much energy as accelerating.

If you like science fiction, I recommend reading The Algebraist by Iain M Banks - this mechanic features frequently in the story and is quite a core limitation in the galactic civilisation he describes, which results in some interesting outcomes.

Sounds cool, I'll check it out. Thanks for the recommendation!

I know it's a game, but Kerbal Space Program really helped hammer home a lot of the things I guess I understood in the abstract. Delta-V is Delta-V. Every change in velocity (ignoring external influences) is cost out of the same budget. Braking is just accelerating in the opposite direction.

Before I played KSP I had a lot of hilarious misconceptions about the physics of rocketry and space travel.

On my first attempt to get into orbit, I loaded up my rocket with as many engines and boosters as I could strap on and fired the rocket straight up. The rocket breached the atmosphere but to my surprise the rocket came straight back down. I literally thought that launching rockets into orbit meant shooting them really high until they were "in space" haha. I then quickly learned that being in orbit means having a horizontal momentum that is balanced with the gravitational pull of a given body. This balance produces a curved elliptical trajectory.

And you have to decelerate to catch up to the station in orbit. If you accelerate you will fall back.

That's because if you fire your rockets to accelerate you end up in a higher orbit, i.e. your circle is much bigger, and vice versa, to overtake/catch up you want to fall to a lower orbit. That's also not intuitive to someone with "earth-experience".

This sounds like I'll have to start playing KSP after all. Lots of interesting space facts!

Yet objects in higher orbits like the moon are going way slower around the earth than low earth orbits.

What's the deal with that?

Higher circular orbits are indeed slower than lower ones. The shuttle goes around the earth in 90 minutes, the moon takes 28 days. Geosynchronous orbit, where one orbit takes one day, is in between.

What's confusing about transfers is that elliptical orbits are orbits where your velocity varies based on where you are in the orbit. Your speed is highest at perigee and lowest at apogee. In fact, your speed at perigee will be higher than the speed of an object in a circular orbit at that altitude. The reverse is true at apogee; your speed will be slower than an object in a circular orbit at the same altitude.

So if you and your docking target are both in circular orbits, and you increase your speed in order to catch up, then you will simultaneously be raising the other side of your orbit, half a turn away. The difference at your current location will be minute, but will grow every continuously over that half a turn. If you're docking in the next few seconds it won't be a problem (the docking mechanism will handle a small offset, and it equalizes your speeds provided the difference isn't too much). If you're still several minutes away, however, you might find that the extra radial velocity means that you fly over the top of your target.

If you miss and continue to follow your new orbit you will indeed find that your overall time to orbit is now higher than your target's. It may be many orbits before you again have a close approach with your target, although if the difference in speeds is small then it may appear from your perspective that you are in fact orbiting your target.

This is exactly the point grandparent was attempting to explain.

The moon is going much, much faster than satellites in LEO. But it has so much farther a trip, it takes orders of magnitude longer to complete an orbit.

If you're behind a satellite in orbit and want to catch up, counterintuitively you have to decelerate. This slows you, but puts you in a smaller orbit. The effect of the smaller orbit greatly outweighs any loss in absolute velocity, so you catch up to your target.

Without any assessment of the math, I'll point out that being in a higher orbit means two things:

(1) Your speed is higher; you cover the same distance in less time.

(2) The distance around the earth is larger than it would be if you were lower down.

That's very interesting, but I don't completely understand.

> I literally thought that launching rockets into orbit meant shooting them really high until they were "in space"

It makes sense that having a high horizontal speed keeps you in orbit if orbit is your goal, but with enough firepower you should be able to make it to a point where you don't fall backwards anymore, right? Or at least where falling is negligible enough that you can stay in that spot for a while.

That is possible, yes. Just rarely useful - it all depends where you're trying to go. If you head straight up with enough velocity to leave Earth's sphere of influence, you end up in orbit around the Sun, not the Earth. And in that case, you inherit Earth's velocity relative to the Sun, so you don't just fall straight back down into the Sun. (because 'head straight up' was relative to the Earth, not the Sun).

From the point of view of the game, several of my little green men can attest that coming home from a solar orbit is rather difficult (energy-intensive, time-intensive, or both)

If the plan is to end up in Earth's orbit, you don't want to shoot so far as to leave our sphere of influence. Earth's gravity drawing you towards it is a key component of orbit. But the other component is having enough horizontal (relative to the surface) to keep missing it - and that's what was missing from the parent's "what goes up, must come down" flights.

Yes, you will eventually reach an altitude where the gravitational pull of a given body is negligible. At that point you will have escaped it's orbit.

Would it be possible to drop off a much lighter package at the destination and use a solar sail to slow it down? We don't need to slow down a big interstellar engine when we get there; let it keep falling into space, while the instrument package slows down.

It could even have a smaller engine of its own; it'll take a lot less time to slow down something that's 10-1% the weight of the entire package.

A solar sail can add or subtract around ~60km/second which seems fast, but this is looking at 13411km/sec so not really that useful.

That doesn't seem right; that would mean it would take about four minutes to slow to zero.

Speed is distance / time, acceleration and deceleration is distance / time ^2. If you accelerate at say 9.8 m/s^2 for 1 second that's 9.8 m/s/s * 1s = 9.8m/s aka if you fall for 1 second your going 22 miles per hour and still being pulled down at 9.8m/s/s.

Using a star a solar sail can change it's speed by ~60km/sec by accelerating much much slower than that for months. The problem is the further from a star get the lower your acceleration. To maximize final speed you basically want to get as close to the star as you can in a huge elliptical orbit and then on your final pass just keep accelerating. Or do the reverse on the way back.

Is that clear?

I don't understand the units here. Surely a solar sail causes acceleration or deceleration, so its contribution would be in m/s^2, not m/s.

60km/s not the acceleration. It's the total contribution to delta-V over a long period of time.

Why is there a total amount of change in velocity possible from solar sails? Shouldn't it depend on the mass of the thing being accelerated, amount of time under acceleration, etc?

The further from the sun you get the less acceleration you get from the sun. So, 4x the acceleration = 2x the speed because your accelerating for 1/2 the time and 1,000,000x acceleration = 1,000x the speed.

Further, the input power is fixed from the sun so the absolute best you could do would be 100% sail that's a single atom thick. However, that sail would be quickly destroyed by the sun and other things. Make an ultra thin sail with a tiny little cargo Say 98% weight of the sail and 2% cargo and dropping the cargo below that does not help much. You end up getting around 60km / second which is really really fast 1km/s = fast sniper bullet. But, we are comparing it to something going 4% of light speed which is insanely fast.

Would it double? Couldn't they use the gravity of Alpha Centauri to pull it into orbit?

Not out to five decimal places, at those speeds you would actually run into a fair amount of stuff in the near vaccume of space and you lose some delta V escaping the Suns gravity well etc. But relative to all the unknowns it's about a 200 year trip vs a 100 year trip. If everything else worked about as well as expected.

Another consideration is if rocket stops working after 10 years on a fly by you still eventually fly by the general area, but if you want to stop well that's just not happening.

These guys want to reach Alpha Centauri in 20 years


Well, according to the wiki article you cited, the 100 years figure is assuming a 3 order of magnitude improvement in propulsion technology. We've barely advanced since the project was conceived, so I'd say it'd probably be closer to 100,000 years instead.

It took 9.5 years for New Horizons to reach Pluto (just 32 astronomical units distant). At that rate, it would take a probe more than 80,000 years to cross the 271,000 AUs separating Earth from this exoplanet.

That presumes a constant velocity. Any extra-solar ship would certainly need a different acceleration profile. Ion engines come to mind: low acceleration, but constant for a very long time. If you're going a far distance, they're ideal.

While you have a valid point, here's what it works out for ion engines.

Assume you want 1,000 years to Alpha Centauri and your ship boosts the entire way. (It'll take a few pictures as it flies through the system.)

  s = 1/2 a t^2
  a = 4.22 light years * 2 / ((1000 years)*(1000 years)) = 8 E-5 m/s/s
Sounds great, right? But the final speed is:

  v = a * t = 4.22 light years * 2 / (1000 years) = 2.5E6 m/s = 0.8 c (!)
Let's say your engine has a ISP of 8000 s, which is about 10x better than anything we have.

Then from the rocket equation you'll need a dry weight (payload, engines, etc.) to propellant ratio of

  e^(2500000 / (9.8 * 8000) = 70,579,094,313,732 = 71 quadrillion
  eg, 1 gram of spacecraft requires 70 kilotons of propellant
But wait, there's more! It takes power to push 70 kilotons of propellant backwards at 80,000 m/s.

  E = 1/2 * m * (v^2) = 2E20 Joule
  Since E = m c^2
   => total annihilation of over a ton of antimatter
Almost certainly more as that assumes 100% efficiency. And of course that extra ton isn't propellant, so it hits you in the rocket equation.

Note: I am not a rocket scientist and likely made mistakes. Even if I'm off by a few zeros, it's enough to show that an ion engine isn't going to get something to Alpha Centari within a few centuries.

Oops! There's a couple of missing zeros. The 2.5E6 m/s = 0.8 c should be 0.8% c. That mistake doesn't continue to the rest of the calculation.

Unfortunately, not ideal enough. But there are other options: https://www.youtube.com/watch?v=EzZGPCyrpSU

I don't think there are electronic or software components today that can withstand interstellar space travel for thousands of years, even at the distance of proxima centauri.

Distance is irrelevant when it comes to space travel. Unless an object is within the vicinity of the Oort cloud, we stand very little chance of actually leaving this solar system in a controlled fashion.

Well, the key is to not take thousands of years about it. Honestly, I doubt we will ever have thousand year missions, unless our lifespans commensurately grow, as humans don't have a great track record when it comes to long term projects.

There are ways, but they are expensive and not without risk. Nuclear pulse propulsion (orion) can yield vast acceleration, and fusion rockets, while unproven, are feasible. You then have less exotic options such as laser driven light-sails and nuclear powered ion drives.

If we were prepared to throw half a trillion bucks at the problem we could see a flyby in our lifetimes - but without a damn good reason, it won't happen.

I wonder if the James Webb Space Telescope will be able to image it since it's so close.

I don't think JWST could even resolve it from its parent star, judging by [0]. They're too close together.

Proxima Centauri has a luminosity of 0.0017 suns [1], so Earth-like conditions would occur at ~sqrt(0.0017) au = 0.04 au, an apparent angular separation of arcsin(0.04 au / 4.2 light years) = 0.03". JWST wants star-planet separations at least an order of magnitude larger [0], ideally with very hot, infrared-bright planets.

E-ELT should be able to resolve them, according to [2]. Page 7 gives values for a very similar scenario: an Earth-like planet around an M dwarf (like Proxima), 6 pc away (5 times farther), at an angular separation of 0.015" (half as wide). In this scenario, E-ELT could image the star and planet as separate points, and take useful spectroscopic measurements of the planet's light. For it example it could detect a spectral line of oxygen (O2) in 4 hours of exposure time.

[0] http://nexsci.caltech.edu/workshop/2016/NIRCam_Planets_and_B...

[1] https://en.wikipedia.org/wiki/Proxima_Centauri

[2] https://www.eso.org/sci/meetings/2014/exoelt2014/presentatio...

Can anyone "do the maths" on this. I don't feel well enough equipped, but I'd love to know the feasibility of this, or what we would be able to image if anything.

A planet the size of the earth at the distance of proxima centauri is about 6.582⋅10^-5 arcseconds wide (a 20'000th of an arcsecond)

angle(in degrees) = size / distance / pi ⋅ 180°, take times 3600 for arc seconds


I didn't find a easy round number for James Webb, but Hubble's angular resolution is 1/10th of an arcsecond, at least a factor 2000 too low to even resolve the planet's disk as one pixel.


http://jwst.nasa.gov/faq_solarsystem.html#angularresolution says:

> The specification is that the telescope is diffraction limited at 2 μm, which means a Strehl ratio of 0.8 and a wavefront error of 150 nm rms. With a 6.5 m telescope, 1.22 λ/D = 0.077 arcsec at 2 μm. The smallest pixels (NIRCam 0.6-2.5 μm) are just 0.034 arcsec. But a lot of the wavefront error is due to imperfect alignment of the parts, and it's possible to do better for a small part of the field of view.

A telescope can still detect something smaller than its resolution. We can see stars with our eyes, after all. Even astronauts in space, where there is no atmosphere to move the light around. Here are some of the best attempts at direct imaging with a telescope: http://www.planetary.org/explore/space-topics/exoplanets/dir...

The problem is the planet is dim and close to a bright star. If this planet is 1 AU out, then there's a 0.77 arcsec between it and the the star, which would make it easier to distinguish the two.

Quantum telescopes should get us beyond the diffraction limit


"Webb's angular resolution, or sharpness of vision, will be the same as Hubble's, but in the near infrared. This means that Webb images will appear just as sharp as Hubble's do. Webb will have an angular resolution of somewhat better than 0.1 arc-seconds at a wavelength of 2 micrometers (one degree = 60 arc-minutes = 3600 arc-seconds). Seeing at a resolution of 0.1 arc-second means that Webb could see details the size of a US penny at a distance of about 24 miles (40 km), or a regulation soccer ball at a distance of 340 miles (550 km)."


Moving matter around takes far too long for star travel. We need to figure out how to assemble machines from long distances using light. (Something like https://en.wikipedia.org/wiki/Optical_tweezers)

Why not just launch a seed factory and send the instructions to build via light?

That seems a hell of a lot more practical and cost effective

Also more possible haha. It's all fun and games until you predicate the success of your project on trying to collimate a laser beam from here to alpha centauri, so that it can manually assemble a (dielectric!) robot one piconewton at a time.

That sounds pretty impossible assuming that we're aiming for a planet with an atmosphere. Maybe if the planet has a moon, and then all we have to do is remote-construct a machine that can both do our research and make the small hop down to the surface?

This is extremely exciting! The million dollar question is of course "how Earth-like?". If the answer is "very" we might actually have found a viable place for an off-world colony!

Getting a probe there in a reasonable time is entirely feasible if can figure out how to keep the darned thing accelerating without stupendous fuel requirements. I'm confident we'll figure that out sooner or later.

Keep dreaming. If we cannot even understand or live in accordance with the life support systems of what is literally the most habitable and human-friendly planet in the Universe then there's no way in hell that we will be able to successfully colonize a lifeless planet.

Stop thinking about outer space when we are in the middle of an anthropogenic mass extinction here on planet Earth.

> literally the most habitable and human-friendly planet in the Universe

Well, we don't know that it's the most habitable; it's just the most habitable we know of. And huge sections of it are completely unsuited for human life. 2/3rds of it will kill a human being in a matter of hours. Anything above or below, let's call it 50 degrees latitude is inhospitable without technology, and the equator isn't a great place to live, either.

It almost certainly is the most habitable planet for life as we know it today because we and our environment co-evolved to match the conditions. Just because 2/3 of the planet (not sure how you arrived at the number / what parts are in the denominator) doesn't meant it is meaningless to what allows us to live here (e.g., oxygen and food supply from the oceans).

OTOH I disagree with the notion that we have to have a full grasp on how to sustain us on this planet before we can make other planets inhabitable for us. I think there are things we will only learn if we try. I am not saying that trying right now is feasible, though.

I agree that we have an extremely low probability of finding a planet that could host our biology as well as Earth's given that we evolved here, but I refute your absolute claim.

Invasive species don't co-evolve to exploit the ecosystems they stumble upon. They happen upon a ecological space where there are incredibly low hanging fruit that they are better equipped at taking advantage of. Imagine space-faring humans stumbling upon a world where the oceans resemble that of the Earth's as they once existed in the 1500's--a world densely populated with an abundance of marine life, ripe for harvesting.

While there is some probability that biochemistry is not 100% universal (ie. alien metabolites could be toxic to humans or cause an immune response), I imagine there are scenarios where the same set of molecules and polymers are used. The principles of biochemistry are going to be similar in a lot of places in terms of how energy is exploited and harnessed.

I think it's incredibly more likely that humans will never be interstellar travelers en masse than to say that Earth is the optimal naturally occurring human habitat in the universe. As far as we know there is an infinite space full of limitless planets. Plenty of biological solution space with which to theoretically work / exploit.

I'm going to assume this is a serious comment and that you are attributing global warming as the cause of the feared extinction.

In that case I think you should dial it back a bit. Humans are pretty adaptable. Even assuming the changes in temperature and sea-level are on the high end of predictions, we will adapt. People will move, infrastructure will be built, and so on. Mass extinction certainly isn't going to happen from global warming as we understand it today.

You might argue there there will be mass casualties, but that is also uncertain, difficult to quantify, and is quite a different thing than 'mass extinctions'. Of course, mass casualties due to weather has and is already a problem -- it isn't something that is uniquely associated with global warming.

> Mass extinction certainly isn't going to happen from global warming as we understand it today.

I assume the parent is referring to the https://en.wikipedia.org/wiki/Holocene_extinction which is certainly happening, and has been for thousands of years.

Climate change is a rather recent contributor, albeit a very serious one of which we've only glimpsed the beginning.

This opinion really bothers me. The absolute best way to maximize chances of humanity's survival is expand off this rock. Improving survival chance here is of little consequence compared to essentially ORing the likelihood of survival over the independent existential threats.

I vaguely recall a quote saying that if an Earth-like planet was ever found around Proxima Centaury, an international project to build an interstellar probe would begin immediately.

Well, we'll see.

Can't wait for the "small furry creatures from Alpha Centauri" memes.

It's probably Trisolaris

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact