New question: why does it need to sustain acceleration? Can we just build a gigantic launcher to shoot it off at a million miles an hour, like a projectile? I mean it’s not coming back, anyway…
Two difficulties: one is that the velocity achievable by your launcher is going to be limited by the acceleration that can be tolerated by the vehicle and launcher, and the other is slowing down at the other end.
The first is more of a matter of economics and practicality, at least if we assume that you have something more capable than chemical propellants. (If you made just a really, really long barreled low friction gun, the projectile would eventually outrun the expanding propellant, and thus could only achieve the propellant expansion velocity.) If you could make an electromagnetic linear accelerator to any length you choose, you could continuously accelerate the vessel to the limit of its structural capability and your power supply. And the more force you apply, the shorter it need be for the same resultant velocity. Of course, you’ll have to account for the reaction forces of the magnets on the railgun structure, so there are some practical limits to how high the acceleration can be in addition to not turning your payload or passengers into jelly. Some wiseacre is going to come along now and suggest that we can just rap the accelerator around a body like the Moon and run the projectile around in a loop continually accelerating it until it achieves the desired speed. This is true in concept (and how modern high power particle accelerators work), but the fact is that you have a resulting radial acceleration component that now becomes massive at high rotational speeds, which will both crush your payload and tear your accelerator up from its moorings. Accelerating a payload of useful mass in such a way would require structural strength way exceeding any real material. Here is an old thread discussing the topic.
The other problem is decelerating at the far end; you are going to have to include sufficient fuel and a propulsion system, unless you just plan to zip through with a handful of blurry photographs as you zip by planets at literally astronomical speeds. This is going to take a lot of fuel, though again, if you can institute some kind of eletromagnetic drag you can reduce the requirements in exchange for lengthening the transit interval. Ultimately, though, this isn’t plausibly with extant or near-future technology.
Stranger
I think you answer your own question:
One way of adding speed, or build up velocity, is to use sustained acceleration. If you keep pressing down on the accelerator, you will increase speed continuously. To use your question as an example: It would be the same as using your gigantic launcher, continuously pushing to boost an object from zero to million miles an hour velocity. This “pushing” is sustained acceleration.
Given that we’d know very little about what the probe would find there, I’d think we’d need more than that - we’d need an AI capability that is nowhere close to existing yet.
And you are correct that primitive != reliable. I’ve got numbers, which I can’t share, alas.
Absolutely. The probe would need to be capable of examining an initial overview of the system, categorizing worthwhile targets for exploration, evalutaing the scientific value versus risk of damage or failure, maximizing mission capability in terms of fuel and lifespan, troubleshooting and self-repair, and preprocessing all the the raw information into something refined enough that it is worth transmitting in the narrow datastream back to Earth; in short, the jobs done by teams of dozens of highly trained and experienced scientists, engineers, and mission controllers, and without any external input or review. Just this part of the problem, separate from the difficulties of getting there and transmitting information back, is a daunting technical challenge that exceeds our current ability and experience. It is all we can do to get probes onto the surface of Mars, or in orbit of Jupiter and Saturn to explore those systems, and we can get feedback and provide correction to those vehicles with only a modest delay.
Stranger
The Economist Magazine once discussed ultra-light spacecraft. They would be tethered to solar sails and powered by lasers cannons which would remain in this solar system. Thus, the need for propellant is minimized.
This webpage discusses Starwisp, which uses a 1 kilometer sail weighing 16 grams, powering a 4 gram spacecraft. The whole thing weighs less than an ounce, and is pushed by beamed microwaves.
According to the article, “All it really takes is the desire and the commitment to a few decades of hard space engineering work and our first interstellar probe could be heading to the stars within our lifetimes.” Though I’d take that opinion with a heap of salt, I’ll note that the paper in question was presented to the AAAS in 1986. [1]
[1] AAAS: American Association for the Advancement of Science? Or American Academy of Arts and Sciences?
Out of curiosity, how long would this signal take to get home? Is it beamed at the speed of light so any information is only 5 years old by the time we get it, or could this data be very well obsolete by the time we even receive it?
Well, it seems unlikely that data on e.g. the number of planets, their atmospheric and surface conditions etc. would become obsolete within any reasonable timespan. Also bear in mind that the probe will have spent several lifetimes getting there. Of all the challenges involved in interstellar probes, having to wait a few years while the data crawls home is pretty much bottom of the list.
I don’t think we have AI good enough to make fully automated anal probes.
Give back as good as we get!
Yes, any radio, any light, any EM signal at all will travel at the speed of light in free space.
We might be able to build a probe that could visit (ominous pause)…the Nemesis Star! (dah dah daaaaah!)
I’m not sure that this is correct: we would already know the habitable zone before launching the probe, plus we can detect the presence of some major planets by the wobble they cause in the stars they orbit. Equally, finding planets within a star system is relatively trivial: take a photograph when pointing at the star and take another a day or two later. If something has moved, it’s not a star. Perhaps, though this might be something for a path-finder probe, with other probes following it by a year or so, so they can adjust trajectories in time.
Can I also ask for expansion on one topic? My though is that for the bulk of the journey, the probe would be off - completely unpowered - and then at the appointed time something would switch it on. This way there would be no issue of failure through use. Do we have that that technology? Some sort of switch operating through radioactive decay, perhaps? Or perhaps a solar cell - once it gets sufficiently near the target star, it would generate sufficient energy to switch on the rest of the probe.
A “4 gram spacecraft” is nothing more that a message in a bottle; there is no way that you could build a useful payload of that mass with anything like conventional technology, much less protect it from ablation, impact, and radiation. Now, a payload that is critically self-replicating and self-organizing, i.e a Von Newmann “Universal Assembler” or Bracewell probe might a possibility, albeit one requiring the capability of autonomous molecular assembly that is way into the realm of science fiction-y “grey goo” at this time. (See Greg Bear’s Queen of Angels or Neal Stephenson’s Diamond Age for examples.) And the amount of power required for the Starwisp is nothing short of incredible; given the difficulty we have in even keeping a space station powered and supplied in Low Earth Orbit, I’d say that maintaining a multi-gigawattt orbiting laser or maser is well beyond current capability.
Having spoken with a couple of mission planners and a number of engineers at NASA Jet Propulsion Laboratory regarding a variety of planetary exploration missions, I have at least a small bit of insight as to what it takes to successfully execute a program of this type. It is one thing to lay out a mission plan, but the plan usually evolves significantly even after the mission is launched and on its way, based upon operational difficulties with the equipment, new information and insights about the target, et cetera. For instance, the Voyager 2 spacecraft, like its sister, was originally intended only to visit Jupiter and Saturn in a regrettably abbreviated version of the original Grand Tour mission to visit all outer planets; however, by the time Voyager was approaching Jupiter, the mission operators had enough experience from Voyager 1, and faith in the beyond-design-limits reliability of the on-board systems, to plan a gravity assist around Saturn to take them to Uranus and Neptune, completing nearly all of the Grand Tour objectives for chump change. (I suspect they held the possibility of doing this as pocket kings until the rousing success of Voyager 1 gave them a full house in aces.) However, this took the extensive judgment of experience programmers, engineers, orbital ballisticians, and planetary science Ph.Ds to make the most of the tiny window of opportunity to thread the needle, and this is using information about the movements of the planets that we can determine from Earth observations down to a few kilometers. We may have, relatively soon, software capable of making those kinds of trades on a limited open-ended basis, but we’re not there yet, and we certainly don’t have the kind of synthetic cognition capable of making the complicated judgements of what mission is scientifically most worthwhile for the risks it poses, which is why we pay Ph.Ds the [del]big[/del]reasonably-good money.
You would not want a probe to be completely unpowered; for one, if may need to make course corrections, and the further away it does this the less energy it takes. It would also be nice to have some kind of telemetry feedback from it so you don’t waste a couple hundred years of effort on a probe that got smashed up in a random collision (however unlikely) passing through the Oort cloud. The energy to do this would be prohibitive given current power sources–the radioisotope thermoelectric generators used for interplanetary missions with durations of less than a decade would be wholly inadequate for interstellar transit–and so you’d be stuck not really knowing whether it was working or not. Of course you would want to keep many systems powered down, both to conserve energy and reduce wear, but the fact is building any complex and delicate electrical or mechanical system intended to work after a century or more without maintenance is way beyond our experience.
If you did have some kind of feasible Starwisp system as discussed above, the smart thing to do would be to build thousands of them (the energy requirement is the same and at the light weight you could boost hundreds of them per rocket launch) and send them out knowing that at least a few would survive. But again, we’re going on about virtually indistinguishable-from-magic technology compared to extant capabilities. Something like this will probably be the way that we explore other star systems–by proxy, rather than running around in red spandex with phasers set to ‘stun’–but it’s not going to be for a long time yet.
Stranger
You can’t do interstellar travel if you’ve got to pack your own propellant, for all the reasons Stranger mentioned. We’d need a breakthrough, such as using the Casimir force or some other property of space itself to propel our ship.
But we’re much better off building gigantic telescopes to explore other star systems - and that’s what we’ll be doing for the next several hundred years at least.
I was thinking that being powered down would be a necessary trade-off. Correct me if I’m mistaken, but not even an active nuclear reactor will last several hundred years? But would a powered down reactor? After several hundred years, could you remove the moderating elements and generate electricity?
Well, for a conventional reactor you’d actually want to add the moderator to increase neutron capture. (The moderator slows fast neutrons down to “thermal” speeds where it can more readily be captured by fuel elements, making an otherwise subcritical pile achieve criticality, i.e. self-sustaining neutron production.)
It would require some significant engineering to develop fuel elements that would be reliable after a century or more in storage without decay and problems arising therefrom (i.e. internal cracking and material phase transitions that can change the neutron-absorbing properties of actinides); not impossible, but modern fuel elements are designed to be used within a few years of manufacture (although they may be in the reactor for a couple of decades, depending on the design). A modular reactor, like the PBMR might be well suited to this application, allowing you to run at low power and then easily scale up, and has relatively few moving parts.
A better option yet might be to use a subcritical reactor with fissionable rather than fissile fuel, so natural decay isn’t a major issue. However, commercial power producing subcritical reactors are currently nonexistent and would require fine control, so that’s still somewhat beyond existing technology.
Another problem with a nuclear reactor (or any other conventional thermal energy source) is radiating away waste heat. A nuclear reactor produces a large amount of unrecoverable heat-energy. (The large beaker-shaped structures you see adjacent to many nuclear power plants are cooling towers, which are necessary to pre-cool water coming out of the outer loop before dumping it back into the environment.) Since your only effective way of getting rid of accumulated heat in vacuum would be to radiate it away (evaporative cooling, while effective, would mean that you’d have to carry extra coolant that would be a consumable) is to radiate it away. For any reasonable size of plant this would mean a big radiator in addition to other structure.
Stranger
What about a really long length of wire?
The wiki page on Starwisp provides a payload of 80 grams, not 4 grams as my other webpage asserted, apparently based upon more recent work by Landis (2000).
Landis (1995) discussed a similar concept with an 8 kg payload. He speculated that the sail itself could serve as a sort of mirror, with the payload using adaptive optics for interpretative purposes. The sail might also provide a dish reflector for communications.
There’s also the matter of powering the space laser:
My conclusion: I see little to contradict Stranger’s POV.
Q: 80 grams is 2.8 ounces. Might that be a useful payload?
For golf? It’s a bit excessive; the regulation golf ball is 1.62 oz For an interstellar probe? Not so much.
Stranger
You’ll also have all sorts of high energy particles smashing into the submicron structures, and unless you test and reconfigure from time to time you might wake up with so many failures that the reliability systems won’t be able to get anything running, even if you don’t run into a comet.