@S1-A.ARPA,@MIT-MC:jim@TYCHO.ARPA (05/23/85)
From: jim@TYCHO.ARPA (James B. Houser) Assume for a moment that we were not in bugetary hard times and could afford an attack of conspicuous consumption. Could a reasonable interstellar one-way unmanned probe mission be designed which would give some legitimate science data in addition to the usual ego stuff. Because of the constraints it will be important keep things simple. The parameters? 1. Assume only near term technology and assets. For example, no space station. You are allowed two shuttle payloads if necessary - doing it in one gets bonus points. 2. Must be ready to launch within 5 years. 3. Assume a target star 20 light years away. 4. Intial data must be received from target star system within 100 years. An interesting question is what collection strategy to use. You could adopt a cometary orbit or try to look for planetary body etc. Minimum requirement is a fly-through at less then .05C mean. Remember that your communications lines have a 40 year turnaround. May be allowed a 10% extension on time limit given a high grade justification. 5. Maximum cost will be 1 billion in current dollars. It may be competing in Congress with a submarine base in Arizona so the cheaper the better. 6. Worship the KISS principle. This thing has to go a long way on it own. Occurred that this would be an interesting thought experiment. You can assume the target is in whatever direction you like if it will help. Prime issue may be the choice of propulsion scheme. Another question is what instruments to carry, preferably few and simple. Any thoughts???? PS "It can't be done" is not an acceptable answer!!! -------
henry@utzoo.UUCP (Henry Spencer) (05/26/85)
> PS "It can't be done" is not an acceptable answer!!! It is probably the only answer right now! The BIS Daedalus study concluded that a less-demanding mission (15% of c rather than 20%, with encounter taking place at full interstellar velocity rather than after deceleration to 5% of c) was possible given the following improvements on today's technology: 1. He3-D fusion engine. 2. Propellant collection from the atmosphere of Jupiter (there is no other good source of He3!) 3. Advanced self-maintaining computers with software that could plan and conduct the final encounter without any help from humans. The Daedalus probe weighed something like 40000 tons at launch, by the way. You won't fit that in one shuttle load! More specifically... > 1. Assume only near term technology and assets. For example, no > space station. You are allowed two shuttle payloads if necessary - doing > it in one gets bonus points. The major problem, as you point out elsewhere, is propulsion. With the current and foreseeable technology, this means mass... lots of it. The BIS study, assuming better-than-current technology, still had a total-to- payload mass ratio of about 100:1. Two shuttle payloads just will not hack it to get a substantial payload to the stars. Note that one shuttle payload is just sufficient to get the Galileo probe to Jupiter! > 2. Must be ready to launch within 5 years. NASA cannot launch *any* major mission that quickly. NASA knew how to launch a space station nearly 20 years ago; the Skylab technology is that old. The ETA for the space station was ten years, and is already slipping further into the future. Personally I think a space station could be launched almost at once if the project was headed by some nasty person who was authorized to bypass NASA bureaucracy and normal purchasing procedures, with a firm objective of getting hardware flying ASAP rather than maximum sophistication and gosh-wow factor, but one might as well wish for wings. > 3. Assume a target star 20 light years away. The BIS assumed Barnard's Star, 8 LY away, and found it hard. > 4. Intial data must be received from target star system within 100 > years. At 20 LY, this gives 20% of c. The BIS set this as their original goal, but found it so hard that they slipped the specs to 15%. More significant was their rationale for the original speed: getting results within 40 years provides continuity within a human lifetime, because people who were young staff members at launch time get to see the results come in. This is psychologically important to both the staff and the people who vote on the funding. > An interesting question is what collection strategy to use. You could > adopt a cometary orbit or try to look for planetary body etc. Minimum > requirement is a fly-through at less then .05C mean. The propulsion requirements are so fierce that even the BIS very quickly opted for an undecelerated flyby as the only feasible mission. > Remember that your > communications lines have a 40 year turnaround. This means that for all practical purposes the mission must be 100% automatic; there is no realistic prospect of ground control. We are not really up to that unless you are willing to be quite unfussy about the nature of the flyby. Don't forget that communication over distances of 20 LY is very hard. You will need plenty of power, which is a problem in itself. There is *no* self-contained power source now existing that will give useful output after 100 years. > 5. Maximum cost will be 1 billion in current dollars. It may be > competing in Congress with a submarine base in Arizona so the cheaper > the better. NASA probably cannot launch anything anywhere near so ambitious for this kind of money. The Viking mission cost over $1e9, as I recall, and that was a couple of orders of magnitude less fancy. (It was also paid for in early-70's dollars.) Again, I suspect serious cost reductions if radically different management approaches were adopted, but the vastly greater mission complexity more than cancels those gains. > 6. Worship the KISS principle. This thing has to go a long way on > it own. Probably too long a way. Self-repair is almost certainly needed, and that kills simplicity right off the bat. Remember that NASA cannot spend lots of money on a mission that has a good chance of failing; they can't bet unless it's nearly a sure thing, politics being what they are. If one were willing to accept a high chance of failure (i.e. launch several to have a reasonable chance of useful return), I suspect one could dispense with self-repair, given limited objectives for encounter observations. The odds of keeping a fusion rocket engine, or the equivalent, operational for that long without on-board maintenance would also seem minimal, so deceleration is out for yet another reason. Some of the stiffest life requirements in the world are those of the phone companies; they only specify a 40-year life, and this assumes human maintenance. Sorry to be so negative, but it really is a fiercely hard problem. The BIS was optimistic about our ability to solve it (without some of the constraints you impose) before too long. I am actually more optimistic than they were, because the propulsion assumptions they made are now sounding a bit conservative in some ways. For example, Robert Forward says that antimatter propulsion requires no serious breakthroughs and could be a cost-effective alternative to H2-O2 for in-space propulsion (note that lifting quantities of H2-O2 from the ground is expensive) very soon. But I don't think "5 years" and "2 shuttle payloads" are viable constraints right now. -- Henry Spencer @ U of Toronto Zoology {allegra,ihnp4,linus,decvax}!utzoo!henry
broehl@wateng.UUCP (Bernie Roehl) (05/29/85)
Hmm...
To get first data back in ~100 years, and given that 40 of those years
is the time it would take the data to make the trip back, you have 60 years
to get to a star 20 ly away. Thus, you must be travelling at .3c, average.
However, the speed in the target system must be reduced to 0.05c; this is
a non-trivial task!
Doing the whole thing by 1990 for less than $1B is, to say the least,
extremely challenging.
Personally, I don't think it's possible. However...
The most promising form of propulsion is probably the use of ground-based
lasers. You would use one shuttle flight to put the probe itself in orbit,
along with a *very* long-lasting nuclear power source and self-repairing
on-board systems (don't laugh; serious research has been done in this area).
The second shuttle flight would carry a modified Centaur stage and a *huge*
solar sail.
You mate the two in earth orbit, and use the Centaur to take the thing on
a trajectory past some massive body (e.g. Jupiter) to give it more of a kick
(or, more accurately, redirect its velocity vector to a more useful direction).
One possibility is to use Jupiter to cancel the craft's orbital velocity
relative to the Sun, and use the Sun's gravity for even more of a slingshot.
All this maneuvering would take a number of years, but the end result would
be a craft leaving the solar system with a reasonable velocity (still nowhere
near c of course, put respectable nevertheless).
By the time all of this is done, we may very well have powerful enough
ground-based (or space-based?) lasers that we can start using them to propel
the craft further. The craft would unfurl its sail, and we would focus the
laser on it, providing thrust. Since lasers produce a highly collimated
beam, there is no real 1/(r squared) problem.
In theory, you can eventually get the thing moving at a pretty good clip;
after all, you get a small but continuous acceleration for ~60 years.
(remember of course that a ground-based laser would not be able to "see" the
probe continuosly, but it wouldn't be that many years before space-based
lasers would supplant the ground-based one(s)).
The big problem is slowing the probe down when it arrives. The best way
is to have a laser on the destination end, but in this case that seems a
little unlikely. Other than that, no solution seems immediately obvious.
Of course, I would hope that in the 100 years it would take to get data
back from the probe, we will have reached the point where the probe itself
is little more than an historical curiousity. Personally, I want to be
there when the probe arrives...
--
-Bernie Roehl (University of Waterloo)
...decvax!watmath!wateng!broehlal@aurora.UUCP (Al Globus) (05/29/85)
> > Assume for a moment that we were not in bugetary hard times and could > afford an attack of conspicuous consumption. Could a reasonable interstellar > one-way unmanned probe mission be designed which would give some legitimate > science data in addition to the usual ego stuff. The British Interplanetary society has designed such a craft, unfortunately, I can't remember what the called it. > Because of the constraints > it will be important keep things simple. The parameters? > > 1. Assume only near term technology and assets. For example, no > space station. You are allowed two shuttle payloads if necessary - doing > it in one gets bonus points. > Might as well use the Space Station. It'll make the job easier and you won't get anything near ready to launch before its available. > 2. Must be ready to launch within 5 years. Not a prayer. Almost nothing gets launched within 5 years of conception, even very straightforward build-another-one-just-like-the-last one spacecraft. The only exception to this, I believe, are some communication satellites. > > 3. Assume a target star 20 light years away. > > 4. Intial data must be received from target star system within 100 > years. An interesting question is what collection strategy to use. You could > adopt a cometary orbit or try to look for planetary body etc. Minimum > requirement is a fly-through at less then .05C mean. Remember that your > communications lines have a 40 year turnaround. May be allowed a 10% > extension on time limit given a high grade justification. > > 5. Maximum cost will be 1 billion in current dollars. It may be > competing in Congress with a submarine base in Arizona so the cheaper > the better. No chance whatsoever. The shuttle launches will cost you $140 million alone, and most of your cost will be engineering salaries for design and construction at $60 - 100 an hour. > > 6. Worship the KISS principle. This thing has to go a long way on > it own. > Don't forget a lot of redundancy.
eugene@ames.UUCP (Eugene Miya) (05/30/85)
> > Assume for a moment that we were not in bugetary hard times and could > > afford an attack of conspicuous consumption. Could a reasonable > > interstellar > > one-way unmanned probe mission be designed which would give some legitimate Unmanned missions with the exception of Voyager have not caught the imaginations of men. Voyager did so only because of the timing of events such as Cosmos, KCET's telecast Voyager coverage, and the diversity of the photos returned by Voyager. You could probably assemble a set of questions like "what planets have we landed" which would measure the impact of these missions and find lots of confusion. Another problem with earlier missions was that many photos returned like Mariner Mercury, early Mars pictures, and so on had cratered surfaces looking like the moon. This is hard to excite the brain. "If you've seen one moon-like planet, you've seen them all?" > > 2. Must be ready to launch within 5 years. > > Not a prayer. Almost nothing gets launched within 5 years of conception, > even very > straightforward build-another-one-just-like-the-last one spacecraft. The > only exception to this, I believe, are some communication satellites. I personally think this is a sad bit of NASA. It would be nice to be able to respond to short lived phenomena such as cometary encounters. (hint, hint) > > 3. Assume a target star 20 light years away. > > > > 4. Intial data must be received from target star system within 100 > > years. An interesting question is what collection strategy to use. You > > could > > adopt a cometary orbit or try to look for planetary body etc. Minimum > > requirement is a fly-through at less then .05C mean. Remember that your > > communications lines have a 40 year turnaround. May be allowed a 10% > > extension on time limit given a high grade justification. > > > > 5. Maximum cost will be 1 billion in current dollars. It may be > > competing in Congress with a submarine base in Arizona so the cheaper > > the better. > > No chance whatsoever. The shuttle launches will cost you $140 million alone, > and most of your cost will be engineering salaries for design and > construction at $60 - 100 an hour. Again, I think a sad commentary. Space flight is expensive in terms of engineering resources. It might be useful if we had a class of expendable special purpose boosters. > > 6. Worship the KISS principle. This thing has to go a long way on > > it own. > > Don't forget a lot of redundancy. Important. You cannot stepwise refine hardware once it's in flight. I thought this question was interesting from the stand point of materials science and autonomous vehicles. We really don't know how well our earthly materials will withstand the rigors of deep space. We have learned a lot from the last two pairs of deep space missions. How would you handle a craft hit by an astroid before leaving the solar system? Would you give up or have the robot try and repair itself. Sounds expensive. Lots of interesting, fun questions (sorry, can't answer some for $1G). Wish I had more time. --eugene miya NASA Ames Research Center {hplabs,ihnp4,dual,hao,decwrl,allegra}!ames!aurora!eugene @ames-vmsb.ARPA:emiya@jup.DECNET
john@x.UUCP (John Woods) (05/30/85)
> From: jim@TYCHO.ARPA (James B. Houser) > > science data in addition to the usual ego stuff. Because of the constraints > it will be important keep things simple. The parameters? > 1. Assume only near term technology and assets. For example, no > 2. Must be ready to launch within 5 years. > 3. Assume a target star 20 light years away. > 4. Intial data must be received from target star system within 100 > years. > Occurred that this would be an interesting thought experiment. > You can assume the target is in whatever direction you like if it will help. > Prime issue may be the choice of propulsion scheme. Another question is > what instruments to carry, preferably few and simple. Any thoughts???? > My initial thought is that it may be better to wait: within 100 years, perhaps within 50 years, we will be able to do much better than the speeds currently available (you mentioned .05C, but to get to a start 20LY away in 100 years, you need at least .2C average speed, anyway!). If in 50 years we can develop the ability to get a probe to that star in 40 years, you're 10 years ahead by waiting 50 years! Not that I'm averse to the idea, but it seems that just puttering around the Solar System for another few decades will be enough to get ready for a high-class interstellar probe. Think we could interest Proxmire in a Shuttle ticket? Hey, to keep it cheap, it could be one-way!-) -- John Woods, Charles River Data Systems, Framingham MA, (617) 626-1101 ...!decvax!frog!john, ...!mit-eddie!jfw, jfw%mit-ccc@MIT-XX.ARPA "MU" said the Sacred Chao...
henry@utzoo.UUCP (Henry Spencer) (05/30/85)
> ... Since lasers produce a highly collimated > beam, there is no real 1/(r squared) problem. Alas, not so. Lasers do not eliminate the inverse-square problem, they merely postpone it. Keeping a beam focused on a lightsail over interstellar distances is still a major problem, although solutions are known (albeit ones that involve very large structures, which would have to be space-based). Note also that producing any noticeable amount of thrust with a lightsail requires tremendous laser power output. Don't be misled by thinking that it only has to match sunlight; sunlight is (speaking *very* roughly) a kilowatt per square meter. When the sail gets large, the necessary laser power gets huge. Note also that lasers are inefficient, maybe 15-20% at best. -- Henry Spencer @ U of Toronto Zoology {allegra,ihnp4,linus,decvax}!utzoo!henry