Subject: Space-tech Digest #121 Contents: Magsails (tension, etc) (10 msgs) Ocean-based tethers (8 msgs) Wanted: historical info on DC-X predecessor designs (2 msgs) Gerard K. O'Neill passes away (2 msgs) ------------------------------------------------------------ Date: Mon, 27 Apr 1992 23:54 EDT From: "GORDON D. PUSCH" Subject: Re: Magsails To: roberts@cmr.ncsl.nist.gov, space-tech@cs.cmu.edu From: John Roberts : > >> Let the ground-loop's flux be established while the magsail-loop is >> *normal*. Cool the magsail-loop below T_c; the gound-loop's flux >> linking the magsail-loop then becomes ``frozen'' --- it can't change. > > Based on my recent reading of things I didn't really understand in > Encyclopaedia Brittanica, I'm not sure type II superconductors can be > relied on to do this ... > Uh, *sort* of ... Strictly speaking, superconductivity is a purely "DC"-phenomenon; for changing currents, there are losses due to the EMF's accelerating both "normal state" *AND* "superconducting" electrons --- with concommit resistive effects; flux vortices break free of "pinning" sites and drift through the lattice, inducing a back-EMF, etc., etc. But even for RF-frequencies, they're still *DARN GOOD* conductors ... :-) One of the principle differences between type-I's and type-II's is that in type-II's, above the "first" critical field-strength, H_c1, flux begins to penetrate into the bulk of the superconductor, in the form of ``flux-vortices'' full of "normal" metal threaded by a single flux-quantum; by H_c2, the vortices occupy the entire volume of the super-conductor, rendering it "normal." (By contrast, type-I's go "normal" all at once when they hit H_c.) Type-II's still ``freeze'' the flux linking them --- except that flux-quanta can ``tunnel'' through the ring. Fortunately, a flux quantum is *VERY* small ... :-) (BTW, the primary problem w/ the high-T_c superconductors iss that their ``vortex pinning-strength'' is quite small; it's still a topic of active research as to whether they have a flux *lattice* or a flux *liquid* in them ... their H_c's are fine (ca. 3 *M*Gauss !!!), but unless the J_c's of bulk materials can be improved to the values seen in thin-film samples, so the vortices don't break loose and drift, the whole "magsail" idea may be ``right out'' ... :-( > I believe the high-temperature superconductors discovered > thus far are all type II. > Yes, so far they are all type-II. >> Now: rapidly reverse the polarity of the ground loop --- preferably via >> inductive switching to a second concentric loop of opposite sense, >> so (almost) no losses occur. Viola! The magsail springs into the sky!!! 8-D > > Since the energy present is represented by the magnetic field of the > original polarity, that might be rather difficult - sort of like reversing > the direction of your car by bouncing it off a stone wall. :-) > Well, I didn't say iot would be *easy* 8-). Seriously, this is not so much of a problem; the trick is in matching the coils to the pulse-forming network. I caution you that my field is beam-dynamics, *NOT* pulsed-power engineering, but a crude idea of how it *might* work is to imagine each coil to be connected via switches to a capacitor: SW1' SW2' |---*-----X----*----X-----*----| C | __|__ | C L1 C X SW1 C _____ SW2 X C L2 C | | | C |---*----------*----------*----| Switch SW1 is intially *closed*; all other switches are ``open.'' At time t=0, the ``crowbaring'' switch SW1 opens, while ``transfer''switch SW1' closes; current begins to flow into the central capacitor (confusingly labeled "C" in the diagram; not to be confused with the coils, "CCC" :-). Half an LC-period latter, the charge on C is a maximum, while the current through L1 is zero. Now close SW1' and open SW2'; current begins to flow through L2. Half an LC-period latter, the charge on C is zero, while the current through L2 is a maximum; now SW2' opens, while SW2 closes, ``crow- baring'' L2 at I_max. Of course, the above is *grossly* over-simplified, high-power ``opening'' switches are *very* hard to come by, etc., etc. ... In accelerator work, we usually use saturable-core reactors nowadays instead of switches; this "magsail" has got me toying around w/ a superconducting equivalent ... Gordon D. Pusch ------------------------------ Date: Wed, 29 Apr 92 17:45 GMT From: CSG0669@VAX2.QUEENS-BELFAST.AC.UK (Dale Amon) To: space-tech@cs.cmu.edu Subject: Re: Magsails Sender: mnr@DAISY.LEARNING.CS.CMU.EDU The primary remaining technological hurdle to a testable magsail is getting close to solved. Zubrin requires Hi-TC wire that is flexible enough to be 'inflated' to its full circular size of a few hundred km circumference. A Japanese group is now making wire in the 100-1000m range. It is probably easier to get from there to 200 km than it was to get from 0 to 100 m. Dale Amon PS: So when do we try it out? :-) ------------------------------ Date: Thu, 30 Apr 1992 18:39 EDT From: "GORDON D. PUSCH" Subject: Re: Magsails To: CSG0669@VAX2.QUEENS-BELFAST.AC.UK, space-tech@cs.cmu.edu Dale Amon writes: > > The primary remaining technological hurdle to a testable magsail is > getting close to solved. Zubrin requires Hi-TC wire that is flexible > enough to be 'inflated' to its full circular size of a few hundred km > circumference. > > A Japanese group is now making wire in the 100-1000m range. It is probably > easier to get from there to 200 km than it was to get from 0 to 100 m. > I spoke to Dana Andrews yesterday; he agrees that tensile strength is *not* the show stopper right now --- getting high-T_c wire of sufficient quality, with a J_c *comparable to **thin-sample** J_c* is the problem. Dale: how is the J-c of this stuff you are talking about? > PS: So when do we try it out? :-) I dunno; I doubt 200 klicks of High-T_c cable will fit in a G.A.S.... :-T (Dr. Andrews was, by the way, quite amused to hear how much fuss this five- year-old idea has suddenly kicked up; he also said he had *no idea* Zubrin had written his article until he open his _Analog_ to read on a plane ... He wishes he could "join the fun;" he doesn't think he can get to the net... To Dani Eder: maybe you can help him find us? :-) RE: the tension problem: they couldn't find any formulas for the tension(!), so they made a crude estimate: they assumed (if I understood him right --- he's going to send me some reprints) that it probably couldn't be *MUCH* higher than the ``pinch-effect'' pressure in an infinite wire, so that's what they used. He also finds the ``logarithmically divergent'' toroidal tension the Landau & Lifschitz formula predicts to be as absurd as I do ... Paul Deitz and I have been corresponding on the ``tension problem'' privately. I hope to take a good, close look at it over the weekend ... I have also spoken w/ a number of my colleagues, and nobody knows the answer, but everybody thinks it's nuts --- that L&L and I have probably mis-applied our ``energy argument,'' and we're not calculating what we *think* we are ... and I agree with them :-). Given that the radial force is basically a small residual due to the inhomogeneity of the poloidal field across the cross-section of the wire, and that the inhomogeneity *does* vanish as the major radius goes to infinity, it seems almost certain that the tension *MUST* vanish in that limit, as I intuitively first believed ... the question now is: *how fast* ??? I'll get back to y'all on it, OK? Gordon D. Pusch ------------------------------ Date: Thu, 30 Apr 92 18:39:48 -0400 From: dietz@cs.rochester.edu To: PUSCHG@crl.aecl.ca Subject: Re: Magsails Cc: space-tech@cs.cmu.edu I want to say that I *don't* think the logarithmically divergent tension is ridiculous. The derivation in L&L seems reasonable to me Given that the energy stored for a given current and wire cross section goes as r log r, wouldn't the virial theorem require that tension goes as log r? Paul F. Dietz dietz@cs.rochester.edu ------------------------------ Date: Thu, 30 Apr 1992 22:33 EDT From: "GORDON D. PUSCH" Subject: E/M tension, and principle of ``virtual work'' To: deitz@cs.rochester.edu, space-tech@cs.cmu.edu ... Just ran across a copy of Fano, Chu, and Adler's _Electromagnetic Fields, Energy, and Forces_. They have the following to say on the principle of ``virtual work'' [emphasis and square-bracket insertions mine]: ``... the force or torque on a *rigid* body can be computed by requiring the *work done by [external forces on] it* for a small displacement plus the corresponding increase of [field] energy stored in the *system* to be equal to the energy supplied (if any) by external electric sources...'' They emphasize that their subsequent derivation of the principle of ``virtual work'' from Poynting's theorem applies to *rigid* bodies and *static* fields, because in this case, the *motional* EMFs resulting from quasi-static displacements are *total differentials*. I hypothesize that the problems w/ L&L's and my derivation are the following: 1) Non-rigid displacement --- it isn't even a *scale* transformation a-la the classical mechanics Virial thm. (for which a precise analog could be derived). (Paul: BTW, what *precisely* do you mean when you write ``Virial thm?'' I've only encountered this term RE: the proportionality of avg. kinetic to avg. potential energy for orbiting bodies. I've been *assuming* you meant the same as ``virtual work,'' but it occurs to me that I might be mistaken ... ) 2) We assume no external sources --- yet speak of change of energy at ``constant current'' or ``constant flux.'' However, every time I try to *sketch* this problem (e.g., to show how one could *measure* it in a lab) I draw constant-current sources, or masses to exert constant tensions, etc. In other words: *external sources* of current, force, etc. I *don't* think L&L and I took all the sources of energy into account! 2) Equating change in field energy to mechanical work --- see external sources cited in item (2). RE: item (2) and (3) (as if (1) alone weren't fatal): last night, I rederived the ``tension'' ... but *this* time, I assumed constant *flux-linkage* --- the logic being, w/out external sources of EMF, the flux linking a superconducting loop *CAN'T CHANGE*; this seems more reasonable than constant *current*, as L&L (and I initially) assumed. I got *exactly* the same tension as before...**UP TO AN OVERALL MINUS SIGN** [is negative tension a **pressure**?!? %-[. I honestly believe that there's something *VERY* important being left out of the energy-balance. There probably *IS* an ``energy method'' derivation for this problem; but I *DON'T* think it's the one I'm using ... :-( Gordon ------------------------------ Date: Fri, 1 May 92 11:43:20 -0400 From: dietz@cs.rochester.edu To: PUSCHG@crl.aecl.ca Subject: Re: E/M tension, and principle of ``virtual work'' Cc: space-tech@cs.cmu.edu > RE: item (2) and (3) (as if (1) alone weren't fatal): last night, > I rederived the ``tension'' ... but *this* time, I assumed constant > *flux-linkage* --- the logic being, w/out external sources of EMF, > the flux linking a superconducting loop *CAN'T CHANGE*; this seems more > reasonable than constant *current*, as L&L (and I initially) assumed. > I got *exactly* the same tension as before...**UP TO AN OVERALL MINUS SIGN** > [is negative tension a **pressure**?!? %-[. Eh? Let's try this. Linked flux = L I, so at constant flux I is proportional to 1/L. The stored energy goes as 1/2 L I^2, which is proportional to 1/L. If we let L be proportional to a Log[8a/b] (correct up to a term of O(a)), the derivative of stored energy w.r.t. a comes out to be, for a >> b, something like - Log[8a/b] I^2 + lower order terms. This is negative, so there is a *tension* on the ring (it wants to expand to reduce the stored energy). Paul ------------------------------ Date: Fri, 1 May 1992 13:40 EDT From: "GORDON D. PUSCH" Subject: Re: E/M tension, and principle of ``virtual work'' To: dietz@cs.rochester.edu, space-tech@cs.cmu.edu Paul Dietz write: > >> RE: item (2) and (3) (as if (1) alone weren't fatal): last night, >> I rederived the ``tension'' ... but *this* time, I assumed constant >> *flux-linkage* ... >> I got *exactly* the same tension as before...**UP TO AN OVERALL MINUS SIGN** >> [is negative tension a **pressure**?!? %-[. > > Eh? Let's try this. Linked flux = L I, so at constant flux I is > proportional to 1/L. ... > If we let L be proportional to a Log[8a/b] (correct up to a term of O(a)), > the derivative of stored energy w.r.t. a comes out to be, for a >> b, > something like - Log[8a/b] I^2 + lower order terms. > > This is negative, so there is a *tension* on the ring (it wants to > expand to reduce the stored energy). > Ummm ... doggone, you're right. Still, you must admit that it *IS* the negative of the ``constant current'' result (derivative of 1/L rather than L brings in a minus sign, as you showed) --- so it should be *L&L's* loop that's under pressure; what gives? ... Ack! Here's the problem: L&L *defined* the ``free energy'' to *differ by a sign* between the constant-current and constant-flux cases. I suppose there's some thermodagnabic reason for it ... Now you know why I hate thermodagnabics :-( Look: *forget* energy-arguments for a minute --- they only work if one includes *ALL* the energy-storage mechanisms (and gets the signs right... :-( Besides, as Dr. Lee-Whiting pointed out to me, we're probably calculating the integrated *Maxwell* stress-tensor, *NOT* the *mechanical* stress (assuming a *non-rigid* variation calculates *ANYTHING* meaningful). So do something simpler: apply Biot-Savart to a cylindrical wire. The current density J is entirely *axial,* the magnetic induction B is entirely *azimuthal,* so (J \cross B) is entirely *radial*. Now ... *WHERE IS THE TENSION **COMING** FROM ?!?!?* HOW IN THE NAME OF NEWTON AND MAXWELL CAN THERE BE A ***TENSION*** IN THE WIRE, WHEN THERE ARE *NO* *ZIP* *ZILCH* *ZERO* ***AXIAL FORCES*** ?!?!?!? WHAT IS THE *PHYSICAL CAUSE* OF A TENSION WITH A *NON-EXISTENT SOURCE* ?!?!?!?!? I am utterly convinced that there is something *wrong* with this energy- derivation. I will *NOT* accept the result unless and until I get the *SAME* result by actually integrating the J-cross-B force for a physically reasonable current distribution. I have found a paper giving a closed-form expression for the field of a toroidal current, I will calculate the radial component of the J-cross-B force density, and I shall report back the result --- I hope --- by the end of next week. The integrals are pretty awful --- associated Legendre functions of the second kind with half-odd order, but they look similar to some integrals I found while calculating the fields of our magnets. Gordon D. Pusch P.S.-- Dr. Lee-Whiting has just called me; he has worked out the hoop-stress on a toroidal current of *rectangular* cross-section, using Biot-Savart (Graham Lee-Whiting is a *very* mathematically competent physicist!). He has obtained an integral which is *finite*, and which he believes he can prove will vanish in the limit that a/b goes to infinity; he is currently working on said proof. He doesn't think there should be a qualitative difference between a rectangular and circular cross-section ... :-T Furthermore, a rectangle is what Andrews and Zubrin want, *anyway* !!! He says the next step is to find the falacy in Landau and Lifschitz's ``proof''... I'm confident he will, too. I'm still going to work out the circular cross-section case though --- I need the practice ... 8-) I'll keep you posted /gdp ------------------------------ Date: Fri, 1 May 92 14:07:19 -0400 From: dietz@cs.rochester.edu To: PUSCHG@crl.aecl.ca, space-tech@cs.cmu.edu Subject: Re: E/M tension, and principle of ``virtual work'' Gordon writes: > So do something simpler: apply Biot-Savart to a cylindrical wire. The > current density J is entirely *axial,* the magnetic induction B is > entirely *azimuthal,* so (J \cross B) is entirely *radial*. Now ... > *WHERE IS THE TENSION **COMING** FROM ?!?!?* HOW IN THE NAME OF NEWTON > AND MAXWELL CAN THERE BE A ***TENSION*** IN THE WIRE, WHEN THERE ARE > *NO* *ZIP* *ZILCH* *ZERO* ***AXIAL FORCES*** ?!?!?!? WHAT IS THE > *PHYSICAL CAUSE* OF A TENSION WITH A *NON-EXISTENT SOURCE* ?!?!?!?!? > I am utterly convinced that there is something *wrong* with this energy- > derivation. I will *NOT* accept the result unless and until I get the > *SAME* result by actually integrating the J-cross-B force for a physically > reasonable current distribution. But we're not talking about a *straight* cylindrical wire: we're talking about a *curved* wire. Let's apply the Biot-Savart law to show that the outward force on a current element is at least order of (log (a/b)) / a^2 (this will imply tension of at least order log(a/b)). We want to determine the total outward force per unit length of wire on a collection of current filaments at some given angle along the loop. Call this position Z. The contribution of the parts of the loop near to this filament are difficult to calculate, so we'll ignore them. (I suggest they as well will cause a net outward force on Z, but I do not calculate this.) The parts of the loop we will integrate over is the part at angle at least angle (b/a)^(1/3) away from Z. These points are at least b^(1/3) a^(2/3) away from Z, so as b/a-->0 we should be justified in treating the collection of current filaments at the points as single filaments of zero thickness at the wire's center. If we do this, the B-S law gives an outward force on the wire proportional to the integral of (I^2/a) cosec(theta) from theta = (b/a)^(1/3) / 2 to pi/2. Let t = (b/a)^(1/3)/2. Now, the integral of cosec(theta) is log ((1-cos(theta))/sin(theta)), so this is proportional to (I/a)^2 log sin(t)/(a-cos(t)), which is approximately (I^2/a) log 2(a/b)^{1/3}. This is order of I^2 log (a/b) / a, as desired. Intuitively what is happening is that the integral in the B-S law on a loop with b=0 is divergent. But in the actual physically real situation, the current in the wire is spread over a distance of order b, so the integral gets cut off. Cutting off the integral naively will only approximate the correct answer, but at least it's of the right order. Paul F. Dietz dietz@cs.rochester.edu ------------------------------ Date: Mon, 4 May 92 12:09 GMT From: CSG0669@VAX2.QUEENS-BELFAST.AC.UK To: space-tech@cs.cmu.edu Subject: Re: Magsails Gordon asked for the reference on superconducting wire: Frederick Myers, 'Superconductors in Japan', Science 255, 1078-1079, (1992) Kenichi Sato of Sumitomo Electric Industries have made 'good quality wires 100 meters long' and 'in one or two years we expect to be making a wire that is nearly 1000 meters long'. A bismuth based superconducting compound is sheathed in a silver tube and rolled flat. In the resulting composite wire the superconductor carries the current, while the silver coating protects the superconductor from degrading and gives it extra flexibility. Sata thinks the wire will soon be ready to be marketed as a cable for high-current power supplies or as a winding for electromagnets. The article goes on to note that Such wires are still plagued by flux creep... but Shoji Tanaka, director of the SRL, reports that a new fabrication technique called melt-powder melt growth (MPMG) can yield a YBCuO supercon. peppered with nonsuperconducting impurity particles that efficiently trap the magnetic flux. The technique recently enabled SRL workers to increase the material's current carrying capacity several-fold, to more than 100KA/cm^2 at 77K... Well, the problems aren't solved, but the progress seems to be quite steady. And considering the economics of getting practical wire, I suspect there will continue to be solid funding in this area by forward looking companies. Dale Amon ------------------------------ Date: Mon, 4 May 1992 15:44 EDT From: "GORDON D. PUSCH" Subject: Re: Magsails To: CSG0669@VAX2.QUEENS-BELFAST.AC.UK, space-tech@cs.cmu.edu From: Dale Amon > > The technique recently enabled SRL workers to increase the material's > current carrying capacity several-fold, to more than 100KA/cm^2 at 77K... > Urrrrrr... still not good enough; it's an order of magnitude below the value Andrews and Zubrin are assuming. Still, it's progress ... Thanks, Dale! Gordon D. Pusch ------------------------------ Date: Tue, 28 Apr 92 09:01:35 CDT From: eder@hsvaic.boeing.com (Dani Eder) To: dietz@cs.rochester.edu Subject: Re: Tethers Cc: space-tech@cs.cmu.edu Yes, a sub-synchronous tether can be built with existing materials. Very roughly, the material strength you need scales linearly with the delta-V you want the tether to perform. You need 2-million psi material to do a synchrounous tether. With 1-million psi\ carbon fiber (state of the art), you can do about a 4 km/s tether, or about 1/2 orbital velocity. You want the bottom end of a non-synchronous tether to be just above the atmosphere, to minimize drag losses. More generally, there is a whole family of tethers that ranges from a 1-km length on up, where the bottom end stays just above the atmosphere, and the top end and center of mass move higher. As the center of mass moves upward, the bottom end moves slower and slower. The synchronous tether is the special case where the bottom end has zero velocity. If you build even larger tethers, then the bottom end moves BACKWARDS. Dani ------------------------------ To: space-tech@cs.cmu.edu Subject: Re: Tethers Reply-To: munck@STARS.Reston.Unisys.COM Return-Receipt-To: munck@Stars.Reston.Unisys.COM Date: Tue, 28 Apr 92 10:05:02 -0400 From: Bob Munck George William Herbert says: > The much-quoted problem of the ground end of a >tether really isn't one. ... You can drop your tether down in >the ocean quite easily, ... Of course, it doesn't have to come down in a single place. Having it split at say 10,000km and come down 20-30 degrees N and S of the equator would have the advantage of leaving equatorial LEO unobstructed. Bob Munck ------------------------------ From: henry@zoo.toronto.edu Date: Tue, 28 Apr 92 14:04:57 EDT Subject: Re: Tethers (moving base of) To: space-tech@cs.cmu.edu >> ...horizontal forces... something has to oppose >> them if you don't want the base wandering around. > >Just what scale of motion should we picture here? If it's >order 10 meters a minute, a floating base with NO anchor >might be just the ticket. Just let it slosh around! ... The time scales are long and motions are slow... but they're not just random oscillation around a specific point. There is systematic drift as well. Firm anchoring really is needed. Henry Spencer at U of Toronto Zoology henry@zoo.toronto.edu utzoo!henry ------------------------------ From: henry@zoo.toronto.edu Date: Tue, 28 Apr 92 14:12:23 EDT To: space-tech@cs.cmu.edu Subject: Re: Tethers >... one big advantage of a fixed base is that it allows the >payload going up and the payload going down to be unbalanced - the difference >in angular momentum goes into speeding up or slowing down the rotation of the >Earth, rather than tearing the elevator away from the planet or causing it >to fall to the ground. Any platform that is reasonably massive (so its weight puts the cable in tension to some degree) and is anchored against horizontal motion will suffice to handle the momentum transfer, although of course there will be limits that have to be observed. A floating base clearly is going to have to be big, if only so that it can ignore wave motion even in fairly severe storms. It might well be necessary to accommodate some range of (slow) motion between the cable base and the platform, for the sake of tides and other changes in surface level. Personally, I agree with the people who think that a land base is less hassle. I don't think the ocean base is impossible, but it has a lot of extra problems. Henry Spencer at U of Toronto Zoology henry@zoo.toronto.edu utzoo!henry ------------------------------ Date: Tue, 28 Apr 1992 15:12 EDT From: "GORDON D. PUSCH" Subject: Re: Tethers To: space-tech@cs.cmu.edu A reason no one seems to have given so far for anchoring a tether is that **GEOSYNCHRONOUS TETHERS ARE UNSTABLE**!!! Every calculation I've ever seen indicates that (for essentially the same reason the ringworld is unstable) a tether over a certain length represents an energy *saddle-point*, not a minimum; it's therefore necessary to anchor it firmly at both ends, lest it fall ``inwards'' or ``outwards.'' I believe I once proved this for a simple ``dumbell'' configuration in orbit; I *know* I've seen it proved for *uniform* tethers in an _Am.J.Phys._ article ... and I'm *pretty* sure Charles Sheffield said it was true for *tapered* tethers in one of his ``popular'' articles on ``beanstalks.'' Does anyone know of a counterexample ??? Gordon D. Pusch ------------------------------ Reply-To: davidsen@crdos1.crd.ge.com Date: Tue, 28 Apr 92 09:11:56 EDT From: davidsen@crdos1.crd.ge.com To: space-tech@cs.cmu.edu Subject: Re: Tethers > Note, though, that there is one reason to prefer a good solid anchor for > a beanstalk: at most points along the equator, there will be small -- > well, relatively small -- horizontal forces due to the lumpiness of Earth's > gravitational field and outside perturbations, and something has to oppose > them if you don't want the base wandering around. A floating base needs > either anchors on the bottom or a lot of propulsive power. That's probably not a fatal flaw, assuming a base point relatively far from land, a sea anchor to hold velocity down would probably do. And if you got some cumulative error you could trim the resistance of the sea anchor with a few ports, so you drift faster when you head the way you want to go. For those who might not know, a sea anchor is just a parachute modified to work in a higher density and viscosity fluid. ------------------------------ Date: Tue, 28 Apr 92 14:13:38 PDT From: gwh@lurnix.COM (George William Herbert) To: space-tech@cs.cmu.edu Subject: Ocean Tether Touchdown It seems like I under-described how you would drop a beanstalk into the ocean successfully. Here's what I had in mind, in more complete detail. The tether itself does not have to be anchored to the platform. In fact, this is perhaps a disadvantage. The tether would actually be tethered to a convenient spot on the ocean floor; a large, heavy anchor would be dropped in place, then it would be tied into the bedrock below the ocean floor using piles. All of this is known and used (and not too expensive) technology from the Oil Rig industry. A couple of million tons force could be handled by a system I sketched out earlier today based on some north-sea platform bases that have been done before. I don't think that at least the initial tether will pull harder than that 8-) Now, we've succeeded in tying the tether down. It goes up through the ocean, which has the side effect of helping to damp motion (though for corrosion reasons, you might want to build a seperate structure that goes from the seafloor to above the water, where the actual "orbital tether" is tied in. That way you don't expose the tether directly to seawater. On the other hand, diamond fiber probably is corrosion resistant 8-) You never know.) Once the structural problem is solved, we move on to the systems problem. Specifically, there has to be a cargo transfer pad for the elevator on/in the tether, and probably some support equipment there too. This can be on a floating (if you like) or legged platform, built right next to or even around (that might be a bad idea with a floating platform) the tether or tether connector. Something the same size/shape/ general configuration of a large oil production platform, with additional helipads and some more space to dock water vessels, would do just fine. Those platforms run around $250-600 million apiece; presuming about the same for the cable anchor (as a quick estimate) gives us an upper bound over just over a billion dollars. And remember, these are proven technologies, not something we have to slowly and painfully develop. -george william herbert gwh@lurnix.com gwh@ocf.berkeley.edu ------------------------------ Date: Tue, 28 Apr 92 15:26:39 CDT From: ssi!lfa@uunet.UU.NET (Louis F. Adornato) To: uunet!cs.cmu.edu!space-tech@uunet.UU.NET Subject: Sub-Synchronous Tethers - a crazy idea > From uunet!hsvaic.boeing.com!eder > Yes, a sub-synchronous tether can be built with existing materials. > Very roughly, the material strength you need scales linearly with > the delta-V you want the tether to perform. You need 2-million > psi material to do a synchrounous tether. With 1-million psi\ > carbon fiber (state of the art), you can do about a 4 km/s > tether, or about 1/2 orbital velocity. You want the bottom end > of a non-synchronous tether to be just above the atmosphere, > to minimize drag losses. I'm sure that such a tether wouldn't be able to lift anything (you'd want _some_ margin between the needed and available strength, right?), but I have this dumb idea I just can't get out of my head... A sub-synchronous tether, with the terminal normally carried above the atmosphere, and embedded conductor to allow reboost against the geomagnetic field. At the high end is a counterweight on a winch. At the low end is an active shock absorber system (the _mother_ of all bungee cords...). In use, the end is lowered into atmosphere, and a plane in flight snags the hook (which will be approaching at around 8,000 mph; a tricky maneuver, but not _completely_ impossible, especially if the hook had some aerosurface control and computer guidance). The shock absorber limits acceleration on the payload to 4gs, stretching 203 km (126 mi for us North Americans) in just over 100 seconds (at which point the payload is up to tether speed). As soon as the hook makes contact, a solar power station at the midpoint starts pumping power into the conductor, generating boost off the back EMF, and powering a winch on the counterweight. Once the payload is up to tether speed, the shock absorber is retracted, the payload hooks onto an electric "truck" (to use a railroad term), and rides up the tether to the midpoint (where it'll be in free fall). I haven't really given any thought to the reentry process. The shock absorber poses an interesting design problem. How about a reel of Kevlar fiber turning a generator that pumps additional power into the tether (although this smacks of perpetual motion to me)? Or a kevlar reel and an ablative brake? Or a hydraulic cylinder and a pulley arrangement? Go ahead, shoot holes in it, I have on my flame-retardant undies. Lou Adornato | "Sure, the cow may have jumped over the Supercomputer Systems, Inc | moon, but she burned up on reentry" Eau Claire, WI | The secretary (and the rest of the company) uunet!ssi!lfa or lfa@ssi.com | have disavowed any knowledge of my actions. ------------------------------ Date: Tue, 28 Apr 92 17:22:37 -0500 From: pgf@srl01.cacs.usl.edu (Phil G. Fraering) To: space-tech@cs.cmu.edu Subject: Wanted: historical info on DC-X predecessor designs The title just about says it all.... BTW, anyone out there wonder how well the DC-Y design could be scaled down, or you could have a DC-Whatever type set up for like 3000 kg? Phil Phil Fraering pgf@srl05.cacs.usl.edu 318/365-5418 "...Norwegian Independence Day, when the Norwegians rose up, and threw off the yoke of Danish oppression." - Garrison Keillor ------------------------------ Date: Tue, 28 Apr 92 18:01:23 -0500 From: ewright@bach.convex.com (Edward V. Wright) To: pgf@srl01.cacs.usl.edu, space-tech@cs.cmu.edu Subject: Re: Wanted: historical info on DC-X predecessor designs >BTW, anyone out there wonder how well the DC-Y design could >be scaled down, or you could have a DC-Whatever type set up for >like 3000 kg? During the 1970's, Gary Hudson designed an entire family of Phoenix launch vehicles, the smallest of which was sized to put a Get-Away Special-sized (200 lb.) payload into orbit. ------------------------------ Date: Mon, 4 May 92 15:04:35 CDT From: ssi!lfa@uunet.UU.NET (Louis F. Adornato) To: uunet!cs.cmu.edu!space-tech@uunet.UU.NET Subject: Whereis O'Neil Last I heard, Jerrard O'Neil was at Princeton doing modeling of a lunar based electrical mass driver. For those who have been living in a cave for the last 30 years (or who can't translate my probable misspelling of his name), O'Neil was the one who legitimized the idea of huge cylindrical colonies, built out of lunar material, in orbit around the earth-moon L3 and L5 points. Does anyone know where he is now, if he's still working on the mass driver models, and (hopefully) his net address? Lou Adornato ------------------------------ Date: Mon, 4 May 92 16:51:46 EDT From: Mark.Maimone@A.GP.CS.CMU.EDU To: space-tech@CS.CMU.EDU, ssi!fla@UUNET.UU.NET Subject: Gerard O'Neill > From: higgins@fnalo.fnal.gov (Bill Higgins-- Beam Jockey) > Newsgroups: sci.space,sci.astro,sci.physics > Subject: Gerard K. O'Neill passes away > Date: 30 Apr 92 07:59:17 GMT > > Gerry O'Neill has passed away. After hearing the sad news from Larry > Boyle of the Chicago Space Studies, I went hunting and found that Bob > Summersgill had already posted an obituary to SEDS-L, which I take the > liberty of quoting here. > > O'Neill's ideas had tremendous power to inspire people and rekindle > hopes for the future of spaceflight that had grown dim in the > post-Apollo slump. He will be missed-- but his influence on a > generation of space advocates will echo for decades. > > O~~* /_) ' / / /_/ ' , , ' ,_ _ \|/ > - ~ -~~~~~~~~~~~/_) / / / / / / (_) (_) / / / _\~~~~~~~~~~~zap! > / \ (_) (_) / | \ > | | Bill Higgins Fermi National Accelerator Laboratory > \ / Bitnet: HIGGINS@FNALB.BITNET > - - Internet: HIGGINS@FNAL.FNAL.GOV > ~ SPAN/Hepnet: 43011::HIGGINS > ======================================== > Date: Tue, 28 Apr 1992 16:20:15 EDT > Reply-To: "Interchapter Communications for SEDS" > Sender: "Interchapter Communications for SEDS" > From: Bob Summersgill > Subject: Dr. Gerard K. O'Neill > > Dr. Gerard K. O'Neill died yesterday, 4/27/92 in Princeton, NJ. He was > 65 years old. Dr. O'Neill was the founder and President of the Space Studies > Institute, Professor Emeritus of Princeton, founder of the GeoStar Corp., > and O'Neill Communications, Inc. He is the author of four books > including *The High Frontier* and *2081*. He created and popularized > the idea of colonies in free space, designed and built the Mass Driver, > and designed the device that allows particle accelerators to fire into > each other. He was a recipient of the SEDS's Arthur C. Clarke Award. > > His death was the result of a seven year fight with leukemia. A memorial > service will be held at the Princeton Chapel, May 26th at 10:00am. > > > - Bob Summersgill | Taxes are not raised for the benefit of > XE605C @ GWUVM - Bitnet | the taxed. -- Lazarus Long > > ------------------------------ End of Space-tech Digest #121 *******************