# SPACE BRIDGE SHORT

85 views

### bria...@news.delphi.com

Jul 25, 1994, 10:06:56 PM7/25/94
to
PARTIAL SPACE BRIDGE

We've seen discussion and calculation of the material and
economic realities of building a space bridge, a super-strong
cable stretching from earth into earth orbit.

What about a partial space bridge? This bridge would reach
partway to orbit, but stop short. A conventional rocket
would presumably take the payload the rest of the way.

Most of the work a rocket does when lifting off from earth is
spent in the lifting fuel and battling the thick lower
atmosphere. Lifting the rocket by space bridge only half
way to orbit would seem to reduce to cost and energy
required to get to orbit by much more than half.

The benefit of constructing a partial space bridge would
seem to be that the requirements for materials strength can be
relaxed, making the project more achievable in the near term.

How tall a space tower could we realistically build, now?
And how much would it cost per pound to get into orbit,
assuming the bridge were used to maximum capacity?

:

### Alan Anderson

Jul 28, 1994, 4:25:53 PM7/28/94
to
In <311r40\$d...@news.delphi.com>, bria...@news.delphi.com (BRIAN...@DELPHI.COM) writes:
>What about a partial space bridge? This bridge would reach
>partway to orbit, but stop short. A conventional rocket
>would presumably take the payload the rest of the way.

Problem: what would hold your "bridge" up? "Beanstalks" use a cable under
tension to connect the ground to a mass moving at faster than orbital velocity,
which is only possible at a greater altitude than synchronous orbit.
Your partial bridge would likely both collapse under its own weight and
just plain fall over.

>Most of the work a rocket does when lifting off from earth is
>spent in the lifting fuel and battling the thick lower
>atmosphere. Lifting the rocket by space bridge only half
>way to orbit would seem to reduce to cost and energy
>required to get to orbit by much more than half.

Not quite. As it turns out, most of the energy required to reach orbit is
spent in increasing the *velocity* of the satellite, not the altitude.

>The benefit of constructing a partial space bridge would
>seem to be that the requirements for materials strength can be
>relaxed, making the project more achievable in the near term.

Again, not quite. It's provably impossible to build a self-supported tower
to orbital altitude because there just aren't any materials strong enough
to hold up their own weight at the sizes required. (Of course, it used to
be provably impossible to go faster than the speed of sound, or for a
bumblebee to fly...)

>How tall a space tower could we realistically build, now?

I'll defer to materials engineers here, but my guess would be a small multiple
of the height of the World Trade Center.

>And how much would it cost per pound to get into orbit,
>assuming the bridge were used to maximum capacity?

Approximately as much as it does now.

=============================================================================
Alan Anderson || If they put a bunch of cattle in orbit,
(Ham Radio WB9RUF) || would it be the herd shot 'round the world?
My views may not necessarily be those of Delco Electronics or its management.

### Albert-Jan Brouwer

Jul 29, 1994, 2:52:28 PM7/29/94
to
Alan Anderson (ande...@kosepc01.delcoelect.com) wrote:
: In <311r40\$d...@news.delphi.com>, (BRIAN...@DELPHI.COM) writes:
: >How tall a space tower could we realistically build, now?

: I'll defer to materials engineers here, but my guess would be a small
: multiple of the height of the World Trade Center.

The space-transport-list derives a scale height of 53 km for some
advanced carbon composite. This makes sense because the WTC does
concrete supports whose density is also at least a factor of two
larger than that of the composite.

With sufficient tapering it should therefore be (barely) possible
to build a tower so that its top sticks above the atmosphere
(say 200 km). As pointed out this does not buy you much as a
rocket launching platform. However, one unique thing is that you
have a vacuum to play with; you can make things go fast without
friction losses.

My favorite application would be a tether launcher; spin a counter-
balanced tether on a vertical axis at the top of the tower. Using
advanced materials and sufficient tapering it is possible to
construct tethers whose tip sweeps around at serveral km/s. Now
all you need to do is tie your payload to the tip, spin up the

The tether sweeps in a horizontal plane. If the tip moves at
8 km/s or more, the payload should stay in orbit instead of
plunging down into the atmosphere.

For practical tether radii (r < 1km) the centripetal acceleration
at the tip (v^2/r) will be rather extreme (thousands of g's), this
limits the applicability to launching rugged stuff.

---
Albert-Jan Brouwer EMail : ajb...@rulhm1.LeidenUniv.nl

### Lance Purple

Jul 29, 1994, 9:52:28 PM7/29/94
to
A few years ago, Robert Forward proposed a "space fountain" tower,
consisting of sections of evacuated pipe with magnets spaced
every so often. A powerful magnetic coil on the ground sends a
constant stream of tiny iron rings flying up the pipe; the
support magnets slow down the stream, and by Newton's second
law, the stream supports them (like a ping-pong ball balanced
on top of a fountain of water).

The "fountain" can be built from the ground up, and no superstrong
materials are needed. We could start design and construction tomorrow.

Of course, there are the following engineering problems:

- You MUST keep the launch coil powered AT ALL times, or the
whole tower comes crashing down!
- You need a MASSIVE foundation, or else the launch coil will
slowly drill itself downward into the earth's mantle.
- The launch coil and support magnets will make unbelievable
amounts of EM noise, so nobody in your hemisphere can

We could get around the EM noise problem by putting it in the
middle of the Pacific ocean (nobody is going to want to live
east of the thing anyway!).

Forward thinks it can be built for under a trillion dollars.

### Henry Spencer

Jul 30, 1994, 8:34:22 PM7/30/94
to sci-spa...@uunet.uu.net
In article <lpurpleC...@netcom.com> lpu...@netcom.com (Lance Purple) writes:
>A few years ago, Robert Forward proposed a "space fountain" tower,
>consisting of sections of evacuated pipe with magnets spaced
>every so often. A powerful magnetic coil on the ground sends a
>constant stream of tiny iron rings flying up the pipe; the
>support magnets slow down the stream, and by Newton's second
>law, the stream supports them...

Actually, I don't think this idea was original with Forward (or if it was,
it goes back more than a few years). People at places like Livermore have
looked into the notion in some depth.

>Of course, there are the following engineering problems:
>
>- You MUST keep the launch coil powered AT ALL times, or the
> whole tower comes crashing down!

Not a big deal. The circulating stream itself stores a massive amount of
energy. If the system is designed with reasonable margins, and proper
attention to controlled reconfiguration, it can ride out quite substantial
power interruptions.

>- You need a MASSIVE foundation, or else the launch coil will
> slowly drill itself downward into the earth's mantle.

Actually, to a first approximation the foundation required is the one
that would be required to support a conventional building of the same
size. The dynamic-support technique doesn't add any extra requirements
there. Of course, the whole point of the exercise is to build ultra-tall
structures, so some attention to foundation design will be needed.

>- The launch coil and support magnets will make unbelievable
> amounts of EM noise, so nobody in your hemisphere can

I would expect that careful design of the hardware should be able to
--
"We must choose: the stars or the dust.| Henry Spencer @ U of Toronto Zoology
Which shall it be?" -H.G.Wells | he...@zoo.toronto.edu utzoo!henry

### Russell Stewart

Jul 31, 1994, 4:04:33 PM7/31/94
to sci-spa...@uunet.uu.net
Lance Purple (lpu...@netcom.com) wrote:
: A few years ago, Robert Forward proposed a "space fountain" tower,

: consisting of sections of evacuated pipe with magnets spaced
: every so often. A powerful magnetic coil on the ground sends a
: constant stream of tiny iron rings flying up the pipe; the
: support magnets slow down the stream, and by Newton's second
: law, the stream supports them (like a ping-pong ball balanced
: on top of a fountain of water).

My only question is, how much iron would be used? Assuming I understood the
concept right, wouldn't that add up to quite a bit?

Also, I don't think I understand how payloads would be lifted through it.

--
____________________________________________________________________
|Russell Stewart | "Dear sir, |
|dia...@rt66.com | I object to your assertation that |
|Albuquerque, NM | the Royal Navy is a haven for cannibalism!|
| | It is well known that we now have the |
| | problem relatively under control." |
| | -Monty Python |
|_______________________|____________________________________________|

### John McCarthy

Jul 31, 1994, 11:27:49 PM7/31/94
to
In article <31h04h\$3...@mack.rt66.com> dia...@RT66.com (Russell Stewart) writes:

Lance Purple (lpu...@netcom.com) wrote:
: A few years ago, Robert Forward proposed a "space fountain" tower,
: consisting of sections of evacuated pipe with magnets spaced
: every so often. A powerful magnetic coil on the ground sends a
: constant stream of tiny iron rings flying up the pipe; the
: support magnets slow down the stream, and by Newton's second
: law, the stream supports them (like a ping-pong ball balanced
: on top of a fountain of water).

My only question is, how much iron would be used? Assuming I understood the
concept right, wouldn't that add up to quite a bit?

Also, I don't think I understand how payloads would be lifted through it.

Hmm. I'd like a reference to the Forward publication. My
recollection is that I proposed the scheme in 1982 and Rod Hyde of
LLNL did a detailed design and the relevant computations. The work
wasn't published, because of difficulties with stability and
protecting the tower from space junk. Hyde may have discussed it at a
conference. Forward is responsible for so many forward looking ideas
that, like Oscar Wilde, he may be credited with things he ought to have
said.

Our scheme involved circulating the magnetic rings and getting almost
all the energy back when the rings came down. The tower was to have
been built with its base on Baker Island.
--
John McCarthy, Computer Science Department, Stanford, CA 94305
*
He who refuses to do arithmetic is doomed to talk nonsense.

### Lance Purple

Aug 1, 1994, 10:59:13 PM8/1/94
to
I dug up my old copy of "Future Magic"; Forward does attribute the
"Space Fountain" idea to Roderick Hyde. Sorry for the oversight.
He cites a design which weighs 9000 tons, and requires 14 GW to run
(total stored KE is 7 terawatts!), with an annual payload capacity of
6 million tons to GEO.

A variant I once though of uses electron beams instead of iron rings,
and uses small magsails in place of the support magnets. I'll post
if I ever finish doing the arithmetic (to avoid the doom of talking
nonsense); but intuition says it would be much lighter...

### Markus Freericks

Aug 4, 1994, 3:40:24 PM8/4/94
to sci-spa...@uunet.uu.net
I don't believe that dynamic structures will ever be feasible.

How could the safety of such a structure be ensured? I imagine any kind of
tower to be a high-throughput installation, in the order of >1 cargo pod
lifted to orbit per hour. In the long run, collisions between cargo and the
flying rings are bound to happen; how big a region has to be kept evacuated
around the tower to be safe from falling rings? Also, what about weather,
eathquakes etc?

Skeptically,
Markus

--
Markus Freericks m...@cs.tu-berlin.de +49-30-314-21390
TU Berlin Sekr. FR 2-2, Franklinstr. 28/29, D-10587 Berlin (Germany)
"Inertia makes the world go 'round."

### David Goldschmidt

Aug 3, 1994, 10:59:33 PM8/3/94
to
lpu...@netcom.com (Lance Purple) writes:

Could you get the energy back from an electron beam the same as you could
from the iron rings? I take it Forward must somehow tap the current caused
by the iron rings as they fly by for power, and then do the same thing
in a far more dramatic fashion to catch them when they fall, letting them
drop free for most of the tower and catching them at a few hundred G at the
bottom. [am I totally misinterpreting the concept? could be]
Incidentally, another poster said that the foundation wouldn't have to be
much bigger than a building of comparable size. I don't believe this;
the entire tower does rest on the foundation indirectly; if you like, you
need an immense foundation to support the ring gun, on which the tower
rests.
How fast would these rings be going? Unless the track was frictionless,
coriolis (sp?) forces could be a problem. The net force on the tower would
not be very great (rings going up cancel those going down), but the rings
would be pressed quite hard against the side of the tower tube.
Quick and dirty figuring (hence probably wrong) says if you fire them
up at 1 km/sec, they get accelerated sideways a little more than 7 g's. Pretty
hard to make that a frictionless track with that kind of force; and wouldn't
they have to go faster than 1 km/sec to reach GEO, even if they weren't being
slowed down along the way by a kajillion magnets?
Maybe the fountain can be a curve, following the coriolis force.. but how
would you get the rings back?
Dave Patterson
my real address is dpat...@bnl.gov, although I will probably get
mail sent here.

Hmm. I no longer believe net lateral force would cancel either.. just how
are we getting the rings back?

### Henry Spencer

Aug 6, 1994, 11:16:54 PM8/6/94
to sci-spa...@uunet.uu.net
In article <31rcoo\$5...@news.cs.tu-berlin.de> m...@cs.tu-berlin.de (Markus Freericks) writes:
>I don't believe that dynamic structures will ever be feasible.
>How could the safety of such a structure be ensured?

The same way you certify a fly-by-wire airliner: with lots of redundancy,
working up in easy stages to make sure the technology is well understood,
and a great deal of care.

>I imagine any kind of
>tower to be a high-throughput installation, in the order of >1 cargo pod
>lifted to orbit per hour. In the long run, collisions between cargo and the

>flying rings are bound to happen...

The rings go up and down in shafts within the tower; they are not visible
externally and are not a hazard to anything outside.

### Henry Spencer

Aug 6, 1994, 11:30:11 PM8/6/94
to sci-spa...@uunet.uu.net
In article <1994Aug4.0...@Princeton.EDU> i...@atomic.princeton.edu (David Goldschmidt) writes:
>...I take it Forward must somehow tap the current caused

>by the iron rings as they fly by for power, and then do the same thing
>in a far more dramatic fashion to catch them when they fall, letting them
>drop free for most of the tower and catching them at a few hundred G at the
>bottom. [am I totally misinterpreting the concept? could be]

It's more dramatic than that: the rings are decelerated on the way up,
and *accelerated* on the way down. They don't drop free; they are *pushed*
downward, contributing upward thrust on the tower on both the upward and
the downward trips. And yes, you do need to do the same in reverse, at much
higher acceleration, at the bottom.

> Incidentally, another poster said that the foundation wouldn't have to be
>much bigger than a building of comparable size. I don't believe this;

>the entire tower does rest on the foundation indirectly...

So? That's what I said. The foundation is supporting the weight of the
tower; the rings are involved only in transferring the forces from the
upper levels to the foundation. The foundation doesn't care whether the
transfer is done by flying rings or by steel girders, except that the
steel girders add weight themselves. The foundation is, to a first
approximation, the same one you would need to support a conventional
building of the same mass.

> How fast would these rings be going? Unless the track was frictionless,

>coriolis (sp?) forces could be a problem...

There is no track; the rings are in free flight. Necessarily so, since
the velocities are too high for sliding contact or bearings. All forces
involved are conveyed electromagnetically.

> Maybe the fountain can be a curve, following the coriolis force.. but how
>would you get the rings back?

I believe the situation is symmetrical enough that the return path follows
the outward path. Yes, you do get curvature of the paths, especially if
the tower is not at the equator (yes, it *is* possible).

Aug 11, 1994, 4:21:09 PM8/11/94
to sci-spa...@uunet.uu.net
In article <Cu5AG...@zoo.toronto.edu>, he...@zoo.toronto.edu (Henry Spencer) writes:
> In article <31rcoo\$5...@news.cs.tu-berlin.de> m...@cs.tu-berlin.de (Markus Freericks) writes:
> >I don't believe that dynamic structures will ever be feasible.
> >How could the safety of such a structure be ensured?

<stuff deleted>

>
> >I imagine any kind of
> >tower to be a high-throughput installation, in the order of >1 cargo pod
> >lifted to orbit per hour. In the long run, collisions between cargo and the
> >flying rings are bound to happen...
>
> The rings go up and down in shafts within the tower; they are not visible
> externally and are not a hazard to anything outside.

Now I _know_ I'm not conceptualizing this right. What are these shafts made of, and
what are they resting on? Also, I haven't seen a description of how cargo is
going to move up this structure (or are we talking about extremely sophisticated
artforms)?

Aug 11, 1994, 6:01:05 PM8/11/94
to sci-spa...@uunet.uu.net
Going back to the original post...

In article <1994072820...@gmlink.gmeds.com>, ande...@kosepc01.delcoelect.com (Alan Anderson) writes:
> In <311r40\$d...@news.delphi.com>, bria...@news.delphi.com (BRIAN...@DELPHI.COM) writes:
> >What about a partial space bridge? This bridge would reach
> >partway to orbit, but stop short. A conventional rocket
> >would presumably take the payload the rest of the way.
>
> Problem: what would hold your "bridge" up? "Beanstalks" use a cable under
> tension to connect the ground to a mass moving at faster than orbital velocity,
> which is only possible at a greater altitude than synchronous orbit.
> Your partial bridge would likely both collapse under its own weight and
> just plain fall over.

Ahhhhhh, but what if we built a half tower from the _far_ end? As it happens,
I've done some research on this...

>
> >Most of the work a rocket does when lifting off from earth is
> >spent in the lifting fuel and battling the thick lower
> >atmosphere. Lifting the rocket by space bridge only half
> >way to orbit would seem to reduce to cost and energy
> >required to get to orbit by much more than half.
>
> Not quite. As it turns out, most of the energy required to reach orbit is
> spent in increasing the *velocity* of the satellite, not the altitude.
>

Right. Boost stage involves:
a) getting the vehicle up to (at least) apogee altitude,
b) bringing the vehicle up to orbital speed,
c) making sure that "a" and "b" are fulfilled at the same instant, and
d) keeping the vehicle supported above atmosphere while you do this

Now, the orbital velocity at around 150 miles is around 17,000 mph. Given the
specific thrust of hydrogen/oxygen (4.6 s?), it takes a _lot_ less fuel to do
"a" (height) than "b". "d" is a function of your maximum acceleration (which
is why the Saturn 5 had a 5G max boost), and "c" is your control system. As it
turns out, the last few mph is a _lot_ more expensive than the first few miles.
I once worked out the numbers from using a catapult off the top of a mountain,
the results were _not_ impressive. The best place to shave off fuel needs is
at the far end.

> >The benefit of constructing a partial space bridge would
> >seem to be that the requirements for materials strength can be
> >relaxed, making the project more achievable in the near term.
>
> Again, not quite. It's provably impossible to build a self-supported tower
> to orbital altitude because there just aren't any materials strong enough
> to hold up their own weight at the sizes required. (Of course, it used to
> be provably impossible to go faster than the speed of sound, or for a
> bumblebee to fly...)
>

I used the X-15 as a model, assuming that we could, with the aerospace advances
made since the '60s, at least acheive the same performance figures today with a
larger craft (ok, we _might_ be able to acheive the same performance figures in
20 years if we're lucky and keep the gov't out of the loop...). Anyway, a trip
to the library (remember those? they're what people without internet access use
as a read-only file archive) gave me a max altitude of 50 miles, and a ground
speed somewhere around 1300mph (less than a 10th of the velocity needed to get
into orbit, but 30% of the shuttle's usual altitude of 150 miles).

Now, even though 17000 mph is the minimum orbital speed RELATIVE TO A FIXED POINT
OF REFERENCE, the orbital period increases with orbital altitude, so the speed
RELATIVE TO THE GROUND decreases, until you get to the point where the orbital
period is equal to the Earth's rotational period, and the ground speed is 0 (GEO).
I hauled out an envelope and came up with an orbital period of 38350 s being
equivalent to a ground speed of 1300 mph. This corresponds to an orbital
altitude of 516 miles, or about 466 miles above the max X-15 altitude. Someone
please confirm this math, as I don't think I took the rotational speed of the
surface (~1000mph at the equator) into effect.

The following is _not_ supported by any simulation or computational model,
but is the result of my qualitative analysis:

Suppose we dropped a tether from a fixed installation in orbit at 516 miles,
and extended it down to 50 miles. A computer controlled probe (complete with a
reaction control system) is attached to the far end. Along comes the 6:30 X-15
from White Sands, and somehow hooks itself on the tether. The center of mass
of the station/X-15 system would suddenly "drop" by an amount determined by the
difference between the masses of the two elements. The new center of mass would
be below the orbital altitude for the velocity of the system, so the entire
system would _rise_. Of course, the total energy of the system would be less
than that of a body in the initial orbit, so the final altitude of the center of
mass would be lower than the initial altitude of the station (there's no such
thing as a free launch). The energy deficit could be made up with some high-isp
system, like an ion engine array, while the tether was being winched up, so that
eventually the entire system would be back where the station started.

Now for the problems:
1) A conical taper of a kevlar cable has a maximum length of 382 miles,
based on a tensile strength of 420,000 psi and a density of .052
lb/in^3 (sorry for the idiotic units). This leaves us about 84
miles short. The deficit might be made up by changing the flying
part of the system, and it certainly a lot better than the 24,618
mile deficit we get for kevlar at GEO.
2) I suspect that, at "docking", the far end station will experience
deceleration, and the near-end will accelerate (both relative to
the velocity vector). In other words, I think the entire assembly
is going to start to pinwheel. I haven't determined the severity
of this effect, but it _will_ increase the tension on the tether,
and I suspect that it'll change the effective density (as in
"weight" per unit volume), thereby reducing the maximum tether
length. This rotational effect _may_ be insignificant, or it
be the dominant force, in which case the whole idea goes down
the tubes. Of course, a high enough spin rate might be useful
for adding velocity to the near-end, and since it'll be farther
out from the CM than the far end, you might be able to accelerate
a payload without bringing the far end to below re-entry velocity.

Lou Adornato | Co-author of the incredibly popular "Lindsay", "David",
ADC Telecommunications | and "Katherine" versions of the human genotype.