Talk:Space fountain/Archive 1

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 1

Pellets

Is there any information on what size and material theorists image these pellets being? 119 09:27, 23 August 2005 (UTC)

Tower stability

How do you deal with wind on the atmospheric parts of the fountain?

How do you deal with the Coriolis effect? Won't this have the effect of slowly toppling the tower? njh Fri May 6 08:13 2005 (UTC)

I'd have to check the papers for the official compensation mechanisms, but as far as I can see, as long as the toppling time is much larger than the time it takes the pellets to make one cycle, you can apply a lateral force to the pellet stream to keep the station stabilized, and apply a corrective force to the stream at the ground station. The limit is that you can't deflect the stream too much without having to change the geometry of the ground station. In practice, you'd probably set up the stream geometry to let you compensate for known forces from the start (possibly resulting in widely-separated base stations for outgoing and incoming streams, depending on the tower configuration). The time scale for correction for ground-to-geosync is about 3 hours (more if your pellets are close to minimum speed, less if they aren't). Intermediate-level platforms shorten this timescale. If someone who's read the original papers on the fountain can describe how the authors planned to stabilize it, that would be very useful. --Christopher Thomas 13:15, 6 May 2005 (UTC)
Update - I've worked through some of the problem specifics, and it looks like you could stabilize a tower using just the top platform and the base station. The particle streams aren't fired pointing directly at the tower and the base station, instead following hyperbolic or elliptical arcs (free-fall paths influenced by Earth's mass, above or below escape velocity, respectively). In a non-rotating reference frame, the upwards stream looks like it's fired on a hyperbolic path starting at the ground station's position at time t1, ending at the platform's position as of time t2. The downward arc is similar, starting at the platform's position as of time t2 and ending at the ground station's position as of time t3 (different from the station's position at time t1). In a rotating reference frame in which the ground station and platform appear stationary, it looks like you're firing projectiles at an angle that exactly compensates for the warping caused by the Coriolis pseudoforce. There are many possible arcs that can connect the ground station and the platform, using different particle speeds; by varying the angle and speed, you can apply force to the platform in any desired direction in the plane of Earth's rotation. If you use two loops instead of one, you can correct in three dimensions (though the two ground stations need to have substantially different latitudes for this). This is mainly important if you're trying to stabilize a platform that's not quite in the equatorial plane. This was a fun problem to work through; thanks for bringing it to my attention :). --Christopher Thomas 22:12, 7 May 2005 (UTC)
There is a simpler solution. Lets assume that we have a cable that can guide these high velocity particles. Say the particles are magnetic and that the cable sheath uses the "null flux" guiding mechanism that is used in some maglev trains. The high speed particles means losses are low. The beam will then apply magnetic pressure on the cable walls thus guiding the beam and cable, and the cable is also under tension for stability reasons. But we can also have the return cable, that is the cable with the beam travelling in the opposite direction, mechanically attached to the other cable. So if we consider a small length of "rope" we have two sections of guide cable with the beams travelling in the opposite directions. In this case the Coriolis force will cancel out since in each cable the force from one beam is the same but in the opposite direction to the other beam.Delt0r 12:49, 27 September 2005 (UTC)

Minor Rewrite of Design Section?

"The projectiles reach the bottom of the tower with almost the same speed that they had when they were launched, losing a small amount of energy due to inefficiencies in the electromagnetic accelerators and decelerators in the tower."

Anyone feel like doing the math to prove only a minor loss? Pellet is launched, upward force applied to tower to slow pellet with drag, store energy for later. Use a large magnet to expel the last of pellet's kinetic energy into upward force for the tower. How can the downward acceleration on the pellet (and thus upwards on the tower) provide a power even close to equal at the bottom again? Won't you have massive inefficients in the pellet to electricity and then electricity to pellet transfers? Even with the use of superconductors? If not, this should be explained in a clearer fashion.--Talroth 04:34, 2 December 2006 (UTC)

I have a photocopy of the original Starbridge papers

Keith Lofstrom [1] here, inventor of the launch loop [2]. Seeing this article reminded me that I have a few file boxes of papers from the 70's and 80's, when all this stuff was new and a lot of enthusiasts were working on it. I can provide lots more information.

I have a bad photocopy of Rod Hyde et al's papers, the original 30 Jan 1981 paper and the 2 April 1982 addendum. I have quite a few other papers from the era, too. I am heading out of town for a week, but when I get back I can scan them and put the images up on the launch loop server.

The first published paper in the general area of momentum storage launchers (my term) or dynamic structures (Bob Forward's term) would be the Roger Arnold and Don Kingsbury articles about The Spaceport in the November and December 1979 issue of the Analog Science Fiction Magazine. This is an orbiting mass driver that captures payloads. Immediately upon seeing the November Analog, I went up to visit Roger Arnold in Kent, Washington - he was working at Boeing at the time. I had been working on a long "flying cable" (tether) system (I presented it at the Orycon 1 science fiction convention in 1979), and Roger's rather elaborate (and high gee) capture system intrigued me. We had a bit of an argument about "frozen spinach launchers", but parted the best of friends. I wonder where he is now?

The launch loop was born out of the flying cable - the problem with all these systems is stability and control, and I decided I needed a rigid fixed surface to work against for lateral stability. For example - if you are aiming flying rings at a cm-accuracy target 40,000 Km away, and transit time is an 5000 seconds, you need to measure lateral and rotational velocities within 2 micrometers/second (about one part in 2E-10 compared to breech velocity). If you are measuring between two points a kilometer apart, you need to measure displacements of 200 nanometers, and know with exteme precision where the datums are on every individual ring. Of course, with rings you need to control 6 velocities and 6 positions to this positional accuracy. And the rings that miss ... Yikes!

So, assuming a planet as a source of basic stability and reaction mass, and a grid of laser interferometers to make up for the fact that planets are actually rather mushy things (tides, earthquakes, and all that), and realizing that there were dozens of other problems with vertical launchers, I went horizontal.

The first thing I came up with was something like Ken Brakke's Skyrail (L5 News, July 1982) or Paul Birch's Orbital Ring (JBIS, Nov 1982) but I abandoned that in mid-1980 as having most of the same problems, as well as being impossible to build from the surface. Since I only needed a 2000Km launch path for 3-gee-to-escape, I chopped off the useless extra 38,000Km and made a closed loop. Hence, the launch loop. It evolved from there, and resulted in a submission to the AIAA Advanced Space Propulsion competition in April 1981. That was followed by a short paper in the November 1981 American Astronautical Society Reader's Forum, another short non-technical paper in the August 1992 L5 News (Ken and I learned about each other from this, and this let Rod Hyde know about both of us), and a presentation at the April 1993 L5 Conference in Houston. I presented just after Eric Drexler presented his first public paper on nanotechnology. We were both mobbed by enthusiastic fans in the hallway for hours after the presentations. That night, Eric and I watched the Soviet Mir space station pass overhead from the roof of the hotel, and discussed Ricardo's Law of Comparative Advantage in a world of big, slow humans and quick, cheap nanointelligences. I also met Bob Forward there, which began a long friendship. Robert Heinlein came to my room party ...

In May 1983, I met with Ken and Paul at the 6th Princeton Space Manufacturing Conference. None of our papers were accepted, but we took over a classroom for a couple of days afterwards to formalize some analysis conventions and do some calculus together. There we explored the "two stream instability", with Paul becoming convinced that we could never make it work. I argued that with damned good measurement ("many wavelengths of light"), and smart active control ("only 14 millimeters per microsecond"), and frequent (100 Km apart) cables to the ground to shed correction forces, taking out the instabilities was quite doable (what Bob Forward called "A mere engineering detail").

I had been working with Stan Schmidt at Analog since September 1982 on yet another short non-technical paper, which appeared in the December 1983 Analog magazine (my first paid sale of writing). This connected me to science fiction author Dean Ing, which connected me to Leik Myrabo, which finally got me a slot at the July 1985 AIAA Advanced Propulsion conference, where I could finally publish some damned equations!

This was all like pushing rope - work was getting busy (I was managing a design team), and I was getting approximately zero outside help. I came to the depressing conclusion that this wasn't going to happen in my lifetime unless I earned enough money to build the damned thing myself. Note, as of July 2005, I haven't made the money thing happen yet. Sigh.

- added by on User:Klofstrom 2005-07-11T21:38:04 (- (User) WolfKeeper (Talk) 23:07, 3 December 2007 (UTC))

Awesome! Thanks for all this detail, I'll try to work as much of it as I can into this and related articles (I wrote the original version of this based solely on Forward's article in Indistinguishable From Magic). Since Wikipedia's got a policy against original research that this first-hand account might fall under, would it be possible for you to copy and paste some of this information on the history of the concept's development onto your launch loop page? That would provide an external source that I could reference.
In return, I promise to pledge a portion of the vast profits I'll no doubt one day earn from my editing work here on Wikipedia to the construction of a space fountain. :) Bryan 03:05, 20 July 2005 (UTC)

The article is severly undermined by no mention of 'pellet' possibilities

I read the whole article expecting at some point to see what type of pellets were candidates... but found no mention. Without a discussion of potential projectile types it might as well be... 'space fountain will be held up by pixies'

Any discussion of projectile types would also need to mention 'wear and tear' and maintenance on the projectiles. If they are a beam of protons, say for example, then some mention of how they would be injected/removed as needed. Or are they proposed to be little ball bearings? I dont know... and I read the article. WalrusLike (talk) 21:29, 16 March 2008 (UTC)

They're pellets, like ball bearings or magnets. They have to be magnetic for the containment and momentum exchange to work.- (User) WolfKeeper (Talk) 21:45, 16 March 2008 (UTC)
Protons are charged and so can be manipulated magnetically.... but whatever.... the article is nonsense without mention of possibilities.... otherwise its held up by pixies. If you go 'ball bearings' then you need to consider impacts, wear and tear, dust, fragmentation, insertion and collection methods... a whole bunch of ignored issues.WalrusLike (talk) 10:59, 23 March 2008 (UTC)
Protons have a charge/mass ratio that is far, far too high. In other words they would repel each other, and would be impossible to constrain to transfer momentum around. They're also in no way, shape or form, 'pellets'.- (User) WolfKeeper (Talk) 19:18, 23 March 2008 (UTC)

Paul Birch hyperlink

The Paul Birch hyperlink is to basketball player Paul Birch. This is clearly the wrong Paul Birch. --- Well, I have removed that hyperlink; there is currently no entry for Paul Birch (which is a shame).Eburacum87.102.28.2 12:24, 2 February 2007 (UTC)

I connected it to the now extant Paul Birch page.- (User) WolfKeeper (Talk) 19:25, 23 March 2008 (UTC)

Criticism

What is the value of achieving high altitudes at *suborbital* velocities?

What prevents the "turnaround" magnetic field from coupling with the Earth's magnetic field and subjecting the station to a translational or rotational force?

What prevents the station's "turnaround" field, when coupling with the particle stream, from generating torque instead of fully bending the stream? Isn't the station simply a commutator in a giant electric motor? —Preceding unsigned comment added by 72.179.47.11 (talk) 14:43, 28 January 2009 (UTC)

Foundation

I have been reading on the available material on the launch loop, and notice that no reference is made to the foundations anywhere. As I understand it, the flow of the pellets inside the launch loop, will prevent the structural components from crumbling under the the weight of the components stacked on top of them as it will effectively neutralize the weight of the structures above it. However, ultimately, this weight will have to be transferred in its entirety to the supporting ground. It seems to me that by that time, an enormous load needs to be transferred. The only solution I can think of is having multiple layers of "active towers" of similar design as the launch loop, transmitting the load downwards. Each layer would have an increasing number of towers until such a point as the stresses transferred to any foundation material would be sufficiently low for those to support them. This requirement of a pyramidal system of support structures could turn out to be a substantial cost addition. Has any work on this been done? —Preceding unsigned comment added by Juaninse (talkcontribs) 21:49, 20 February 2008 (UTC)

No, there's no massive load at any point; due to the curvature of the launch sections it works out at about 1-2 tons per meter; which is well within the strength capabilities of foundations. Basically, by changing the curvature you can adjust the stresses acting downwards, you can spread them over a long distance. In addition you can choose the footprint over which the 1 tonne force can be applied- if you pour concrete you can make the load spread out sideways.- (User) WolfKeeper (Talk) 22:43, 20 February 2008 (UTC)
Thank you for the clarification. In that case, maybe that part needs to be addressed in the article, to help clear all misunderstandings. —Preceding unsigned comment added by Juaninse (talkcontribs) 03:32, 22 February 2008 (UTC)

Reentry?

Things burn up in reentry because they are moving at super sonic speeds relative to the atmosphere and the slowing down to terminal velocity (if they do not burn up first). Parts falling from a critical failure of a space fountain would fall like anything else reach terminal velocity after speeding up from zero. No part of the fountain would burn up. 74.194.24.44 (talk) 03:23, 14 December 2009 (UTC)

Popular culture reference

Do we really need a "popular culture" section for this article? There's only the one reference to today's xkcd comic, which is more or less a passing mention and not immediately visible at that. I'm not removing the section myself as I just semi-protected the page, but I don't think it really needs to be there. Hersfold (t/a/c) 19:25, 28 January 2009 (UTC)

That reference is why I hate wikipedia. —Preceding unsigned comment added by 192.12.184.2 (talk) 19:51, 28 January 2009 (UTC)

I think that the very fact that the page has undergone numerous edit/unedit actions with regards to the mention on xkcd proves that it is in fact a popular culture item, and should be allowed to remain as such. --Webmaren (talk) 21:09, 28 January 2009 (UTC)

No, the reference should not be there. It is certainly not worth mentioning because it adds no value to the article and is barely related. Please remove. 163.181.251.10 (talk) 21:42, 28 January 2009 (UTC)

It's pointless promotion of xkcd. I love the comic and read it every other day as soon as it's released, but it's not a notable enough mention to merit being mentioned. Should we also add an "In popular culture" to the Stick figure article? How about AIDS? Or ASUS Eee PC. We can't simply form wikipedia around the content of webcomics.
As for Webmaren's comment, it proves nothing of the sort. All it proves is that there are enough fanboys out there that see Wikipedia as some sort of playground/advertising board and not a trusted source of information. Greggers (tc) 22:15, 28 January 2009 (UTC)
Actually, all it proves is that Randall M. basically told all his readers to go look up Space Fountain on Wikipedia today. For that reason, it's probably important enough right now simply because it explains the article locking and the fact that the world's eyes may be on the concept more today than usual. It's a culturally relevant current event, but it won't be sufficiently important as an historic one. Remove the mention of xkcd once Friday's comic replaces the current one. 24.85.145.58 (talk) 22:44, 28 January 2009 (UTC)
I agree with this. I don't think the fact that space elevators were mentioned in the alt text of a particular webcomic strip is worthy of note. That is, I can't see it being of interest to anyone a year from now. If it were the main subject of a particular popular xkcd strip, like sudo, then maybe ought to mention it.
But it's true that the reference might serve as an explanation for why the subject is currently popular and the article locked, and I suppose there's no harm in leaving it temporarily for that reason. Electrolite (talk) 23:09, 28 January 2009 (UTC)
For better or for worse, xkcd seems to be one of the world leaders in driving suddenly spikes in bizarre search terms and spontaneous Wikipedia vandalism; if someone googles "space fountain" in an attempt to find out why it's suddenly topping the hot search terms, Wikipedia should have their answer. DustFormsWords (talk) 23:01, 28 January 2009 (UTC)


The funny thing is xkcd has parodied the very fact that "in popular culture" sections are tacky. Now what? Does xkcd imply that xkcd shouldn't be included? If so, then why were you listening to what a webcomic had to say in the first place? Wokka-wokka-wokka. --Bobak (talk) 01:17, 29 January 2009 (UTC)

We could toss it in the xkcd article 72.73.238.150 (talk) 04:39, 29 January 2009 (UTC)

The fact that it shows up in the alt-text does make it relevant and proves that it is worthy of being searched. Most of the time, the alt-text is punch line of the joke that the comic is about. In this case, Space Fountain actually was more humorous. I'm not sure why the people who maintain this article care, because to me it just means more people are looking at it. Just because you get credited in some online web-comic doesn't vandalize the site. Just simply means that your topic is now more publicly known. A lot of people read xkcd... the "owners" of this page should be thrilled! Iheartxkcd (talk) 18:43, 29 January 2009 (UTC)

AS SEEN ON
xkcd

I don't wish to start some sort of argument, Iheartxkcd, but I thought I'd pick you up on a couple of points:

  • The fact that more people are reading the article makes no difference to us. The fact that it provides those who do with a reliable source of information warms our hearts.
  • Whilst this topic was the subject of the joke, you can't simply tag every article in Wikipedia with a big ol' "As seen on xkcd" sticker.
  • Nobody "owns" Wikipedia articles. Except Jimbo Wales, who we all obviously look up to like some sort of god. Greggers (tc) 19:52, 29 January 2009 (UTC)
You can't tag every article in Wikipedia with an "As seen on xkcd" sticker, but you could make a page titled "List of Wikipedia pages featured on xkcd." Then, when Randal points the teeming masses to a page, you could simply put a link to the list at the top rather than having different sets of Wikipedia editors on different Wikipedia pages dealing with the same issue over and over. Guy Macon 15:33, 28 February 2011 (UTC)

I don't know I like it. -134.50.203.20 (talk) 23:38, 29 January 2009 (UTC)

Pointless comment -> It's not the 'alt' attribute. It's the 'title' attribute. :) --Ajwittenburg (talk) 21:57, 30 January 2009 (UTC)

Why not in service already?

The tone of the article suggests these things should be in service already. There should be some discussion of why they're not. Are the obstacles technical, political, or something else? --Vaughan Pratt (talk) 07:22, 29 October 2011 (UTC)

Lateral force?

Does the top bend also creates a lateral force on the tower? --TiagoTiago (talk) 16:00, 7 November 2011 (UTC)

Change in lead

I changed this section: "...also continually threatens to cause instant, catastrophic destruction if the containment systems fail"

To this: "...also continually threatens to cause the collapse of the tower if the containment systems fail"

As a) the body claimed a collapse would not be instant and b) "collapse of the tower" is a more factual and straightforward description of what would happen than "catastrophic destruction". — Preceding unsigned comment added by 86.167.228.19 (talk) 12:38, 28 May 2013 (UTC)

Penis

Does anybody else think the illustration of the Hyde design for the tower in the article looks suspiciously like the outline of a penis & scrotum, & that that's just perfect, considering the whole idea? RubyQ (talk) 14:06, 14 October 2016 (UTC)

Bose Einstein what now?

I just deleted these sentences:

Another option for space fountain construction would be a Bose–Einstein condensate. Lasers simulate a magnetic field in a BEC, which might be ideal. Instead of using pellets, a laser is used.

This assertion is unsourced and doesn't make any sense to me. Whoever wrote this seems to have thought that using a BEC ... somehow ... would enable the pellets to be replaced with a laser, but I don't see how, and anyway BECs can only exist in tiny volumes at temperatures close to absolute zero, which isn't what we have in a space fountain.

--Elwoz (talk) 03:42, 1 December 2017 (UTC)

Picture of the Space Fountain

Request: additional diagrams depicting initial activation, stages of erection, as well as deactivation, and return to flaccid state. —Preceding unsigned comment added by 97.115.143.18 (talk) 20:45, 6 January 2011 (UTC)

I'm wondering if the picture in this article was meant as a joke or if it's just accidentally suggestive. —Preceding unsigned comment added by Robobogle (talkcontribs) 21:38, 28 January 2009 (UTC)

This is exactly why I checked the talk page on this. Is there any reason whatsoever that the space fountain would have a scrotum? --98.216.50.36 (talk) 06:04, 29 October 2010 (UTC)

Mentioned on Reddit: http://www.reddit.com/r/wikipedia/comments/ftdyr/well_thats_quite_a_diagram/ Guy Macon 15:20, 27 February 2011 (UTC)

That image is 100% an intentional penis of exact proportions. 76.181.233.121 (talk) 10:37, 2 March 2016 (UTC)

"A thing's a phallic symbol if it's longer than it's wide". Last stanza of [[ http://www.metrolyrics.com/psychotherapy-lyrics-melanie.html Psychotherapy]] by folksinger Melanie Safka. KeithLofstrom (talk) 03:26, 6 May 2018 (UTC)

Not fictional?

GliderMaven suggested in a recent edit summary that this thing is not fictional. If it actually exists, wouldn't we have heard about it? Wouldn't it be a good idea to say in the article where on Earth it is (with some reliable references – of which, oddly enough, there don't seem to be any). On the other hand, if it's just some silly made-up nonsense, is there any reason to have an article on it? Justlettersandnumbers (talk) 22:13, 14 December 2018 (UTC)

You're being a WP:DICK. By that logic I'd like to know where a circle is. An actual circle. Where the radius is exactly constant at every point in its infinitesimally thin line. If not, if nobody can point to one, then clearly the circle is fictional, and (sic) by your argument, wikipedia is not allowed to contain fiction???
None of that is how Wikipedia works. Space fountains aren't fictional, they're proposed real structures and there's a reference in the article, and anyway, even if it was entirely fictional Wikipedia even allows articles on fictional objects and ideas. Seriously, stop being such a dick. GliderMaven (talk) 22:46, 14 December 2018 (UTC)
Stop redirecting this to random other topics please. This concept is neither a space elevator nor an orbital ring. A redirect that leads readers nowhere is worse than no article at all. If you think this article shouldn't exist then nominate it for deletion but don't make these silly redirects. --mfb (talk) 23:49, 14 December 2018 (UTC)
Delighted to oblige: Wikipedia:Articles for deletion/Space fountain. Justlettersandnumbers (talk) 00:04, 15 December 2018 (UTC)
The concept is not fictional in the sense that it is a serious proposal for space propulsion. It is being taken seriously by scientists. It is mentioned in the abstract of this paper from Acta Astronautica for insance. It is discussed in a bit more detail in this paper from Journal of Propulsion and Power but it's behind a paywall. It's covered in a number of book sources [3][4][5]. I didn't get a chance to comment at the AFD, but there's definitely enough RS out there to support a standalone article. SpinningSpark 23:38, 3 January 2019 (UTC)

Efficiency?

If a space fountain needs a devices that shoots pellets straight up to the end of the tower, wouldn't it be easier to just use that device and no tower to lift cargo into space? What's the benefit of "adding a tower" to the pellet shooting machine? Is it efficient? How fast would those pellets have to be at the start for the whole process to work and some reasonable height? Paede 11:58, 10 Apr 2005 (UTC)

In theory, the starting speed is as low as you like (for as short a tower as you like). In practice, the bending magnets or what-have-you will likely assume a certain speed range for pellets being deflected, meaning that to inject new pellets into the system, you'll need a gun adjacent to the main deflecter that accelerates them to a few km/sec, to replace pellets that have been lost from an already-working system. This makes initial startup a bit trickier (you might do it by slowly populating the loop with full-speed pellets until the momentum transfer is enough to lift the tower top off of the construction framework). As for why you'd have cargo ride the stream instead of being the stream itself, there are several reasons. If pellet speed his higher than escape velocity (allowing an arbitrarily high platform), the accelerating forces felt when initially injecting the pellet and when deflecting at each end are immense (think "thousands of gravities"). While you could launch raw materials that way, you probably wouldn't want to launch most other types of cargo. Riding the stream, cargo can be launched at any acceleration by varying the strength with which it couples to the stream. Individual pellets are also likely small, to make injection more practical. Cargo riding the beam can be heavier; while it needs the same amount of energy imparted to reach a given height or a given final velocity, this energy can be put in over a longer time, for lower power required. Lastly, if the platform is relatively low (e.g., outside the atmosphere, but far below geosynch or possibly even LEO), the pellet stream can move at a speed well below orbital speed (the platform isn't in orbit; it's stationary relative to the surface, supported by the pressure required to deflect the stream). Launching cargo on this trajectory would just cause it to fall back to earth. A platform at or below LEO altitudes would be useful for tourism, observatories, or even another launcher (using fountains as support pylons for a Lofstrom launch loop's housing, or for a very long fragile-cargo magnetic accelerator). In short, the accelerator mechanisms are only good for one type of cargo being sent in one direction (small pellets that can withstand thousands of gravities of acceleration, being launched nearly vertically at or above escape velocity), while a fountain allows many different types of cargo to be sent on different types of path, depending on how the fountain is configured and what's on top. --Christopher Thomas 13:35, 10 Apr 2005 (UTC)
Also, the tower provides a tube of vacuum for the pellets to travel through so there's no problems with air resistance. Launching cargo straight from Earth's surface would require it to be encased in a lot of heat shielding, since it would go up like a meteor in reverse, and would produce powerful sonic booms that could have a nasty environmental impact. You'd also need a rocket on board to circularize the package's orbit, which could get expensive if you're launching thousands of packages (a low-orbit space elevator would need to provide transverse thrust for packages leaving it too, but this could be done with reusable tugs or with an accelerator on board the top station. Bryan 16:40, 10 Apr 2005 (UTC)

Thanks for your replies. I understand the sonic boom problem, but this could be already be solved with a 30km or 40km high tower - above it, there is virtually no air resistance anyway. My problem still remains: A pellet-shooting machine that can shoot stuff thousands of km up is an invention that is quite difficult to achieve. Isn't the whole concept of "Space Fountains" flawed because the pellet-shooting machine would make 95% of the encountered difficulties, and once you've got a working pellet shooting machine, there is no need for the surrounding tower above 20km? This is still science, not pure fiction, right? Paede 03:35, 12 Apr 2005 (UTC)

  • Check my reply again. A gun-type launcher has strong limits on what you can do with it, that are relaxed if you instead put a platform at the top and more slowly lift objects to to the platform by coupling them to the fast-moving stream. This is why the idea of a space fountain (or launch loop or space elevator, for that matter) is proposed. You correctly note that you don't need a physical shroud around the fountain once you're above the atmosphere, but that has little relation to the question of why you'd use a fountain in the first place.

    As for building the gun, light gas guns that presently exist can already launch projectiles at the required speeds. See space gun, supergun, and light gas gun for more information. In practice, something resembling a coilgun would be more likely to be used as an injector, due to higher rate of fire. The main engineering challenge for building a space fountain is likely to be the bending magnets that redirect the stream of pellets at the ends of the loop, not the pellet injector.--Christopher Thomas 19:49, 12 Apr 2005 (UTC)

    • Thanks, that clarifies everything! Paede 00:44, 13 Apr 2005 (UTC)
    • A "supergun" is not a coherent concept in the real world outwith science fiction and video games, as it is just a vague and woolly addition of "super-". large-calibre artillery is, however. Uncle G (talk) 09:36, 15 August 2019 (UTC)

zero velocity relative to the ground?

Re: "One downside of the space fountain is that it does not provide orbital speed on its own; payloads released from the top have zero velocity relative to the ground", geosynchronous satellites have zero velocity relative to the ground and they say in orbit just fine. --Guy Macon (talk) 17:34, 27 March 2021 (UTC)

I've removed the sentence. It was originally added without a source here, and the only potential source doesn't mention that as a downside. –dlthewave 18:26, 27 March 2021 (UTC)
If you go to GEO you lose the cited advantage that it can be much shorter than a space elevator (2000 km and 40-200 km are discussed as examples in the references). You can't have both. Reference 1 explicitly mentions that you still need to accelerate the payload in case of the shorter options. But I don't care enough about that sentence to start another discussion. --mfb (talk) 12:05, 28 March 2021 (UTC)