Talk:Space elevator

Graphs of optimal profiles

Someone anonymously deleted two graphs which I added that show the optimal cross-section profile for two different values of specific strength. These graphs are not original research. They are based on the equations that were already in the article, and which are referenced. The anonymous editor wrote as an edit comment,

"Undid the addition of a large amount of WP:OR about FBL. Much of it is dubious such as the required FBL being related to planet radius, and the misleading use (in the graphs) of material with specific strength corresponding to FBL of one earth radius, as if that was significant, which is isn't. Undue weight on FBL and incorrect info about it at that."

FBL apparently means "free breaking length". The text, before my edit, already explained that the free breaking length has to be on the order of the Earth's radius, so that's what I used for the first graph. For the second, I used a free breaking length ten times shorter, as was already in the text. The graphs show that what the text said is correct -- if the free breaking length is similar to the earth's radius then the idea is feasible, but if it's one-tenth then the idea becomes ridiculous. I am reverting his reversion, and if he wants to dispute what I say, let him do it here rather than starting an edit war. Eric Kvaalen (talk) 09:21, 28 January 2019 (UTC)

"...are not original research. They are based on the equations...". That's WP:OR. As for the free breaking length having to be on the order of earth's radius, that's meaningless. The only meaningful distance is the 36,000 km geosync radius.

I removed your changes, per WP:BRD, when controversial matter is added and reverted, it is not added back in until discussion achieves consensus. Reinstating it is simply edit warring. Tarl N. (discuss) 18:16, 28 January 2019 (UTC)

I left a message on the IP's talk page, hopefully they'll see it and participate in this discussion. Tarl N. (discuss) 18:20, 28 January 2019 (UTC)

A more specific complaint about your edit would be WP:SYNTH (which is a part of OR). Specifically, Do not combine material from multiple sources to reach or imply a conclusion not explicitly stated by any of the sources. Certainly any conclusion about the free breaking length in relation to earth's radius qualifies as synthesis, because I haven't seen that in any reference cited here. Tarl N. (discuss) 19:06, 28 January 2019 (UTC)

I agree, Tarl N.. One or two things Eric Kvaalen added, such as minor grammar edits for flow, were not dubious, but they were bundled in with a lot of other stuff that was. 40.117.60.104 (talk) 01:45, 30 January 2019 (UTC)

@Tarl N.: I don't see why you call making graphs from known equations "original research"!

Are you interested enough in this subject to discuss with me whether "the free breaking length having to be on the order of the earth's radius is meaningless", or are you only interested in debating whether making graphs is original research? If you are interested, we can discuss the question. (By the way, as I said, that stuff was already in the article. I did not add it.) If you think it's inappropriate to discuss this on a talk page, then we can discuss it on my user page, or even by e-mail, but I think it would be better to discuss it here where other interested persons can read it.

Eric Kvaalen (talk) 13:21, 31 January 2019 (UTC)

Per WP:CALC: Routine calculations do not count as original research, provided there is consensus among editors that the result of the calculation is obvious, correct, and a meaningful reflection of the sources. Basic arithmetic, such as adding numbers, converting units, or calculating a person's age are some examples of routine calculations.. What you are doing is well beyond that description.

To paraphrase a certain movie "show me the sources!" If you have a source that says anything about the radius of the earth (as such!) relating to the free breaking length, we can look at that. If you came to a conclusion that 6600 km (or something) for some reason is meaningful, unless that is because of the distance from center of the earth to the surface, then that distance is completely unrelated to the earth's radius and using that term is providing a misleading implication. Either way, you are engaging in WP:SYNTH unless you have a source that explicitly says something about the radius of the earth relating to the free breaking length.

We've butted heads before; it's not clear a discussion is going to be worthwhile. My view is that you have a tendency to misunderstand what you read on Wikipedia and build on your misunderstanding. The only way to overcome that is to provide sources for your assertions. Tarl N. (discuss) 13:35, 31 January 2019 (UTC)

EK, The "known equations" are just there to satisfy the pedants who like showing off how much they know about equations. It's not actually that critical to describe the largely-indiscernible detail that's already there, let alone the (clearly lovingly prepared I'll credit you) detail given in the graphs. This isn't a book about space elevators. It's enough to say "tapering helps a lot because X" IMHO. The equations themselves are just an idealized starting point for a design anyway.
Free breaking length is the wrong parameter to use in calculations or to dwell on for very long. It's really only good for illustrating the concept of high specific strength from a ropemaker's mentality. Each term can be converted to the other, but specific strength is what's used. FBL suffers from needing to be "translated" between a constant-g field for which the parameter applies and the g-proportional-to-(1/r^2)-compensated-for-centrifugal actual effective field. 40.117.60.104 (talk) 01:58, 1 February 2019 (UTC)

Space elevator optimal cross section profile for characteristic length equal to Earth's radius. The top of the curve corresponds to the geostationary altitude. The area under the curve gives the ratio of the total mass of the cable to the mass that it is able to lift from the earth's surface.NOTE: This graph has been removed from the article for being unref'd original research. It also implies a one-to-one relationship between required free breaking length and planet radius, which is incorrect. 40.117.60.104 (talk) 07:49, 2 February 2019 (UTC)

Space elevator optimal cross section profile for characteristic length equal to 1/10 of Earth's radius. Notice that the area under the curve is now over 100,000.NOTE: This graph and its caption has been removed from the article for being unref'd original research. 40.117.60.104 (talk) 07:49, 2 February 2019 (UTC)

@Tarl N.: I'm putting the graphs here to aid the discussion. Lots of people make graphs and put them on Wikipedia. These graphs are elementary computations. For instance, in the first graph, the value at altitude/Lc=5 is given by

$\exp(1/1*(1+0.00346/2-1/6-0.00346/2*6^{2}))$

Please check that this indeed equals about 2.17 as shown. I think we can trust Excel actually. Shall I send you the Excel file?

You seem to believe that I am the one who put in that stuff about the earth's diameter. You're beatin' the wrong horse! But he happens to be right, as I'm sure I can convince you if you are willing to listen.

I don't have the same recollection of our previous exchange as you.

I also disagree with Mr Anonymous. These equations are very important, for they can be used to find the minimum mass of the space elevator in order to lift a given mass, using a cable of a given specific strength.

Eric Kvaalen (talk) 19:47, 1 February 2019 (UTC)

The point is that our task on Wikipedia is not to read stuff placed here and further elaborate on it. Our task is to find material published in reliable sources, and describe it here, making clear where it came from. This article in particular is a nightmare of uncited gobbledygook, to the degree that very little of it is verifiable. The improvement this article needs is citations for all the assertions made. At least some of which are probably simply false.

As for your calculations, please read WP:CALC again: Routine calculations do not count as original research, provided there is consensus among editors that the result of the calculation is obvious. Why you are coming up with characteristic length in terms of earth's radius is far from obvious, and well into WP:SYNTH. If you claim that assertion was already there, and you don't understand how it got there, that itself is a problem. It should not be further propagated without a citation. Tarl N. (discuss) 01:37, 2 February 2019 (UTC)

@Tarl N.: So you say that $\exp(1/1*(1+0.00346/2-1/6-0.00346/2*6^{2}))$ is not a routine calculation? I do not suppose that you can do that calculation in your head, so in that sense it's not "obvious", but still, do you not have a calculator? Eric Kvaalen (talk) 07:05, 2 February 2019 (UTC)

It's time to knock it off, Kvaalen. You're clearly thrilled about the work you've done and you want everyone to see how wonderful it is. But, anyone can see the graphs on old versions of the article, that's not the problem. No need to plaster it here. Plastering it here adds nothing except to show off some clean artwork (but bad information) and to drag out a lost argument. You're playing WP:ICANTHEARYOU, which is disruptive. We've said over and over that the first chart misleads because it suggests an incorrect relationship to planet radius. Pay attention. Stop your stupid quibbling. Cut your losses and go home. 40.117.60.104 (talk) 07:30, 2 February 2019 (UTC)

I reduced the size of the bad graphs so as to not imply undue legitimacy. They're still zoomable for close examination. 40.117.60.104 (talk) 08:29, 2 February 2019 (UTC)

The discussion at WP:DRN is closed as failed - I wasn't aware that stating I didn't think the dispute was amendable to mediation would close it, but in retrospect, it makes sense. The options remaining, as I understand it, are WP:3O and WP:RFC. And, of course, WP:DISENGAGE. Tarl N. (discuss) 23:12, 4 February 2019 (UTC)

Earth's radius

Sorry for my late arrival ... let's straighten out some facts first. Biem (talk) 07:58, 8 February 2019 (UTC)

On of the comments above has been "The only meaningful distance is the 36,000 km geosync radius" : I can't agree with that. The reason is the last equation in Space elevator#Cable section section, that states :

$\Delta \left[\ln(S)\right]={\frac {\rho }{\sigma }}g_{0}r_{0}\left(1+{\frac {x}{2}}-{\frac {3}{2}}x^{\frac {1}{3}}\right)$

Though the notations are not the same ones, this last equation is documented by ref_40, "The physics of the space elevator", but incorrectly refers to "equation 7" where it really should be "equation 8", the next one. I do agree that ""Routine calculations do not count as original research, provided there is consensus among editors that the result of the calculation is obvious"", and the notational changes should be obvious enough.

The equation states that the differences in section can be expressed by a constant factor ($\scriptstyle {\frac {\rho }{\sigma }}g_{0}r_{0}$) multiplied by a correction factor in x (1+x/2+...), meaning that to the first order of approximation, the relevant factor is the constant. It mentions both r₀ (Earth radius) and r₁ (geosync radius) in its left side, but on the right side the geosync radius has disappeared. The constant factor only depends on the earth's radius, and other factors such as rho, sigma and g.

The only "traduction" has been to replace the geosync radius (which is a computed value) by its value, once again given by ref_40, which eliminates the expression of geosync radius and expresses the x correction factor directly with known constants - ground gravity, earth radius and and rotational speed. This is basic algebra, "routine calculation", a variable substitution that allows a change of focus : the x represents the fraction of ground gravity that is relieved by rotation on the equator, meaning that it is really a small factor most of the time.

So in fact, the equation means that "the only meaningful distance is the planet radius and its ground gravity" as far as diameter variations are considered.

That way of seeing things is essential when looking at other planets and moons ; it explains why, for instance, it is much easier to make a space elevator on the Moon than it is on Earth. This is why the formula is expressed with respect to x instead of geosync radius : comparaison is easier that way, which is usefull for comparaisons in section "Extraterrestrial elevators". And actually, Earth is the worst case of all telluric planets.

Biem (talk) 07:58, 8 February 2019 (UTC)

Free breaking length

Free breaking length is defined as the maximal length for which a cable can sustain its own weigh, under constant gravity. It is also mentioned in ref_40 equation_8, under the name "characteristic length", which has the same definition. Equation_8 shows that the taper ratio directly depends of the ratio between free breaking length and Earth radius.

The equation obviously shows that when the free breaking length equals the Earth radius, their ratio reduces to one, only the correction factor survives, therefore the taper ration ~ e = 2.71828... Which is straightforward.

So then, " If you have a source that says anything about the radius of the earth (as such!) relating to the free breaking length, we can look at that." - Well, that's the one, where Earth's radius is explicitly stated and related to the free breaking length. As you can see, there is no original research in that part. So please, do look at that, and judge the graphs according to their straightforward dependency to that documented equation.

Biem (talk) 07:58, 8 February 2019 (UTC)

OK, well, if you want real OR you may consider this : "when the free breaking length is such as $\scriptstyle L_{0}>r_{0}(1+{\frac {x}{2}}-{\frac {3}{2}}x^{\frac {1}{3}})$ (ie, Earth radius times correction), then a cylindrical cable reaching the ground from geosynchronical orbit can sustain its own weigh" (that is to say, beyond that resistance, profile optimisation is not needed any more). Funny thing, when you come to think of it : the cable strength needed to go from geosync to ground with a cylindrical cable is actually a free breaking length of Earth's radius, to a small factor approximation. This is because the important technical factor is ground gravity, not directly rotation speed. But I haven't seen anything like that stated clearly anywhere. This needs doing some maths (OR, OK) that will be left as an exercice. Biem (talk) 10:38, 8 February 2019 (UTC)

Is a graphic representation OR ?

Basically, the graphic is a plot of excel computations of the formula, for various altitudes :

$\Delta \left[\ln(S)\right]_{r_{0}}^{r_{1}}=-{\frac {\rho }{\sigma }}\Delta \left[{\frac {GM}{r}}+{\frac {\omega ^{2}r^{2}}{2}}\right]_{r_{0}}^{r_{1}}$

The interesting point being to show the effect of various material characteristics, synthesized as their free-breaking length. So, basically, to me :

• Plotting the graphic for a given ρ/σ value is straightforward and cannot be qualified as original research.

Which values should be taken? this is suggeted by the other equation :

$\Delta \left[\ln(S)\right]={\frac {r_{0}}{L_{0}}}\left(1+{\frac {x}{2}}-{\frac {3}{2}}x^{\frac {1}{3}}\right)$

which shows that the taper ratio varies eponentially with the r0/L0 ratio. If this ratio is lower than one, the taper ratio is not significantly dependant of the tensile resistance ; otherwise it varies exponentially and can attain hudge values.

• Choosing r0/L0 = 1 and 10 are logical choices to illustrate the point that just one order of magnitude actually makes a huge difference.

Biem (talk) 10:27, 8 February 2019 (UTC)

Request for comment

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

Should the graphs which I tried to add be allowed? See Talk:Space elevator#Graphs of optimal profiles. Eric Kvaalen (talk) 09:23, 6 February 2019 (UTC)

• Oppose No is the short answer. They are original research, not backed up by external review or taken from established and peer reviewed source material. I add maps, diagram and chart right left and centre. I'm doing diagrams for the Red Orchestra groups at the moment, about seven of them, but not a single of them is orginal research. All of them are established groups that have been written about extensively for over 70 years and are well documented. So when I create the charts they will be exactly as written. I would never consider creating something on my own and publishing it. It is not possible and not acceptable by consensus. scope_creep^Talk 11:17, 6 February 2019 (UTC)
• Comment If there is dodgy charts on this article, they should be removed. scope_creep^Talk 11:20, 6 February 2019 (UTC)

scope_creep, how is it original research? The formula I graphed is in the references as well as in the article. In what way does a graph need to be "backed up by external review" or "taken from peer reviewed source material"? As I have said above, lots of people make graphs of functions and add them to Wikipedia articles. Eric Kvaalen (talk) 14:23, 6 February 2019 (UTC)

• Allow per comments above. The statements that the formulas are OR is erroneous (see discussion above, references is equation 8 given in the article Aravind, P. K. (2007). "The physics of the space elevator" (PDF). American Journal of Physics. 45 (2). American Association of Physics Teachers: 125. Bibcode:2007AmJPh..75..125A. doi:10.1119/1.2404957.), & the graphical work cannot be considered as OR. A choice of parameter "Free Breaking Length = Earth radius" is natural & is the one relevant to illustrate the point, so the graphics are OK. Biem (talk) 10:42, 8 February 2019 (UTC)
• Change : Furthermore both graphics are useful to make the reader understand how a small variation on tensile strength can have huge consequences on the cable feasibility. But the graphics should really be limited from ground to geosync, there is no need to pull them further since the length beyond geosync is arbitrary and the real problem is the minimal diameter variation to get up there (beyond that the mass can be said to be "the same order of magnitude", which is quite enough to clarify the point). Biem (talk) 11:52, 8 February 2019 (UTC)

Thanks, Biem, for the support. I do think it's useful to extend the graph beyond the geostationary altitude, because by using the optimal profile for quite a distance above that the total mass will be less. If one were to put a counterweight just above the geostationary altitude it would have to have a very large mass, going to infinity as the distance above geostationary goes to zero (because the centrifugal force minus the gravitational force goes to zero in that case). If one continues with the optimal profile a thousand miles higher and then puts a counterweight, the total mass will be less. If one carries on 10,000 miles, it will be even less, and so on. The minimal mass is approached as the cable (or thread) goes ever higher and higher, following the optimal profile, before a counterweight is used. The total mass is a very important factor, because even for the case where the free breaking length equals the radius of the earth, the total mass will be at least about 30 times the mass which one desires to lift, and when the free breaking length is one tenth the radius of the earth, it's tens of thousands of times more. The integral under the whole optimal profile gives the minimal mass, so I think it's better to have graphs which go out to where the cross section becomes small and the rest of the integral towards infinity is negligible. Eric Kvaalen (talk) 13:20, 8 February 2019 (UTC)

OK, if you want to reason on the total mass involved, that is a correct approach. Biem (talk) 15:33, 8 February 2019 (UTC)

• Oppose As per previous discussions. The free breaking length and radius of the earth are not meaningful comparisons. Without an WP:RS, that entire section should probably be excised. Tarl N. (discuss) 21:44, 8 February 2019 (UTC)

Tarl, Biem has gone ahead and explained why the radius of the earth is relevant (as I proposed to do but you declined to hear it). See #Earth's radius. Meanwhile I have noticed that Arthur C. Clarke, (1979) "The space elevator: 'thought experiment', or key to the universe?" makes this point. Eric Kvaalen (talk) 09:05, 9 February 2019 (UTC)

Clark talks about 4960 km for his "escape length" figure, not earth's radius. That's the key point - using the radius itself is absolutely meaningless in this context, and misleading in that it implies the size of the planet is somehow itself involved. But I've spent so much time arguing in past interactions with you about stuff you clearly don't understand, that at this point, unless you have a reliable source which says what you intend to put in the article, I'm not interested. Tarl N. (discuss) 09:17, 9 February 2019 (UTC)

No it's not meaningless, as Biem has explained. Anyway, you can't complain about the parameters I use in my graphs. If they happen to show that the mass becomes enormous if the FBL is 1/10 the earth's radius, but not if it equals the earth's radius, well, that's not my fault. We editors are not obliged to hide things from the readers if they can't be found in the literature (though as I say, Arthur C. Clarke says basically the same thing). Eric Kvaalen (talk) 09:26, 9 February 2019 (UTC)

There is no point in saying that "Clark has not mentioned the fact" (who cares) the point is that Earth's radius is explicit in the reference mentioned : in equation 8, R is the Earth's radius, and Lc is the free breaking length. Clearly you have not read the section #Earth's radius above, or have not understood it. Your saying that "The free breaking length and radius of the earth are not meaningful comparisons" is blatantly false, obviously you have not seen the meaning of the $\scriptstyle {\frac {\rho }{\sigma }}g_{0}r_{0}$ factor. Hint : the $r_{0}$ in that formula is Earth's radius. You could do exactly the same maths for the planet Mars or Venus, and come to the same conclusion : the resistance needed for a cable on Mars is that which gives it a free breaking length of Mars' radius, under Mars' ground gravity. Yes, it implies that the size of the planet is somehow itself involved, just face it, there is no rejecting it by talking nonsense "about stuff you clearly don't understand". Biem (talk) 09:43, 9 February 2019 (UTC)

Your "equation 8" doesn't say anything about minimumFBL=planetRadius. Doesn't even solve for minFBL. Citing it for that is total bunk. 40.117.60.104 (talk) 01:25, 10 February 2019 (UTC)

First of all, he didn't say "minimum" FBL -- the question is whether the earth's radius is relevant. Second, it's not bunk at all. What he's saying is that the equation says that the ratio of earth radius to FBL enters into the equation in an exponential. So it's very important. If that ratio goes up to 10, say, then the mass of the contraption becomes ridiculously high. Eric Kvaalen (talk) 06:51, 10 February 2019 (UTC)

Nice. You're very smart. To bad you're not published in a reliable ref. 40.117.60.104 (talk) 10:43, 10 February 2019 (UTC)

Arguing on this is exactly why we have a prohibition on WP:OR. Find me a WP:RS that talks about space elevator materials in terms of free breaking length and Earth's radius. Yes, you can come up with an equation that uses earth's radius. That does not mean it's meaningful, and attempting to make it so is WP:SYNTH. Tarl N. (discuss) 10:00, 9 February 2019 (UTC)

??? Isen't the source quoted a reliable one ? 2A01:CB04:11E:6200:2059:6185:ED3:D9CA (talk) 10:29, 9 February 2019 (UTC)

It's not original research to make a graph, or two graphs. Eric Kvaalen (talk) 10:07, 9 February 2019 (UTC)

It is if the graphs are a major misinterpretation, and are presented with newly invented normalization and scaling that confuses even experts. 40.117.60.104 (talk) 01:32, 10 February 2019 (UTC)

In what way is it a "misinterpretation"? And the "normalization" as you call it is simply a device to make it so that the integral under the curve equals the minimum ratio of the mass of the elevator to the mass of the object to be lifted. That makes it very clear to a mininally-educated reader that the idea of a space elevator is bonkers if the free breaking length is only 637 km. And you can't come along and say "No, don't make a graph that clearly shows that! It's original research!" (Anyway, we're not the first to realize that. It's said by others that the free breaking length has to be quite long. Edwards doesn't give a free breaking length, but he does say that that tensile strength has to be 100 GPa, which gives a free breaking length of about the earth's radius.) Eric Kvaalen (talk) 06:51, 10 February 2019 (UTC)

You misread WP:CALC, and you are misusing uncited material to boot. As to the source being reliable, read WP:RS. Clarke's commentary (an address to a congress) falls under WP:PRIMARY. And even so, he does not talk about the free breaking length in terms of the earth's radius. Tarl N. (discuss) 13:51, 9 February 2019 (UTC)

Also, you added a comment about Edwards elevator requiring tensile strength of 100 GPa. Please give a source for that assertion. Tarl N. (discuss) 13:58, 9 February 2019 (UTC)

I've already said that the calculations are routine. Nobody complains when people make graphs of formulas and put them on Wikipedia! The source for Edwards is right there in the text (I did not take it out). Tarl_N., do you really not understand that the earth's radius is the relevant distance, or are you just saying that we can't let users see that because we haven't found a published "secondary" source that says it explicitly? Eric Kvaalen (talk) 06:51, 10 February 2019 (UTC)

Tarl_N. : Stop avoiding the precise answers given to your questions. You asked "Find me a WP:RS that talks about space elevator materials in terms of free breaking length and Earth's radius." Now the reference mentioned in the article talks about space elevator materials in terms of free breaking length and Earth's radius (equation n°8). Please acknowledge. Biem (talk) 14:26, 9 February 2019 (UTC)

Equation 8 does not talk about free breaking length. It talks about the taper ratio of a space elevator. It uses the earth's radius because it starts at the earth's radius and goes outwards (hence the exponential). It shows the taper increasing from the surface to the top of the elevator, does not show a releationship between free breaking length of a uniform cable and the radius of the earth. Tarl N. (discuss) 14:45, 9 February 2019 (UTC)

Tarl_N. : you are getting beyond WP:Faith. Let's do it slowly. Equation 8 reads :

${\frac {A_{g}}{A_{s}}}=\exp \left({\frac {R}{2L_{c}}}\left[2-3\left({\frac {R}{R_{g}}}\right)+\left({\frac {R}{R_{g}}}\right)^{3}\right]\right)$

• What do you think Lc represents ?
• What do you think R represents ?
• Does the comment on next page, which goes "Note how an increase in the characteristic length Lc leads to a faster than linear decrease in the taper ratio.", qualify as a discussion of space elevator materials ? If not, explain why.
• Does that publication (not equation) qualifies as "a WP:RS that talks about space elevator materials in terms of free breaking length and Earth's radius" ? If not, explain why.
• Given that R<<Rg, what can you say of the term between square brackets ?
• Given that the term between square brackets is ~ 2, how would the taper ratio varies when R is less than Rc ? and when R is greater than Rc ?

Well, in short : do you really understand what you are reading ? And could you prove it by reading my answers and answering to the point, without making your objections a randomly moving target ?

Biem (talk) 16:41, 9 February 2019 (UTC)

Biem, it's the idea that minFBL = planetRadius (one-for-one!) that's the bone of contention here, not whether planet radius appears in any equation for min FBL (or "Lc"). Both factors will of course appear in many equations. You just haven't established that your "equation 8" relates at all to the idea (wrong idea) that min required FBL equals a planet's radius one for one. Maybe the two of you are trying to prove two different things? I don't know. But, if you're trying to prove "minFBL = planetRadius", well it's hard to see how you can do that with your "equation 8", unless you plan on some major refactoring of it. Even if that was possible, it would be serious WP:OR, WP:SYNTH, and whatnot. 40.117.60.104 (talk) 02:06, 10 February 2019 (UTC)

I have just answered that, higher up. Saying that we can't use the fact (even on a talk page) that an exponential of a fairly large number is very high is like forcing stupidity. Eric Kvaalen (talk) 06:51, 10 February 2019 (UTC)

Nobody ever said FBL = planetRadius. The point is simply that given a planet radius and its ground gravity, the material resistance needed is at least of that order of magnitude. how would the taper ratio varies when R is less than Rc ? and when R is greater than Rc ? If the FBL is significantly lower you simply can't do it, if it is significantly higher a cylindrical cable will do and taper is irrelevant. So, indeed, minFBL ~ planetRadius as far as the order of magnitude is concerned. Thus, the choice of FBL ~ planetRadius and FBL ~ 10 x planetRadius are relevant choices to illustrate the point. I would have chosen e^(√10)~24 and e^10~22000 for a more dramatic comparaison of a 5dB change. Biem (talk) 07:45, 10 February 2019 (UTC)

Biem, I think you mean Lc rather than Rc. And you meant FBL ~ planetRadius/10, not times 10. Eric Kvaalen (talk) 10:41, 11 February 2019 (UTC)

That it includes surface gravity, and is only weakly omega dependent, and is only "of same order of magnitude" to surface radius is good progress. Still, that's a lot of WP:OR unless it's found explicitly somewhere in a reliable source -- with Lc and only only Lc on the left side and not at all on the right side. Even then, it would be probable undue weight on FBL, And, "order of magnitude" is so crazy vague so as there to be no notable correlation at all. Seeing a pattern where there is none is for mystics and conspiracy theorists. As far as relevant choices go , a much better choice would be samples of various extant materials compared with non-existing materials that *would* meet the specific-strength requirements. 40.117.60.104 (talk) 10:29, 10 February 2019 (UTC)

(edit conflict)I don't have to prove anything to you. The equation uses the breaking length (they call it characteristic length), that doesn't mean a plot of a relationship between radius and characteristics length is meaningful. The characteristic length is a result of gravity and tensile strength, the fact that it can be restated in terms of radius and mass doesn't make it meaningful.

Any statement not cited is WP:OR. If it's cited and the reference doesn't actually say what was added, it's WP:SYNTH. Tarl N. (discuss) 16:57, 9 February 2019 (UTC)

You do have something to prove, since you pretend to raise "relevant" questions and ignore the answers. Your refusal means that your opinion cannot be taken into account, you behave as you're just trying to bully out Eric Kvaalen's intervention, which is against the wikipedia principles. So do answer, or please keep away from this debate.

And "Any statement not cited is WP:OR" is just jibberish, since the redaction of an encyclopaedia does involve a redactional work - otherwise, "Any statement literally reproduced is WP:copyvio", which is far worse. Biem (talk) 17:15, 9 February 2019 (UTC)

Tarl, I think it would be a good thing to make "a plot of [the] relationship between radius and characteristic length" (showing the minimum mass versus that ratio), but that's not what I did. Doing that would require integrating under the curves for all the values of the ratio that are used along the abscissa. That really would be in the category of not an obvious or trivial calculation! Eric Kvaalen (talk) 06:51, 10 February 2019 (UTC)

• Oppose Even if there are no OR problems, I question the value of the graph to the typical reader. --Ronz (talk) 16:51, 9 February 2019 (UTC)

  This is a (another) redactional problem, which may be relevant, but not to be discussed under this heading. Biem (talk) 17:15, 9 February 2019 (UTC)

  I'm not sure how to interpret that beyond taking it as evidence that you're not here to work cooperatively with others to build this encyclopedia. Please strikeout. --Ronz (talk) 04:18, 10 February 2019 (UTC)

Thank you, Ronz. I don't know what a "redactional problem" is, but I do agree about the problem of WP:NOTHERE. Kvaalen is clearly "not here", based on constant WP:DONTGETIT and filling up walls of text quibbling about his crackpot graphs. Kvaalen has been beating that horse for two weeks now. I'm not sure yet about Beim. 40.117.60.104 (talk) 11:04, 10 February 2019 (UTC)

That's not how to talk to people! Eric Kvaalen (talk) 06:51, 10 February 2019 (UTC)

• Oppose, I think. Adding a plot created from a published and referenced equation is not original research. But it seems like this plot used parameters (e.g. free breaking length) which are not obvious, specified in the references or a matter of consensus. Selecting those values to make the plot would be original research. Also the values use would have to be reasonable ones, otherwise the plot is not useful and could be misleading. If someone can find a reference for reasonable parameters to put into the equation, then the plot would be fine. It's the unreferenced insertion of those parameters which makes it original research. Fcrary (talk) 20:37, 9 February 2019 (UTC)

But when you make a graph, you have to choose the parameter! The choices I made are obviously relevant because they show that when the parameter is 1, you get a nice graph and a minimal mass "only" about 30 times the mass to be lifted, whereas if you use the value 0.1 you get a graph that goes "sky high" and the minimal mass will be ridiculously high. This was obviously realized by others such as Edwards, even though he doesn't say it in so many words. But anyway, we are not obliged to hide obvious things from the readers. If the graphs show that, then so be it. Eric Kvaalen (talk) 06:51, 10 February 2019 (UTC)

• Oppose. Awful.
  ---The graphs themselves suck. Have any of you actually taken a look at them? It's the graphs that suck, not (necessarily) the equations they're supposedly based on. Take a look at the X scale. It's different for each plot. Kvaalen took it upon himself (in a fit of WP:OR) to "normalize" the altitude, which nobody ever does that way and you will never see in a reliable ref. See if you can figure out what's going on with it. I can't, and I'm a fair expert. Kvaalen clearly isn't an expert yet he's making stuff like this up! Any inclusion of graphs such as these would need to see them re-done in a way that's understandable. And importantly, they shouldn't be in terms of FBL because...
  ---Free breaking length is relatively unimportant. "Don't dwell on FBL". "30 MegaYuris or bust". FBL is nice in an introductory way to show that materials need to be strong and low-density. It's an important concept to know and it's surely relevant to space elevators, so it should be mentioned, especially that 5000km number from Clarke (not 6370km!). FBL is proportional to specific strength, but engineers do engineering using "Specific Strength", not FBL. In fact, the big mantra they use for fun is "30 MegaYuris or bust". They define a "Yuri" as 1 Pa/(kg/m^3). FBL isn't used for a number of practical reasons, probably the main ones being the fact that it assumes 1) a constant gravity strength for the whole of the 5000km, and 2) a constant cross section area -- both of which don't apply. Also, 3) The units are in length which requires oddball conversion constants be thrown in and handled properly just to get to the real meaning that the straightforward units give you for free. The point is, we shouldn't dwell on FBL. Any graphs should present the Specific Strength in its straightforward units as the parameter of concern. No graph should mention FBL because it gives FBL undue weight, implying wrongly that FBL is used more than it is.
  ---Never imply FBL=planetRadius. The graphs use an FBL of 6370 km, Earth's radius. That's because Kvaalen had previously asserted (wrongly, and it was removed) that the minimum necessary (specific strength) is that of an FBL = planet radius. Planet radius is indeed a factor, but not a one-to-one factor as stated by kVaalen. In fact, quite the opposite(!). Larger surface radius by itself decreases the necessary distance needed to dip into the "gravity well", thus decreasing the required specificStrength/FBL. Okay okay, of course, as radius increases, so does surface gravity which would tend to increase required specificStrength/FBL. Then there's rotation rate and planet density that factor prominently into the matter. And, there's all the practical design factors like strengthening for point loads, etc. There's also the Clarke source saying "5000km" which is far from 6370km. The point is, the idea of FBL=planetRadius is wrong. The graphs use FBL of 6370km, wrongly implying some significance of planet radius. We shouldn't do that. That's one of the reasons they were removed. 40.117.60.104 (talk) 00:58, 10 February 2019 (UTC)

Mr 104, you moved my replies from under the portions they apply to to way down under, and accused me of being aggressive for putting them where I did. I think that's unfair. But anyway, here are my replies. Eric Kvaalen (talk) 10:41, 11 February 2019 (UTC)

You're right. Sorry. I saw soon afterward that the comment was formatted in a way that made it look like three separate comments. My fault there with the "booby-trapped" formatting. That's all water under the bridge to me now though. Also, when I ejected the comments (from inside mine) I didn't really know where to put them. I just had to make my best guess. You're right to "fix 'em up" below, and you're justified in being a little indignant too. My apologies.  :-) 40.117.60.104 (talk) 20:39, 11 February 2019 (UTC)

This first one was under your heading "The graphs themselves suck":

The use of altitude divided by free breaking length falls right out of the integral for the (minimal) total mass. I used the same range of altitude in each graph, as you can see from the fact that the maximum (the geostationary altitude) is at the same place relative to the whole range. Using this scale makes it that the integral under the curve which gives the minimal total mass is easy to estimate. To say that we can't use a smart way of graphing something because it might reveal too much is silly. You say you're a fair expert. Can you elaborate? Eric Kvaalen (talk) 06:51, 10 February 2019 (UTC)

This one was under your heading "Free breaking length is relatively unimportant."

I did not say that "minimum necessary (specific strength) is that of an FBL = planet radius". I said, "In order for a space elevator to be realistic, the characteristic length has to be on the order of r₀, the radius of the earth, which is 6377 km." That was removed, but it happens to be true. The factors that you mention such as planet density only come into play through the ratio of FBL to planet radius. There is a relatively minor effect of what we have called x, the ratio of centrifugal force to gravity at the planet surface, and the cube root of x which is the ratio of the planet radius to the "planet-stationary" radius. If you like I will show you that the Clarke value is related to the planet radius (which it is not far from!). Adding strengthening for point loads or whatever will only increase the total mass. The minimum total mass is given by doing the integral which I put in the text (obviously), and that integral corresponds to the area under the curves in my graphs, using the kind of x-axis that I used. Eric Kvaalen (talk) 06:51, 10 February 2019 (UTC)

And this one was under your heading "Never imply FBL=planetRadius."

No, you're wrong. The FBL is relevant, because of the equation (given in Aravind) that shows that it is [Earth radius]/FBL which matters. That is the critical parameter. Eric Kvaalen (talk) 06:51, 10 February 2019 (UTC)

The taper ratio is exp(R_earth/FBL x Correction factor), and the "correction factor" for Earth is ~ 0.778 : this is why Clarke states that a cable should have "sufficient strength to support 5000 km (actually, 4960)" : when FBL = R_earth x Correction factor the tapered cable has an e=2.71828 taper, but which is more, that is the minimal value for which a non-tapered cable would do the trick - this is what he had in mind, otherwise the precision is incomprehensible. Biem (talk) 09:06, 10 February 2019 (UTC)

See my comments lower down. Eric Kvaalen (talk) 10:41, 11 February 2019 (UTC)

• Comment - The section could benefit from graphs, but the ones proposed need to be improved. E.g., the Y-axis needs to be labeled with units and if the two are being directly compared then they need to use the same units and scale (perhaps a log scale would work in this case?). Also, any claims or observations in the description need to be reliably sourced. Sparkie82 (t•c) 07:04, 14 February 2019 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Suggestions

Eric Kvaalen, If I were you, I would not insist on keeping those specific pictures, which have indeed a lot to be criticized against them. Maybe you can draw a new pair of graphics taking into account the comments that have been made :

• The "free breaking length" is a mathematical concept, computed assuming a "constant gravity". For obvious reasons, "constant gravity" is not possible for the lengths involved, so this really just reflects a physical property of the material, not a "real" distance to be measured somewhere. So though a useful comparison figure, the concept itself is potentially misleading and should be given a low profile. So, please, don't mention it.
• The formula given in Space elevator#Cable material is documented by itself, but its consequences are not.
  • It indeed means that to make a space elevator on a planet, the "free breaking length" of the material under ground gravity should be of the order of magnitude of the planet radius (not an exact value). Some contributors seem to feel uncomfortable about that, but that is exactly what the equation means, once the property of the exponential function are considered (and those who don't know those properties should have a brief update - or stay out of the discussion). I think this has been clarified.
  • But this neglects the fact that the corrective factor (the one that takes into account the rotational speed) should be considered (though it won't change the order of magnitude);
  • And that value is but a fuzzy milestone when tapered cables are considered (cables can be made not quite reaching this value).
• If you want a "real sharp value", you must consider a cylindrical cable, not a tapered one, and take into account the corrective factor. This is (to my knowledge) what Arthur C. Clarke has in mind and computed in his adress. The corresponding maths is straightforward (same equation of gravitational field, but instead of a dS/S you integrate a dF, but since the primitive on the right side will be the same, it's yielding the same limit value - those who master the elevator equations will know what I mean). But though not worthy of a scientific article, that is undocumented (as far as I know) and therefore best left unmentioned in the Wikipedia article.

Now, beyond the (relatively irrelevant) discussion around the link between free breaking length and Earth's radius, I think a couple of graphics may help the reader to understand why material characteristics make a crucial difference "around these values". But your choice having been proved polemical, I would suggest this :

• To apply the documented formula documented in the article, and make a graph out of it, obviously is not original research. Indeed, "Adding a plot created from a published and referenced equation is not original research". The problem, if any, lies in the choice of "FBL" parameter in both examples.
• There is no need to insist on taking the Earth's radius as reference free breaking length, since that is not the "sharp value" anyway. Actually, it would be better to take into account the corrective factor. And better yet, since Arthur C. Clarke has given that precise 4960 km value - just use it as a documented value as a starting point. (really, who cares?)
• Though the "surface" scale will obviously be different (unless you use a log diagram?), do use the same scale for altitudes in both diagrams, for comparison purposes.

Well ... that's about it. I'd be glad to answer any question. Biem (talk) 17:53, 10 February 2019 (UTC)

First of all, I will go ahead and show how to calculate the minimal specific strength needed for an untapered cable which descends from the geostationary altitude. First let's define g₀ as $GM/r_{0}^{2}$ where $r_{0}$ is the earth's radius. The total weight per mass is the integral of the gravity minus centrifugal force:

$\int _{r_{0}}^{r_{g}}(g_{0}r_{0}^{2}/r^{2}-g_{0}xr/r_{0})dr=g_{0}r_{0}(1-r_{0}/r_{g}-x/2+xr_{g}^{2}/2r_{0}^{2})$

Since $r_{0}/r_{g}=x^{1/3}$, this simplifies to:

$g_{0}r_{0}(1-x^{1/3}-x/2+x^{1/3}/2)$

or simply $g_{0}r_{0}(1-x/2-x^{1/3}/2).$ So the cable has to be able to hold up the equivalent of $r_{0}(1-x/2-x^{1/3}/2)$ under gravitation $g_{0}$. Actually, the net force at the earth's surface is $g_{0}(1-x)$. So we find that the free breaking length is $r_{0}(1-x/2-x^{1/3}/2)/(1-x)$ or 0.926 times the earth's radius, or 5904 kilometres. Apparently Clarke didn't think about the fact that there is a term of $+x^{1/3}/2$. But anyway this shows that the value is, to a first approximation (when we ignore x), the radius of the earth.

Now, concerning your points, I don't agree that the "free breaking length" should be left out of the discussion. It is found right there in the equation for the taper, where we have $\rho g_{0}/\sigma .$ I'm not sure what you mean by "a real sharp value". Clarke's value is not really sharper than the radius of the earth, and anyway, as I have just showed above, he left out a certain term (so the correct value for an untapered cable is twice as close to the radius of the earth than what he said).

I could perhaps find a way to put both a scale of kilometres and my scale where I divide by L_c, but I really think my scale is needed in order to show how the minimal total mass of the contraption depends on L_c. Although the amount of taper depends mostly on the ratio of L_c to the planet's radius, the minimal total mass of the system also depends on how long the cable is, not just on how thick it gets. (The mass of the counterweight does not enter into the minimal total mass, because the minimal value, or lower limit to be more accurate, is approached as the cable becomes infinitely long and the counterweight mass goes to zero.) This means that if we slowed the planet's rotation (that is, we lower x) the minimal total mass would increase. The scale I used for the abscissa allows the viewer to immediately get an idea of the size of the integral that gives the minimal total mass.

Would the other editors agree to my graphs if I used some other values of L_c, but values which would show that the total mass sky-rockets when the value drops significantly below the radius of the earth, or Clarke's number?

Eric Kvaalen (talk) 10:41, 11 February 2019 (UTC)

My, my, my... You've got the integration wrong. Please check and correct it before any further development. Yours, Biem (talk) 17:49, 11 February 2019 (UTC)

Hi guys. I think this is good progress. Just a few comments:
1) I'm assuming, Biem, that by "cylindrical" you mean "constant cross-section area", with the actual shape of the x-section area being unimportant to the strength problem.
2) In an ideal of "presentation goodness", I suggest the height axis start at the center of the earth (to show relative scale compared to earth), then proceed upward graduated in km with zero being at the surface. The distance between zero and GEO should be the same on all graphs (if there's more than one).
3) I'm not actually opposed to going beyond GEO as long as the height scale is consistent between graph(s). 40.117.60.104 (talk) 21:16, 11 February 2019 (UTC)

Just thought of: Make the graph(s) in the .svg format so they're easily editable. That way, nobody can complain about nit-noids, they can just fix them. 40.117.60.104 (talk) 21:23, 11 February 2019 (UTC)

Statement above: My, my, my... You've got the integration wrong. Please check and correct it before any further development. This is exactly why we have a prohibition on OR. I'm not opposed to adding graphs, I want to see the citation for the formula. Evidently what's going on here is well beyond the WP:CALC standard of calculation is obvious, correct, and a meaningful reflection of the sources. How many screenfuls have we now dedicated to what should have been a trivial discussion of "where is the WP:RS documenting this interpretation?" Tarl N. (discuss) 21:42, 11 February 2019 (UTC)

Biem, please explain why you say I have the integration wrong. Tarl N., there's nothing wrong with talking on the talk page about whether an equation is correct or not, or about whether a cited source is correct or not. This business about so-called "original research" is not about talk pages. Do you see something wrong with my integration? If so, speak up. In any case I'm goin' to bed. Eric Kvaalen (talk) 21:51, 11 February 2019 (UTC)

That's the point. We're not here to sanity check your calculations - that stage of publication is done in WP:RS. Publishing reviews are what define them as reliable. What Wikipedia editors should do when reviewing article changes is ensure that additions reflect what the cited sources say. A wikipedia article takes an expert to write, but must be left in a state where a librarian not expert in the subject matter can maintain it against arbitrary and potentially incorrect changes. When we have disagreement about the correctness of symbolic integration, we're well beyond the "obvious" stage. That's long been a problem for this article, but we don't need to make it worse. Tarl N. (discuss) 22:08, 11 February 2019 (UTC)

I think if Kvaalen wants to make plots, let him make plots. If they're ultimately found to be based on complex integrations, done by him, then reviewed/corrected by peers on this page -- just like they do in real research -- well then maybe that will serve as an excellent example of what "original research" really is. It will be useful and enlightening if only for that reason. Who knows? Maybe he'll find a good cite for it. That would transform it all from "ugly pre-OR" to "additional support for the non-dubiousness of the source", which would be good. 40.117.60.104 (talk) 23:07, 11 February 2019 (UTC)

Biem, you're right. I got mixed up with too many minus signs. So the correct integral is:

$\int _{r_{0}}^{r_{g}}(g_{0}r_{0}^{2}/r^{2}-g_{0}xr/r_{0})dr=g_{0}r_{0}(1-r_{0}/r_{g}+x/2-xr_{g}^{2}/2r_{0}^{2})$

Since $r_{0}/r_{g}=x^{1/3}$, this simplifies to:

$g_{0}r_{0}(1-x^{1/3}+x/2-x^{1/3}/2)$

or simply $g_{0}r_{0}(1+x/2-3x^{1/3}/2).$ So it's true what I said, that Clarke forgot a term, but I had it with the wrong sign. And it's true that to a first approximation (ignoring x) it's the planet radius.

Note that if one used this minimal free breaking length in an untapered design the contraption would be unable to lift any weight. As soon as you attach something to the bottom of the cable, it will break at the geostationary altitude.

But as I said, there's nothing wrong with doing an integration on a talk page. That is not what the "original research" law is about. (I don't like using such a high-falutin' term for a simple integration, but never mind.)

So now, what about my question? Do you agree to my graphs if I use some other values of L_c, but values which would show that the total mass sky-rockets when the value drops significantly below the radius of the earth? (I don't like the idea of using the Clarke value because that was for an untapered design, whereas my graphs are talking about tapered designs.)

Eric Kvaalen (talk) 03:23, 12 February 2019 (UTC)

Since we seem to have an inherent disagreement over this being WP:OR, WP:SYNTH or entirely acceptable, I've asked for help from WP:NOR WP:ORN. See Wikipedia:No_original_research/Noticeboard#Space_elevator. Tarl N. (discuss) 03:33, 12 February 2019 (UTC)

I repeat, the "original research" law doesn't apply to talk pages. Nor does the thing about "synthesis". Eric Kvaalen (talk) 06:38, 12 February 2019 (UTC)

Talk pages are for discussion of how to best maintain the article Let's keep it on that. Going over new work like this is better done somewhere else. As of now, this kind of talk would be significant WP:OR if it were it to be introduced into the main article. Discussing it here when there's no sign of a source means you're discussing something that has nothing to do with the article, and would be immediately reverted if it was introduced without a source. A talk page is not a place for the educational enlightenment of the editors (apart from short digressions), nor is it a place for peer review like this. If anybody wants to participate in a salon, please do. Talk amongst yourselves, somewhere else! 40.117.60.104 (talk) 04:54, 12 February 2019 (UTC)

Mr 104 (I wonder whether you are actually Skyway from the way you talk), we've been talking here about whether the Clarke value has something to do with the earth's radius since back on the 9th of February, and it was Tarl who contested that. Even you have said that it is "far from" the radius of the earth, and that the radius of the earth has nothing to do with what the free breaking length should be. And then when I prove that the free breaking length for an untapered cable (what Clarke tried to give) has to be more than a value which to a first approximation is the earth's radius, you suddenly say we can't talk about that on a talk page!

Now can you two get back to my question? What values of the free breaking length do you think I should use?

Eric Kvaalen (talk) 06:38, 12 February 2019 (UTC)

It might look like I made a change in position there, and I did only remember that talk page rule after I made the "let him make plots" post. But, the reality of the situation is that it's better for you if you do keep the talk page for talking about the article. Instead, you're actually doing your WP:OR right here in front of everybody -- with peer review and all. (previously IP...104)RFinlay72 (talk) 06:08, 13 February 2019 (UTC)

What do your reliable sources use? Tarl N. (discuss) 06:44, 12 February 2019 (UTC)

Come off it, Tarl. A person can make a graph without having to have a "source" for the values of the parameters! Eric Kvaalen (talk) 10:58, 12 February 2019 (UTC)

The result sounds correct this time. As I said above, the integration of the right hand part is in fact the same as the one when the cable is tapered, so you should end up with the same correction factor and the same limit value.

Now, since the discussion here is not about the value of the free breaking length, but the parameter of a graph, I would suggest to use the 4960 km value given by Clarke in his conference, which is not far from the one just computed, and use the conference as a source. As long as it is of the same order of magnitude (the difference is a few percent, as far as I recall) the profile won't differ significantly from one value to the other ; and the point is not to build a cable with precisely that resistance, but to show the dramatic effect of having but one tenth of it. Biem (talk) 12:22, 12 February 2019 (UTC)

Biem, I'm not sure I understand -- what do you mean by "the difference is a few percent"? Between what and what? Eric Kvaalen (talk) 13:59, 12 February 2019 (UTC)

In the $\scriptstyle \Delta \left[\ln(S)\right]={\frac {r_{0}}{L_{0}}}\left(1+{\frac {x}{2}}-{\frac {3}{2}}x^{\frac {1}{3}}\right)$ : With x~0.0035 (see article) the correction factor is ~ 0.774. Earth's equatorial radius is ~6378km, so the computed limit value for the free breaking length would be ~ 4937km. The difference with Clarke's 4960km is 0.5%, so it is essentially the same thing (difference under rounding errors for x). Hence : take the documented value in Clarke's speech, since it is available, documented, and won't change significantly your graphics. Biem (talk) 20:08, 12 February 2019 (UTC)

Well, Biem, I could use 4960 or 5000 for one of the two graphs. But actually I think it's rather silly because Clarke didn't say that that was the free breaking length which should be used -- he only said it was the minimum which would allow a normal (untapered) cable (with no load attached at the bottom) to hang from the geostationary altitude to the earth's surface. He goes on to say that one can actually use a material whose breaking length is a fraction of this, even chewing gum! (But that that would probably give a mass larger than the mass of the universe...) He also mentions Kevlar with a breaking length of 200 km and says that also gives an astronomical total mass. Anyway, we need two graphs, one with a reasonable free breaking length (I mean one that gives a reasonable total mass), and one with a somewhat lower free breaking length which gives a very high value for the minimal total mass. What do Tarl N. and 104 say? Eric Kvaalen (talk) 08:25, 13 February 2019 (UTC)

On WP:ORN Eric suggests he now wants to post a graph of Aravind eq(7) Aravind (By the way, that has a DOI, we should add it to the citation - the more formal url (albeit behind a paywall) is https://doi.org/10.1119/1.2404957 ). I have no objection to that as long as it's in SI units. I'm not sure what's interesting about that graph, it's a simple exponential curve, but the main question I'd have is what parameters are to be used for ρ and T (density of cable and strength). Tarl N. (discuss) 16:55, 13 February 2019 (UTC)

Well, then, Eric Kvaalen : DO use it. I can't see what is holding you.

1. The whole wikipedian world is telling you that unless there is a documented reason for choosing a particular value, it will be considered as OR.
2. We have demonstrated in this talk page that the 4960 value quoted by Clarke is the one you should have chosen in the first place, being the actual "free breaking length" needed for Earth (your first approach actually was very approximative).
3. This value is documented by Clarke's speech, so that specific value is OK with respect to OR criterium.
4. Though the choice of that parameter is (indeed, between us) relevant, there really is no need to say you are using the limit value of a cylindrical cable when computing the profile of a tapered cable, unless you want to get caught into endless discussions.

So, please, present your new graphics, document the choice as being that of Clarke's address, and do end this tiresome discussion. And please take into account the criticisms that have been made to the graphic formats when designing the new ones : this part is not essential for your demonstration, there is no harm in yielding that point to your critics - it just will make your good will more obvious.

I do hope this will put a satisfactory end to the debate... Biem (talk) 15:16, 14 February 2019 (UTC)

Minor comment on the above; Clarke's speech does not strictly qualify as WP:RS. It's WP:PRIMARY, and it shows in some of the errors he made. On the other hand, in this article, using it as a reference is still an improvement. Tarl N. (discuss) 20:46, 14 February 2019 (UTC)

Here are some published values... Aravind gave Lc=T/(rho*gc) implying that Specific Strength (T/rho) = gc*Lc. With gc = 10m/s2, Clarke's Lc of 5,000,000m would equate to a Specific Strength of 50 MegaYuri's (MPa/(kg/m3)), which is in the right ballpark. The context of "Clarke's material" of 50MY ("5000km") is constant cross-section area ("cylindrical"), no safety factor, and no additional load carrying. Adding a taper ratio >1 adds a safety factor and load-carrying capacity for that same "Clarke material". ISEC has used the phrase "30 MegaYuri's or bust"; they expect that a material of 30 MY (3000km) would enable a cable with a not-too-large taper, a load carrying capacity, and a safety factor of 2 or 3. Edwards gave 100 "GPa" which is 77 MY or 7700 "km". To summarize, some key "notches" for Specific Strength that have been used by some published sources are:
--ISEC (~2011): 30 MY, converts to 3000 "km"
--Edwards (~2002): 77 MY, published as "100GPa" per 1300kg/m3, 77 MY converts to 7700 "km"
--Clarke (1979): 50 MY, published as "5000 km"
As always, I suggest staying away from putting things in terms of "Lc" ("FBL") for the reasons I've given before. As I've shown above, it's pretty simple (likely not OR) to convert to Specific Strength (in MYuri or MPa/(kg/m3)) which is the modern consistent term and unit for that property. "Yuri" is a new term for Pa/(kg/m3) (or N/(kg/m) ), it's not really "official", so it might be better to just spell it out as "Pa/(kg/m3)" or equivalent. (previously IP...104)RFinlay72 (talk) 21:17, 14 February 2019 (UTC)

RFinlay72, I think the best would be to use the ISEC value of 30 MJ/kg for the "decent value" graph and half that for the graph which is meant to show how the minimal mass would then be much higher.

Tarl N., you say "Eric suggests he now wants to post a graph of Aravind eq(7)". I haven't changed what I'm suggesting. It's the same equation as what I graphed before, which I put in the text (which you reverted) and which was already in the text higher up in a slightly different form.

I will put the altitude in km. But the cross section is relative to the value at ground level. It's unitless.

Also, I don't see why you say "I'm not sure what's interesting about that graph, it's a simple exponential curve". It's not a simple exponential curve, it's the exponential of a function involving a constant, an inverse, and a quadratic.

I disagree with Sparkie82 who says the two graphs have to have the same scale.

Now, I would like to know whether you two consider it "original research" to say that the total mass of the cable with this profile would be

${\begin{aligned}{\text{total mass}}&={\frac {mg_{0}\rho }{\sigma }}\int _{r_{0}}^{r_{max}}\exp \left[{\frac {r_{0}}{L_{c}}}\left(1+{\frac {x}{2}}-{\frac {r_{0}}{r}}-{\frac {x}{2}}{\frac {r^{2}}{r_{0}^{2}}}\right)\right]dr\\&=m\int _{r_{0}}^{r_{max}}\exp \left[{\frac {r_{0}}{L_{c}}}\left(1+{\frac {x}{2}}-{\frac {r_{0}}{r}}-{\frac {x}{2}}{\frac {r^{2}}{r_{0}^{2}}}\right)\right]d{\frac {r-r_{0}}{L_{c}}}\\\end{aligned}}$

To me it seems just a totally obvious fact that the mass is the integral of the cross section with respect to the altitude, and I don't think that should be called "research" at all, original or not! Nor is it "research" to replace $g_{0}\rho /\sigma$ with $1/L_{c}.$ That's just the definition of $L_{c}.$ The point of the restriction against original research is to prevent spurious "facts" from getting into Wikipedia, not to stop people from making remarks such as "these graphs show that the minimal mass is much higher if the specific strength is 15 MJ/kg rather than 30 MJ/kg". We're not here to remove any statement from Wikipedia for which we don't have an explicit reference, no matter how obvious the statement is — to "dumb down" Wikipedia. That fact (about 15 versus 30, or similar) is what lies behind ISEC's slogan of "30 MegaYuris or bust!". We may not have an explicit reference saying "ISEC says that 15 MegaYuris would make the system have a huge mass whereas 30 MegaYuris would give a reasonable system", but that's what ISEC thinks (correctly).

Eric Kvaalen (talk) 09:42, 18 February 2019 (UTC)

@Eric Kvaalen: If you want a graph of Aravind Eq. (7) for Earth, there are three variables: Strength, Density and Height (T, ρ and r). Pick two and graph the third. The part I find interesting from this equation is the taper ratio from surface to geosynch for particular materials. If you want to fix density and graph taper vs. strength at geosynch, that's interesting and obvious from the equation. If you want to fix density and strength, and graph taper vs height, that's also obvious from the equation, although less interesting. The numbers you are feeding in and getting out should be obvious in the graph, hence my request that they be labeled in SI units. If you want to add labels (e.g, locations of earth surface and/or earth geosynch orbit, or strengths of steel or kevlar), that's good too - easily documented.

As for total mass, unless you can find someone who has published such an equation, I'd be reluctant to say it follows under WP:CALC exemption. The history of this discussion suggests that mistakes and misunderstandings are easy, and to avoid another thirteen screenfuls of discussion, I'd prefer to stay away from anything that isn't easily citeable. Tarl N. (discuss) 21:45, 18 February 2019 (UTC)

There's no need to treat the tensile strength and the density as two parameters because it's only the ratio of them that matters. The important thing in determining whether a space elevator is feasible is the total mass, not the ratio of cross section at geostationary altitude to the cross section at ground level. The total mass depends on the integral which I wrote above (on Feb. 18). This shows that it's not just the cross section ratio function (the exponential term) that matters but also the specific strength itself. A weaker material necessitates a thicker cable at all altitudes. That's why the total mass ratio is equal to the exponential term integrated over $r/L_{c}$. Do you really not see that the total mass is given by this integral? I think you have a sufficiently high level of math to see that it's the truth. It doesn't take a high level. Eric Kvaalen (talk) 09:19, 20 February 2019 (UTC)

Indeed, you could use specific strength as strength/density (in SI units), and plausibly claim it under the obvious and correct exemption. But it's not all that interesting, because all the relevant materials have densities within an order of magnitude of each other. And it's getting dicey at that point. Certainly the fancy construct you proposed above is well beyond the limit. Given the huge amount of text in this discussion (and how far back I have to scroll backwards just to answer each subsequent missive from you), I'm not inclined to say that even that falls under the exemption. If you want to graph Aravind eq(7), do so. If you want to get fancier, find an RS that's published the fancier equation you want to graph.

By the way, can you close the RFC? It's not doing any good at this point. Tarl N. (discuss) 02:42, 21 February 2019 (UTC)

@Eric Kvaalen: Since you're ignoring the discussion and going ahead and modifying the OR in Space_elevator#Cable section equations, can you explain where the ${\frac {3}{2}}x^{\frac {1}{3}}$ term comes from? Somehow two terms with different powers of ratios of $r_{0}$ vs $r_{1}$ disappear into that term: $-{\frac {GM}{r_{1}g_{0}r_{0}}}=-{\frac {r_{0}}{r_{1}}}$ and $-{\frac {\omega ^{2}r_{1}^{2}}{2g_{0}r_{0}}}=-x{\frac {r_{1}^{2}}{2r_{0}^{2}}}$. The transformation of those two terms into $-{\frac {r_{0}}{r_{1}}}-x{\frac {r_{1}^{2}}{2r_{0}^{2}}}=-{\frac {3}{2}}x^{\frac {1}{3}}$ is certainly far from "obvious and correct" - x doesn't even have a term of r1, so there must be a magical ratio there somewhere. It is in my target for things that needs to be cleaned up. Even if that term is explainable, it is far from maintainable (if someone changes it, how can a random reviewer argue?), and the subsequent equation, which is marked as citing Aravind eq(7), isn't eq(7), it's depending on that transformation. Tarl N. (discuss) 06:29, 28 February 2019 (UTC)

@Eric Kvaalen: Ping? I think this particular OR is simply wrong, and using bogus equations to add calculated numbers is a net negative. Any comment before I revert the changes? Tarl N. (discuss) 07:40, 2 March 2019 (UTC)

It's just a matter of using the definition of the geostationary radius. In the subsection Apparent gravitational field we have the equation

$r_{1}=\left({\frac {GM}{\omega ^{2}}}\right)^{\frac {1}{3}}$

So

$r_{1}^{3}\omega ^{2}=GM$

and

${\frac {\omega ^{2}r_{1}^{2}}{2g_{0}r_{0}}}={\frac {GM}{2g_{0}r_{0}r_{1}}}={\frac {r_{0}}{2r_{1}}}.$

The other day I decided to put back in just the innocuous improvements of the writing and the "minor grammar edits for flow" which RFinlay said were not dubious, and he anyway reverted them, calling it "original research"! What do you think of that?

Eric Kvaalen (talk) 21:17, 5 March 2019 (UTC)

Ah. Got it. The transform, while not the incorrect failure I thought, is far from obvious, and is thus unmaintainable (remember the principle - it takes experts to write articles, but they need to be maintainable by less-than-experts once the experts have moved on). And it's not needed, we could use Aravind[7] directly, particularly since that's the equation it cites. No need to carry out that transform. That section needs re-writing to better conform to published sources. Tarl N. (discuss) 21:25, 5 March 2019 (UTC)

I think it's much better to put in a bit of explanation. The "no original research" thing is meant to prevent people from putting in research results that haven't been checked and might not be correct. It's not meant to stop articles from being written in a more understandable way! That's why I want to know what you think about RFinlay reverting this edit of mine. Eric Kvaalen (talk) 12:08, 10 March 2019 (UTC)

@Eric Kvaalen: You're mistaken. The NOR prohibition is to ensure articles can be verified against the sources they claim to cite - this is a fundamental maintenance issue. In wikipedia, much like in science, merely being right is not sufficient for inclusion. There is an exclusion for "obvious and correct" calculation, which this certainly wasn't. The specific prohibition from NOR: This includes any analysis or synthesis of published material that serves to reach or imply a conclusion not stated by the sources. To demonstrate that you are not adding OR, you must be able to cite reliable, published sources that are directly related to the topic of the article, and directly support the material being presented. (my bolding)

This particular transform is unnecessary and it took weeks to track down what the heck was going on - it needs to come out, not be further elaborated on. If you can find someone who has published based on that transform, by all means, cite it. Or find someone who will publish research you write based on it. But leaving something in there which will mystify future reviewers is not appropriate - we've had edit wars in the past where someone has made changes to obscure equations under the assertion that "it's wrong", and because the sources didn't agree with either the original version or the new version, consensus was not achievable. We need to ensure that any equations stated are readily trackable to published sources. Tarl N. (discuss) 20:09, 10 March 2019 (UTC)

I've taken out the OR of that section. What it needs is a table showing cross-section taper values for various materials (pointing to specific strength article), and potentially two graphs, one showing how the strength of material changes the taper, and the other showing the shape of the taper along the radius. I'll work on those today. Tarl N. (discuss) 20:42, 10 March 2019 (UTC)

Tarl, you still didn't answer my question about RFinlay's reversion of my latest edit.

I see that you have extensively edited the article. Some of your edit is improvement, but I'm glad that as a Wikipedia reader I saw the article in the state it was in before! It was more informative and explanatory, and gave insights that the present version avoids.

For instance, you give the formula for the "taper ratio" using a number, 4.8479758×10⁷, which, besides having too many digits and missing its units, is not nearly as meaningful as using the distance given by dividing this by g and then using free breaking length instead of specific strength in Nm/kg.

You need to give a reference for the value of 62,000 MPa for predicted tensile strength. I don't agree with RFinlay that saying "(predicted)" is enough of a caveat. As the reference says which you had put there, it ain't gonna happen. Defects will be present, and you have to have this kind of tensile strength everywhere, not just inside a particular nanofibre. Does anyone think we will be able to make single fibres that are tens of thousands of miles long? (And with no defects?)

I think it's pointless in your graph to start at the earth's centre. The equation wouldn't even be correct inside the earth! And you don't give the value of specific strength that you used. (I can see that you used 3911, but it should be stated.)

Back in January I added this:

"To achieve a characteristic length of 6000 km, the tensile strength needs to be very high. Even if the density is as low as 1 g/cc, the tensile strength would need to be about 60 gigapascals. A cable one square millimetre in diaameter made of a material with a characteristic length of 6000 km would be able to lift (under standard gravity) 6 cubic metres of the same material. Whatever the density (and the cross section), if the payload to be lifted into space weighs one tonne, then every 6 metres of the "cable" at ground level would have a mass of one gram, more of a thread than a cable. (We see that the artists' conceptions in this article are thus not realistic.)"

That was reverted (along with everything else I did) by RFinlay (anonymously at the time). I think it's a useful and important point. Do you agree with me that all the "artist's conceptions" in our article are unrealistic?

Eric Kvaalen (talk) 08:06, 24 March 2019 (UTC)

Right. Lessee:

• Taper ratio too many digits. Feel free to round or truncate to taste. I ask you not remove the note, since that's the only tie between Aravind eq.7 and the value, it's unmaintainable without the note.
• Taper ratio no units. It's dimensionless. It's a ratio of areas, all units cancel out.
• Carbon nanotubes. The value 62000 isn't predicted, it was measured at microscopic sizes, which the original discoverers knew was meaningless in bulk. But it gets a huge amount of popular press, so I included it, with the caveat. I'll remove the line, it's misleading as is (they are not predicted to have tensile strength of 62 GPa in bulk). Comment to RFinlay, yes, yarns and ropes depend on van der Walls forces, you won't get 62 GPa from van der Walls, you need covalent bonds. But it's moot, that line is gone.
• Free breaking length... The entire discussion of free breaking length is meaningless in this context. This is not a uniform cable in a uniform gravity field. The equation in question is Aravind eq.7, which deals with density and tensile strength (or inverse of specific strength).
• As for the artists conception, since this entire structure is hallucinatory, I can neither agree nor disagree with the artist renderings.

Tarl N. (discuss) 21:41, 24 March 2019 (UTC)

Oh, final point. The graph I added starts at the Earth's surface (left edge is earth's surface, vertical bar is geosync). The units on the graph are in terms of distance from the center, since the equation uses R in terms of distance from the center. The caption wording was modified by RFinlay. Tarl N. (discuss) 21:44, 24 March 2019 (UTC)

Reprise

@Tarl N., RFinlay72, and Andrew Swallow: I just came back to this talk page after a month. To respond briefly to Tarl, the number I said was missing its units (now rounded to 4.85×10⁷) is not the taper ratio. It has units of J/kg or whatever. It was RFinlay who started calling your 62000 "predicted". I agree with you that it's totally unrealistic for a real cable. (I don't know why you mention van der Waals forces to RFinlay -- did he mention them?) I still disagree with you that free breaking length is "meaningless in this context". As we saw (above), the free breaking length has to be similar to the earth's radius. I see that your graph has been removed, I don't know why. I think it would have been clearer if you had grid lines at the different x values, and I don't see why you used metres!

Finally, back to my question about the artists' impressions. Do you three agre with me that all the artists' impressions we have in this article show "cables" or "towers" that are much too thick? See what I wrote above (March 24).

Eric Kvaalen (talk) 20:17, 26 April 2019 (UTC)

By the time it reaches the Earth's surface the cable is as thin as a cotton thread. You may be able to use it to sow a button on. However depending on the taper value for the material it could be thousands of time wider at the Geosynchronous Orbit point. An alternative implementation increases the number of threads say every 1000 km. until GEO Andrew Swallow (talk) 01:06, 27 April 2019 (UTC)

@Andrew Swallow: Yes but most of the artists' impressions show the thing near the earth's surface. Anyway, in order to be realistic the total mass has to be something reasonable. If you have a cable that's quite thick for tens of thousands of miles, it would have a huge mass and it would not be practical to launch all that mass into space. That's why (as I tried to show with my graph, see above) it's very important to have a free breaking length that is similar to the earth's radius, say 6000 km. Otherwise the total mass is enormous. Eric Kvaalen (talk) 05:09, 27 April 2019 (UTC)

(edit conflict) @Eric Kvaalen: Lessee:

• Yes, indeed, the units of 4.72e^7 are in g r^2/r, or m^4/kg/s m^2/s^2.
• RFinlay claimed on his talk page that nanotubes will be able to support increased loads by holding onto each other with Van der Walls forces.
• As for free breaking length and radius of the earth. The only relationship between them is numerical coincidence. Consider an earth with the same mass and rotation period, but with the density of Saturn. The radius would be about double, but the taper would be essentially unchanged (ever so slightly changed, because the cable would be about 6000 km shorter). So, the radius is not a critical variable in that equation, which is why I so vociferously objected when you first claimed a relationship.
• As for free breaking length itself, it is a quantity defined for a uniform cable in a uniform gravitational field, which is not what we have here. So the entire relationship is spurious (and uncited).
• The materials section should simply be deleted. I had added a value for the best real material we have (Carbon fibre), and carbon nanotubes with a citation why the 60,000 value is unreasonable in bulk. He decided to delete the citation and double the specific strength, which moves it from fantasy to truly hallucinogenic.
• As for the artistic impressions, I disagree with Andrew - the thinnest it will be is a cable sufficient to support the payload. That's going to be at least a few millimeters. By the taper will transform it from a cable into a wall once you get significantly above the atmosphere. Tarl N. (discuss) 05:35, 27 April 2019 (UTC)

.

This section was long enough as it was, and it was done. Let's not add anything more to it. I suggest start a new section if there's something new to say. Endless scrolling is "Ugh!". Nice new section is "Aaaah! Sweet relief!".  :-) RFinlay72 (talk) 15:55, 27 April 2019 (UTC)

As an aside, my brainfault in the units is inexcusable; We know the constant divided by specific strength is unitless, so the constant has to be in units which cancel out to be dimensionless. Specific strength is m^2/s^2, so the units of the solved constant for earth geosynch taper have be be in m^2/s^2. Which will matter if anyone wants to calculate taper in furlongs per fortnight. Tarl N. (discuss) 23:32, 27 April 2019 (UTC)

Graph

Graph - would this be better? Tarl N. (discuss) 22:34, 24 March 2019 (UTC)

tape graph

Here are some things I'd like to see for such a graph:
1) The curve should start at zero altitude. It's distracting to show the curve rendered for R values below Earth's surface. It's not like a hole is dug so the cable can continue to the center. Also, gravity stops following the inverse square law below the surface, so the equation is invalid there. If you just don't show the curve for R below the surface then you don't have to explain anything.
2)The scale would be sweeter IMO if zero was set at the surface, so the scale shows "altitude" instead of "radius".
3) If a plot is to be included at all, it should be for a material that would actually be used with a taper ratio that would be doable. I suggest a material of 30MY specific strength. Astronomical taper ratios (shown in the table) are discussed only as an example of why it's clearly impossible to use those other materials. The 10^(very large) numbers for their taper ratios makes that clear without going into detail with it via a plot. It would be undue weight (off topic digression?) to go into that kind of detail for a material that wouldn't actually be used. RFinlay72 (talk) 21:15, 27 March 2019 (UTC)

Using altitude rather than distance from the centre of the Earth requires a more complex equation because the the radius of the Earth has to be added and subtracted at various places. Andrew Swallow (talk) 21:24, 27 March 2019 (UTC)

It's not that complex. It's a matter of making a shifted a scale on the plot, not a matter of substituting Alt for R in the equation or anything needlessly wonky. But, I'm not hard over on that preference. I am fairly hard-over on points 1) and 3) though. RFinlay72 (talk) 21:41, 27 March 2019 (UTC)

@Rfinlay72:. I can start the graph edge at zero altitude (r=6378000 meters), which is what the graph on the article does. I modified this graph to start at zero, so I could have a separate line indicating earth's surface. I'm agnostic about which form of the two graphs would be better, but the labeling should be in radius, not altitude - because the equation as cited uses radius.

As for a more useful material, I disagree. The whole point is that no such materials exist. Not only we don't know of them, but material scientists have good reasons for believing that no such material can exist. So we shouldn't use figures for a hypothetical material we don't even believe is possible. Tarl N. (discuss) 21:38, 27 March 2019 (UTC)

That's not the whole point at all. The point is already made well clear that a space elevator requires new materials. Going on and on and on with details about how bubble gum or dryer lint require large taper ratios -- with plots! -- is digression off topic, and POV. This is not a materials textbook. RFinlay72 (talk) 21:48, 27 March 2019 (UTC)

Predicting the future is forbidden by WP. If you personally are a pessimist about the future of materials, you should keep your POV out of the article. But even then, the article is about an idea, not whether the idea is achievable or not. Stay on topic. RFinlay72 (talk) 21:54, 27 March 2019 (UTC)

The point is that you placed a description of an "advanced material" with specific strength precision down to five decimals. That strongly implies a non-hypothetical. It's not about whether I'm pessimistic or optimistic, it's about a sense of reality. If a space elevator is ever constructed, it will be done with real materials, and we have no evidence that any materials like your "advanced material" can even exist. And there is evidence (which I had cited, and you removed) that the particular 5-decimal precision material you are describing cannot exist. Tarl N. (discuss) 01:34, 28 March 2019 (UTC)

I picked that number only because the taper ratio was already computed for it. I didn't want to recompute the ratio for say 30MY. I saw that the number was ugly, but I just expected someone else to fix it. It didn't occur to me that it would convey that other meaning. Whoops! :-)
Regarding "real materials": Saying "we have no evidence that..." is an expectation you hold based on your experience, knowledge, and constitution. Right? Some others with more knowledge think otherwise, still others with more knowledge agree with you. I actually have no expectation about the arrival of materials technology. It comes or it doesn't. It has no impact on how we go about describing the idea. To digress onto "predicting the future" by giving projections about what will or won't happen is prohibited for good reason. To actually take a side and push the idea that something won't happen makes it additionally WP:POV. Loading up the text with examples of materials that wouldn't work -- and then making plots of them -- is just that. RFinlay72 (talk) 05:46, 28 March 2019 (UTC)

What are the conditions for that equation?

What are the conditions for that equation anyway? Does it give a profile only for a cable that can just barely hold itself up? No safety factor? No load capacity? If so, it's only useful for showing one bound: for a given material the taper ratio needs to be at least X. But, once you decide you want a safety factor or load, the profile shape given by the equation would no longer apply. What does the source say about the equation's applicability?
If the equation is only good for establishing that one bound, then we should make that clear in the text. Also, if it's only good for that, then making plots of profile shape with it would only be the shape for a no-safety-factor, no-load-capacity cable -- which would never be considered to actually build, and so would be off topic. RFinlay72 (talk) 22:21, 27 March 2019 (UTC)

I looked more closely at Aravind. He doesn't mention safety-factor or load. However, he doesn't have A-sub-s as zero. So, there's something there for load and safety-factor, but what? He uses Lc, which implies Tmax. He's not dividing it by a safety-factor. In the real world, a safety-factor would be used of course. It would be accounted for by dividing the specific strength by it (or Lc or Tmax), and then seeing how everything else washes out. If we did that however, it would be WP:OR.
Aravind does use a non-zero A-sub-s, which can be used for a load and/or a safety-factor. He doesn't elaborate on it. There's no indication he gave any thought beyond that A-sub-s couldn't be zero. RFinlay72 (talk) 23:43, 27 March 2019 (UTC)

The conditions for the equation are that tension = cross-section times tensile strength at all locations along the cable. That includes the bottom of the cable, at the earth's surface. If the tension at the surface is (e.g.) 980 kN, that presumes the bottom of the cable is being pulled down by either a 980 kN spring, or 100 tons of weight, applied externally. They both result in 980 kN of tension.

If that tension is a weight, the weight can be placed at different locations on the cable, without causing any difference in tension above the location of the weight (actually, slightly reduced as gravity decreases above the surface). Below the weight, the tension is reduced by those 980 kN.

The equation does not include safety margins. It's easy to do, but Aravind didn't do it. Tarl N. (discuss) 01:28, 28 March 2019 (UTC)

By the way, if there's no safety margin, then a cable based on it will be broken by breathing on it. That's something to think about.  :-)
Does Arvin give that 980 KN in the paper? I don't see any mention of, or and purposeful reason for that 980KN tension at the surface. "Going there" ourselves would be WP:OR.
Truth is, taper ratio and shape are functions not just of Lc (specific strength), but also of safety factor, designed load carrying capacity, etc.. The Aravin equation doesn't do any of that. Arvin seems to only gets his taper ratio by assuming a cable goes to r=0 (or something like that, it's hard to tell), then measuring the area at the surface and dividing the Geo-area by it. We can after-the-fact compute a range of possibilities for safety margin and load, but Aravin didn't do that. It's clear he only meant it to show that taper helps, to give estimate of taper ratios, and to show that some materials are unworkable. He muddled around everything else. The equation just isn't reliable for anything else -- including plots. I'm inclined even to say the equation isn't reliable enough for inclusion at all because its "givens" aren't clear in the source. RFinlay72 (talk) 05:07, 28 March 2019 (UTC)

No, the 980 kN was a "for example" I introduced. But he makes clear that that tension = cross-section times tensile strength. By the way, if there is zero tension at the surface end, $A_{s}=0$, and thus $A_{g}=0$, so we already have such a space elevator in place - zero cross-section all the way to geosynchronous orbit. Tarl N. (discuss) 06:03, 28 March 2019 (UTC)

That's an interesting idea. Putting that "null case" aside for the moment, a cable with no load carrying capability that can just barely hold itself up (safety factor of one) will have a zero x-section area at the surface and any non-zero finite area for Ag. In that case, taper ratio would be infinite and it just couldn't be the measure of how you would shape and size the profile for a material of given Lc (Sp. Strength). Instead, some factor such as (difference-in-geopotential)-per-(double-area) would define the "steepness" of exponential growth of area (or something like that). Also in that case it would be impossible to make a shape with constant stress, right? -- because the stress is zero at the base and non-zero at geo. Actually, maybe not -- if the base was "0" and the bottom of the cable was "0+", maybe you could define an impossibly slender shape approaching "0+" from the top that has constant stress. That would be an interesting (but moot) calculus problem because any real cable would need a design load and a safety factor.
Too bad Aravind didn't specify a safety factor or load. He didn't actually make his assumptions clear, which makes his equation unreliable in my opinion. We really can't go on implying that a particular material always implies a particular taper ratio because it doesn't. Those other parameters are needed. And, we aren't allowed to figure them out for ourselves, they need to be clear in the ref -- we also can't figure it out because Aravind doesn't give enough info for it. I'm not hard-over on removal of the equation, but I don't think lots of results should be computed by it -- such as many materials in a table or any plot. Doing so implies more reliability than the equation actually deserves. RFinlay72 (talk) 07:57, 1 April 2019 (UTC)

You're misunderstanding the equation. At every point, surface area times tensile strength equals tension (max tension). If a tether is designed for (e.g., 980 kN) at the bottom, but that 980 kN tension is not placed on the tether at the bottom, then at every point along the tether, there is an excess of 980 kN of strength above what the cable itself requires. If you want a safety margin, you design that in by adding (max) tension at the bottom, which isn't used. Tarl N. (discuss) 14:51, 1 April 2019 (UTC)

Interesting analysis there. That tension at the bottom could either be a result of an unstated safety margin, or an unstated load capacity, or some of each. Aravind doesn't say anything about either of those things, so we can't know without analysis by us -- original analysis by us. Your analysis isn't includable because it's WP:OR, but also because it assumes things not given in the ref.
Any analysis on the equation (and its supporting text in the ref) to try to extract those assumptions is WP:OR and not includable. But, we may certainly do that kind of analysis on a reference to determine its reliability. If we were to do that analysis, I'll bet both/all parameters (Safety factor, bottom tension, etc.) couldn't be extracted explicitly. At best, you'd have to assume one or more of the parameters and then fit the other(s) to the curve. All that original analysis to make sense out of something that doesn't make sense is some serious WP:OR, so it wouldn't be includable -- but it would serve well if only to show how unreliable the equation really is!  :-)
And yes, I am misunderstanding the equation -- because Aravind provides insufficient information as to his assumptions. Nobody can understand it without those assumptions, which are unstated. Unstated equals unreferencable. RFinlay72 (talk) 21:58, 1 April 2019 (UTC)

Since Aravind doesn't divide Lc by a safety factor, it might be reasonable to infer a zero safety margin. And, if the tension at the bottom really is easy and straightforward to infer, then maybe a statement saying "This equation is for zero safety margin and a maximum load of (bottom tension)." would be okay. It's a real stretch for WP:OR though, and we'd have to verify (to ourselves) that the equation (size and shape) really does fit those assumptions by re-doing the calculus. RFinlay72 (talk) 22:25, 1 April 2019 (UTC)

The equation is very simple - cross section times tensile strength = tension at every point along the cable. It's the meaning behind that which seems to be a problem. For me, it's trivial to observe that there is tension on the bottom of the cable, and that this tension can be some combination of lift and safety margin. If you don't like that, that's fine - and avoiding any further mention of the meaning of the equations is fine. But don't remove Aravind's equations because you think he didn't think of something that he likely thought was too obvious to state. Tarl N. (discuss) 01:00, 2 April 2019 (UTC)

I'm not threatening to remove the equation, but it does need to be qualified for it to stay. As it stands, the equation makes it look like Lc is the only thing that determines profile shape and taper ratio. It's not clear at all that Aravind thought such matters "too obvious to state". What's obvious to me is that Aravind didn't know what he was talking about. Other sources mention that taper ratio and shape can't be solved for on Lc alone. What's required is a design -- which includes safety factor and load capacity (among other things). It's really outrageous (and unreliable) for him to assert taper_ratio vs. Lc like that without clarifying the other parameters. He bothered to clarify "constant stress", but not load or safety factor? Qualify the constraints on the equation without undue WP:OR and it can stay. No problem. Read the 1April 22:25 post I made, I backed off from a hard line there. RFinlay72 (talk) 22:15, 2 April 2019 (UTC)

Evidently, the entire section on cable materials didn't merely need to be rewritten, it needs to be entirely removed. I did remove the blatantly false value for carbon nanotubes. Tarl N. (discuss) 01:44, 3 April 2019 (UTC)

I just undid the CNT removal because it's right out of Aravind and cited too. But actually, I would be fine with removing the whole section. RFinlay72 (talk) 02:10, 3 April 2019 (UTC)

We don't need to do our own research selecting materials and computing TR for them when a table of them is sitting right there in the same source the equation is from. RFinlay72 (talk) 02:14, 3 April 2019 (UTC)

The citation for carbon nanotube strengths is This is the oft-cited 63 MPa strength. Aravind is quoting an NIAC space elevator report three years later, I don't know where they would have gotten a value of 130 Mpa, there was no hint of higher values, and it was understood at the time that microscopic whiskers always tested much higher than bulk materials. Tarl N. (discuss) 04:13, 3 April 2019 (UTC)

That's fair. I'm okay with adding that number in addition to the Aravind number. I suggest notes tagged on to each of the CNT numbers to describe their differing provenance and the fact that one of them is for multi-walled. RFinlay72 (talk) 06:05, 3 April 2019 (UTC)

References

Incidentally, please don't add stuff on breaking lengths. The use of breaking lengths was what started this whole fiasco, and is completely irrelevant. Breaking lengths are descriptions of of uniform cables in uniform gravity fields, neither of which apply to an elevator. As best I can tell, the introduction of breaking lengths was OR to allow further obfuscation. Tarl N. (discuss) 04:16, 3 April 2019 (UTC)

I don't get it.  :-) I added conversions of Lc's to Specific Strength where it seemed apt -- in an attempt to begin to frame things less in terms of Lc and more in terms of Specific Strength. If I say "Lc" on this page, it's only as a linear analog for "Specific Strength" because it's easier to type!  :-) RFinlay72 (talk) 06:08, 3 April 2019 (UTC)

A practical space elevator needs to carry a payload and have a safety margin but an equation without them is still useful because it gives a lower boundary. If the cable is weaker than these figures the space elevator will self distruct. An enhanced equation can contain safety margin and several payloads. Andrew Swallow (talk) 11:02, 3 April 2019 (UTC)

I'll note that the way cables are specified with a safety margin is something like "need it capable of supporting 10 tons, so design for 15 tons". The equivalent of that falls out of the Aravind equations by increasing $A_{s}$, although that's not stated. Tarl N. (discuss) 21:26, 3 April 2019 (UTC)

Safety margins significantly increase costs and regulators frequently get involved with so need their own paragraph in the high level requirements. Andrew Swallow (talk) 22:47, 3 April 2019 (UTC)

Aravind used SafetyFactor=1 (SafetyMargin=0). So yes, his numbers are lower bounds for taper ratio. However, he did the whole thing in a screwed up way that can arbitrarily result in taper ratios starting anywhere from what he gave to infinity (for the same material). He did mention how to adjust for a safety factor, but he didn't think in terms of load. His cables have tension at the surface, but only because he extends his "free standing tower" to the center of the earth and the tension at the base is just happenstance from the mass of the imaginary section of his cable below the surface.
He makes two naive (dumb) assumptions:
1) He assumes a "free standing tower" (as he defines it) is necessary, which it isn't. A free standing tower as he defines it is an academic idealization with little utility in the real world.
2) The bottom of his tower goes to the center of the earth, which is arbitrary. His free standing tower" itself doesn't need to have a bottom at the center of the earth.
-- If instead of bottoming out at the center of the earth, his cable bottomed out at say halfway from the surface to the center, his taper ratio would be much larger.
-- If it bottomed out at the surface, his taper ratio would be infinite.
So, that equation is the result of an analysis that gives widely differing answers depending on a critical parameter (where the bottom is), a parameter that can be just about anything, giving just about anything for a result. The author didn't even discuss it clearly. Now that's the definition of "unreliable".
RFinlay72 (talk) 22:40, 4 April 2019 (UTC)

Gravity drops to zero at the centre of the planet so using the centre of the Earth as the end point is valid. Although an end at the surface of the Earth would be realistic. Andrew Swallow (talk) 23:14, 4 April 2019 (UTC)

Huh? You're first sentence is nonsequitur. The first part is true, but the second part doesn't follow from it (aside from it also being false). Your second sentence is obvious (duh) :-). FYI, when Aravind's derivation "goes to the earth center", he's talking about an imaginary section of his cable below the surface who's weight is manifested in the real world as the tension at the ground-level anchor point. RFinlay72 (talk) 07:52, 5 April 2019 (UTC)

If you want to categorize a reliable source as unreliable, you have to find either a reliable source which categorizes it as such, or a reliable source which contradicts it. You can't just say "I don't like his assumptions" and categorize it as unreliable. Tarl N. (discuss) 01:40, 5 April 2019 (UTC)

Horse hockey. If that were so, any published pile of malarkey, just by being ignored by everyone, would meet the reliability requirement for inclusion. A ton of videos about alien technology on Youtube are now admissible!
Your "argument" for reliability would be stronger if you actually made one instead of just contradicting. Wikipedia doesn't use rote rules like you suggest, it relies on editors' reason and discussion to determine reliability. So, make some actual reasoned and knowledgeable arguments yourself and maybe you'll have a fair point. You tell us why he's reliable. Go ahead, do the math. I did. RFinlay72 (talk) 07:52, 5 April 2019 (UTC)

Artists' impressions

@Tarl N., RFinlay72, Andrew Swallow, and Biem: At the request of RFinlay I am starting a new section to continue the discussion which I restarted a few days ago higher up (which I have now put under the subheading #Reprise).

In answer to Tarl's first contribution of yesterday:

• I don't know why you wrote 4.72 instead of 4.85. You now have the units right, and they are the same as what I said (J/kg).
• The similarity between the needed free breaking length and the radius of the earth is not a numerical coincidence. It comes from the formula which you deleted:

$\Delta \left[\ln(S)\right]={\frac {\rho }{\sigma }}g_{0}r_{0}\left(1+{\frac {x}{2}}-{\frac {3}{2}}x^{\frac {1}{3}}\right),$

where S was what you have called A (the cross-sectional area), r₀ is R (planet radius), and $x=\omega ^{2}r_{0}/g_{0}\approx 0.0035$ is the ratio between the centrifugal force on the equator and the gravitational force. To put this in another way, we can use another parameter, let's call it y, which is $y\equiv x^{1/3}=R/R_{g},$ the ratio between the planet's radius and the "geostationary radius" (though "geo" might not apply if it's not the earth). Then the the change in cross section is given by:

$\Delta \left[\ln(A)\right]={\frac {\rho }{\sigma }}g_{0}R\left(1+{\frac {y^{3}}{2}}-{\frac {3}{2}}y\right)$

As long a y is not too large, we see that to keep the cross section small we have to have

${\frac {\sigma }{\rho g_{0}R}}\approx 1.$

In your example of a planet with the mass and rotational speed of Earth but a smaller density, so that the radius is twice as big, the needed free breaking length will be about twice as much as for the real earth, but since the surface gravity g will be one quarter, it means the specific strength would only need to be half as much as for the real earth. Since y would be twice as much, then we wouldn't need the ratio of free breaking length to planet radius to be as close to 1 as for the real earth to get the same cross section ratio. I don't know how you came to the conclusion that the taper would be about the same (with the same cable material).

• As for deleting the materials table, I think the more you delete from this article the worse it becomes.
• The thickness of the cable at ground level should be on the order of 1 square millimetre. That would be enough to lift about 6 tonnes. The thickness at geostationary altitude would be a few square millimetres. That's why I say that the artists' impressions are all wrong.

Eric Kvaalen (talk) 12:56, 28 April 2019 (UTC)

The free breaking length is just an alternative to T which is the stress the cross-section area can bear without yielding (N·m−2=kg·m−1·s−2), its elastic limit. Because the human mind is used to dealing with lengths the free breaking length is much easier to understand but in official equations physics sticks to the basic units. Andrew Swallow (talk) 17:20, 28 April 2019 (UTC)

It's not "just" that. It appears in the equation I have just mentioned. Eric Kvaalen (talk) 11:30, 29 April 2019 (UTC)

You will find T and free breaking length are linear. Andrew Swallow (talk) 15:44, 29 April 2019 (UTC)

Actually, breaking length serves as a proxy for Specific strength, not tensile strength. All I can say at this point is that any comments about free breaking length in relation to the earth's radius need to be cited from a reliable source. I just don't have the energy to engage in another WP:WALL argument. In answer to the question that will be asked, no I do not regard symbolic calculus as an "obvious and correct" WP:CALC exemption, because it's so easy to make mistakes and has caused so many arguments on this article (going back to long before Eric Kvaalen arrived on this scene). Basic arithmetic (to solve for a specific cited case), as long as it is clearly explained, probably qualifies for that exemption, but symbolic manipulation to achieve uncited conclusions does not. Tarl N. (discuss) 01:40, 2 May 2019 (UTC)

@Tarl N.: The question is whether the artists' impressions that we give in the article are realistic. If not then we should take them out (or point out that they are totally unrealistic). Do you see that what you said about there being no connexion with planet radius was wrong? Eric Kvaalen (talk) 17:08, 3 May 2019 (UTC)

Since we disagree on reality, there isn't much point. You think that taper would be millimetres, we're not in the same universe.

As for the radius issue, when you find a reliable source where someone describes space elevator tapers in terms of free breaking lengths of uniform wires in a constant gravitational field, and that the taper is described in terms of the planetary radius relating to that breaking length, then you can cite it. Heck, submit an article to a peer-reviewed journal and get it published. But Wikipedia is not the place for publishing new conclusions. Tarl N. (discuss) 19:34, 3 May 2019 (UTC)

@Tarl N.: I'm not talking about putting something in the article. I'm talking about taking stuff out of the article, and whether you see that your statement "As for free breaking length and radius of the earth. The only relationship between them is numerical coincidence" is false. Eric Kvaalen (talk) 06:09, 4 May 2019 (UTC)

If you're talking about removing the artists impression that shows two narrow cables, fine by me. I'm time-constrained this week (finals week, a bitch when you're in your 7th decade), I don't have time for discussion. By the way, your ping didn't work (I didn't get notified), I just happened to notice it go by on my watchlist. Looking at the edit history, you added it after your signature was already in place, so Wiki never processed it. It's during the processing of ~~~~ that pings get processed, and if paragraph's signature has already been converted, it doesn't process the ping. Tarl N. (discuss) 00:14, 9 May 2019 (UTC)

@Tarl N.: Thanks for the tip.

There are four "artists' impressions" in our article, all of them showing cables that are so thick that the whole apparatus would have a mass of tens of thousands of tonnes! (I estimate at least 1 kg per metre.) As I said, one would only need a cross section of a square millimetre or so. That would give a total mass of something like 30 times the payload mass, so maybe 30 tonnes.

You still haven't answered my question about the relationship between free breaking length and planet radius.

Are you taking finals or giving finals?

Eric Kvaalen (talk) 15:52, 12 May 2019 (UTC)

If it isn't going into the article, the discussion doesn't belong here, per WP:NOTFORUM. At this point, I have no interest in arguing with you beyond the simple question of "which source are you citing?" Tarl N. (discuss) 06:16, 14 May 2019 (UTC)

You seemed quite interested when you thought you were right. Eric Kvaalen (talk) 14:09, 8 June 2019 (UTC)

Artist impressions are okay

The artist impressions are fine. They're well within the bounds of accuracy expected of such things. Engineers have a wide range of potentially workable ideas, and artists have a wide range of interpretations of those ideas. It's par for the course for "artist impressions". Regarding how thick a cable would be, that depends on the design and the time period, not on original research/design done on a talk page. Also, consider that a cable would be made thicker and thicker over time because bigger is better for safety and utility. So yeah, Kvaalen thinks it's too thick, but it should be thick. RFinlay72 (talk) 19:17, 12 May 2019 (UTC)

I forgot: Since the cables are "ribbons", there's no way to discern their thickness dimension from an illustration. I imagine a cable-ribbon would have a constant width to accommodate climbers so they wouldn't have to adjust their "gage" as they went along. It's the thickness that would taper as a function of height. RFinlay72 (talk) 22:49, 12 May 2019 (UTC)

Here are the four artists' impressions, so we can discuss this more easily:

One concept for the space elevator has it tethered to a mobile seagoing platform.

A seagoing anchor station would also act as a deep-water seaport.

A conceptual drawing of a space elevator climber ascending through the clouds.

Space Elevator with Space Station

In the first one we can see that the cable, or ribbon, is much bigger than needed for the size (or weight) of the platform, since in reality it would only require a cross section of 1 square millimetre to lift about 6 tonnes. In the next one (with the sailboat), we don't see the load, but the cable or ribbon is very wide, so the mass of the cable would be enormous. The third picture has the same problem as the first — the cable is much too big for the load. And the last picture is the worst of all. It shows a huge cable. All four of these would have a total cable mass of tens of thousands of tonnes, or even millions of tonnes in the case of the fourth one.

And to use the argument which has been cited many times above, there's no peer-reviewed reference for any of these! They're all "original research"! And wrong to boot.

Eric Kvaalen (talk) 14:09, 8 June 2019 (UTC)

Think about this. We're talking about a construct which would be expensive (not to over use the "astronomical" adjective, but that kind of expensive). So it will need a lot of traffic to be economical. The speeds along the cable are almost certainly going to be subsonic, to avoid shockwaves creating undue stress along the cable. For subsonic elevator, let's conservatively assume 300 m/s (not knowing the speed of sound in whatever material). So, 36 million meters divided by 300 meters per second is 33 hours. That's a total of 262 trips per year, assuming the cable is used purely for ascent to Geosynch orbit, or a total of 1500 tons lift per year. We can achieve that life capacity today with a few launches of Falcon Heavy, at far lower cost.

I have no idea how much traffic would be needed to justify a space elevator, but it's more than 1500 tons per year. Which means a cable capable of supporting more than 6 tons of payload. Tarl N. (discuss) 18:56, 8 June 2019 (UTC)

Cute, but let's keep the discussion limited to the article. We don't want to give walls-of-texters even more to blather on about, eh? Anyway, isn't it about time to archive this talk page? This page is an extremely tall wall. The bottom 30% or so is from one wall-of-texter. The rest is mostly crazy-old and completely inactive. RFinlay72 (talk) 05:27, 19 June 2019 (UTC)

Auto-archiving header added. It should archive a bunch of this page into archive 9 the next time lowercase sigmabot comes by. Tarl N. (discuss) 05:32, 19 June 2019 (UTC)

I removed the auto archiving, which is brainless and error-prone. Manual archiving is easy enough and it forces at least a little bit of human attention for each case. I'm fine with what the auto system did though. Someday soon I hope we will be able to archive this abominable section!  :-) RFinlay72 (talk) 21:42, 20 June 2019 (UTC)

If a conversation hasn't been touched in 180 days (I figured a half-year was enough), it's clearly done and not actively being used to improve the article. Tarl N. (discuss) 02:24, 21 June 2019 (UTC)

The cables in the first three pictures may be flat and very thin, not cylinders, so this may be an accurate representation. I agree that the Bonnestel-style last painting has nothing to do with any potential reality.Spitzak (talk) 19:52, 9 July 2019 (UTC)

Inappropriate sourcing

I removed some content with terrible sourcing, another user has reinstated it a couple of times. Here's the problem:

The organization published two issues of a peer-reviewed journal on space elevators called "CLIMB"<ref name="isec" /><ref>{{cite web | url=http://www.spaceelevator.com/2012/01/first-issue-of-the-space-elevator-journal-released.html | title=First Issue of the Space Elevator Journal Released | work=The Space Elevator Reference | date=20 January 2012 | first=Marc | last=Boucher | accessdate=2 June 2012 | deadurl=yes | archiveurl=https://web.archive.org/web/20120513064307/http://www.spaceelevator.com/2012/01/first-issue-of-the-space-elevator-journal-released.html | archivedate=May 13, 2012 | df=mdy-all }}</ref><ref>{{Cite web | url=http://www.isec.org/index.php/store/climb-the-space-elevator-journal | title=CLIMB - the Space Elevator Journal}}</ref> and a magazine "Via Ad Astra".<ref>{{Cite book|url=https://www.worldcat.org/oclc/1020867745|title=VIA AD ASTRA - VOL 1|last=ISEC.|date=2015|publisher=LULU COM|isbn=132964123X|location=[Place of publication not identified],|oclc=1020867745}}</ref>

ISEC also conducts one-year studies focusing on individual topics. The process involves experts for one year of discussions on the topic of choice and culminates in a draft report that is presented and reviewed at the ISEC Space Elevator conference workshop to allow input from space elevator enthusiasts and other experts. Study Reports are usually published early the following year, to date these are as follows :<ref>https://isec.org/isec-reports/</ref>

Both spaceelevator.com and isec.com are self-published for the purposes of this content. "X did Y, source, website of X saying X did Y" is bad practice for a topic where fiction masquerades as fact a lot of the time. These claims need sourcing that meets the Wikipedia trifecta of reliable, independent, secondary. These don't meet any of the tests.

The following are self-published books on lulu.com. Lulu is a vanity press. In fact, it is pretty much the vanity press. In each case the "reference" is the sale spage at lulu.

• 2010 - Space Elevator Survivability, Space Debris Mitigation
• 2012 - Space Elevator Concept of Operations
• 2013 - Design Consideration for Space Elevator Tether Climbers,
• 2014 - Space Elevator Architectures and Roadmaps
• 2015 - Design Characteristics of a Space Elevator Earth Port
• 2016 - Design Considerations for the Space Elevator Apex Anchor and GEO Node
• 2017 - Design Considerations for a Software Space Elevator Simulator
• 2018 - Design Considerations for the Multi-Stage Space Elevator

References

These books are basically sales pages for self-published material, and we don't include those. Guy (Help!) 21:52, 7 August 2019 (UTC)

You can buy them, that is probably sufficient evidence for the claim that they published them. I question the relevance of these things, however, and support the version without it until sufficient importance can be demonstrated from independent sources. --mfb (talk) 01:26, 8 August 2019 (UTC)

They're only "published" in the sense that they were made public, just like posting on Twitter is a way to "publish" your thoughts.

They're not "published" in the traditional sense that an established, respected organization has judged them worthy and then paid for the privilege of reselling them. Lulu is a vanity press. They'll print anything you pay them print. ApLundell (talk) 02:30, 8 August 2019 (UTC)

Exactly. There is a high bar to including Lulu books as a source. Visit the website and look for books on known bullshit - the illuminati, for example. On second thoughts, don't. Your brain won't thank you. Guy (Help!) 07:27, 8 August 2019 (UTC)

I noticed that there are free PDF versions of all of the articles on the ISEC website [1], so at least it isn't paywalled. Still, this is self publishing, and likely not WP:RS. I would recommend that editors sympathetic to ISEC set up a project at Wikiversity exploring space elevator options, and perhaps consider making content available on Wikibooks if it has instructional value and an open license. Lsparrish (talk) 18:32, 9 August 2019 (UTC)

Good call. Lulu.com is not a reliable source. I can prove this because I just published a book on lulu.com titled Lulu.com is not a reliable source for any Wikipedia article. If that isn't enough to convince someone, I don't know what will. --Guy Macon (talk) 20:42, 9 August 2019 (UTC)

'0 results for "Lulu.com is not a reliable source for any Wikipedia article" ' :( Anyway, looks like consensus that it doesn't belong in the article. --mfb (talk) 22:48, 9 August 2019 (UTC)

It's a homeopathic book. By that I mean not a book about homeopathy, but a book where the dilution of the text with non-text is so great that the book no longer contains any characters from the original text. Alas, the process of homeopathic dilution also removed the title, author name, copyright notice, etc., allowing multiple unscrupulous authors to copy my work. See Empty book and Intentionally blank page for multiple examples. Even Wikipedia uses it for pages such as redlink. The WMF was agreed^{[Citation Needed]} to compensate me for all such uses at a rate of one dollar per word, payment to be made homeopathically...   :)   --Guy Macon (talk) 02:04, 10 August 2019 (UTC)

Guy Macon, That is fucking hilarious. Even if it's bollocks. Guy (Help!) 11:12, 10 August 2019 (UTC)

seeking clarity of meaning.

the fifth paragraph in the section discussing climbers has a confusing line in it. The paragraph is pasted below, with the relevant statement bolded. Basically, my concern is that an inverted pendulum is dynamically unstable, and the behavior that I think is trying to be described is one that is quite stable, and will be pulled back into vertical (radial?) alignment by the centrifugal force acting on the counterweight. That said, I'm not confident enough to just go and change it.

"The overall effect of the centrifugal force acting on the cable would cause it to constantly try to return to the energetically favorable vertical orientation, so after an object has been lifted on the cable, the counterweight would swing back toward the vertical like an inverted pendulum.[51] Space elevators and their loads would be designed so that the center of mass is always well-enough above the level of geostationary orbit[54] to hold up the whole system. Lift and descent operations would need to be carefully planned so as to keep the pendulum-like motion of the counterweight around the tether point under control.[55]"23.30.241.78 (talk) 01:40, 4 September 2019 (UTC)

Inverted in the sense that the mass is at the top, but swinging like a pendulum because the mass is at a lower potential in rotating coordinates. I called it "a bit like a pendulum" now. --mfb (talk) 09:56, 4 September 2019 (UTC)

Such elevator is nice but absolutely ineffective

Because of changing in angular velocity, climber applies euler force to the cable. Even simple model shows that such force try to change orbit of counterweight and "position" of cable and can be compensated (but not completely) only if center of mass of elevator is higher and much bigger than position and mass of climber and mass of cable. It means that center of mass of elevator must be in counterweight, not as shown on the picture. It also restricts mass of climber. Moreover, for fully compensation of coriolis effect you need the same rockets as for launching "climber" for the same orbit. So, why we need cable?178.140.161.126 (talk) 08:08, 22 September 2019 (UTC)

See [2], [3] and [4]. --Guy Macon (talk) 09:03, 22 September 2019 (UTC)

You don't need a rocket and the center of mass can't be in the counterweight. Anyway, see Help:Talk pages for the scope of talk pages. Personal speculations are not part of it. --mfb (talk) 11:05, 22 September 2019 (UTC)

Well, it is much easier to blame another one in "personal speculation"... Any object in space has one or another orbit because the centrifugal force is balanced by gravity. Centrifugal force depends on orbital velocity. And any space agency uses rocket to accelerate this object to this velocity and so to launch object to the orbit. So, you have no profit up to geostationary orbit. Coriolis force applied by climber must be compensated by rocket (or by thruster) or mass of climber must be much less than mass of counterweight. In another case coriolis force will be too big and system goes below geostationary orbit. 178.140.161.126 (talk) 20:59, 23 September 2019 (UTC)

Repeating wrong claims doesn't make them right. --mfb (talk) 00:09, 24 September 2019 (UTC)

A Commons file used on this page or its Wikidata item has been nominated for deletion

The following Wikimedia Commons file used on this page or its Wikidata item has been nominated for deletion:

• Tsiolkovsky.jpg

Participate in the deletion discussion at the nomination page. —Community Tech bot (talk) 17:14, 30 October 2020 (UTC)

Undid wonky effort to talk about taper in intro.

IP 2600:100F:B049:782D:6AD9:1AF6:4EBE:ADE0 introduced some effort to discuss taper in the intro. It was done awkwardly. It was strictly true, but the way it was done implied that there's an expectation or desire to use a constant cross section, which really isn't true. It also implied that for a tapered cross section we *do* have the materials, which also isn't true. I removed it for that main reason -- that it was worded badly as to create those untruths. But also, even if it was worded better, there's also the problem of bringing too much complexity into the intro. The issue of straight vs. tapered is probably not high-level-summary enough for the intro section. For whatever reason, Mfb is trying to prevent reversion to the original wording. So, discuss...! MartyWu (talk) 18:18, 13 December 2020 (UTC)

"To construct a space elevator on Earth, the cable material would need to be stronger and lighter (have greater specific strength) than any known material." This statement is wrong and needs to go, no matter how long it might have been in the article. Even if you think Kevlar is impossible instead of just completely impractical, carbon nanotubes are a known material. We just don't know how to make very long cables out of it.

Suggestion: Available materials are not strong enough to make a space elevator practical. Carbon nanotubes (CNTs) have been identified as material that could lead to a practical design.[2][4] Other materials considered have been boron nitride nanotubes, and diamond nanothreads, which were first constructed in 2014.[5][6]

Tapering is an essential part of every proposed space elevator on Earth, I think it should be mentioned in the lead. We can add it to the previous paragraph: This structure is held in tension between Earth and the counterweight like an upside-down plumb bob. (existing sentence) followed by: The cable thickness is adjusted based on tension, it has its maximum at a geostationary orbit and the minimum on the ground.

--mfb (talk) 19:41, 13 December 2020 (UTC)

While you were making your comments above I was adding a possible solution. Check it out. By "known materials", I think the intent is to mean macroscopic materials that can actually be made. CNT molecules have the needed specific strength, but none have yet been spun into a fiber that does. MartyWu (talk) 20:05, 13 December 2020 (UTC)

(edit conflict) First, nobody has ever seriously proposed an space elevator design without taper. It is worth mentioning lower down but not in the lead.

Second, the idea that carbon nanotubes are strong enough is dubious at best.

• Nanotubes Might Not Have the Right Stuff
• Carbon nanotubes too weak to get a space elevator off the ground
• A space elevator is possible with today’s technology, researchers say (we just need to dangle it off the moon)
• Carbon Nanotubes Can't Handle a Space Elevator
• Why we'll probably never build a space elevator
• People Are Still Trying to Build a Space Elevator

The only thing that stops "dubious" becoming WP:FRINGE is the possibility that some unknown technology will, in twenty years or so, produce a new material that is strong enough. --Guy Macon (talk) 20:18, 13 December 2020 (UTC)

I had added taper-mention in the intro to appease Mfb. Whether it's mentioned in the intro isn't that important, just that it shouldn't imply undue presence in designers' "thought-space" of constant cross section the way it did. There are some designs that do use constant cross section to advantage though, so it shouldn't be dismissed with so much gusto. MartyWu (talk) 21:20, 13 December 2020 (UTC)

All the pessimistic articles refer to Zhu et al which quotes 50-60 GPa from Yakobson et al. space.com in particular is oddly specific saying you need 62 GPa (61 is not enough?). But Yakobson et al reveals where that number comes from on page 332: You need 63 GPa for... an untapered cable. Add a few rounds of sloppy journalism and you get claims that CNT are not strong enough for a space elevator just because they are not strong enough for an untapered one. And, as you said already, no one wants to build that anyway. --mfb (talk) 23:56, 13 December 2020 (UTC)

Do you have a reliable source that specifically says that the required GPa figures being used are invalid and presenting an allegedly more accurate figure that is lower that the strength of the strongest know materials? Or do you only have WP:OR? --Guy Macon (talk) 01:41, 14 December 2020 (UTC)

There is no hard minimum, it's just a matter of acceptable taper ratio. That means the only specific number you find is one that's not relevant. Yakobson explicitly writes that his number is for an untapered space elevator, it should be obvious that it doesn't apply to tapered designs. There are plenty of sources (already in the article) that discuss how CNT are viable. There is e.g. this paper (arXiv) with an equation for the taper ratio - equation 6. Plug in 63 GPa (and the density of CNT) and you get 3.4, plug in 30 GPa and you get 13, plug in 15 GPa and you get 180. The source explicitly discusses a ~30 GPa cable in several places. They argue that current designs have insufficient safety factor, but that's just an argument for larger tapering ratios. --mfb (talk) 02:22, 14 December 2020 (UTC)

More WP:SYNTHESIS and WP:OR We have no use for the conclusions you have drawn from the above source. What the source actually says is:

"All these approaches unequivocally suggest that the megacable strength will be reduced by a factor at least of ~70% with respect to the theoretical nanotube strength, today (erroneously) assumed in the cable design. The reason is the unavoidable presence of defects in so huge a cable. Preliminary in-silicon tensile experiments confirm the same finding. The deduced strength reduction is sufficient to place in doubt the effective realization of the space elevator, that if built as designed today will certainly break (in the author's opinion). The mechanics of the cable is also revised and possible damage sources discussed."

You can't start with a source that says "it will break" and use that as a citation to support a claim like "They argue that current designs have insufficient safety factor".

Please, if you reply to this, leave out your calculations. Leave out your conclusions. Do not imply any conclusion not explicitly stated by the source you are citing. Nobody cares about your original research. Stop annoying us with it. --Guy Macon (talk) 07:32, 14 December 2020 (UTC)

It doesn't say "it will break". It says "if built as designed it will break". Which is a very different statement. If the author would conclude that it will break - no matter what - then that would be interesting. But they don't do that, for a good reason.

You are asking for sources that say it can be built with CNT. We have plenty of them in the article. You are asking about sources specifying a minimum breaking strength. There is no such thing. No one cares enough about the few sloppy journalist pieces you linked above to explicitly refute them, but they all reference an untapered design, so they are clearly not relevant. So what else could I possibly provide? Note that plugging in numbers into simple referenced formulas should be covered by WP:CALC and this is a talk page not an article. --mfb (talk) 02:19, 16 December 2020 (UTC)

Did you try actually reading WP:CALC? It says "Routine calculations do not count as original research, provided there is consensus among editors that the result of the calculation is obvious, correct, and a meaningful reflection of the sources. Basic arithmetic, such as adding numbers, converting units, or calculating a person's age are some examples of routine calculations." Plugging in numbers into simple referenced formulas is explicitly forbidden. We do not trust you or any other Wikipedia editor to choose the correct formula to plug the numbers into (with the specific exceptions listed above).

You say "no one cares enough about the few sloppy journalist pieces you linked above to explicitly refute them" but that is not a valid reason to not present what is in the sources. Find a source that specifically supports your claims or they don't go in the article. --Guy Macon (talk) 04:20, 16 December 2020 (UTC)

An exponential is only minimally more complex than division. Did I use too many examples for your taste, should I have only used the 63 GPa one? I used the formula for illustration, not to change the article.

As I said repeatedly, we already have multiple of these sources in the article. Bradley C. Edwards (#13 and #2 which should be merged), Smitherman for NASA (#14, needs an archive link like this), and Price referencing Smitherman (#16, archived), universetoday (#23), ... But for some reason you keep ignoring them. Some popular science articles referencing an article that explicitly says it's for an untapered design are absolute truth and everyone else saying CNT are good can't be mentioned. --mfb (talk) 05:26, 19 December 2020 (UTC) PS: Looks like many sources need archived versions.

@Guy Macon: Does the non-reaction mean you have no objections to fixing the lead? --mfb (talk) 22:52, 29 December 2020 (UTC)

I implemented my earlier suggestion as no one objected for two weeks. --mfb (talk) 02:07, 4 January 2021 (UTC)

@IP (Guy Macon?): What you call "fringe" is literally every actual study, and many of them are cited in the article already. The only sources that invent this "impossible" myth are secondary sources misunderstanding the primary sources (or copying from other secondary sources), as documented above. There is no research concluding it's impossible. This is purely an invention by people who have no expertise whatsoever but need to report about it anyway. --mfb (talk) 01:51, 5 January 2021 (UTC)

The Economist, Smithsonian Magazine, New Scientist, Space.com, Popular Science, and New Scientist all specifically say that a space elevator is impossible with any currently known material, while of course acknowledging that in the future some new material that is strong enough may be invented. See WP:CRYSTAL.

You keep telling lies about sources supposedly saying that a space elevator is possible with known materials, but you never quote where they say that because none of them do. All of your sources either talk about possible future materials that may be suitable, or they simply describe some material without explicitly saying anything about space elevators. See WP:SYNTH: "Do not combine material from multiple sources to reach or imply a conclusion not explicitly stated by any of the sources. Similarly, do not combine different parts of one source to reach or imply a conclusion not explicitly stated by the source."

Go ahead. Name the source that says that a space elevator is possible with currently known materials. Quote the exact words where the source specifically says that. You can't do it, can you? 03:38, 5 January 2021 (UTC)2600:1700:D0A0:21B0:9DC2:2F96:B54A:8270 (talk)

First of all, stop inventing quotes. The NewScientist article does not contain "According to their calculations, the cable would need to be twice as strong as that of any existing material including graphite, quartz, and diamond." Adding that again is blatantly wrong and borders on vandalism. That sentence is in the space.com article.

Second, you confuse existing with known. We can't produce long CNT fibers yet, but that doesn't make the CNTs unknown. We can't build an elevator with available materials. That's not in question. Calling the design studies "speculations" and sloppy secondary sources (citing the primary studies with opposite claims) "conclusions" is absurd POV.

A design study "here is how a space elevator can be built" doesn't need to have a single well-quotable sentence. The whole study is the statement that it can be done. And the study even shows how it arrived at that conclusion. Citing a single sentence from it doesn't add anything useful.

What do you think of the following option? Remove the invented quote, combine the second and third sentence to "Some sources have concluded that future advances in carbon nanotube (CNTs) manufacturing can lead to a practical design[2][7][8] while others have questioned this result.[9][10][11]" That way we have the same sources as in the current version and readers can draw their own conclusions. --mfb (talk) 05:30, 5 January 2021 (UTC)

What part of "Do not imply a conclusion not explicitly stated by any of the sources." are you having trouble understanding? If you continue to remove citations to reliable sources because you don't like what they explicitly state and prefer your interpretation of other sources, implying a conclusion not explicitly stated by those sources, you are likely to be blocked from editing Wikipedia.

Also, The article says "Reference 10: "Christensen, Billn (June 2, 2006). "Nanotubes Might Not Have the Right Stuff". Space.com. Retrieved January 3, 2020. "recent calculations by Nicola Pugno of the Polytechnic of Turin, Italy, suggest that carbon nanotube cables will not work... According to their calculations, the cable would need to be twice as strong as that of any existing material including graphite, quartz, and diamond." I just confirmed that that quote is in [5]. Please stop engaging in personal attacks such as false accusations of misquoting. And even if you were right, a simple mistake such as putting a quote in the wrong citation should be dealt with by putting it where it belongs, not accusing other editors of wrongdoing. 07:02, 5 January 2021 (UTC)2600:1700:D0A0:21B0:9DC2:2F96:B54A:8270 (talk)

The wrong quote is in the NewScientist reference, as I wrote explicitly. It's fine in the space.com reference but that's not the only place where you put it. I'm not implying any conclusion that's not stated by any source.

Can you comment on my recent text suggestion please? --mfb (talk) 09:15, 5 January 2021 (UTC)

A search on the phrases "any existing material including graphite, quartz, and diamond" and "recent calculations by Nicola Pugno" only finds those phrases in one place. I am trying to decide whether you are making an honest mistake or deliberately trolling.

I have no objection to "Some sources have concluded that future advances in carbon nanotube (CNTs) manufacturing can lead to a practical design while others have questioned this result" or any other reasonable rewording. What I object to is [A] removal of reliable sources and [B] implying a conclusion not explicitly stated by the source you are citing. The fact that you refuse to acknowledge that this is a Wikipedia policy is troubling, but the "Some sources have concluded that future advances in carbon nanotube (CNTs) manufacturing can lead to a practical design while others have questioned this result" wording appears to be supported by the sources. Please be careful about violating WP:CRYSTALBALL by writing about company plans that were announced years ago and never mentioned since. All such material should be removed from the article unless there is a reliable source that say they are still working on the project. No mention of the current company web page is a very bad sign.

Also troubling is your claim "I'm not implying any conclusion that's not stated by any source." you misquoted Wikipedia policy, which is "Do not imply a conclusion not explicitly stated by any of the sources." You already rejected that Wikipedia policy with "...doesn't need to have a single well-quotable sentence. The whole study is the statement that it can be done." That's just another way of saying "not explicitly stated by any of the sources." 13:21, 5 January 2021 (UTC)2600:1700:D0A0:21B0:94CA:87AA:D065:C566 (talk)

How can you find that sentence only once? Here, I removed it from the wrong place. I posted where exactly it is, I can't see how it would have been difficult to find.

Looks like we found a compromise for the phrasing, good. I implemented that.

I did not quote any Wikipedia policy, so trivially I cannot have misquoted one. I could go into more detail, but generally I find this meta-discussion bizarre and unnecessary. You spend so much time on discussing who used which word where on the talk page instead of focusing on what we should write in the article. --mfb (talk) 11:16, 6 January 2021 (UTC)

International Space Elevator Consortium (ISEC)

This organization has a website and I think that website should be listed under external links. Entwhist (talk) 23:09, 25 March 2021 (UTC)

I was going to say that it is a blog that has no proposed solution to the problem that no known or predicted material is strong enough, but reading the "Reducing the Strength Requirement" section of https://www.isec.org/researchsummary I realized that the page is a blog that actually has a proposed solution to the problem that no known or predicted material is strong enough.

I don't think there is enough coverage in reliable source to justify an external link (see WP:ELNO) but their "support the lower parts of the tether from the earth’s surface using the momentum of objects traveling very fast in a vacuum" idea is certainly something to keep an eye on. If they actually build a small-scale working prototype instead of just blogging about it I expect that they will get a lot of coverage in the press. --Guy Macon (talk) 23:39, 25 March 2021 (UTC)

That's a space fountain/space elevator hybrid. The "problem that no known or predicted material is strong enough" only exists in a few sloppy popular science articles. --mfb (talk) 15:13, 27 March 2021 (UTC)

"This is extremely complicated. I don't think it's really realistic to have a space elevator. It would be easier to have a bridge from LA to Tokyo than an elevator that could take material into space." --Elon Musk, quoted in that notoriously "sloppy"^{[Citation Needed]} source, BBC[6] --Guy Macon (talk) 17:30, 27 March 2021 (UTC)

Guy whose company dominates the rocket market dismisses something that could put his company out of business. Top quality source. But note that he doesn't say it's impossible, or that there wouldn't be any material with the required strength. --mfb (talk) 11:34, 28 March 2021 (UTC)

But if it was feasible, wouldn't they be seriously working on it, instead of saying that it's implausible or impractical? I'm also not saying that Musk is always right, but he's not the first to doubt it, Guy points at some other sources below. I'm not sure I'd call the idea a fringe theory, but it's a hypothesis about future technology that currently amounts to science fiction, with no realistic candidate, something the article should certainly clarify for readers... —PaleoNeonate – 01:42, 29 March 2021 (UTC)

Just to be clear, I am not calling space elevators a fringe theory. I am saying that the view that a material currently exists that is strong enough to construct a space elevator is a fringe theory. --Guy Macon (talk) 05:05, 29 March 2021 (UTC)

It's not feasible with currently available materials, the article makes that clear enough already. No one questions that either. Useful CNTs on a large scale are not a currently available material. --mfb (talk) 13:25, 29 March 2021 (UTC)

Fringe Theory

I retract the "predicted materials" bit I wrote about before. It turns out that there are people willing to predict all sorts of things that never actually happen, and some that predict things that do happen as technology advances -- but you rarely know which is which ahead of time.

As for the strength of materials needed:

• "As outlandish as it seems, a space elevator would make getting to space accessible, affordable and potentially very lucrative. But why it hasn’t happened yet basically boils down to materials—even the best of today’s super-strong and super-lightweight materials just still aren’t good enough to support a space elevator."[7] --Smithsonian Magazine
• "Unfortunately, that structure would not only have to be up to 100,000 kilometres long – more than twice Earth’s circumference – it would also have to be able to support its own weight. Quite simply, there is no substance on Earth that has such properties."[8] --BBC
• "The chief obstacle is that no known material has the necessary combination of lightness and strength needed for the cable, which has to be able to support its own weight. Carbon nanotubes are often touted as a possibility, but they have only about a tenth of the necessary strength-to-weight ratio and cannot be made into filaments more than a few centimetres long, let alone thousands of kilometres."[9] --The Economist
• "A single out-of-place atom is enough to cut [the carbon nanotube] strength by more than half... The tubes’ strength is a result of their atomic structure, with walls made from just a single layer of carbon atoms... efforts to spin multiple nanotubes into a practical large-scale fibre have only produced ropes with strengths of 1 [gigapascal]... a single atom out of place, turning two of the hexagons into a pentagon and heptagon was enough to cut the ideal strength of a CNT to 40 [gigapascals], with the effect being even more severe when they increased the number of misaligned atoms... just one misplaced atom is enough to weaken an entire CNT fibre, and since nanotube manufacturing processes are flawed at the moment, you will inevitably end up with a bad tube in your fibre... That’s bad news for people who want to build a space elevator, a cable between the Earth and an orbiting satellite that would provide easy access to space. Estimates suggest such a cable would need a tensile strength of 50 [gigapascals]"[10] --New Scientist
• " Laboratory tests have demonstrated that flawless individual nanotubes can withstand about 100 gigapascals of tension; however, if a nanotube is missing just one carbon atom, it can reduce its strength by as much as thirty percent. Bulk materials made of many connected nanotubes are even weaker, averaging less than 1 gigapascal in strength. In order to function, a space elevator ribbon would need to withstand at least 62 gigapascals of tension. It therefore appears that the defects described above would eliminate carbon nanotubes as a usable material for a space elevator cable. "[11] --Space.com
• "Researchers at the Hong Kong Polytechnic Institute figured out that even a single atom out of place in the structuring of a carbon nanotube reduces the strength of the structure by dozens to factors. And that’s a serious problem, since it’s very hard to create large-scale sections of carbon nanotubes without a single flaw. Obviously, ensuring that a space-elevator sized volume of carbon nanotubes is totally flawless would take…well, it would be rather time consuming."[12] --Popular Science

So, while some future advance in science might make a space elevator possible (if someone in 1800 said that no known engine technology had the power-to-weight ratio needed for heavier-than-air flight they would have been entirely correct, but roughly a hundred years later someone did it and less than a hundred years after that we landed on the moon) the idea that it is possible using current technology is a WP:FRINGE view. --Guy Macon (talk) 14:06, 28 March 2021 (UTC)

No one claims we would have the right material today that can make a space elevator practical, why do you spend so much time discussing this? The claim is that CNTs can be strong enough if we find a way to manufacture them with a high quality on a large scale. They don't need to be atom-by-atom perfect, but better than what we can do today outside of microscopic samples. --mfb (talk) 20:13, 28 March 2021 (UTC)

Yes, they do need to be atom-by-atom perfect - citations above stating that a single mis-placed atom weakens the tube to uselessness. Worse, exposure to radiation (either gamma or charged-particle impacts as found in the Van Allen belts) will make perfect nanotubes into imperfect nanotubes, weakening them continuously as they remain exposed to space. The statement "no known material exists" would be accurate, because CNTs are demonstrably not usable. Tarl N. (discuss) 20:35, 28 March 2021 (UTC)

Mfb, you are the one who has been WP:BLUDGEONING this page with claims like

• "Add a few rounds of sloppy journalism and you get claims that [carbon nanotubes] are not strong enough for a space elevator"[13]

and

• "You are asking for sources that say it can be built with [carbon nanotubes]. We have plenty of them in the article."[14]

And let us not forget you removing

"Other sources have concluded that [carbon nanotubes] will never be strong enough"

from the article.[15]

"Sloppy journalism" appears to equal "citations to reliable sources that disagree with Mfb's original research".

Carbon nanotubes are not strong enough for a space elevator. The view that they are is a WP:FRINGE view. --Guy Macon (talk) 00:23, 29 March 2021 (UTC)

We cannot produce CNTs on a large scale with relevant strength today (1). No one questions that. You keep adding sources to back that claim for some reason. CNTs with relevant strength are strong enough to build an elevator - if we can produce them on a large scale in the future (2). Every scientific source says so, and plenty of other articles do so as well. A few secondary sources misquote the scientific studies or simply copy misunderstandings from others, which is unfortunate. For some reason you seem to interpret sources for (1) as references against (2) as well. And you keep ignoring all sources that disagree with your view.

@Tarl: No, not uselessness. Just much weaker than the theoretical maximal strength. All the discussed strengths, even the most pessimistic estimates, are sufficient to make a space elevator possible. --mfb (talk) 13:22, 29 March 2021 (UTC)

WP:NOTFORUM section 3. What's the purpose of this section? I see no proposal to change the article in any way. --mfb (talk) 18:24, 29 March 2021 (UTC)

It is in response to your repeated advocacy of a fringe theory in general, and your removal of "Other sources have concluded that [carbon nanotubes] will never be strong enough"[16] from the article in particular. Which was, I might add, a change to the article -- and not a good one. Feel free to stop pushing your fringe theory and other editors will stop disagreeing with you. --Guy Macon (talk) 19:53, 29 March 2021 (UTC)

So you are saying this section is not intended to improve the article in any way? That's exactly what WP:NOTFORUM is about. You are the one who keeps restarting this discussion for no reason. From my side this was done in January. --mfb (talk) 23:14, 29 March 2021 (UTC)