Author Topic: Elon Musk: glass geodesic domes  (Read 238513 times)

Offline Ionmars

  • Full Member
  • ****
  • Posts: 1702
  • North Carolina, USA
  • Liked: 663
  • Likes Given: 1817
Re: Elon Musk: glass geodesic domes
« Reply #340 on: 11/22/2016 02:18 pm »
Spreading out the load seems a good idea in general.  How about a robotic sealer that tracks along a group of six panes and sits down slowly as it has a large valve to allow a softer action.

Edit/

   This same bot could be used on the outside to clean off dust, otherwise it might end up being the little bot that never gets used.  :o
A replacement panel for a leaking or blown-out panel should approach the damage area from the inside of the dome rather than the outside. This way the rush of air to the outside would draw the replacement panel towards the old panel frame. This rather than fighting against the current.

The outside bot could pass the replacement panel edgewise through a frame opening to the inside and insert from there.
Edit: This may require a thin wire loop embedded in the middle of each panel. Then the outside bot could be put to work during construction. Once the framework was built, the outside bot could install every panel.
« Last Edit: 11/22/2016 02:31 pm by Ionmars »

Offline LMT

  • Lake Matthew Team
  • Senior Member
  • *****
  • Posts: 2351
    • Lake Matthew
  • Liked: 424
  • Likes Given: 0
Re: Elon Musk: glass geodesic domes
« Reply #341 on: 11/22/2016 03:45 pm »
SpiderBot

Spreading out the load seems a good idea in general.  How about a robotic sealer that tracks along a group of six panes and sits down slowly as it has a large valve to allow a softer action.

Edit/

   This same bot could be used on the outside to clean off dust, otherwise it might end up being the little bot that never gets used.  :o

"SpiderBot, SpiderBot.  Does whatever a SpiderBot does." 

Ref.

« Last Edit: 12/14/2016 06:35 pm by LMT »

Offline TripD

  • Full Member
  • ****
  • Posts: 872
  • Peace
  • Liked: 851
  • Likes Given: 677
Re: Elon Musk: glass geodesic domes
« Reply #342 on: 11/22/2016 03:45 pm »
@Ionmars   Yah it doesn't show perspective very well, but the view shown is actually on the inside.  Sorry about the confusion.

Offline Ionmars

  • Full Member
  • ****
  • Posts: 1702
  • North Carolina, USA
  • Liked: 663
  • Likes Given: 1817
Re: Elon Musk: glass geodesic domes
« Reply #343 on: 11/23/2016 07:21 am »
@Ionmars   Yah it doesn't show perspective very well, but the view shown is actually on the inside.  Sorry about the confusion.
Sounds good.
Skunkworks will still have a challenge to transfer this technology from an airship to a dome on Mars; particularly where  trusses are employed between panels. The machine couldn't slide over the surface without bumping into ridges. No doubt you will find a good solution and I hope you will have the chance to develop it. :)

Offline Ionmars

  • Full Member
  • ****
  • Posts: 1702
  • North Carolina, USA
  • Liked: 663
  • Likes Given: 1817
Re: Elon Musk: glass geodesic domes
« Reply #344 on: 11/23/2016 08:47 am »
This post is a follow-up to my previous Reply #313 in this thread. Here I present a better sketch of the modified dome structure and a further modification. The idea is to employ a series of “super-hoops” that are over-designed to resist tension forces. In this way we can design the paneled areas to resist the tensional forces between these hoops and not for the total force against the entire dome. Note that the hoops over the center of the dome are the largest and therefore have been made thicker (stronger) as compared to those near the periphery.

The new change is that the panel coverings between hoops are not flat, but are dome-shaped. This is analogous to a domed roof over a tunnel, but here the domed tunnel is draped over the larger dome structure. The reason is that smaller domed roofs can resist greater tensional force. In this example the distance between hoops is arbitrarily set to 40 meters, a distance that a domed-roof could span with a good safety margin. The result is a large dome of about 230 meters.
« Last Edit: 11/23/2016 08:48 am by Ionmars »

Offline Rei

  • Full Member
  • ****
  • Posts: 540
  • Iceland
  • Liked: 332
  • Likes Given: 161
Re: Elon Musk: glass geodesic domes
« Reply #345 on: 11/23/2016 10:59 am »
This post is a follow-up to my previous Reply #313 in this thread. Here I present a better sketch of the modified dome structure and a further modification. The idea is to employ a series of “super-hoops” that are over-designed to resist tension forces.

Just like I wrote last time:

Quote
That design will be vulnerable to torsion and lateral stress.  A geodesic by design resists torsion and stress equally from all directions.

That doesn't make them perfect.  From a mass perspective, the optimal shape involves trusses.  Geodesics are simpler to build than truss structures, but the bending moment of inertia is proportional to the thickness of a beam to its, what, third power?  Haven't done a FEA run in half a year so I'm a bit rusty  ;)  But basically if you want to maximize stress and minimize mass, you have a heirarchy of structural members, with small numbers of those of the greater thickness, leading to greater numbers of less thickness, and so forth, all the way to the glazing.

Of course, your lowest mass structures are pressure-supported.

The last time I designed a greenhouse, it was an isogrid, about twice as long as it was wide, so I tried using the asymmetry to allow me to focus reinforcements predominantly on the narrow axis.  It didn't work, even in that "easier" case.  The shape tries to twist or lean and overload the weaker elements on the long axis.

The same will happen to you here, but even moreso.  If you try to put all of your strength on one axis, it'll transfer to the weak elements and overstress them.

Also: it's nice and easy to draw a bunch of random elements.  It's a whole different ballgame when you have to actually join them together in a buildable fashion.  And account for the strength of the joints in your FEA run.
« Last Edit: 11/23/2016 11:05 am by Rei »

Offline Rei

  • Full Member
  • ****
  • Posts: 540
  • Iceland
  • Liked: 332
  • Likes Given: 161
Re: Elon Musk: glass geodesic domes
« Reply #346 on: 11/23/2016 11:15 am »
On the upside: unlike on Earth, most of your stress on Mars is self-loading.  Your wind / snow and even dust loads will be very small.  And of course gravity is lower.  Only in cases where you want to pile regolith on top do you have to deal with heavy loadings per square meter.

Anchoring will be a pain, however.  Maybe earth screws?  Better pick an area that's not shallowly underlain by bedrock...

Just thinking about the environment... I'm betting that corrosion between dissimilar metals is going to be huge.  Normally your panels basically act like charge collectors even on Earth, accelerating corrosion between, for example, an aluminum frame and stainless bolts, or whatnot.  But on Mars they'll be far more effective at it.  Every last element that contacts every last other element better be very resistant to galvanic corrosion, w/proper isolation.

(I know some people have talked about CF-framing.... I've never heard of anyone making a CF-framed greenhouse, so I have no clue how well that'd actually work.  They make fiberglass greenhouses (I never have), which are good in some regards, but vulnerable to UV and are flammable.  But CF with an advanced binder might be able to work around that (at significantly greater material cost - but that's typical for space applications)

Offline LMT

  • Lake Matthew Team
  • Senior Member
  • *****
  • Posts: 2351
    • Lake Matthew
  • Liked: 424
  • Likes Given: 0
Re: Elon Musk: glass geodesic domes
« Reply #347 on: 11/23/2016 02:01 pm »
Storm Corrosion (Of Love & Emotion)

Just thinking about the environment... I'm betting that corrosion between dissimilar metals is going to be huge.  Normally your panels basically act like charge collectors even on Earth, accelerating corrosion between, for example, an aluminum frame and stainless bolts, or whatnot.  But on Mars they'll be far more effective at it.  Every last element that contacts every last other element better be very resistant to galvanic corrosion, w/proper isolation.

Yeah, I recall Stan-1967 joked about setting up an electroplating shop.  ISRU++.  (Hi Stan.)

Cathodic protection would seem necessary.  I think magnesium anodes should protect common frame metals:  titanium, aluminum, iron (with ISRU), whether above ground or underwater.  Carbon fiber also, I guess, judging from C electronegativity; though carbon fiber suffers underwater for other reasons.

Magnesium would be sacrificed rapidly at the anodes.  I suppose it could be reprocessed, or else replenished relatively easily from basaltic sand, via for example the Calera CO2 sequestration process, which one might use for Ca/Mg fertilizer production anyway.



« Last Edit: 12/14/2016 06:34 pm by LMT »

Offline meekGee

  • Senior Member
  • *****
  • Posts: 14158
  • N. California
  • Liked: 14046
  • Likes Given: 1392
Re: Elon Musk: glass geodesic domes
« Reply #348 on: 11/23/2016 02:54 pm »
On the upside: unlike on Earth, most of your stress on Mars is self-loading.  Your wind / snow and even dust loads will be very small.  And of course gravity is lower.  Only in cases where you want to pile regolith on top do you have to deal with heavy loadings per square meter.

Anchoring will be a pain, however.  Maybe earth screws?  Better pick an area that's not shallowly underlain by bedrock...

Just thinking about the environment... I'm betting that corrosion between dissimilar metals is going to be huge.  Normally your panels basically act like charge collectors even on Earth, accelerating corrosion between, for example, an aluminum frame and stainless bolts, or whatnot.  But on Mars they'll be far more effective at it.  Every last element that contacts every last other element better be very resistant to galvanic corrosion, w/proper isolation.

(I know some people have talked about CF-framing.... I've never heard of anyone making a CF-framed greenhouse, so I have no clue how well that'd actually work.  They make fiberglass greenhouses (I never have), which are good in some regards, but vulnerable to UV and are flammable.  But CF with an advanced binder might be able to work around that (at significantly greater material cost - but that's typical for space applications)
Because his is more of a giant pressure vessel than a greenhouse.  And the bigger it gets, the harder it is to build.

ABCD - Always Be Counting Down

Offline Oersted

  • Member
  • Senior Member
  • *****
  • Posts: 2897
  • Liked: 4098
  • Likes Given: 2773
Re: Elon Musk: glass geodesic domes
« Reply #349 on: 11/23/2016 03:47 pm »

Domes sound fancy, but what about good old fashioned pyramids? To me it seems much easier to work with (and produce) flat surfaces.

A geodesic dome is composed of flat surfaces and very few different elements. That is the beauty of it.

Offline Ionmars

  • Full Member
  • ****
  • Posts: 1702
  • North Carolina, USA
  • Liked: 663
  • Likes Given: 1817
Re: Elon Musk: glass geodesic domes
« Reply #350 on: 11/23/2016 07:28 pm »
...
...
The last time I designed a greenhouse, it was an isogrid, about twice as long as it was wide, so I tried using the asymmetry to allow me to focus reinforcements predominantly on the narrow axis.  It didn't work, even in that "easier" case.  The shape tries to twist or lean and overload the weaker elements on the long axis.

The same will happen to you here, but even moreso.  If you try to put all of your strength on one axis, it'll transfer to the weak elements and overstress them.

Also: it's nice and easy to draw a bunch of random elements.  It's a whole different ballgame when you have to actually join them together in a buildable fashion.  And account for the strength of the joints in your FEA run.
Thanks for sharing your experience. My intuition and some of the earlier posts were wrong. So as the size of dome increases you should only increase the thickness and strength of all members equally. Eventually the members are so thick as to crowd out any usable glass panes. Se la vie, Elon.

Offline meekGee

  • Senior Member
  • *****
  • Posts: 14158
  • N. California
  • Liked: 14046
  • Likes Given: 1392
Re: Elon Musk: glass geodesic domes
« Reply #351 on: 11/23/2016 10:07 pm »
...
...
The last time I designed a greenhouse, it was an isogrid, about twice as long as it was wide, so I tried using the asymmetry to allow me to focus reinforcements predominantly on the narrow axis.  It didn't work, even in that "easier" case.  The shape tries to twist or lean and overload the weaker elements on the long axis.

The same will happen to you here, but even moreso.  If you try to put all of your strength on one axis, it'll transfer to the weak elements and overstress them.

Also: it's nice and easy to draw a bunch of random elements.  It's a whole different ballgame when you have to actually join them together in a buildable fashion.  And account for the strength of the joints in your FEA run.
Thanks for sharing your experience. My intuition and some of the earlier posts were wrong. So as the size of dome increases you should only increase the thickness and strength of all members equally. Eventually the members are so thick as to crowd out any usable glass panes. Se la vie, Elon.

Actually the struts can grow linearly in width, and linearly in thickness, giving you what your need w/o reducing the fraction of "open sky".

However, as you grow in diameter, you start paying a lot for unused volume.  You then start looking at partial domes, so at least you're not wasting skin, but structurally, you don't get get a break.  a 50 m diameter 120-dome and 180-dome requires the same thickness.

So this really moves you towards a large number of smaller units, and if you're in the 5-10 m range, full spheres start gaining an advantage compared to just domes.

So I'm predicting a combination of both:  up to 50 m diameter partial domes, and up to 10 m diameter full spheres.

For larger enclosed volumes - subtropolis-type excavations.  Maybe even in salt deposits.


ABCD - Always Be Counting Down

Offline Rei

  • Full Member
  • ****
  • Posts: 540
  • Iceland
  • Liked: 332
  • Likes Given: 161
Re: Elon Musk: glass geodesic domes
« Reply #352 on: 11/23/2016 10:20 pm »
...
...
The last time I designed a greenhouse, it was an isogrid, about twice as long as it was wide, so I tried using the asymmetry to allow me to focus reinforcements predominantly on the narrow axis.  It didn't work, even in that "easier" case.  The shape tries to twist or lean and overload the weaker elements on the long axis.

The same will happen to you here, but even moreso.  If you try to put all of your strength on one axis, it'll transfer to the weak elements and overstress them.

Also: it's nice and easy to draw a bunch of random elements.  It's a whole different ballgame when you have to actually join them together in a buildable fashion.  And account for the strength of the joints in your FEA run.
Thanks for sharing your experience. My intuition and some of the earlier posts were wrong. So as the size of dome increases you should only increase the thickness and strength of all members equally. Eventually the members are so thick as to crowd out any usable glass panes. Se la vie, Elon.

Not exactly.  Again, I've mainly dealt with plastic, but with polycarbonate panels, if you want longer spans, you simply use thicker panels.  Increase the panel thickness and you increase its bending moment of inertia, proportional to the third power of the thickness.  Now, there are practical issues to take into account; if the gap between the two layers gets too large, it loses insulative ability because gas moves around too readily, so you start having to add more layers.  In addition to the number of layers having light-absorbing properties, each additional layer also has an inherent degree of scattering/reflection.  So for example, here in Iceland, which is rather low light by Earth standards, but with relatively mild winters and cheap heating, double layer is usually optimal, which caps you off at about 12mm panels.  But in a place like Minnesota where light is more abundant but heating problems more challenging, you'd probably want at least three layers, if not more.

With your structural elements, the same scaling factors apply.  Increase the thickness a little bit on a given beam axis, the moment of inertia increases dramatically on that axis.  So this is what you have to work against the problems of increasing spans, wherein your stress increases proportional to the third power and your deformation to the fourth power.  Now, your mass increases a lot as you start increasing the dimensions of a beam on each axis, so what you do if that becomes too much of a problem is you instead make trusses, which dramatically increases the effective thickness of the "beam" (now truss).  But of course that comes at a the cost of a lot more joints, which is more engineering work, construction work, complexity, etc.  And on and on it goes.

Basically, there's countless options when it comes to construction, and lots of tradeoffs.  So you define your parameters - what sort of dimensions you want, what sort of spans, what your environmental conditions are, what your construction methods will be, etc, and then you work through progressive iterations to find what is - as a whole - the optimal solution.

I'll point out that I haven't done any work with structures where you have an elevated internal pressure.  As a general rule, optimized structures in tension tend to be lighter than ones in compression, so it's understandable why NASA has done a lot of looking into inflatable greenhouses.  A lot of such papers I've seen for "inflated" structures these days tend to focus on plastic films (of a variety of types) reinforced with a high tensile matrix, often most vectran.  Your typical non-biaxially oriented plastic may have a couple dozen MPa tensile strength (higher for more crystalline plastics, lower for non-crystalline, in general).  A biaxially oriented one may be in the ballpark of 200MPa or so.  But high strength unaxial fibers are several GPa.  So obviously you want the stresses born by fibers, but with broader sheeting in-between.  And generally you want a multlayer sheet because each type of layer provides different properties - tensile strength, elastic modulus, gas permeability (often different between water vs. other compounds), UV resistance, abrasion resistance, fog resistance / antifouling, transparency in different parts of the spectrum, and so on.  And these parameters often vary tremendously - permeability differences can be across many orders of magnitude, for example.  Then there's issues of manufacturing, bonding, packaging (foldability), and on and on...  :)

« Last Edit: 11/23/2016 10:26 pm by Rei »

Offline LMT

  • Lake Matthew Team
  • Senior Member
  • *****
  • Posts: 2351
    • Lake Matthew
  • Liked: 424
  • Likes Given: 0
Re: Elon Musk: glass geodesic domes
« Reply #353 on: 11/24/2016 02:46 am »
No(de) Limits

Noodling a basic Thistle Dome GSA: 

Abaqus CAE pops out a structural module with ballpark properties.  But a full Thistle Dome with hundreds of modules and fine mesh / realistic stresses will blow past the Abaqus Student Edition limit of 1000 nodes. 

Question:  Anybody seen or heard tell of a comparable structural shareware with unlimited nodes?

Computing time isn't a problem.  Mesh oversimplification is a full-stop problem, on even a tiny dome segment.

« Last Edit: 12/14/2016 06:34 pm by LMT »

Offline oiorionsbelt

  • Full Member
  • ****
  • Posts: 1766
  • Liked: 1190
  • Likes Given: 2685
Re: Elon Musk: glass geodesic domes
« Reply #354 on: 11/25/2016 11:34 pm »
There has been a lot of discussion about anchoring these domes but as they will be pressurized, how would they be sealed where the dome meets the regolith?

Offline lamontagne

  • Senior Member
  • *****
  • Posts: 4266
  • Otterburn Park, Quebec,Canada
  • Liked: 3838
  • Likes Given: 716
Re: Elon Musk: glass geodesic domes
« Reply #355 on: 11/25/2016 11:47 pm »
There has been a lot of discussion about anchoring these domes but as they will be pressurized, how would they be sealed where the dome meets the regolith?
Either the wall is buried a few feet down,or there is an airtight membrane brought in towards the center of the floor.

Offline oiorionsbelt

  • Full Member
  • ****
  • Posts: 1766
  • Liked: 1190
  • Likes Given: 2685
Re: Elon Musk: glass geodesic domes
« Reply #356 on: 11/26/2016 12:00 am »
There has been a lot of discussion about anchoring these domes but as they will be pressurized, how would they be sealed where the dome meets the regolith?
Either the wall is buried a few feet down,or there is an airtight membrane brought in towards the center of the floor.
Could a "ridgid floor" solve both sealing and anchoring issues?

Offline oiorionsbelt

  • Full Member
  • ****
  • Posts: 1766
  • Liked: 1190
  • Likes Given: 2685
Re: Elon Musk: glass geodesic domes
« Reply #357 on: 11/26/2016 12:02 am »

Offline Aussie_Space_Nut

  • Full Member
  • **
  • Posts: 280
  • South Australia
  • Liked: 130
  • Likes Given: 430
Re: Elon Musk: glass geodesic domes
« Reply #358 on: 11/26/2016 12:40 am »
I wanted to try and get my head around these Geodesic Domes so I decided to model one up. I chose a "Kite" shape as only 1 panel shape is reguired for a sphere, or 3 panel shapes for a half dome. (The Kite panel and both its LH and RH half.)

My thinking was a flat pack dome/sphere.

I found that the Kite panel became too large, IMHO, for large 100 metre domes.

I think though they could be used to finish either end of a "Tube" style dwelling. But even then the interface between the half dome and the tube becomes a problem due to the "facets" of the dome meeting the smooth curve of the tube.

In the end I think the whole Geodesic idea is great for "Cool Looking Architecture" here on Earth but for Mars is simply impractical. Not saying it cant be done, I just dont think it can be done easily. Happy to be proved wrong though!


Offline Aussie_Space_Nut

  • Full Member
  • **
  • Posts: 280
  • South Australia
  • Liked: 130
  • Likes Given: 430
Re: Elon Musk: glass geodesic domes
« Reply #359 on: 11/26/2016 12:45 am »
For my second dome attempt I tried panels which all had the same edge length. A Triangel, Square and Pentagon plus their LH & RH half panels. My conclusions were same as before. The panels just get too big to handle.


Tags:
 

Advertisement NovaTech
Advertisement Northrop Grumman
Advertisement
Advertisement Margaritaville Beach Resort South Padre Island
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
1