Building a large scale modular spacecraft in space?
Discussion
Eric Mc said:
One of the original mission scenarios suggested for Apollo (and the initial front runner - favoured by Von Braun and his team) was to launch the Command Service Module on one Saturn V and launch additional fuel tanks on a second Saturn V 90 minutes apart. The two elements would dock in earth orbit and head off to the moon as a single craft. This original scenario had the Command Service Module fitted with legs and the entire craft would descend down to the lunar surface.
Buzz Aldrin in 'No dream is too high' says that NASA engineer John Houbolt had the idea to make the CSM and LEM two units.It wasn't actually Huboult's idea but he championed the concept and essentially put his job on the line in an effort to get it considered as an option. He made himself very unpopular at NASA for a while - until Von Braun suddenly decided that Lunar Orbit Rendezvous was the only way forward. Suddenly, he and his team were heroes.
This is a fairly well known picture of him explaining the concept around 1962
He died only two years ago.
This is a fairly well known picture of him explaining the concept around 1962
He died only two years ago.
slybynight said:
I think he means - assembly of an interplanetary craft in orbit - ISS isn't going anywhere but down! It certainly makes all that sort of thing a bit more possible, but why would you bother? Mars is about the only realistic place to go isn't it? Colossally difficult (don't forget the habitat you'd need to take with you/have pre-deployed there) and what's the point? Other than some sort of life-boat insurance against us completely destroying this planet.
Well quite, that's exactly why humans are driven to explore. The human race would have died out if the guy who said "hey, I wonder what's over that hill" had instead gone back inside the cave. The things you learn along the way (whether it's going to Mars, or simply looking through a telescope) have payback for "problems on Earth". I wish I could find an essay where Asimov wrote about how Astronomy had led to various discoveries with direct relevance to life today - it was highly surprising how many scientific discoveries had been triggered by pointing a telescope up and wondering what was "out there".
When Orion starts heading to the moon and beyond on a regular basis (from about 2020) it is planned that it will be docked to a habitation module.
The module will either be launched separately or perhaps, like Apollo, as part of an enlarged SLS booster/Orion combination. As can be seen in the image, the Orion spacecraft will need to dock nose first with the habitation module. If the combination is launched together on one SLS, then the Orion will have to perform a transposition and docking manouever similar to what Apollo used to do.
The module will either be launched separately or perhaps, like Apollo, as part of an enlarged SLS booster/Orion combination. As can be seen in the image, the Orion spacecraft will need to dock nose first with the habitation module. If the combination is launched together on one SLS, then the Orion will have to perform a transposition and docking manouever similar to what Apollo used to do.
You'd think that in these days of computer design/simulations etc they'd be able to ask a computer 'How to get three men to Mars?', feed in all the parameters and it would think up the design of the perfect/most efficient/cheapest rocket to do it, calculate the exact cost and print out the spec.
There would probably be a drop-down menu 'Required probabilty of success: 100, 95, 90' etc.
There would probably be a drop-down menu 'Required probabilty of success: 100, 95, 90' etc.
Eric Mc said:
What types of scenarios would a computer come up with that haven't been thought about already?
No idea. The theory is that instead of humans batting various ideas about for decades and getting nowhere, the computer would simply tell them what was best. Take it or leave it.There are lots of useable options - no one of which are "best". It depends on all sorts of variables. Generally, regarding Mars, you have a two year gap between launch windows - due to the relationship between earth and Mars as they each orbit the sun. The size and weight of the item you want to land on the surface dictates the size of the rocket you will need to get the basic machinery off the earth and provide enough braking fuel to drop you into Mars orbit when you get there.
All the variables will have been looked at by those who are planning such missions and eventually the methodology that suits their needs best will be chosen.
All the variables will have been looked at by those who are planning such missions and eventually the methodology that suits their needs best will be chosen.
Simpo Two said:
Eric Mc said:
What types of scenarios would a computer come up with that haven't been thought about already?
No idea. The theory is that instead of humans batting various ideas about for decades and getting nowhere, the computer would simply tell them what was best. Take it or leave it.And the modelling - which will be based on the basic physics of orbital mechanics (known for 500 years).
The ways to get to Mars using current and available technology are well known and understood. We've been doing it with unmanned probes for 51 years.
The new bits that need to be learned are how to drop much larger objects safely down onto the surface, protecting the humans on the way to and from the planet and protecting them on the surface.
None of these are insurmountable problems and don't need computers to be solved.
The ways to get to Mars using current and available technology are well known and understood. We've been doing it with unmanned probes for 51 years.
The new bits that need to be learned are how to drop much larger objects safely down onto the surface, protecting the humans on the way to and from the planet and protecting them on the surface.
None of these are insurmountable problems and don't need computers to be solved.
reminds me of the apocryphal tale of the early days of computing. Time was booked on the latest supercomputer to analyse retail data. everything was fed in to the machine in the hope of finding new efficiencies/untapped profits.... supply logistics, materials cost, household budgets, the end result the machine came up with was..... " Have christmass every week."
Id imagine the mars planner would come up with something similar. "Organise slaves to shovel uranium into the nuclear reactor engines
Id imagine the mars planner would come up with something similar. "Organise slaves to shovel uranium into the nuclear reactor engines
I don't want to come across as a dismissive luddite on this. But look at what has happenned to manned planetary exploration since Apollo....As soon as the great unwashed realised that there was no Jetsons like holidays for them on the moon, They lost all interest in funding it. I'd rather someone went for it BIG STYLEE. Shipping stuff up to Mars that could be used for tunnelling/colonisation etc. If its just a couple of people taking a small step for man, I think most people will have turned over to Big Brother before they even get there.
Simpo Two said:
xRIEx said:
Sorry, I thought your first post was a joke. You know computers don't think, right?
I did say 'feed in all the parameters'. It's all down to maths at the end of the day.You could, I suppose, try to use some genetic algorithms (*) to "evolve" a spacecraft design.
The difficulty with that is there is a large gap from the basic modelling done in most academic studies of genetic algorithms (akin to "Game of Life") versus building a million-component rocketship.
Ultimately at the moment the best designers are still human.
(*) https://www.doc.ic.ac.uk/~nd/surprise_96/journal/v...
The difficulty with that is there is a large gap from the basic modelling done in most academic studies of genetic algorithms (akin to "Game of Life") versus building a million-component rocketship.
Ultimately at the moment the best designers are still human.
(*) https://www.doc.ic.ac.uk/~nd/surprise_96/journal/v...
xRIEx said:
A computer program can simulate the relative merits of a a hemispherical nose compared to a conical nose of a rocket, but it can't decide, "well, how about I try a hexagonal pyramid nose and see if that's better than the two initial options." It takes humans to imagine the possibilities before the computer can be programmed to model the scenarios.
I see it as a kind of triangle, like Ohm's Law, namely that if you know two out of three things you can work out the third. In your scenario you have the nose shape and the atmosphere and calculate the aerodynamics; I see a possibility to feed in atmosphere and aerodynamics, and the optimum nose shape is computed. Meantime I'll settle for being 'ahead of my time' 
PS I should add that I am greatly in favour of the human element - if used constructively.
Building a concrete plant on the moon might be an option:
http://www.nss.org/settlement/nasa/spaceresvol3/cp...
All the materials are there: plenty of aggregate, water ice (https://en.wikipedia.org/wiki/Lunar_water) and high alumina compounds to form the cement. Sections could be 3D-printed using the base materials extracted from the lunar surface. They could then be transported using re-usable rockets and assembled in the lunar orbit, all entirely remotely. All you'd need to send from earth are the 3D printers, compressed air for entrainment and perhaps carbon fibre strands for reinforcement. All are lightweight in proportion to the mass of material they could generate on the lunar surface. That's where re-usable rockets can help out, by ferrying components for assembly in lunar orbit. Moving the thing about is another matter entirely though..
http://www.nss.org/settlement/nasa/spaceresvol3/cp...
All the materials are there: plenty of aggregate, water ice (https://en.wikipedia.org/wiki/Lunar_water) and high alumina compounds to form the cement. Sections could be 3D-printed using the base materials extracted from the lunar surface. They could then be transported using re-usable rockets and assembled in the lunar orbit, all entirely remotely. All you'd need to send from earth are the 3D printers, compressed air for entrainment and perhaps carbon fibre strands for reinforcement. All are lightweight in proportion to the mass of material they could generate on the lunar surface. That's where re-usable rockets can help out, by ferrying components for assembly in lunar orbit. Moving the thing about is another matter entirely though..
Edited by 30v on Sunday 24th July 22:03
Max_Torque said:
Potentially stupid question: How does one work out the stress calcs for a craft that is built, and only operates in, zero G? I assume presurisation loads will far out weigh any others??
I guess so, but even the ISS has to boost its own orbit from time to time, and the fact it is so big means different parts of it are in different effective orbits (by only a little) so this needs corrections too. not suer what the mass of the ISS is but it would have to cope with a little force to boost it.As for building a modular spacecraft, yes thats what we have already done several times.
Theres a certain minimum size/weight to make it worthwhile though, otherwise all you end up with is the weak and useless bits (docking lamps etc).
AFIk SpaceX's plan for manned Mars is one huge space ship weighing about 100 tons ( Mars lift off weight).
Gassing Station | Science! | Top of Page | What's New | My Stuff