SpaceX: Mars Colonial Transporter "MCT" -- Speculation (not Raptor)

[meekGee]

Well, if you've set a power source on the surface, it now works continuously.  Might not make sense on year 1, but over 10 years, should it does.

Also, the infrastructure to maintain the field (if solar) might not have to grow linearly with the field.

Anyway, I'm personally pro-nuclear.  I'm addressing solar since it might go down that way in spite of what I think... 


Flight wise, imagine that you have a cargo carrier (distinct from the manned MCT) that drops off the payload and free-returns some 3 years later, so can be used on the next² launch window.

You start with one cargo MCT, and enlarge your fleet every launch opportunity - you can hit 25 launches in 4 launch windows - and you have an ever-growing fleet at your disposal.

I don't think 1000 tons is out of the question at all.  And if the vehicles are reusable, the next 1000 tons will take even less time, since you're not starting with a single vehicle.

You'll need precision landing of course, but relatively little robotics - most of the work will be done by the first crew.

[Joel]

If the base relies on shipment of basic supplies from Earth there is no way it will survive. It must be pretty much self-contained. Solar cell manufacturing has made huge advances the last couple of years. I don't think it's unreasonable at all to set up robotic solar cell production, especially for a simple (but less efficient) technique such as thin film solar cells.

As I said, why not sponsor an X-prize for doing so?

[meekGee]

If the base relies on shipment of basic supplies from Earth there is no way it will survive. It must be pretty much self-contained. Solar cell manufacturing has made huge advances the last couple of years. I don't think it's unreasonable at all to set up robotic solar cell production, especially for a simple (but less efficient) technique such as thin film solar cells.

As I said, why not sponsor an X-prize for doing so?

I'm pretty familiar with solar technologies.  I think producing the cells locally a) will happen much sooner than people think, and b) will happen much later than you think 

As for an X-prize - these things are not magic bullets for solving everything...  If there's a demonstrated need (e.g. SpaceX issues a bid for it) then you don't need a prize.  Either it can be done or not.  It's a lot faster for Elon to go to Mia Sole or First Solar and ask them how much it will cost to miniaturize and automate one of their lines, and what such an automated line will look like.   I don't think the answers will be conducive to moving forward with the project though, not in the first few years. 

A big enabler for many of these things will be locally-produced glass.  But even that is further down the line.  The first couple of waves will have to bring the power equipment with them.

One of the nice things about a nuclear reactor, btw, is that there is a lot less assembly required.  So even if its mass ends up being comparable, you don't have to do all the earthwork of installing 20,000 m2 of panels.  It "comes in a box".

[llanitedave]

I get a sinking feeling every time I read these discussions about the massive amounts of cargo required for infrastructure development, for power supplies, for propellant manufacture...  And then doing the whole thing with chemical rockets.  No wonder the costs are so high!

I'd be willing to delay the implementation of these ideas by a few years if we could devote that extra time and expense to developing an efficient, high-isp solution to the transport problem.  That would result in significant mass savings, reduce launch costs, and have cascading effects in increasing practicality and exploration potential in all other areas.

I'd have thought by now that our ion drives would be able to commonly surpass the "sheet of paper" thrust levels that we had in the Deep Space 1 and Dawn missions.  I'm looking forward to the success of VASIMR on the ISS, but I think we need to devote more resources to making this kind of technology available and common.  Ultimately I don't care whether the power source is solar or nuclear, but we need to get these things working in a practical, production sense.

I know it's not in the bailiwick of Space X to develop these kinds of systems for its own missions, but I'd like for this kind of technology to be available to them "off the shelf".

[IRobot]

Elon probably has an open architecture to allow a future high-isp stage, but I think his plan is to go ahead with what we got that works and is cheap.

[meekGee]

I get a sinking feeling every time I read these discussions about the massive amounts of cargo required for infrastructure development, for power supplies, for propellant manufacture...  And then doing the whole thing with chemical rockets.  No wonder the costs are so high!

If you think that's depressing, try to follow through on the mass requirements of a nuclear electric drive.

The problem is that the dV to Mars is not large enough that the gain from a high-ISP but high mass/thrust propulsion system is worth it.

Once you factor in propellant production on Mars, and basically IMO it's going to be chemical all the way.

[Joel]

I'm pretty familiar with solar technologies.  I think producing the cells locally a) will happen much sooner than people think, and b) will happen much later than you think 

<snip>

A big enabler for many of these things will be locally-produced glass.  But even that is further down the line.  The first couple of waves will have to bring the power equipment with them.

In a first iteration you could ship thin film solar cells from Earth and use locally produced glass. According to  http://telecom.esa.int/telecom/www/object/index.cfm?fobjectid=13406, a thin film solar cell blanket can weigh as little as 0.240 kg/m2. Assuming 500 W/m2 on Mars and 5 % daily average efficiency that translates into about 10 kg/kW. Or about 3 tonnes for the hypothetical 300 kW "MCT" refueling system. That's not too much, if my calculations are correct...

[Robotbeat]

10kg/kW is high TRL. Heck, the ultraflex arrays (basically the ones that landed on Mars with Pheonix) can get to 100-150W/kg (or 6-10kg/kW, if you prefer), and they're being used on the advanced Cygnuses, among other things. They're supposedly capable of being deployed at 1g, Mars is less than that.

(ISS's arrays are very inefficient by modern standards, FWIW. There's been enormous progress in the last 30 years or so since ISS's array technology was developed.)

If you build the structure on Mars (or just roll them out on the surface), 1000W/kg (or 1kg/kW) or better is possible.

(all figures assume 1AU insolation. Mars has about half as much.)

[Joel]

10kg/kW is high TRL. Heck, the ultraflex arrays (basically the ones that landed on Mars with Pheonix) can get to 100-150W/kg (or 6-10kg/kW, if you prefer), and they're being used on the advanced Cygnuses, among other things. They're supposedly capable of being deployed at 1g, Mars is less than that.

Yeah, you're completely right. Checking the specifications of UltraFlex  http://www.atk.com/wp-content/uploads/2013/05/UltraFlex-2012.pdf gives something like 4.5 kg/kW for almost off-the-shelf panels. Assuming half of Earth's solar irradiation and 30 % availability (reasonable?) gives something like 10 tonnes of solar arrays to be able to get enough power to produce the propellant for a one shot return trip in ~18 months. That sounds completely reasonable to me. So no need to ship a nuclear reactor, monster cargo deliveries or produce solar arrays in-situ in first iteration.

[meekGee]

10kg/kW is high TRL. Heck, the ultraflex arrays (basically the ones that landed on Mars with Pheonix) can get to 100-150W/kg (or 6-10kg/kW, if you prefer), and they're being used on the advanced Cygnuses, among other things. They're supposedly capable of being deployed at 1g, Mars is less than that.

(ISS's arrays are very inefficient by modern standards, FWIW. There's been enormous progress in the last 30 years or so since ISS's array technology was developed.)

If you build the structure on Mars (or just roll them out on the surface), 1000W/kg (or 1kg/kW) or better is possible.

(all figures assume 1AU insolation. Mars has about half as much.)

Just looked up the PDF.  Fair enough - I was using "terrestrial technologies" since I ruled out non-concentrating TJ cells without thinking enough   

(Using TJ cells with no concentration is extremely expensive. Even at 1000x concentration, TJ cells go for 0.20 $/Watt.  If the modules cost 100 $/Watt, that's $100M for your MWatt plant.  But in this context, it's affordable.)

Still:

A) The one place they were designed for in-atmosphere use, they only had to be designed for 90 days.  multi-year durability will make them heavier again.  These ones look like they can be picked up by the wind outright.

B) The numbers are at the module level.  You still need the framing and foundation.  But let's suppose each of these is stuck on a lightweight rod that's stuck in the ground - as simple as possible. If the foundation is 1 kg/m2, and you get 100 W/m2, that's another 10 kg/kW

So I stand corrected - I ruled out TJ cells without thinking.

Doing the math again - 20 kg/kW, and 1.5 MW field, 30 tons.  Give or take.

Definitely the better option.  Sheeesh.   Amazing what you can get when the hardware price's not a consideration...


This actually makes other components weigh more then the solar field, and so I should re-visit those estimates since I discounted them since they were lighter than my solar field estimate.

I think ISRU plant and habitats were the next two heaviest items.

[meekGee]

10kg/kW is high TRL. Heck, the ultraflex arrays (basically the ones that landed on Mars with Pheonix) can get to 100-150W/kg (or 6-10kg/kW, if you prefer), and they're being used on the advanced Cygnuses, among other things. They're supposedly capable of being deployed at 1g, Mars is less than that.

Yeah, you're completely right. Checking the specifications of UltraFlex  http://www.atk.com/wp-content/uploads/2013/05/UltraFlex-2012.pdf gives something like 4.5 kg/kW for almost off-the-shelf panels. Assuming half of Earth's solar irradiation and 30 % availability (reasonable?) gives something like 10 tonnes of solar arrays to be able to get enough power to produce the propellant for a one shot return trip in ~18 months. That sounds completely reasonable to me. So no need to ship a nuclear reactor, monster cargo deliveries or produce solar arrays in-situ in first iteration.

30% availability (8 hours a day) is too much.

Non tracking solar will get about 4 hours of peak-equivalent per day, if tilted correctly.

[A_M_Swallow]

Tracking can be added to later systems.  The frame, base and supports can be made from ISRU materials like sintered regolith.  The motors, controller, photo-diodes, wires and batteries can be brought from Earth.

[sojourner]

Why not use a solar furnace type of plant?

You could make the basic components (mirrors, stands for the mirror, etc) in situ and then ship in just the hi-tech components from Earth.

[meekGee]

I read up a bit more on the Ultra-flex design.

IMO, it is primarily a space-based design.  I'm impressed that it worked on Phoenix, but it could not have lasted very long.  From what I can see, there's basically no protection on the cells.

This is not to say that a variant can't be developed for long-endurance atmospheric service, but it won't have anywhere near the mass density of the in-space model.

I don't have a source for this, it is based on looking at the PDF and at
http://nmp.jpl.nasa.gov/st8/tech/solar_array3.html
and the fact that most of the heritage for this design seems to be for in-space use.

The other thing to consider is wind-blown dust damage.  Mars has a lot of it, and the winds are fast. The only real scratch-resistant front surface that works on Earth is glass. For a panel like this, maybe 1 mm gorilla glass would do, at 2.5 kg/m2 (or 25 kg/kW), and then you need encapsulant gel as a filler behind the glass, which adds some weight too, and a backplane... And then the CTE issues show up, since the temperature swing is large.

So for a colony application, even at the module level, I'd back away from saying it can be done for 10 kg/kWatt, even if that's what was achieved for Phoenix.

---

Maybe they can come up with a very thin per-cell SiO coating instead.  I don't know.  It's hard, since at least on Earth humidity creeps in through any channel, and a huge amount of effort goes into module encapsulation.  It also took many years of research to arrive at solutions, and the solutions are heavy (not that it's a major consideration for terrestrial systems)

On Mars, you might have additional damage mechanisms such as CO2 freezing inside the cells, etc.

[meekGee]

Why not use a solar furnace type of plant?

You could make the basic components (mirrors, stands for the mirror, etc) in situ and then ship in just the hi-tech components from Earth.

Because the industry that produces these components requires a lot of power (Which you don't yet have) and a lot of industrial infrastructure (which you don't yet have).

You're talking about mining ore and running a foundry.  Or mining sand and making glass.  (Which in order to be transparent, has to be very pure, in comparison with what you'll get just by melting any old sand you find, so requires more infrastructure to separate and purify)

It will happen, but it just can't happen quickly.

[aero]

How do Batteries for over night electricity compare with fuel cells? With fuel cells at least you wouldn't run out of electricity until you had consumed all of the stored fuel you had previously made. Aren't fuels less massive than batteries?

[Patchouli]

How do Batteries for over night electricity compare with fuel cells? With fuel cells at least you wouldn't run out of electricity until you had consumed all of the stored fuel you had previously made. Aren't fuels less massive than batteries?

For short term storage on a small scale batteries win out but long term ie storing enough power for days or weeks fuels cells win out.

Storing enough energy for a colony using batteries to out last a Martian dust storm would be an exercise in futility.

Here fuel cells or nuclear power are your only viable options.

If you're making enough fuel to send a lander back to Earth then you already have most of what you need for a fuel cell based standby power system.

[neveroddoreven]

Everyone seems to be focusing mainly on methane when ISRU is brought up, but I would think that LOX would be a bigger concern due to it comprising a larger portion of the propellant mass and needing to be stored at cooler temperatures. Would be easier to produce the LOX using the various oxides available in the Martian soil (Al₂O₃, MgO, CaO, SO₃, SiO₂, ect...) rather than the atmospheric CO₂? Does anyone have any idea what the least power intensive method of production would be?

[guckyfan]

Everyone seems to be focusing mainly on methane when ISRU is brought up, but I would think that LOX would be a bigger concern due to it comprising a larger portion of the propellant mass and needing to be stored at cooler temperatures. Would be easier to produce the LOX using the various oxides available in the Martian soil (Al₂O₃, MgO, CaO, SO₃, SiO₂, ect...) rather than the atmospheric CO₂? Does anyone have any idea what the least power intensive method of production would be?

If you produce Methane from local water and CO2 you produce the correct amount of O2 in the same process as byproduct out of the water and CO2 used. So no separate production required.

If you want to produce O2 for breathing in the habitat, it is easiest to split CO2 into CO and O. That process requires the least energy input. That be the easiest process too, if you want to produce storable fuel, 2CO and O2 to burn to produce heat and electricity at night.

[Noborry]

... but I would think that LOX would be a bigger concern ...

Mars has plenty of easily-processed Oxygen, in the convenient form of Co2 gas. All that is needed to make LOX is some plumbing and a heap of energy.

Making methane is very much more difficult. The problem being sourcing hydrogen. There simply isnt enough of it in the atmosphere, in any form, to use as feedstock for fuel-making. One would need to either import the H2 (all the way from earth!!!), or mine it as water-ice from the surface.
Mining is *much* more complicated than simply using the atmospheric gasses.