SpaceX aims big with massive new rocket | Hacker News Reader

SpaceX aims big with massive new rocket

flightglobal.com

117 pointssomeperson49 comments

If we're playing 'guess-the-acronym', I'm going to depart from something Mars-based and suggest that the 'C' is to do with 'Cryogenic', and that now that they are more confident with engines they might be switching to a harder, but much higher performance, fuel combination like liquid oxygen with liquid hydrogen.

Whereas LOX+Kerosene has a Specific Impulse (crudely the amount of 'go' you get from a pound of propellent) of about 340s, LOX+Hydrogen engines have an Isp of about 450. That's really a huge increase in terms of what you can launch with a given size of rocket.

Tom Mueller, the head of propulsion at SpaceX, previously worked for TRW (the company that did the US's ICBMs in the 50s) and while there he worked on a LOX+Hydrogen engine, the TR106 ( http://en.wikipedia.org/wiki/TR-106 ), which was a beast.

It's harder to work with - both propellants are very cryogenic - but any rocket engine designer will be very aware that LOX+Hydrogen is hovering at the top of the evolutionary ladder looking mighty attractive for those that have the resources to take it on - a position Space X find itself in as they become more mature and heavyweight. They're already looking at it for their replacement Upper stage vacuum engine for Falcon 9 - the Raptor ( http://en.wikipedia.org/wiki/Raptor_(rocket_stage) ).

Also, almost any Mars mission would involve a lot of heavy lifting into low earth orbit to construct a vehicle that would transport people there and back. A space vehicle that can support several people for say 2 years, and move itself to and from Mars, is going to by way bigger than anything that can be launched by a single rocket all at once. A big heavy-lifter whose engines have a much improved Isp (which rather relaxes the expensive mass saving measures you need in the rest of the rocket to launch a given payload, or alternatively lets you launch a much heavier payload for the same vehicle) is consistant with a mars road map.

I doubt they would undertake something like this just for Mars - developing rocket engines is hecka expensive. SpaceX got started by inheriting most of the cancelled FastTrac project from Nasa ( http://en.wikipedia.org/wiki/Fastrac_(engine) ) and indeed still outsource their turbopumps (the hardest bit of an engine) to Barber Nichols, who did the design for Fastrac. That was a great move on their part - starting with a blank sheet would have been madly expensive and very slow - a working baseline engine design is a very high-dimensional optimisation problem, before you even get to cutting metal. My point is, if they're now starting clean-sheet engine design, I'm sure they must have in mind some way for it to pay for itself because it's a hugely costly business.

We live in interesting times!

Hydrogen brings absolute oodles of troubles with it. For upper stages, where it is much more critical to minimise the wet weight, that trouble is generally worthwhile. For lower stages, not so much -- it's generally FAR more cost-effective to just build a heavier stages which will do the same job, but with lower ISP. In the past, SpaceX has always made the (correct, IMHO) decision to optimise the cost-efficiency of their rockets, rather than the propulsion efficiency, and hopefully they won't stray from that path.

(They've said in the past that they're working on a LH2 upper stage already, so if you see them building up a cryogenic capacity, that would explain why).

Also, I kinda miss the days when they could get away with calling their next generation rocket the "BFR", and no true geek needed to even ask what that stood for.

Indeed, as I say it's very much harder to work with, but it's worth a revisit with 2012 technology, I think. I also am involved with oxygen+hydrogen rocket propulsion research but can't really discuss it much here. But things have come on a bit in the last 30 years :)

Re: 'BFR', well I have a few hypotheses about 'MCT' in a similar vein but I reasoned they might be a little bit blue for HN!

Edit: A better reason for first stage hydrogen occured to me over a cup of tea:

> "SpaceX has always made the (correct, IMHO) decision to optimise the cost-efficiency of their rockets, rather than the propulsion efficiency, and hopefully they won't stray from that path."

I think the economics change when you try and go reusable in the way that they are. Lox Hydrogen engines and tankage might be more expensive, which is obviously a factor in an expendable vehicle, but if you reuse the vehicle, and have to keep enough fuel back to land, as SpaceX plan, I think actually the equation goes far more back towards optimising for propulsion efficiency in order to get the cost efficiency. The launcher cost metric is basically dollars per kg into orbit, and for a given reusable rocket, lox+hydrogen will get many more kgs into orbit than lox+kerosene - you need to keep much less fuel back in reserve to land the various rocket stages, and the extra fuel you can use on the ascent is also a lot more thrusty per unit mass. I wouldn't be surprised, assuming you sell at a dollar/kg price, if lox+hydrogen ended up being much more profitable in reusables than lox+kerosene, everything else being equal. I have not done the sums though.

I admit to having a bit of an attitude about LH2; for 25 years after they developed the SSME, NASA maintained an attitude that if "it ain't hydrogen, it's crap". No non-hydrogen developments were funded during that time, and not coincidentally, no new first-stage engines ever reached flight status. Until SpaceX came along and developed its own hydrocarbon engines, we were reduced to buying hydrocarbon engines from the Russians. During that time I developed a heck of a bias towards hydrocarbons -- but I have to grant that it's certainly possible that advances in LH2 have balanced out the negatives since then.

You're correct that as you go reusable, the economics shift away from capital costs to operating costs. However, it seems to me that when you consider first stages from a systemic perspective, it's not obvious that it's any cheaper. Yes, Hydrogen gives you much more impulse, and so for a given degree of propulsion, you can use less fuel by weight. However, LH2:

1. Costs 4x as much as kerosene by weight

2. Is only 11% as dense as kerosene, requiring much larger and heavier (and highly insulated) tankage to hold it, and larger and heavier turbopumps, combustion chambers, and everything else that puts weight on a rocket. So while your wet weight may still be less, your dry weight is generally quite a bit higher. T/W is always lower than for hydrocarbon systems.

3. This much greater bulk increases your drag losses if you're operating in an atmosphere; lower T/W increases your gravity losses.

4. Although hydrogen doesn't suffer from coking on the combustion surfaces the way that hydrocarbon engines do, it embrittelises (is that a word?) everything it touches, making between-flight inspections a much more crucial and delicate process. There are hydrocarbon engines which have been fired thousands of times between rebuilds, for hundreds of hours of firing time. Show me a hydrogen engine is even 10% as durable.

Of course when you're in orbit, neither drag nor gravity loses matter, and the nature of the Rocket Equation dictates that every ounce of gross weight really matters. So LH2 makes undoubted sense there. But the exact inverse is true when you're launching off the ground. Unless some fundamental breakthroughs have occurred which I'm not aware of (always possible), I will remain very surprised if SpaceX is pursuing cryogenic first stages.

All valid and I'm enjoying the discussion but I would address the points as:

1) Sure at the moment, and it will be difficult to match economies of scale with the oil industry, but the amount of research money going into hydrogen production at scale is enormous. It's only getting cheaper, and while 'The Hydrogen Economy' has been promised since the 70s and hasn't yet arrived, it's definitely coming. I'm sort of thinking about this debate in a 10+ years out sense rather than right now, I admit.

2&3. Sure, but that only becomes a smaller problem as rockets get bigger (i.e. very heavy launchers such as this one being proposed) and the cube/square law works in your favour (i.e. tank volume increases faster than frontal area to which the drag is proportional). T/W is often limited anyway by payload constraints rather than rocket engineering constraints, especially in manned systems. Doesn't matter if the vehicle can do 5G off the tower, the people inside it can't.

4) 'Embrittels' I would guess? Anyway, it occurs in most metals but certainly not all metals and very certainly not "everything it touches". This is the clue for the way out of this problem. I obviously couldn't show you a hydrogen engine with the same track record as no operational hydrogen oxygen engines have been built with a clean sheet using what we know today. That said, a few are being designed precisely as we speak for very long life operation in reusable vehicles. I don't think the chickens have come in to roost yet.

I'm enjoying the conversation too! Your point about the scaling properties making LH2 more favourable for larger vehicles is something I hadn't properly appreciated in those terms, but is certainly correct. As far as T/W goes, I think that you do need to be able to pull 3G off the tower, and moreover do it with one propulsion system (or cluster of identical systems). Again from a systems-engineering point of view, the point where a launch vehicle design jumps the shark is where you're strapping together completely disparate types of rockets to get the desired effect. I don't think that an LH2 engine has ever actually done this, although the the X-33 designs perhaps could have worked. (I always thought that the Venturestar was emblematic of everything that can go wrong with an LH2-driven design, since it basically turned into a (non-)flying kitchen sink exposition. But the DC-Y was elegant enough; I would have loved to see somebody make an honest attempt at that.)

So basically, it sounds like there are three things required to make a cryogenic first stage/RLV viable:

1. A market for large payloads -- large enough to allow the scaling laws to work out in LH2's favour -- with sufficient frequency to pay off the development costs. As such, this doesn't exist yet, but it's certainly conceivable that someday it could.

2. New types of LH2 engine designs, which are much more durable than SSMEs (which I guess would be the current benchmark of LOX/LH2 durability? Or would that be RL-10s?) Presumably with some substantial revolutions in materials sciences. This may be well underway already (as you allude), but is not public info yet.

3. New, cheaper ways of generating hydrogen. As long as the lowest-cost method is steam reformation, then hydrocarbons will, by definition, be cheaper. If that's the case, then you'd end up in a curious situation where LH2 might only be economical in the middle of the spectrum. In an ELV which is primarily concerned with capital costs, hydrocarbons clearly are more more cost-effective, since they're much easier to develop and build. On the other hand, in an RLV which achieves true airline-like operations -- where the cost of fuel begins to become a meaningful concern, which it currently is not -- then hydrocarbons win again. That might only leave a thin band in the middle where LH2-based designs are the most cost-effective. On the other hand, a new low-cost hydrogen-generation technique (eg., bacteria that exhale it) could change the game.

So yes, I guess I'd echo your sentiment that this is something that could be relevant in a decade, if all three of those items develop in the right way. Definitely promising enough to merit some R&D, but not yet assured or inevitable.

Agree on all counts, especially 2. Funny you should mention the [LOX/LH2 Single Stage To Orbit] DC-Y! If you can only have one stage, Hydrogen is also very attractive. I agree with you that someone should make a proper go of it because it's possible and it makes a lot of sense. It was actually this realisation that prompted me to change career and work where I now do. (I am British, work in propulsion research, especially Oxygen (be it stored liquid or from the air...) with Hydrogen, am interested in reusable space vehicles, and I think wings are A Good Thing, and I live in the city where you did your MBA. As a fellow space enthusiast that's hopefully enough for you to figure it out :D ).

Oho! That is indeed a very legitimate and exciting piece of technology which you're working on! Lucky you!

(Just to be clear, I'm not remotely an actual rocket scientist. I'm a transport planner / urban designer who specialises in Personal Rapid Transit systems, and starts companies on the side for fun; I've just spent an inordinate amount of time in pubs with fairly eminent rocket scientists, and have acquired an Opinion or two along the way. Plus, I need to keep one foot planted firmly in that world, so that once I've made my fortune elsewhere, I can move on to colonising the inner solar system without delay...).

Personally, I've always tended to be more of a VTVL guy, but anyone who can make wings work certainly has my support. In the case of of Reaction Engines / Skylon, I have to confess a bit of skepticism -- not from a technical point of view (the credentials and capabilities of the team are superb), but from a business-plan point of view. I've done various offhand models to get it to work, and the financing costs always kill it. It only works if you presuppose A.) an existing launch market of about 250-500 Skylon-class payloads per year, and B.) that SpaceX or Blue Origin haven't succeeded in lowering launch costs with their flyback boosters etc.

In contrast, SpaceX has a much more bootstrap-able evolutionary path, which allows them to largely avoid financing costs, and to theoretically get their customer prices closer to their marginal costs much more quickly. My feeling has always been that Reaction Engines ought to focus more on the possibilities for aviation, since it is a large enough market that it could absorb the R&D / finance costs much more easily. Once you're turning a profit in that market, then use it to bootstrap Skylon.

All IMHO, of course!

This gotta be one of the most interesting technical conversations that I have seen on HN!! Since you both seems to have a lot of knowledge about propulsion systems, I wanted to ask you about advice on how to get into the field. I am an aerospace undergrad graduating next year. I am thinking of going to grad school (undecided between masters or phd) and propulsion is one of the areas that I would like to work on.

Well, I'm not at all in the field -- armchair quarterbacking it is just an old hobby for me. But if you're in America, my advice would be: go to Space Access every April, and plan on spending your summers in Mojave. That's how most of the people I know have done it.

Wow, great discussion!. May I ask a dumb layman question?, seeing the huge flame that rockets produce (obviously), It wouldn´t make sense to create a kind of statoreactor at the lower part of the rocket to take advantage of all that heat?.Like a kind of after burner, that way it would be possible to use even an greater mass of air to increase thrust (at least for the limited time that the rocket is still in the atmosphere, then drop the "ring" as another stage).

Where does liquid methane CH4 fall in the spectrum between H2 and Kerosene?

Heh, I didn't realise Musk used to do that.

I found a well written and interesting article from that period of time (2005), it's not his first announcement of "BFR" but it's worth a read just to remind ourselves how far SpaceX has come. Also it has some tidbits on his roadmap for the vehicle's development: http://www.thespacereview.com/article/497/1 (Fairly sure he's talking about the super heavy lift, using the Merlin 2 and all)

>Musk said he that while he would like to slow down the rate of growth the company has experienced since its inception over three years ago, he envisions that SpaceX could grow as large as 400-500 people, but no larger. “Companies do change once they grow beyond a certain level,” he said. “Once everyone doesn’t know everyone, the company becomes slightly different.”

SpaceX now has more than 1800 employees.

I'm going to add a link to this most excellent book about rocket propellants called Ignition!: http://library.sciencemadness.org/library/books/ignition.pdf. Foreword by Asimov himself, who was friends with the author. It's an entertaining description of years of work on the topic.

My comment serves no purose other than to say YES. I love this book dearly, it's probably the best written and most enjoyable niche technical book ever written. It's an absolute gem. I high commend it to people as an excellent short read. One of those instances where you say 'why on earth would I be interested in the history of liquid rocket propellents?' but you should read it anyway - it's massively enjoyable, some bits are laugh-out-loud funny, and a tour de force in technical writing and communicating complex ideas to non specialist readers.

It actually got me into rocket propulsion research, having been a signal processing, ML and control theory person by background. May I offer it my highest commendation to the curious HNer and thank you baq for providing a link to it.

We lived in much more interesting times. N1 rocket 40 years ago had better isp. And "Energia" single unit (up to 8 could be equipped in a single complex) had isp of 476s 20 (yes, twenty) years ago. This enabled the USSR to send 20t to Mars or 8t to Jupiter in 1991.

But the USSR falled, and this tech will never be recovered.

The Antares rocket uses, or is about to use, NK-33 engines recovered from the N1 program. The reason these engines are so efficient is that they are closed loop. But that doesn't mean they are overall cost efficient. Of course when you're going beyond orbit the calculations change.

http://en.wikipedia.org/wiki/Staged_combustion_cycle_(rocket... http://www.orbital.com/Antares/

>>But the USSR falled, and this tech will never be recovered.

Ugly fallout of politics. By the way what happened to all the technology and intellectual power?

I mean shouldn't Russia still have it?

http://youtu.be/WmR1QRGeBV4?t=2m7s

Not really politics. Those rockets didn't work very well. (did the N1 ever have a successful flight?)

Russia had plenty of stuff that DID work well, and managed to make a business out of selling launch services.

    Also, almost any Mars mission would involve a lot of heavy lifting
    into low earth orbit to construct a vehicle that would transport
    people there and back.

I keep seeing this assertion but I'm genuinely curious to know why this is the case.

If you're going to assemble the vehicle in LEO anyway before you send it off to Mars why is it so much better to do so by lifting up X pieces for assembly with X huge rockets instead of lifting up X * N pieces on X * N smaller rockets?

Sure you get into engineering problems eventually if your rockets are too small, since any single component can't be larger than those rockets can carry. But if nobody even has a design for such an in-orbit-assembled vehicle, then just starting out by making a really big rocket seems to be a case of putting the cart before the horse.

My guess, and I'm sure someone more knowledgable will chime in, would be efficiencibutton a larger rocket/lift capability- Defintely with the current 'throwaway' model, where whole rocket costs are ~95%+ cf. fuel; but I believe even with re-usable rockets (as they are aiming for) you have significant cost advantage per kg (or pound if you prefer) from a larger rocket.

So, as I see it, less cost, less launches, less messy construction (larger modules/however the beast turns out)

Maybe. Or maybe it just means a larger payload. The cost per kg could end up anywhere with new technology. At least in the beginning it might be much higher.

In the long run, you'd expect carrying heavier payloads to be cheaper per-kg (because you're lifting more payload and less rocket, relatively speaking), but carrying larger payloads to be more expensive (because you need to be using a bigger rocket). Those play against each other to some extent.

In the very short term a new technology is likely to be pricier than the existing tried, tested and refined version (especially when you include R&D costs), how long it stays that way is hard to predict.

Actually, someone came up with a variant of Zubrin's Mars Direct plan that involves in-orbit assembly of pieces launched by existing rockets.

possible payload range of the new rocket is 150-200t to low Earth orbit (LEO). A vehicle of that size would easily eclipse NASA's proposed Space Launch System, which will eventually be capable of launching 130t to LEO

This really allows you to begin to grasp the economics of what Musk et al are doing -

* The Space Launch System has been funded to the tune of $17Bn through to 2017 (Wiki)

* SpaceX got through the Falcon 1 and Falcon 9 development to where we are now (Viable launch system with paying clients) on around ~$1Bn

Even assuming that they have to start the development from scratch, new tooling, new plant and equipment, you could extrapolate out at a stretch that perhaps it would cost a cool $3Bn (And likely well under this) if they are planning on having some form of prototype 'MCT' up and running in around 3 years

so that would be around 1/6th the cost of a bloated government program for an extremely heavy lift capability, potentially re-usable if they are able to complete the engineering of that little monkey, making production of a space elevator within the realms of possibility (and maybe making earth rocket launches redundant!) as well as cheap heavy launch capability to mars.

The solar system is getting a lot smaller, it seems.

> making production of a space elevator within the realms of possibility

Rocketry has little to do with a space elevator, the materials science for a space elevator does not even exist.

you're right, they are completely different fields.

However one of the more likely methods of construction (from this distant point anyway) would be a launch of lots of material, tie it all to some asteroid or other large counterweight that has been put in space a good 90,000+km away, and then unspool it all to the surface, which would require a good and cheap heavy launch capacity

(I suppose you could say if we are going to do such an undertaking as build a space elevator, with all the billions it may entail, then launch costs to get it going will probably be the least of our worries, but as far as I see it, every little bit helps!)

The problem is not the principle of the thing, the problem is that a part of it would have to be made of unobtanium.

Earths gravity is such that you'd need a material for the tether that is stronger than any material that we are currently aware of.

So there is no point in getting any of the mechanisms in place or developed in any detail until there is a material that can withstand the forces in play. Right now, sad but true, an earthbound space elevator is science-fiction.

http://en.wikipedia.org/wiki/Unobtainium

By the looks of it, the future generations on earth will look at our chase for unobtainium in the same we look at alchemists today.

It kind of depends on which experts position paper you read on it - I point you in the direction of

10.1016/j.actaastro.2012.01.008 (Towards the Artsutanov's dream of the space elevator: The ultimate design of a 35 GPa strong tether thanks to graphene, N.M. Pugno)

For someone that is exploring the materials we will likely use in any attempt at an elevator.

Yes, we can't make the materials today. Like, right now. Or even next month, especially in the quantities needed. But we also can't make the rockets that might make it more accessible. And, more than likely, the first space elevator will be built on the moon, not on earth, which will definitely require heavy lift and would probably be a proving ground for the technology and materials.

So yes, science fiction, but so is pretty much everything else that spaceX wants to do.

They always say the future is just around the corner and in this case they might just be true. Or at least, as you can tell, I hope it to be true.

I thought maybe incorrectly that a carbon nanotube based material would provide the required strength? But we currently do not have the technology to build it?

The material doesn't exist, but the materials science certainly does. Carbon nanotubes are theoretically strong enough, and a ribbon made of nanotubes several centimeters long with a good epoxy gluing them together would do the trick, according to a NASA study by Bradley Edwards.

Nobody's achieved it yet, but plenty of people are working on it.

The really exciting thing is that they're working on making things reusable since that seems to be the major barrier to affordable launches. Things like reusable rocket stages: http://spaceflightnow.com/news/n1209/24grasshopper/ and reusable spacecraft: http://en.wikipedia.org/wiki/Dragon_(spacecraft)

I always found it mindblowing that we spent billions on systems that we then destroyed to get things out of our gravity well. I'm really exited about what the future will bring, what a great time to be alive.

"Elon Announces He Might Be Building A New Rocket, Provides Absolutely No Details At All."

6m's pretty big, though. Falcon 9 is 3.66m. Ares V (the space shuttle replacement) is 10m. Saturn V was 10.1m.

I'm really not sure why they're not using RP-1, though. What's the alternative propellant? LH2 has all sorts of problems.

My guess is that they will be using a methane engine.

That's an interesting point.

Methane is sort of halfway between kerosene and hydrogen. No operational space launcher has used it, though, as far as I know, some engines like the venerabe RL-10 hydrogen upper stage work horse have been modified to run on it on a test bench.

Methane is about half-way between Kerosene and Hydrogen in terms of Isp but it's far closer to Kerosene in density, which is critically important for overall achievable rocket stage performance (weight ratios). Liquid methane has about half the density of Kerosene but liquid Hydrogen has about 1/12th the density. Moreover, liquid Hydrogen is a super cryogen whereas liquid methane has a higher boiling point than LOX so handling, materials, and insulation requirements for LCH4 are comparable to LOX.

I was thinking the same thing - either Moon or Mars Crew / Cargo / Colony Transporter. It's an exciting time around these parts, and as I understand it the Merlin concept is pretty scaleable.

Perhaps the new fuel he's speaking of is some sort of hypergolic mixture (Hydrazine etc) that can be used for just about everything once the vessel arrives somewhere without much available oxygen.

A few days before this article, a redditor reported a conversation where Elon Musk tells his friend that Dragon is just an appetizer for what he called "Mars Colonial Transport"... MCT?

http://www.reddit.com/r/space/comments/11bpvi/elon_musk_just...

This was submitted 11 hours ago (http://news.ycombinator.com/item?id=4661364), but only got 9 upvotes so was probably missed by many people. I sincerely believe this is strong Hacker News material that wasn't successful due to an unfortunate submission time, so it's worth resubmitting. This news deserves quality HN discussion.

> Musk declined to say what 'MCT' stands for, and declined to answer further questions on the project.

There's some speculation [1] that MCT stands for Mars Crew Transporter (or Mars Cargo Transporter).

(The comments on [1] are very interesting, speculating on Musk's aims for a rocket that currently doesn't have that much value to private industry)

[1] http://nextbigfuture.com/2012/10/spacex-developing-larger-en...

aaah my dreams are coming true. How exciting, even from this far off speculative point.

> During an April interview, SpaceX president Gwynne Shotwell discussed a project with similar characteristics, describing engines with "more than 1.5 million pounds" of thrust.

For comparison, 6.7MN (1.5 million pounds) was roughly the thrust of Saturn V's F-1 engines (it had 5 on first stage and that many J-2 — at a mere 4.4MN each — on second stage), and it had an LEO payload quite a bit under those asserted here (120t).

Going beyond that means SpaceX is going to compete with the Zenit's RD-171 (~8MN), which is the current holder of "most powerful liquid-fuel rocket engine" title (the shuttle boosters are the holder of "most powerful rocket engine" title at more than 12MN each).

The Merlin page[0] seems to confirm that, suggesting the Merlin 2 is planned for around 8.5MN. That's absolutely enormous.

You still have to wonder how many of these things they'll put on a rocket to get it to push the expected 200t to LEO: the original Energia only lifted 100t to LEO with 4 zenit boosters (so 4xRD-170) and its own engines (4xRD-0120 at 2MN each), there were plans to have a 200t-lifting Energia… by strapping 8 zenits to it

[0] http://en.wikipedia.org/wiki/Merlin_(rocket_engine)#Merlin_2

How many launches per year?

I don't see the economics here. Not with expendable, not with reusable.

Plus, if you have fewer big engines, you can't land on them as their thrust is too high.

SpaceX - because screw FTL, we'll just build a better rocket. ;).

Jokes aside, they're doing really impressive job and I'm very happy to see visible progress in what I thought was a stale industry.

Given Musk's vision for the future, MCT clearly means Mars Colonization Transport.

Are they building Sea Dragon?

Eaoq

Reply on news.ycombinator.com