spacex SpaceX's "BFR/BFS/MCT" Thread -- Facts, Conjecture, Math and Hypothesis

[Unobscured Vision]

Periphery: There's enough of a trickle of information regarding SpaceX's in-development super-heavy lifting vehicle (called the "Big <snip> Rocket", lovingly) and the Mars Colonial Transport (also referred to as the "BFS", in the same vernacular) that we can start compiling some information about its' rumored and not-so-rumored statistics & capabilities.

 

We'll also begin compiling some of the rumored technology that'll make it work, both theoretical and what SpaceX is actually doing in the build-up for BFR's General Information Release which could be as soon as September 2016.

 

 

Images: (L) Falcon-XX/Raptor Concept by NSF-L2 Community --- (R) SpaceX Launch Vehicle Concepts (2010) @SpaceX

 

http://www.nasaspaceflight.com/2014/08/battle-heavyweight-rockets-sls-exploration-rival/

 

The above article is a great place to get familiar with the concepts of what SpaceX want to accomplish with "BFR" -- whether its' final name ends up being "Eagle" or "Falcon XX" is not known yet.

 

More information will be added as it becomes available.

[DocM]

Cargo and tanker BFS versions seem to be no brainers, and what could you do with a 50-100 tonne satellite or big satellite dispenser under cargo door(s)?  Eek!

 

How about prepositioning tankers at Sun-Mars L2 as depots for a manned asteroid belt mission? Ceres? Fly a BFS as a reusable space station/exploration gateway where ever?

 

The implications of a large, reusable, long duration spacecraft like BFS could be immense and SpaceX is talking Mars landings in under 10 years.

[DocM]

Found the original of this pic on Reddit, but it showed a BFR which was way taller what math says is necessary. I cleaned it up and truncated the height as an angle cut since we don't really know what that number is.  

 

What is now depicted is a comparison of Saturn V, BFR and Falcon 9 stage diameters, assuming that BFR's is the rumored 15 meters. Note the little dude at the bottom.

 

 

 

[Unobscured Vision]

9 minutes ago, DocM said:

Cargo and tanker BFS versions seem to be no brainers, and what could you do with a 50-100 tonne satellite or big satellite dispenser under cargo door(s)?  Eek!

 

How about prepositioning tankers at Sun-Mars L2 as depots for a manned asteroid belt mission? Ceres? Fly a BFS as a reusable space station/exploration gateway where ever?

Gosh, we could do a ThinkTank on setting up Operations on Ceres alone. In a lot of ways, aside from the low gravity and distance, it's better than Mars. Solar is a viable option there; but the panels would need to be 3x larger than on Mars to generate the same amount of power. The benefits are no atmosphere or weather to contend with (aside from the Solar kind). Ceres' gravity is so low it wouldn't be anything to get escape velocity -- a Soyuz TMA using its' own thrusters could do it easily -- and maintaining orbitals would be exceedingly easy. The low gravity, however, is a concern too: the trip out there, the Mission duration, then the trip back would might cross the "point of no return", physiologically speaking -- where the body possibly could not readjust to 1:1 "Earth" gravity.

 

Imagine it ... Ceres Base, just mining away at all of that ice. Bigelow Habs, specially crafted for the purpose. You've got all of the Water, O2 and Fuel you'd ever need right there; only thing you're missing is a starter crop and +20% replacement seeds as needed for replenishment and the soil. Use some chemistry to break down the silicates, clays and carbonates, and we've got everything else we'd require.

 

In short, Ceres would make an awesome Gateway Station to the Outer Solar System if not for the question about low-and-zero gravity physiology concerns.

 

Reference: https://en.wikipedia.org/wiki/Ceres_(dwarf_planet)

 

BFR would likely be able to punch that ticket too.  

[Unobscured Vision]

Okay, BFR's Engine Conjecture, Take 1.  Continued discussion from a train of thought that began here; a report that SpaceX's McGregor Test Site is now licenced to carry out engine tests that include:

 

Quote

(from http://www.kwtx.com/content/news/McGregor--City-modifies-SpaceX-rocket-testing-rules-378857891.html)

 

....

"The restrictions on sound are established within three miles of the center point within the City of McGregor.

For a test that exceeds 115 decibels of sound, the city requires a permit that includes payment of $5,000 for each test.

For a test that exceeds 120 decibels, the city requires a permit and a payment of $7,500-per-test.

For a test that exceeds 125 decibels a permit and payment of $50,000 is required, the ordinance says.

The ordinance also limits acceptance tests to 15-seconds or less and says that if noise limits of 125 decibels are exceeded, the test must be curtailed within 3 seconds."

 

"...and all tests involving a rocket motor that generates 2 million pounds of thrust must be conducted during daylight hours."

The special clauses in the agreement wouldn't be there unless there was a specific level of thrust that needed to be approved.

 

We know quite a bit about how SpaceX does things. We know, for instance, that SpaceX does not "long-duration test" any of their hardware while they are in Development Phase -- they save that kind of testing for Public Viewing, and they combine tests to save time and money.

 

And yes, they like to make a bit of a spectacle of these tests too; but that's what we adore about SpaceX -- if they succeed or fail, the Public is witness to it. SpaceX is usually quite forthcoming about results, positive or negative, and they're always willing to go big or fail spectacularly in front of a gaggle of cameras. It's the ultimate in Reality Television.  

 

Better still, we're in a position to know how SpaceX plans to make their Methane-based "Super Engines" behave. We're in a position to make some pretty informed assumptions about how each of these engines will likely function, what the overall stats will be per-unit, and what the probable layout of BFR Block 1 (for lack of a better moniker) should be.

 

We'll proceed in stages, carefully. (See what I did there? ) First up? A couple of specific questions, hypotheticals, and good, solid maths before we start doing the "Lego Layout".

 

Answering Question 1 is important, because it determines the operation of the engines; such as "how hard are we pushing them?", "do we need to push them to maximum or can we get away with what we did on Falcon-9 again?", and so forth. Important information when deciding what the engines really need to be capable of, instead of what their stated capabilities are. So ...

 

Question 1: SpaceX, being so well-versed and comfortable using an Octaweb Design for the First Stage of their Rockets (F9, FH, etc), are masters of the Octaweb design. It has proven reliable, resilient, and a great platform for Reusability as well as Launching. Is it at all likely that SpaceX will scale up this design for BFR, given their familiarity with it?

 

We need to determine if SpaceX is using a scaled-up Octaweb design, and nine engines, on BFR. This will determine BFR's lifting capability!

 

A cursory search on Google reveals some information straight from Elon himself:

Quote

MCT would land 100 metric tons on the surface of Mars. (Musk, Reddit AMA, 1/5/15)

 

The MCT booster would be around 15 million pounds of thrust (Musk, AskMen Interview, 1/2/15)

 

MCT would have a little over 230 metric tons of force per engine (Musk, Reddit AMA, 1/5/15)

 

There would be both a sea level version and a vacuum version of Raptor (Musk, Reddit AMA, 1/5/15)

 

Raptor engines would have a vacuum Isp of 380 (Musk, Reddit AMA, 1/5/15)

 

It's rumored that part of the MCT architecture would include a 15 meter diameter tank (L2 leak, SpaceX subreddit, 12/12/15)

 

Numbers from the "December Leak" (where an alleged SpaceX employee leaked details of BFR's intended capabilities to the wild)

15m radius

236mT to LEO 

15,000,000 pounds thrust (which is 30 raptor engines) 

5-6,000mT for BFR/MCT weight

Now we've got some numbers to work with. And they're incredible if they're true.  But if they are true, it would indicate that they aren't using an Octaweb; but something more like N1's design.

 

Thoughts?

[Unobscured Vision]

Oh -- and they're going with 30 Raptors, at 500,000 lbf each? Merlin 1d++ can do 205,475 lbf; so ...

 

They're sandbagging Raptors' performance. Just like with M1d++ performance, they're being conservative again. Wouldn't be surprised if those Raptors could get up to 700,000 lbf or more.  

[DocM]

Or, the 550 klbf Raptor is a RaptorVac (vacuum) engine for the BFS spaceship, which includes the upper stage, and there's a much larger Raptor for the BFR first stage. A SpaceXer testified before Congress that Raptor scales quite well, so this diversity is possible.

 

If the sea level Raptor is a ~1.67 mlbf engine as once discussed then they could directly transfer the 9 engine Octaweb layout from F9 and have nearly the 15 mlbf projected for the BFR first stage. Much of the avionics should transfer with relatively few mods.

 

For BFS with 550 klbf RaptorVac engines, a 5 engine layout gives it engine-out capability and about 2.75 mlbf of thrust. With right sized tanks that's a honkin' big ∆V for any spaceship. A rocket version of the Tesla S Insane Mode.

[Unobscured Vision]

Yeah, I don't think the number of Raptors are going to be that high per rocket. I was fading out last night, so I thought it best to go to bed and continue the conversation today instead.

 

What they might do is go with a reverse design of the RD-180. What I mean by that is they could have, say, four combustion outlets at 500~700k each that terminate at the Engine Bell which contribute to the overall engine output. They could easily get to the 2-million mark if they do it that way; but it means essentially using four engines per bell. It's like using four Fuel Injectors per Cylinder in a Car Engine -- you'll get massive power output increases at the cost of fuel efficiency; but I'm guessing that we don't really care much about that S1 being very fuel efficient. The efficiency improvements can come later, after they get the engines working and the power curve where they want it.

 

Bottom line? SpaceX can hit the 2-million mark per engine, from an Engineering Standpoint; and it'd be pretty straightforward if unconventional from a Rocket Engine design point-of-view. But it could be done and it wouldn't be that hard.  

[Unobscured Vision]

@DocM, you've peeked my curiosity here.  If SpaceX were so inclined ... using that methodology of four combustion outlets at 1.67 mlbf each fed into a single, massive engine bell ... good grief. 

 

Let's do the math ... 1.67*4 = 6.68 mlbf output at sea level per engine ... that's some unholy levels of power out of a single engine. Now drop nine of those puppies into an S1 in a massively scaled-up Octaweb ... we get a power output of 60.12 mlbf. 

 

Dear Multiverse, that's gonna crack the skin of the Earth. There's no Launch Facility that will handle that kind of power. No where. It wouldn't be safe to be within 25 km of that launch due to the noise alone. To even see the launch, the cameras will have to be remotely-controlled from afar.

 

WOW.

 

So this one could be a "BFR Replacement" Engine concept ... 

 

Remember, folks, this is just an idea I have on how SpaceX can design an engine ... from the Third Circle of Gehenna itself.

[flyingskippy]

Better to derate the engines on the first few launches than to push the engines to their limit and risk failure.

 

SpaceX did right by sandbagging the figures for the M1, gathering data with actual missions,  and then adjusting figures from there. 

 

Data is just as valuable as physical property in this case. Better to know what an engine is truely capable of than to risk pushing it too far and end up a failure. 

 

[bguy_1986]

25 minutes ago, flyingskippy said:

Better to derate the engines on the first few launches than to push the engines to their limit and risk failure.

 

SpaceX did right by sandbagging the figures for the M1, gathering data with actual missions,  and then adjusting figures from there. 

 

Data is just as valuable as physical property in this case. Better to know what an engine is truely capable of than to risk pushing it too far and end up a failure. 

 

They will probably do the same when they re-use the F9 as well won't they?  Or do they think she can handle 100% throttle again? 

[DocM]

To qualify the Raptor they'll have to do it at full throttle. They'll need every bit of its planned full thrust to get the monster booster and spacecraft off the ground. 

 

The post-landing tests will be like they did testing the first landed stage: full throttle. Katie, bar the door.

[flyingskippy]

Is it possible  though that the planned 500klbf is just for the first variant of Raptor? Similar to what they did with all the variants of the M1. 

Edited May 13, 2016 by flyingskippy

[Unobscured Vision]

I'm sure it'll have plenty of room to scale and improve.  

[DocM]

2 hours ago, flyingskippy said:

Is it possible  though that the planned 500klbf is just for the first variant of Raptor? Similar to what they did with all the variants of the M1. 

Until recently I took them at their word that the design optimized at 550,000 lbf and there would be 27-30 in the BFR first stage.

 

This week changed that. McGregor city fathers and SpaceX reached a new noise abatement agreement, and therein was a time window during which SpaceX could test up to a 2,000,000 lbf engine.

 

Literally dropped my effin tablet.

 

Speculation:

 

I'm now open to the idea that the 550,000 lbf Raptor is the USAF upper stage and BFS spacecrafts vacuum engine - RaptorVac if you will - and that the BFR booster engine will one big SOB.

 

About 2 years ago Aviation Week reported this Raptor booster engine would be the size of Saturn V's Rocketdyne F1, anout 1,550,000 lbf.

 

Put 9 of those into an Octaweb and you have a fracking monster. F9 on meth and pcp.

 

We'll see.

[anthdci]

i not sure I'm my number are right but does that means each raptor engine is in the range of 9-10 times as powerful as a merlin at sea level? (Wiki says SES was 165,000lbf) 

[DocM]

The booster M1D is soon to be 171,000 lbf at launch, 190,000 lbf in flight, so 9x an M1D at launch or one engine as powerful as a Falcon 9 first stage. Or more. The McGregor noise agreement said 2 mlbf.

 

We just dunno what Raptors thrust is now. That noise agreement calls many "safe" assumptions into question, and they've sandbagged their performance before.

 

Whatever, BFR will be the most powerful rocket ever. Period. Musk said it'll make Saturn V look small. A beast.

Edited May 13, 2016 by DocM

[Unobscured Vision]

Can't wait to see what they've come up with. Elon wouldn't say it if they didn't already have something they had already simulated on the computers that was workable and do-able.

 

Only thing they've been off about was their timetable for Falcon Heavy (which threw Dragon 2 slightly off-schedule); but we know why that's late -- the Double-Plus Platform Upgrade to SpaceX's current Core Stock. Totally understandable; and they would have been foolish not to pursue those upgrades and impair FH's performance.

 

Bite the bullet now and reap triple the rewards later.  

[Joe2mercs]

A little quick rough and ready calculation helps on the speculation on the BFR starting with the end point of a required '100 tons on Mars' and using the Saturn V missions to the moon for reference.  75% the mass of the lunar module (LM) was fuel; however it carried enough fuel for the ascent module to regain orbit and this we can omit since the intent is to use methane synthesised on Mars using solar or nuclear power.  The LM was about 15 tons.  In orbit, awaiting the return of the ascent module (lunar orbit rendezvous - LOR) was the service module of about 25 tons which contained enough fuel for the return trip.  My guess is that that a trip to Mars will leave some hardware elements in orbit for rendezvous and communications.  Of the 48 tons sent to the Moon only about 7.5 tons actually landed.  The Mars trip will not need a fuelled mother ship in orbit and it will not have to land the fuel needed to take off again. Set against this is Mars extra gravity and the lack of thick atmosphere; the weight penalty of having some winged descent shuttle does not make sense in an atmosphere 1/1000th as dense as that of Earth and if you have to lift it back up again into orbit.  The descent vehicle I am almost sure will be a 'tail sitter' rocket (100 tons with occupants, life support etc.) and that perhaps 30% of its mass will consist of fuel used for descent.  That gives us a starting descent vehicle of about 135 tons in Mars/Hohmann orbit.  The Saturn V lofted 140 tons into low earth orbit (LEO) in order to insert 48 tons on a Lunar trajectory. The delta v required to go to Mars is not much more than going to the Moon, it's just further away. We can therefore reason that 135 tons sent to Mars will require about 400 tons in LEO. The Suturn V launched about 4% of its take off mass in LEO (140 tons vs ~3000 tons).  Elon will employ reusable rockets that put about 2.2% of take off mass into orbit. This gives us lift off weight of 18,300 tons or does it? The Saturn V F1 engines used RP1 with LOX at a ratio of 1 to 2.6 by mass.  The Raptor engine will use Methane to LOX at a ratio of 1 to 3.8 by mass.   All things being equal a methane rocket will be about 33% heavier, due to the extra LOX, and about 77% greater in volume, due to methane's low density, as compared to a kerosine rocket of similar payload capacity.  Yes the Raptor engines will be have a higher ISP and yes the closed cycle design of the Raptor engine will improve efficiency by about 14% over that of the Merlin 1D but the MCT lifter will be still 11% heavier than an RP1 equivalent. So the take off mass is now 20,300 tons!  I think that this is too large for one rocket and therefore we are left with either three flights of 6,800 tons or four flights of 5,100 tons with assembly of the MCT in orbit.  The smaller option, if using nine engines in an octaweb arrangement on a single core, would put the Raptor thrust into F1 territory (~1.5 million lbs)

[DocM]

28 minutes ago, Joe2mercs said:

A little quick rough and ready calculation helps on the speculation on the BFR starting with the end point of a required '100 tons on Mars' and using the Saturn V missions to the moon for reference.  75% the mass of the lunar module (LM) was fuel;

>

>

The Suturn V launched about 4% of its take off mass in LEO (140 tons vs ~3000 tons).  

>

>

Elon will employ reusable rockets that put about 2.2% of take off mass into orbit. 

Recalibrate for much more of the launch mass to LEO and a much larger vehicle and S1 staging low. The BFR S2 is also expected to be integrated into the BFS vehicle, which will be refueled in orbit.

Edited May 23, 2016 by DocM

[Joe2mercs]

27 minutes ago, DocM said:

Recalibrate for much more of the launch mass to LEO and a much larger vehicle and S1 staging low. The BFR S2 is also expected to be integrated into the BFS vehicle, which will be refueled in orbit.

I would be interested to the physics and chemistry that allows much more of the launch mass to LEO and what that would represent in terms of percentage of the take off mass.  Elon has said that expendable rockets struggle to put 4% of take off mass into orbit and he also revealed that his own reusable rockets will struggle to exceed 2% of take off mass to LEO; I generously used a ball park figure of 2.2% due to the economies of scale afforded by large rocket design.  The Saturn V set a very high bar for expendable rockets and Elon will, no doubt set a high bar for reusable rockets.  

[DocM]

Lighter structure, S2 doing the vast majority of the work, different cycles, much higher iSP all adds up. This thing isn't Falcon.

Edited May 23, 2016 by DocM

[DocM]

Further,

 

You say 4% to orbit for F9 without reuse, 2.2% with which vastly overstates the reuse penalty at 45%.

 

Musk estimated the penalty at 30% for a return to launch site landing (RTLS), and 15% for a drone ship (ASDS) landing. RTLS requires 3 burns; braking/boostback (the largest burn of all), reentry and landing. ASDS requires 2 burns; reentry and landing.

 

SpaceX just demonstrated a 3rd variant: suicide-slam, which has a penalty closer to 10% for LEO and an ASDS landing.

 

Now consider a BFS that boosts 5%+ to LEO and has a 10% penalty for a suicide-slam ASDS landing. That's over 4.5% mass to orbit, not 2.2%.

 

 

[bguy_1986]

1 minute ago, DocM said:

Further,

 

You say 4% to orbit for F9 without reuse, 2.2% with which vastly overstates the reuse penalty at 45%.

 

Musk estimated the penalty at 30% for a return to launch site landing (RTLS), and 15% for a drone ship (ASDS) landing. RTLS requires 3 burns; braking/boostback (the largest burn of all), reentry and landing. ASDS requires 2 burns; reentry and landing.

 

SpaceX just demonstrated a 3rd variant: suicide-slam, which has a penalty closer to 10% for LEO and an ASDS landing.

 

Now consider a BFS that boosts 5%+ to LEO and has a 10% penalty for a suicide-slam ASDS landing. That's over 4.5% mass to orbit, not 2.2%.

 

 

Is there a plan to land BFS on an ASDS?  Would have thought that would have been too big to do.

[DocM]

That's one of the obvious trades, along with launching from a platform at sea or an island  because of the severe acoustics it'll generate - it'll have to be 10 miles from anywhere.

 

Edit - size: ASDS is a 300x170 foot barge. MARMAC makes a 400 foot barge that with extensions could be 200+ feet wide.  More than enough.

Edited May 23, 2016 by DocM