BFR and a bit of (hopefully) helpful scepticism

[alexterrell]

First off, I'm a fan of SpaceX. I was cheering their vision in the days of Falcon 1, when many (the majority?) here were dismissing them as here today, gone tomorrow, and not appreciating that space access is always difficult and always expensive.

Now the mood has swung, and we take BFR as a given. We talk about launch costs of below $100/kg like they're already baked in - let's plan our 2024 family vacation in orbit!

I can see that Falcon 9 has been revolutionary primarily because it's challenging business processes and the way of doing things (though First Sage recovery is technologically revolutionary, and I think the main reason that it's being done now and not in the 1970s is software and real time processing capability).

But for BFR - away from Musk's Reality Distortion field:

1. SpaceX claim that their BFS (without booster) will be able to make it to orbit and back with a tiny payload. This is an extraordinary claim as SSTO has been a dream for many decades, and often regarded as unachievable. Innovative companies like Pioneer thought they could almost get there with air refuelling, REL think they can with some form of air breathing, but no traditional rocket. Add to that:
   - BFS is using methane rather than LH2, so should have lower performance (though better costs)
   - BFS has to carry thermal protection systems and landing gear. (Most SSTO was predicated on cheap mass produced throw away rockets)
Any conventional thinking would say "no way". What have they got that the others didn't?

2. Engine reuse. What are they talking? 100 engine re-uses before a major over haul? So what are they doing that the Space Shuttle Main Engines didn't? They were also good for 100 uses, but needed rebuilding every time. Add to that, this is a new engine with a new fuel choice.

3. Thermal protection system: As with engines, the Space Shuttle had problems in this area, with expensive tiles needing a lot of inspection and servicing. Has technology moved on, or are the requirements less than with the shuttle (lower density airframe?). Of course, the requirements for deep space reentry are much higher than for LEO - but we assume SpaceX have it sorted.

4. Safety. It seems there are two ways to prove safety:
   - You analyse it to death, with some form of probabilistic safety analysis (similar to nuclear power plants). That is what Falcon 9 is going through now. But the analysts then insist on an independent launch abort system, which BFS doesn't have.
   - You fly a large number of missions to demonstrate a track record. This is the Soyuz approach. It works, if you have time. But how will BFS demonstrate safety?

5. Remain in space time
Depending on mission architecture, it is assumed that BFS has a long duration survive in space claim. How feasible is this? The shuttle was designed for up to 30 days and had extra-ordinarily complex systems to achieve this. And that added to cost - complexity on complexity. How can BFS avoid this?

The hubris around BFR is a bit like that around the shuttle in the 1970s. It was going to make space flight routine. BFR is also promising to be like the promised shuttle on Viagra, with 5 times the promised launch mass (about 8 times the realised launch mass), and ~1/10th the promised cost (~1/200th of the realised cost). Could it equally under deliver? 

Hopefully SpaceX can overcome these issues and launch BFR/BFS around the same time as SLS  They can prove me wrong like they proved most people wrong with Falcon 9. But how?

And of course, even if, like the shuttle, BFR/BFS ends up 10 times the target price, it's still competitive with every launch system apart from Falcon 9 Block 5 (and New Glenn and Skylon).

[TomH]

Your points are sound. Reasoned skepticism (both spellings correct) is what science is all about.

[sleepy-martian]

1)Going by statements from EM himself ("Earth is the wrong planet for SSTO") , "minimal payload" should be thought of as  close to zero. Most SSTO designs actually did it because of reusability, at least the way I understand it, and of course still wanted to have a useful payload.

2&3)What does BFS have that the space shuttle engines / TPS doesn't have?
Half a century worth of progress in Material Science, simulation capability etc. And a, so far at least, vastly more efficient design process.

EDIT: Engine re-use is demonstrated right now with block 5 and to a lesser extent previous blocks already to that's kind of the thing I doubt the least

I do share your concerns re: Passenger safety

[SimonFD]

Ok, here's my take on your points:

1) SSTO - I think Elon was demonstrating a capability not a programme goal. The Everyday Astronaut has done an excellent video on SSTO and mentions BFS (along with - "Even if it can, why bother?").

2) 100 uses before major overhaul - aspirational within the programme, not an immediate capability. Just as with the Merlin engine the Raptor design will iterate to achieve the goal over the life of BFR/BFS.

3) See 2

4) Yep, with you on that one

5) ISS has been in space for a while. There might be some lesson learnt there....

Again, just my opinion

[su27k]

My optimism comes from:
1. Unlike the Shuttle, SpaceX is standing on 40 years of aerospace (and non- aerospace) technology development, including lessons learned from Shuttle itself. They are free to pick and choose what to use.
2. They also have Falcon 9/Dragon reuse experience behind them. If Block 5 works out, then they already got first stage full and rapid reuse solved, that's 70% of the BFR already well understood and basically risk free.
3. The two stage architecture plus orbital refueling means they have a lot of performance margin, they can afford to miss their performance target and still produce something useful. Then they can do rapid iteration on it like they did with F9.

Also I don't understand the point about safety, unlike the Shuttle, BFR can fly unmanned, so it can prove its safety just like any other unmanned launch vehicle.

Finally about using methane in SSTO, see the discussion here: https://forum.nasaspaceflight.com/index.php?topic=35838.0, this forum predicted the SSTO possibility years before Elon made the comment.

Edit/Off Topic: Interesting to see Robotbeat basically predicted BFS back in 2014, including a fairly accurate estimate for take off mass:

And such a craft would make a KILLER Mars lander/ascent vehicle! Direct-ascent straight to Earth would be no problem, ...

(Might need to scale up to 1500t take-off mass so you would be big enough to house crew for the trip).

[speedevil]

I note the posts:
BFS - how bad can it be? (and still get to Mars with little delay)
BFS - without BFR eats the launch market.

In short, and to sum up, systemic margin.

SSTO was 'impossible' because to do it economically required a payload. BFS-SSTO - if it is possible has a payload fraction likely in the 1% range or below, and likely could operationally do this with some light modifications. (swap out most of the raptors with sea level ones, and add a couple more).

New rocket development - or shuttle - has typically aimed at a 'right sized' rocket - where an underperformance by 30% payload may mean your rocket can't meaningfully fly.

The combination of BFS and its booster give truly enormous margin in the design - if and only if two things work - reuse and on-orbit tanking, and even these can have some caveats.

For example, if your payload to orbit drops from 150 to 50 tons, you need to tranship cargo to your Mars lander to get to 150 tons and you may have more issues doing propellant production on Mars, but your costs only go up by a modest amount for the initial landers.

Or if the heatshield doesn't work in version one, and you need to do a 3km/s retroburn - again, payload around 50 tons, which does not affect commercial payload at all, as they aren't that big.

2022 - well - it's as good a notional date as June 2020 for SLS, and it is useful to not preface discussions which may involve orbital windows with 'or it may happen x years later if the LV is not ready'.

[Torbjorn Larsson, OM]

You are quite good in distorting reality yourself, throwing in things that does not necessarily have to do with system development and capabilities. I would take all these things with a huge pinch of salt; and I would suggest that outsider criticism is rarely helpful (what are the odds).

My analysis would be that BFR is less hubris than building SpaceX and Falcon 1 in the first place. And I think that is what they have noted officially. Though SpaceX/Falcon 1 was nearly lost, so it is a low bar. I'll take it though, especially since it is not my money.

But on to the list anyway, since you brought those points as well:

1.  The raw capability may or may not include yet to be shown TPS, which development has been admitted as "hard". The raw capability or its lack is not a program risk.

2. They are not using liquid hydrogen for one, huge temperature difference, or stressing their many smaller engines, huge "disassembly required" difference. If the engines can run one mission before disassembly, the system works (but will be more expensive).

3. TPS technological readiness, see above. Likely the largest program risk, and admitted so!?

4. Oy; "analysts demand" not demonstrated and irrelevant to what the system demands; not going to happen; stop this nagging. Apples vs pears, heavy mass spaceplanes (Shuttle, BFS) vs light mass capsules; you go down with the traffic airplane (or car, bike, even your fancy shoes) since it is the safest method allowed by the system. There is some irony when people complain about the Shuttle had its passengers strapped to the propellant tanks, since traffic planes strap them between (but yes, there were differences too).

5.  Shuttle was a LEO system, BFS is a interplanetary system; the demands are likely different. And the technical readiness is even further away than the system demand to finalize TPS; not needing closed loop to start with (small crew, large cargo) helps.

What I can see the Shuttle used fuel cells for power, the BFS will have the customary solar cells. The extended duration pack was cryogens for the fuel cells (and added avionics) https://en.wikipedia.org/wiki/Extended_Duration_Orbiter . That the BFS won't need that must count in its favor (disclaimer: I have picked a favorite, and the Shuttle lost so far - big time), so thanks for bringing it up!

[JamesH65]

1. SpaceX claim that their BFS (without booster) will be able to make it to orbit and back with a tiny payload. This is an extraordinary claim as SSTO has been a dream for many decades, and often regarded as unachievable. Innovative companies like Pioneer thought they could almost get there with air refuelling, REL think they can with some form of air breathing, but no traditional rocket. Add to that:
   - BFS is using methane rather than LH2, so should have lower performance (though better costs)
   - BFS has to carry thermal protection systems and landing gear. (Most SSTO was predicated on cheap mass produced throw away rockets)
Any conventional thinking would say "no way". What have they got that the others didn't?

SSTO is possible due to the size of the rocket, and modern materials, but it's not useful. Very unlikely to be used commercially, so ignore it.

2. Engine reuse. What are they talking? 100 engine re-uses before a major over haul? So what are they doing that the Space Shuttle Main Engines didn't? They were also good for 100 uses, but needed rebuilding every time. Add to that, this is a new engine with a new fuel choice.

If you are rebuilding an engine after every use, its not really reusable, is it? How much of that 100 uses ME is original? SpaceX have more experience with multiple use engines than anyone else on the planet. They KNOW how to make rockets. There will still be things to fix during development, but to bet against getting some thing to work well enough for their use cases is foolhardy.

3. Thermal protection system: As with engines, the Space Shuttle had problems in this area, with expensive tiles needing a lot of inspection and servicing. Has technology moved on, or are the requirements less than with the shuttle (lower density airframe?). Of course, the requirements for deep space reentry are much higher than for LEO - but we assume SpaceX have it sorted.

Yup, safe to assume that they know what they are doing.

4. Safety. It seems there are two ways to prove safety:
   - You analyse it to death, with some form of probabilistic safety analysis (similar to nuclear power plants). That is what Falcon 9 is going through now. But the analysts then insist on an independent launch abort system, which BFS doesn't have.
   - You fly a large number of missions to demonstrate a track record. This is the Soyuz approach. It works, if you have time. But how will BFS demonstrate safety?

Falcon 9's development has been mostly fast iteration, not huge piles of analysis. It's got results faster than any other supplier. It having to do more analysis now to get human rating, but there is no better analysis than actually having a returned booster to tear apart and examine. I expect BFS to take the same approach. It'll bee cheap to test it as well, because it's fully reusable. Just need fuel. Whether it need's a LAS? I doubt it, but some people will always want it, even in the face of overwhelming evidence it's not necessary (which will come with testing)

5. Remain in space time
Depending on mission architecture, it is assumed that BFS has a long duration survive in space claim. How feasible is this? The shuttle was designed for up to 30 days and had extra-ordinarily complex systems to achieve this. And that added to cost - complexity on complexity. How can BFS avoid this?

Plenty of theads on here on that. Read them.

The hubris around BFR is a bit like that around the shuttle in the 1970s. It was going to make space flight routine. BFR is also promising to be like the promised shuttle on Viagra, with 5 times the promised launch mass (about 8 times the realised launch mass), and ~1/10th the promised cost (~1/200th of the realised cost). Could it equally under deliver? 

Yes, but we won't know for some years, and to call it out now is premature. No point in concern trolling about it.

Hopefully SpaceX can overcome these issues and launch BFR/BFS around the same time as SLS  They can prove me wrong like they proved most people wrong with Falcon 9. But how?

By doing exactly what they are doing. Seems to work. Since you have been proved wrong already, seems only prudent to be careful when predicting in the future - your track record is 0 for 1.

[alexterrell]

You are quite good in distorting reality yourself,

What reality? There is no BFR at the moment. There is a test engine and prototype fuel tank. And lots of promises that SpaceX can do what no one else has done before, and 10 times better. A bit of scepticism is justified, but countered by a lot of hope.

For human rating, I can't see how NASA will be able to human rate BFS. Of course, that may not be an issue, especially if they can demonstrate 100 flights without mishap.

Shuttle was a LEO system. BFS is a LEO and interplanetary system. It has to do what the shuttle did, and more.

[alexterrell]

I note the posts:
BFS - how bad can it be? (and still get to Mars with little delay)
BFS - without BFR eats the launch market.

In short, and to sum up, systemic margin.

......
For example, if your payload to orbit drops from 150 to 50 tons, you need to tranship cargo to your Mars lander to get to 150 tons and you may have more issues doing propellant production on Mars, but your costs only go up by a modest amount for the initial landers.

Good points - if the payload drops to 50 tons, the cost per kg triples. But the Mars hardware probably costs more anyway.

I think it fits with SpaceX philosophy: Aim really high, and end up high.

I think it comes down to 2 questions:
1. Will BFR/BFS work? Answer probably.
2. Will it work to the margins predicted? Probably not, but it doesn't really matter.

Is there more chance of seeing a BFS launch than getting hold of a Tesla Model 3?

[zodiacchris]

Hmm, it’d be too bad if Nasa wasn’t to man-rate BFS 
Unless the crew consisted of former Nasa astros and non Nasa personel, which would probably work just fine, provided Elon can foot the bill.

Besides, after Shuttle, with 1:70 LOC ratio, and the recent decision not to fly crew on the first SLS due to financial and schedule, rather than LOC concerns, I’m not sure that the human rating criteria are particular impartial or objective...

[spacenut]

I for one, hope it works as planned.  Carbon fiber will make it lighter.  As mentioned SpaceX has landing covered, and engine reuse covered, and very well made engines.  Solar power for long duration missions of course.  They still have to master bringing the whole ship back from space, and land.   I think this will be done many times before allowing people to fly. 

[alexterrell]

Hmm, it’d be too bad if Nasa wasn’t to man-rate BFS 
Unless the crew consisted of former Nasa astros and non Nasa personel, which would probably work just fine, provided Elon can foot the bill.

Besides, after Shuttle, with 1:70 LOC ratio, and the recent decision not to fly crew on the first SLS due to financial and schedule, rather than LOC concerns, I’m not sure that the human rating criteria are particular impartial or objective...

Of course the criteria are flawed. But Shuttle (dangerous) and Soyuz (probably safe) weren't subject to it. (If it had existed in the 1960s, we wouldn't have had the moon landings)

I'm not sure if there is a plan for NASA personnel to fly in the BFS. They are meant to fly in the human rated, but perhaps (checks with Lawyer) less safe SLS. (Because it will have flown a handful of times only before launching crew).

[niwax]

   - BFS is using methane rather than LH2, so should have lower performance (though better costs)

It's not all about ISP. An 85t BFS with 100t cargo and 1100t fuel at 370s ISP gets the same performance as a 140t vehicle with the same cargo and fuel at 420s. SpaceX is the king of mass fraction and from what Elon has been saying those 85t are the upper end of the scale with some margin to go lower.

Besides, after Shuttle, with 1:70 LOC ratio, and the recent decision not to fly crew on the first SLS due to financial and schedule, rather than LOC concerns, I’m not sure that the human rating criteria are particular impartial or objective...

Of course the criteria are flawed. But Shuttle (dangerous) and Soyuz (probably safe) weren't subject to it. (If it had existed in the 1960s, we wouldn't have had the moon landings)

I find it hard to compare the shuttle on just LOC ratios since no other space system had the insane idea to send up 5-7 astronauts for several days to place a satellite or an upper stage in LEO.

[Robotbeat]

1) The secret is that expendable SSTO is really easy to LEO. A F9 booster would get a few tons to LEO. Reusable SSTO is hard, true, but:
HYDROGEN MAKES SSTO HARDER, NOT EASIER. Mass fraction is CRITICAL. Also, SpaceX has already proven the best pumpfed rocket engine thrust to weight ratio EVER and by a large margin. That helps a lot. SpaceX has also demonstrated deep cooling which is a NEW technique operationally. So SpaceX has best mass fraction tech. That’s new.

2) SpaceX has lots of experience with engine reuse already. They test fire Merlin a lot, and even in a nominal mission, Merlin center is relit/reused like 5 times without refurb. The new engine fuel is better due to being volatile and non-coking. Full flow stages combustion also reduces temperatures in some parts of the rocket which makes reuse further easier.

3) is genuine challenge but there have been TPS improvements since the 1970s.

4) safety by using BFR a LOT. Using BFR for everything will mean it may do hundreds or even thousands of flights before ever carrying people up. If they really do that, it’ll be safer than SLS with a LAS, as a LAS only buys you an extra order of magnitude improvement in safety. A LAS itself fails to save the crew every 10 or 20 uses and can add extra failure modes.

5) long duration enabled by solar panels. Shuttle relied on fuel cells.

[envy887]

First off, I'm a fan of SpaceX. I was cheering their vision in the days of Falcon 1, when many (the majority?) here were dismissing them as here today, gone tomorrow, and not appreciating that space access is always difficult and always expensive.

Now the mood has swung, and we take BFR as a given. We talk about launch costs of below $100/kg like they're already baked in - let's plan our 2024 family vacation in orbit!

I can see that Falcon 9 has been revolutionary primarily because it's challenging business processes and the way of doing things (though First Sage recovery is technologically revolutionary, and I think the main reason that it's being done now and not in the 1970s is software and real time processing capability).

But for BFR - away from Musk's Reality Distortion field:

1. SpaceX claim that their BFS (without booster) will be able to make it to orbit and back with a tiny payload. This is an extraordinary claim as SSTO has been a dream for many decades, and often regarded as unachievable. Innovative companies like Pioneer thought they could almost get there with air refuelling, REL think they can with some form of air breathing, but no traditional rocket. Add to that:
   - BFS is using methane rather than LH2, so should have lower performance (though better costs)
   - BFS has to carry thermal protection systems and landing gear. (Most SSTO was predicated on cheap mass produced throw away rockets)
Any conventional thinking would say "no way". What have they got that the others didn't?

This is inaccurate.

Methalox is BETTER for an VTOL SSTO than LH2. It's denser and permits much better TWR.

And one-use SSTO is easy. Falcon 9 boosters/FH side boosters can already do it, with aluminum structures and gas-generator low pressure engines.

2. Engine reuse. What are they talking? 100 engine re-uses before a major over haul? So what are they doing that the Space Shuttle Main Engines didn't? They were also good for 100 uses, but needed rebuilding every time. Add to that, this is a new engine with a new fuel choice.

The SSME was designed in the 1970s. Later blocks had updates for better reusability.

Merlin is reusable without rebuilding. Raptor will be built to the same goals. And FFSC methalox is easier on the turbines than FRSC hydrolox.

3. Thermal protection system: As with engines, the Space Shuttle had problems in this area, with expensive tiles needing a lot of inspection and servicing. Has technology moved on, or are the requirements less than with the shuttle (lower density airframe?). Of course, the requirements for deep space reentry are much higher than for LEO - but we assume SpaceX have it sorted.

This is the major outstanding question for BFS. PICA-X is certainly up to the reentry task, but will it last many flights? And how durable is it on the ground?

4. Safety. It seems there are two ways to prove safety:
   - You analyse it to death, with some form of probabilistic safety analysis (similar to nuclear power plants). That is what Falcon 9 is going through now. But the analysts then insist on an independent launch abort system, which BFS doesn't have.
   - You fly a large number of missions to demonstrate a track record. This is the Soyuz approach. It works, if you have time. But how will BFS demonstrate safety?

Fly lots of flights, obviously. BFR is designed for turnaround and reflight in hours. It could in principle fly more in 3 months than the Shuttle did in 30 years.

5. Remain in space time
Depending on mission architecture, it is assumed that BFS has a long duration survive in space claim. How feasible is this? The shuttle was designed for up to 30 days and had extra-ordinarily complex systems to achieve this. And that added to cost - complexity on complexity. How can BFS avoid this?

The hubris around BFR is a bit like that around the shuttle in the 1970s. It was going to make space flight routine. BFR is also promising to be like the promised shuttle on Viagra, with 5 times the promised launch mass (about 8 times the realised launch mass), and ~1/10th the promised cost (~1/200th of the realised cost). Could it equally under deliver? 

Hopefully SpaceX can overcome these issues and launch BFR/BFS around the same time as SLS  They can prove me wrong like they proved most people wrong with Falcon 9. But how?

And of course, even if, like the shuttle, BFR/BFS ends up 10 times the target price, it's still competitive with every launch system apart from Falcon 9 Block 5 (and New Glenn and Skylon).

Shuttle was limited by power from the fuel cells and reaction control hydrazine, neither of which could be replenished in space.

BFS will use solar panels for power (unlimited endurance) and boiloff from the main tanks for RCS (unlimited, with orbital refueling).

[envy887]

The hubris around BFR is a bit like that around the shuttle in the 1970s. It was going to make space flight routine. BFR is also promising to be like the promised shuttle on Viagra, with 5 times the promised launch mass (about 8 times the realised launch mass), and ~1/10th the promised cost (~1/200th of the realised cost). Could it equally under deliver? 

The biggest difference between BFR/BFS and other SSTO architectures, and even the Shuttle, is margin. Even if SpaceX grossly misses on many cost and performance targets, the result is still a very capable system.

The Shuttle estimated costs were based on a high flight rate to justify highly capitalized manufacturing of the expendable ET and mostly expendable SRBs. Full reuse changes that equation by capitalizing the vehicle itself, not the manufacturing capability, which enables payback at MUCH lower flight rates. Full and rapid reuse drives the cost down even further by capitalizing manufacturing and by streamlining operations.

[dglow]

The biggest difference between BFR/BFS and other SSTO architectures, and even the Shuttle, is margin. Even if SpaceX grossly misses on many cost and performance targets, the result is still a very capable system.

Margin AND full reusability.

[groundbound]

I'll agree with all the people responding that many of the listed risks are not that high.

But the OP missed perhaps the biggest risk: flight rate. Without a high flight rate the costs do not come down nearly as much. And many of the reasons given for why BFR will have a high flight rate are extremely speculative and risky.

[IRobot]

4. Safety. It seems there are two ways to prove safety:
   - You analyse it to death, with some form of probabilistic safety analysis (similar to nuclear power plants). That is what Falcon 9 is going through now. But the analysts then insist on an independent launch abort system, which BFS doesn't have.
   - You fly a large number of missions to demonstrate a track record. This is the Soyuz approach. It works, if you have time. But how will BFS demonstrate safety?

If SpaceX can do the same as with Falcon 9 Block 5, they can launch the same rocket over and over until it fails.